University of Massachusetts Amherst ScholarWorks@UMass Amherst
Doctoral Dissertations 1896 - February 2014
1-1-2001 Expanding the scope of templated macromolecular synthesis in vivo : the incorporation of methionine analogues into proteins in vivo by altering the methionyl-tRNA synthetase activity of a bacterial expression host. Kristi L. Kiick University of Massachusetts Amherst
Follow this and additional works at: https://scholarworks.umass.edu/dissertations_1
Recommended Citation Kiick, Kristi L., "Expanding the scope of templated macromolecular synthesis in vivo : the incorporation of methionine analogues into proteins in vivo by altering the methionyl-tRNA synthetase activity of a bacterial expression host." (2001). Doctoral Dissertations 1896 - February 2014. 1015. https://scholarworks.umass.edu/dissertations_1/1015
This Open Access Dissertation is brought to you for free and open access by ScholarWorks@UMass Amherst. It has been accepted for inclusion in Doctoral Dissertations 1896 - February 2014 by an authorized administrator of ScholarWorks@UMass Amherst. For more information, please contact [email protected].
EXPANDING THE SCOPE OF TEMPLATED MACROMOLECULAR SYNTHESIS IN VIVO:
The Incorporation of Methionine Analogues into Proteins in Vivo by Altering the Methionyl-tRNA Synthetase Activity of a Bacterial Expression Host
A Dissertation Presented
by
KRISTI L. KIICK
Submitted to the Graduate School of the
University of Massachusetts Amherst in partial fulfillment
of the requirements for the degree of
DOCTOR OF PHILOSOPHY
May 2001
Polymer Science and Engineering © Copyright by Kristi L. Kiick 2001
All Rights Reserved EXPANDING THE SCOPE OF TEMPLATED MACROMOLECULAR SYNTHESIS IN VIVO:
The Incorporation of Methionine Analogues into Proteins in Vivo by Altering the Methionyl-tRNA Synthetase Activity of a Bacterial Expression Host
A Dissertation Presented
by
KRISTI L. KIICK
Approved as to style and content by:
David A. Tirrell, Chair
£ ^ rit. OtIa^xIak E. Bryan Coughlin, Memben
Lila M. Gierasch, Menroer
Thomas J. McCarthy, Department Head Polymer Science and Engineering To my grandparents ACKNOWLEDGMENTS
There are many people to whom I am indebted for my undertaking, enjoying, and successfully completing this graduate work. Their influence extends over decades and thousands of miles, and for their support, advice, and encouragement I will forever be
grateful. First, I would like to thank my advisor, David Tirrell. Working in David's laboratories for the past six and a half years has been an incredible personal and
professional journey, and provided opportunities for me to work on truly interesting
scientific problems. His graciousness, sense of humor, respect for scientific inquiry,
personal strictness, and precise thought and articulation will serve as models for my own
behavior, and the gentle guidance he has provided has helped me develop confidence in
my capabilities. For all of those things I am very thankful.
I would also like to thank my committee members for their time and helpful
suggestions not only throughout the scientific portion of my graduate research, but for
their other professional input as well. Bruce Novak was on my committee in its early
stages and contributed many helpful suggestions on potential project directions. Bryan
Coughlin was kind enough to participate on my committee during the final stages of my
research, and his enthusiasm and curiosity about biological polymerizations has been
appreciated. Lila Gierasch introduced me to biochemistry in 1987, and I was delighted to
V have the opportunity to ask her to participate on my doctoral committee after all these years.
For scientific discussions and advice, I am indebted to past and present members of the Tirrell group. A better working environment would be difficult to find, and it is in large part due to these people that my graduate experience was so productive, gratifying,
and fun. In the early days, it was Jody Beecher, Vince Conticello, and Jim Thomas. As a
student at the University of Massachusetts, I was fortunate enough to have colleagues
who have also been good friends: Wendy Naimark, Karen Guhr, Jill Sakata, Scott
Kennedy, Kathy DiZio, Jeff Linhardt, Nandita Sharma, David Flanagan, Sarah Jouppi,
and Karin Rotem. Their friendship, support, and advice have been truly appreciated. Yi
Tang, Marissa Mock, Sarah Heilshorn, Isaac Carrico, Angelika Niemz, Jens Thies, and
Kent Kirshenbaum, among others, have made the Caltech part of the journey a pleasure.
On the scientific aspects of my projects, I owe a debt of gratitude to many people.
Scott Ross and Pratip Bhattachary helped during the NMR characterization of proteins
containing non-natural amino acids. Jan van Hest, Ralf Weberskirch, Helen Blackwell,
and Eliana Saxon kindly provided the methionine analogues that we investigated.
Rebecca Alexander provided many helpful suggestions and conversations, and having a
chance to work with her was an unexpected opportunity to work with an old friend.
Heather Maynard and Sheldon Okada were enormously industrious and helpful in our
vi attempts to modify unsaturated side chains via ruthenium-catalyzed ring-opening metathesis polymerizations. Nagarajan Vaidehi, Deepshikha Datta, and Jeremy Kua contributed to our modeling work and the development of computational protocols to predict methionine analogue/MetRS binding.
On a personal note, I would like to thank some people who had a critical influence
in my life many years ago and are still important in my life today. Richard Person
introduced to chemistry me during my junior year in high school, and it was directly due
to his influence that I was able to discover my interests both in science and in obtaining a doctoral degree. Burnaby Munson and John Burmeister had an important impact on my
undergraduate experience at the University of Delaware and helped me make important
personal and professional decisions. Mike Johnson, my Master's Thesis advisor at the
University of Georgia, offered me a great start in research, and his continuing support of
my career is appreciated. My dearest old friends Jody Beecher, Janine Sullivan, Jennifer
Lastova, and Rosann Kaylor have offered unwavering support over many years and in
many difficult situations, for which I am extremely grateful. And of course, I am deeply
indebted to my family - Mom, Dad, and Karen - for providing me with the sturdiest of
roots and a desire to contribute to the world in a positive way. I am fortunate to have
such a loving and supportive family.
Vll My final thanks go to fiance my Rick Beyer, who has perhaps contributed the most to making the past few years so enjoyable and satisfying. I am grateful for many things about him, including his intelligence, kindness, sense of humor (which he might not believe), values, and enthusiasm for learning new things. By patiently listening
during countless hours of conversation, offering varied perspectives, and constantly encouraging me to express my opinion, he has helped me strengthen my voice. It's an
extremely generous gift, and I will never be able to adequately thank him.
viii ABSTRACT
EXPANDING THE SCOPE OF TEMPLATED MACROMOLECULAR SYNTHESIS IN VIVO:
The Incorporation of Methionine Analogues into Proteins in Vivo by Altering the Methionyl-tRNA Synthetase Activity of a Bacterial Expression Host
MAY 2001
KRISTI L. KIICK, B.S., UNIVERSITY OF DELAWARE
M.S., UNIVERSITY OF GEORGIA
M.S., UNIVERSITY OF MASSACHUSETTS AMHERST
Ph.D., UNIVERSITY OF MASSACHUSETTS AMHERST
Directed by: Professor David A. Tirrell
The in vivo incorporation of non-natural amino acids is controlled by the
aminoacyl-tRNA synthetases (aaRS). The correlation between the incorporation of
methionine analogues into proteins in vivo and activation of analogues by methionyl-
tRNA synthetase (MetRS) in vitro has therefore been investigated. Activation of
methionine analogues 2 - 13 in vitro, assessed via the ATP-PP, exchange reaction,
correlates well with the ability of analogues to support protein synthesis in vivo,
substantiating the critical role of MetRS in the incorporation of methionine analogues
than those into proteins. Methionine analogues with k,,/K,„ values up to 2000-fold lower
ix for methionine (i.e., (homoallylglycine), 2 3 (homopropargylglycine), and 9 (norleucine)) can support synthesis of a target protein (DHFR) under standard conditions of protein
expression employing a conventional bacterial host.
Overexpression of MetRS in a bacterial host was investigated as a method to increase the number of methionine analogues that can be incorporated into proteins in vivo. Equipping a bacterial host with extra copies of the gene encoding wild-type MetRS permits incorporation, under certain conditions, of the additional methionine analogues 4
(cw-crotylglycine), 5 (fran^-crotylglycine), 7 (2-aminoheptanoic acid), 8 (norvaline), 11
(2-butynylglycine), and 12 (allylglycine). Assessment of the level of replacement of
methionine via amino acid analysis and N-terminal sequencing indicates levels of
replacement of approximately 60% to 98%.
In an effort to broaden further the chemical functionality available for engineering
novel proteins, two azido amino acids, azidoalanine (14) and azidohomoalanine (15)
were investigated for their ability to support PPj exchange in vitro and protein synthesis
in vivo. While 14 does not support PPj exchange or protein synthesis, 15 is the most
efficient substrate for MetRS of all the analogues tested to date, with a k^./K^, value that
is 400-fold lower than that for methionine. In vivo experiments confirm that this analogue supports protein biosynthesis, with essentially quantitative replacement of
X methionine. DHFR containing 15 can be selectively modified by Staudinger ligation with triarylphosphine reagents.
These results suggest new strategies for incorporation of non-natural amino acids
via manipulation of the aaRS activity of a bacterial host and provide new opportunities
for the chemical modification of proteins.
xi TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS ^
ABSTRACT IX
LIST OF TABLES
LIST OF FIGURES ^^jj
CHAPTER
1. INTRODUCTION 1
1.1 In Vivo Synthesis of Protein Polymers 1
1.2 Aminoacyl-tRNA Synthetases 6 1.3 Methionyl-tRNA Synthetase 9
1 .4 Preliminary Investigations of Methionine Analogue Incorporation 17 1.5 References 25
2. IN VITRO ACTIVATION OF METHIONINE ANALOGUES BY METHIONYL-tRNA SYNTHETASE 36
2. 1 Introduction and Objectives 36 2.2 Experimental Section 37
2.2.1 Analogue Synthesis 37
2.2.2 Determination of Translational Activity 37
2.2.2.1 Small Scale Protein Expression 38
2.2.2.2 Large Scale Protein Expression 39
2.2.2.3 Protein Purification 40
2.2.3 MetRS Expression and Purification 41
2.2.4 Activation of Methionine Analogues in Vitro 43
2.2.5 Computational ModeHng of Methionine Analogues 44
xii 2.3 Results and Discussion
2.3. 1 Determination of Translational Activity of Methionine Analogues 45 2.3.2 MetRS Expression and Purification 45 2.3.3 Activation of Methionine Analogues by MetRS in Vitro 47 2.3.4 Determination of Protein Yields 49 2.3.5 Computational Modeling of Methionine Analogues 51
2.4 Summary and Conclusions 53 2.5 References
3. ALTERATION OF THE METHIONYL-tRNA SYNTHETASE ACTIVITY
OF A BACTERIAL EXPRESSION HOST 64
3. 1 Introduction and Objectives 64 3.2 Experimental Section 66
3.2. 1 Genetic Manipulation of a Bacterial Expression Host 66
3.2.1.1 Construction of Plasmid pUC 1 9-Nhelink 67 3.2.1.2 Construction of Plasmid pUC19-W305F 70
3.2. 1 .3 Construction of Expression Plasmids PQE9-W305F, pQE15-W305F, and
PQE15-MRS 71
3.2.2 Protein Expression 73
3.2.2. 1 Confirmation of MetRS and DHFR Expression 73
3.2.2.2 Translational Activity of
Methionine Analogues in Vivo 74
3.2.2.3 Determination of Protein Yields 75
3.2.3 Determination of Intracellular Concentrations
of Methionine Analogues 76
xiii 3.2.4 Protein Characterization 77
3.2.4.1 Amino Acid Analysis and N-terminal Sequencing 77 • 3.2.4.2 NMR Spectroscopy 77
3.2.5 Activation of Methionine Analogues by MetRS in Vitro 78
3.2.6 Characterization of GFP Fluorescence 81
3.3 Results and Discussion 83
3.3. Genetic 1 Manipulation of a Bacterial Expression Host 83
3.3.2 Translational Activity of Methionine Analogues in a
Modified Bacterial Expression Host 85 3.3.3 Protein Yields from Modified Bacterial Expression Hosts 88 3.3.4 Incorporation of Additional Methionine Analogues
into Proteins in Vivo 90 3.3.5 Intracellular Concentrations of Methionine Analogues 92 3.3.6 Protein Characterization 93
3.3.6.1 Amino Acid Analysis and N-terminal Sequencing 93 3.3.6.2 NMR Spectroscopy 95
3.3.7 Activation of Methionine Analogues
by MetRS in Vitro 97
3.3.8 Characterization of GFP Fluorescence 100
3.4 Summary and Conclusions 102
3.5 References 129
4. IN VIVO INCORPORATION OF AZIDOAMINO ACIDS 132
1 4. 1 Introduction and Objectives 32
4.2 Experimental Section 133
4.2.1 Computational Methods 133
4.2.2 Analogue Synthesis 133
xiv 4.2.3 Activation of Azidoamino Acids by MetRS in Vitro 1 34 4.2.4 Translational Activity of Azidoamino Acids in Vivo 1 34 4.2.5 GFP Fluorescence 1 35 4.2.6 Modification of Azide Residues via the Staudinger Ligation 1 35
4.3 Results and Discussion 1 37
4.3. 1 Computational Modeling of Analogues 1 37 4.3.2 Activation of Azidoamino Acids by MetRS in Vitro 138 4.3.3 Translational Activity of Azidoamino Acids in Vivo 139 4.3.4 Protein Characterization 140 4.3.5 GFP Fluorescence 142
4.3.6 Modification of DHFR-15 by the Staudinger Ligation 143
4.4 Summary and Conclusions 1 45 4.5 References 160
5. FUTURE RESEARCH DIRECTIONS 161
5.1 Discussion 161 5.2 References 164
APPENDICES
A. LIST OF AMINO ACIDS 1 66
B . LIST OF METHIONINE ANALOGUES 167
C. TOCSY SPECTRA FOR DHFR-5 1 68
D. DNA SEQUENCE OF NHELINKER 1 69
E. DNA SEQUENCE OF METRS NHE I CASSETTE 1 70
F. DNA AND AMINO ACID SEQUENCE OF METRS 175
BIBLIOGRAPHY 178
XV LIST OF TABLES
Table Pager,
2. Kinetic parameters 1 for methionine analogues in the ATP-PP, exchange reaction and protein yields for bacterial cultures supplemented with the analogues 55
3. 1 Intracellular amino acid concentrations as determined by
amino acid analysis IO4
3.2 Amino acid analysis results for DHFR containing methionine analogues 105
3.3 Extent of replacement of methionine by methionine analogues 106
3.4 Kinetic parameters for activation of methionine analogues by MetRS 107
3.5 Comparison of catalytic efficiencies for the activation of methionine analogues
by MetRS and W305F 108
4. 1 Comparison of kinetic parameters for 14 and 15 with those
for methionine, 3, and 9 147
4.2 Amino acid analysis results for DFHR-Met and DHFR-15 148
4.3 Mass spectrometric analysis of tryptic fragments of DHFR-15 149
xvi 1
LIST OF FIGURES
Figure ^ Page^
1.1 Schematic representation of the template nature of in vivo
protein synthesis 20
1.2 Schematic of gene construction and protein synthesis 21
1 .3 General structure of tRNA 22
1.4 The active site of MetRS 23
L5 The structure of MetRS 24
2.1 SDS-PAGE analysis of DHFR synthesis by E. coli strain
CAG18491/pQE15/pREP4 56
2.2 SDS-PAGE analysis of MetRS purification by size-exclusion
chromatography 57
2.3 Activation of methionine and methionine analogues by MetRS 58
2.4 Comparison of kinetic parameters and protein yields for methionine analogues 59
2.5 Electron density maps and negative isopotential surfaces for methionine and methionine analogues 60
3.1 Genetic strategy for increasing MetRS activity in a bacterial host 109
3.2 Schematic of Nhe linker designed to change the identity of the
cohesive ends of the gene encoding W305F to Nhe 1 110
3.3 Restriction digestion results for the construction of plasmid pQE9-W305F 1 1
3.4 Restriction digestion results for plasmid pQE15-W305F 1 12
xvii 3.5 SDS-PAGE analysis of the expression of DHFR and MetRS in CAG18491/pREP4 bacterial hosts 113
3.6 Activation rates of methionine by whole cell lysates 114
3.7 SDS-PAGE analysis of DHFR expression in bacterial cultures
supplemented with methionine analogues 115
3.8 Western blot analysis of the DHFR yield from modified
bacterial host 1 CAG 849 1 /pQE 1 5-MRS/pREP4 116
3.9 Western blot analysis of DHFR synthesis by bacterial expression hosts CAG18491/pQE15/pREP4 and CAG18491/pQE15-MRS/pREP4 1 17
3.10 Yields of purified DHFR obtained from 1 -liter cultures supplemented
with methionine, 2, 3, or 9 118
3. 1 1 SDS-PAGE analysis of DHFR synthesis in bacterial cultures
supplemented with methionine analogues at 500 mg/L 119
3.12 N-terminal sequencing results for indicating occupancy of the initiator site
in DHFR produced in bacterial cultures supplemented with 4 120
3.13 N-terminal sequencing results for indicating occupancy of the initiator site
in DHFR produced in bacterial cultures supplemented with 5 121
3.14 N-terminal sequencing results for indicating occupancy of the initiator site
in DHFR produced in bacterial cultures supplemented with 7 122
3.15 N-terminal sequencing results for indicating occupancy of the initiator site
in DHFR produced in bacterial cultures supplemented with 8 123
3.16 N-terminal sequencing results for indicating occupancy of the initiator site
in DHFR produced in bacterial cultures supplemented with 11 124
of the initiator site 3. 1 7 N-terminal sequencing results for indicating occupancy 125 in DHFR produced in bacterial cultures supplemented with 12
xviii 3.18 "H NMR spectra (599.69 MHz) of DHFR, 5, and DHFR-5 126
3.19 Activation of methionine analogues by MetRS 1 27
3 . 20 Fluorescence spectra for GFP produced from CAG 1 849 1 /pQE3 1 -GFP/pREP4
cultures supplemented with methionine, 2, 3, or 9 128
4. 1 The Staudinger ligation for the chemical modification of azidoamino acids with a triarylphosphine-FLAG reagent 1 50
4.2 Electron density maps and negative isopotential surfaces for methionine,
14, and 15 151
4.3 Activation of methionine and methionine analogues by MetRS 152
4.4 SDS-PAGE analysis of the translational activity of
methionine analogues 14 and 15 153
4.5 Western blot analysis of the expression of DHFR in bacterial cultures
supplemented with methionine or 15 154
4.6 N-terminal sequencing results indicating occupancy of the initiator site in
DHFR produced in bacterial cultures supplemented with 15 155
4.7 Mass spectrometric analysis of tryptic fragments of DHFR-15 156
4.8 Fluorescence spectra for GFP produced from cultures of bacterial host
GAG 1 849 1 /pQE3 1 -GFP/pREP4 supplemented
with methionine or 15 157
4.9 Western blot analysis of DHFR produced from cultures of bacterial host
GAG 1 849 1 /pQE 1 5/pREP4 supplemented with methionine or 15 158
159 4. 1 0 Mass spectrometric analysis of tryptic fragments of DHFR-15
xix CHAPTER 1
INTRODUCTION
1.1 In Vivo Synthesis of Protein Polymers
Although many advances in synthetic polymer chemistry have been made over
the last several decades to provide the polymer chemist with increasing control over the
structure of macromolecules (1-7), none have provided the level of control that is the
basis of the catalytic, informational, and signal transduction capabilities of proteins and
nucleic acids (8). With increasing scientific and technological interest in producing well-
defined architectures and surface chemistries for nanostructure, biomaterials, and
biosensors applications, the need to synthesize precisely engineered molecules with the
control characteristic of proteins and nucleic acids has gained new importance. We have
chosen to focus on the production of protein polymers with novel and useful materials
properties owing to the useful structural, biological, and catalytic activities that are
mediated by this class of macromolecules.
Protein synthesis in vivo is a template-directed polymerization in which
messenger RNA (mRNA) directly encodes cellular DNA information (Figure 1.1). At
corresponding amino the ribosome, mRNA is translated by transfer RNA (tRNA) into a
1 acid sequence. Each is tRNA charged with the appropriate amino acid by a highly selective class of enzymes, the aminoacyl-tRNA synthetases (aaRS). Once charged, the aminoacyl-tRNA is delivered to the ribosome by the elongation factor EF-Tu and
accepted at the ribosomal A site. The translationally active amino acid is then covalently attached to the end of the nascent protein chain.
Harnessing the molecular weight and sequence control provided by in vivo
synthesis should permit control of folding, functional group placement, and self assembly
at the angstrom length scale. Indeed, the biosynthetic method shown in Figure 1.2 has
been used to produce proteins that exhibit predictable chain-folded lamellar architectures
(9-12), unique smectic liquid-crystalline structures with precise layer spacings (13), and
controlled reversible gelation (14). It has become of additional interest to expand the
novel chemical and physical properties that can be engineered into these classes of
protein polymers by the precise placement of non-natural amino acids.
We have focused on in vivo incorporation of non-natural amino acids into proteins
owing to the synthetic advantages it offers with respect to other methods for analogue
incorporation. Introduction of non-natural amino acids can be achieved relatively simply
via solid-phase peptide synthesis. While this method circumvents all biosynthetic
machinery, the multistep procedure is limited to synthesis of peptides less than or equal
to approximately 50 amino acids in length and is therefore not suitable for producing
protein materials. Chemical aminoacylation methods, introduced by Hecht and
2 coworkers (15) and exploited by Schultz, Chamberlin, Dougherty, and others (16-19)
provide a powerful method for the site-specific incorporation of non-natural amino acids.
But because these methods (except in special cases) require the use of cell-free translation protocols that limit protein yields, they are also unsuitable for production of protein materials. Alteration of the synthetase activities of the cell is also possible through the introduction of heterologous synthetases, but has met with limited success
and only permits incorporation of a non-natural amino acid in a site-specific manner (20-
25). The simplicity of the in vivo approach, its relatively high synthetic efficiency, and
its capacity for multisite substitution make it the method of choice for production of
protein materials whenever possible.
While in certain cases it may be desirable to limit incorporation to a single site in
a protein chain (e.g., studies of protein folding, structure, and dynamics), multisite
substitution also has distinct advantages, as such substitution can cause important
changes in protein behavior. For example, incorporation of selenomethionine in place of
methionine has long been known to facilitate protein structure determination by x-ray
crystallography (26-28). The incorporation of fluorinated functional groups into proteins
has imparted to protein films the low surface energy characteristic of fluoropolymers;
contact angles of hexadecane on fluorinated protein polymers (70°) are much higher than
those on unfluorinated controls (17°) (29). Incorporation of trifluoroleucine in place of
leucine also results in increases in the thermal stability of leucine zipper peptides (30);
3 these results may have important consequences for increasing protein stabiUty, improving protein assembly, or strengthening ligand-receptor interactions. Furthermore, alkene functionality introduced into artificial proteins via dehydroproline can be quantitatively modified via bromination and hydroxylation (31). However, the number of amino acids
shown conclusively to exhibit translation^ activity in vivo is small, and the chemical
functionality that has been accessed by this method remains modest.
The incorporation of non-natural amino acids into proteins in vivo requires that the amino acid analogue meet several criteria. First, the analogue must be recognized and
transported across the cell membrane into the cell, either by transport mechanisms
specific for the natural amino acid or by more general transport machinery (32,33). Once
inside the cell, the analogue must also not be degraded. Second, the non-natural amino
acid not only must be recognized by the aaRS during translation, but must also be able to
form a stable aminoacyl-tRNA that is not subject to the aaRS editing mechanisms that
normally prevent the misacylation of tRNA (34). Finally, the non-natural aminoacyl-
tRNA must be an efficient substrate for the elongation factor Tu (EF-Tu) and must also
be accepted at the ribosomal A site.
Perhaps surprisingly, given the extremely high fidelity of protein biosynthesis, it
has been known for decades that a reasonably large number of analogues have been
identified which meet these criteria (35-44). For example, we and others have
demonstrated the ability of the wild-type translational apparatus to use non-natural amino acids with nuorinated (29,38), unsaturated (31,45,46), electroactive (47), and other side chain functions (42,48-50). Strategies that would permit the incorporation of a broader range of non-natural amino acids in vivo, however, would provide additional - opportunities for protein engineering.
The incorporation of non-natural amino acids into proteins in vivo does not appear to be limited by transport into the cell or discrimation by EF-Tu and the ribosome.
Transport processes do not appear to be highly specific (51,52) and hundreds of
analogues have been accepted at the ribosome in in vitro translation assays (19,46,53,54).
Therefore, the in vivo incorporation of non-natural amino acids into proteins appears to be controlled most stringently by the aaRS. Investigations have thus been directed toward understanding the recognition of amino acid analogues by the aaRS in order to expand the novel chemical and physical properties that can be engineered into proteins in
vivo.
5 1-2 Aminoacyl-tRNA Synthetases
Each of the 20 natural amino acids (Appendix A) has a corresponding aaRS that is responsible for catalyzing the covalent attachment of an amino acid to its cognate tRNA in two steps, activation and aminoacylation:
aaRS + aa+ ATP [aaRS:aa~AMP] + PP-
[aaRS:aa~AMP] + tRNA'''' aa-tRNA"" + AMP + aaRS
where aaRS is the enzyme, aa is the amino acid, PPj is pyrophosphate, aaRS:aa~AMP is
the aminoacyladenylate complexed with the enzyme, and aa-tRNA"' is the aminoacyl-
tRNA. Due to the critical role of these enzymes in the synthesis of cellular proteins that control metabolic processes, they have been the subjects of crystallography and
enzymology investigations for decades, and much is known about their structures and
mechanisms of action. To date, the crystal structures of 15 of the 20 aaRS have been
solved and have provided insight into the detailed mechanisms of enzyme action (55-96).
Although the aaRS are structurally diverse, identification of small stretches of conserved primary sequence, combined with determination of similar three-dimensional
structures in many of the aaRS, has led to the division of the 20 aaRS into two distinct
groups of 10 enzymes each (97). The class I aaRS (aaRS for Arg, Cys, Gin, Glu, He,
6 Leu, Met, Trp, Tyr, and Val) display a catalytic center which is built around a nucleotide binding domain called the Rossmann fold (98). The Rossmann fold also harbours two signature sequences, HIGH and KMSKS, that are the basis of the initial division of these aaRS into a common class. The conservation of these sequences has been demonstrated to result from the functional role that the residues play in binding of amino acids, ATP, and tRNA to the aaRS (see discussion for MetRS, below). Indeed, aaRS from both prokaryotic and eukaryotic organisms display the signature sequences characteristic of
the class I enzymes (99).
The class II aaRS (those for Ala, Asn, Asp, Gly, His, Lys, Phe, Pro, Ser, and Thr),
on the other hand, are characterized by a catalytic center that is organized around a seven- stranded antiparallel P-sheet surrounded by a-helices (97,100,101). These secondary
structures build a domain which contains three degenerate sequence motifs that are
conserved in all the class II aaRS, regardless of the origin of the enzyme. The structural
classification of the aaRS classes also appears to be correlated to mechanisms of action.
Where the class II aaRS aminoacylate the 3'-0H group of the terminal adenosine of the
tRNA, the class I aminoacylate the 2'-0H group (97). The similarities in the sequence,
structure, and mechanism within the two classes of enzymes suggest a common ancestor
for enzymes belonging to the same class.
Two functions are conducted by the aaRS: activation of an amino acid and
aminoacylation of the amino acid's cognate tRNA. Correspondingly, all of the aaRS for
7 which three-dimensional structures have been determined demonstrate organization into two domains: an active center domain for recognition and activation of the amino acid and a domain specialized in the recognition of tRNA (99). The recognition of tRNA is
not a limiting factor in the fidelity of protein biosynthesis, as the tRNA are relatively large molecules with several identity elements. All tRNAs are approximately 76
nucleotide residues in length and assume a typical tRNA cloverleaf structure shown in
Figure with 1.3, the amino acid acceptor terminus and the anticodon at opposite ends.
Primary targets for recognition of tRNA include the acceptor stem, the three bases of the anticodon, and the D-loop (99). These targets provide for high selectivity of the aaRS for
its tRNA; for example, the selectivity of E. coli IleRS for tRNA'" is 10^ against tRNAs
for phenylalanine or formylmethionine (34).
The accuracy of amino acid recognition by the active center domain, however, is
of prime importance in ensuring the fidelity of amino acid incorporation into proteins, as
the amino acids are small molecules often with similar side chain structures. While the
discrimination between natural amino acids most certainly involves amino acid residues
contained in the signature sequences (class I) and motifs (class II) of the aaRS, often this
mechanism for recognizing the amino acid is insufficient to prevent the enzyme from
incorrectly forming enzyme-bound noncognate aminoacyl adenylates (misactivation).
Indeed, some of the aaRS (AlaRS, IleRS, LeuRS, and ValRS) are very permissive in the
8 activation of noncognate amino acids, with IleRS and ValRS misactivating 7 and 8 noncognate (but naturally occurring) amino acids, respectively (34).
Despite the frequency with which some amino acids are misactivated, the fidelity of amino acid incorporation into proteins is very high, with error rates (substitution of one natural amino acid for another) on the order 1 of in 10,000 (102). The high fidelity is achieved by editing mechanisms present in some of the aaRS, which occur either by hydrolysis of the misactivated amino acid, or through the deacylation of a mischarged tRNA (34,99,103,104). However, it must be noted that not all aaRS require proofreading for accuracy, and therefore the molecular basis for the specificity of aaRS most often
occurs at the first step of amino acid recognition. Because of the importance of the
amino acid-aaRS interaction, efforts to manipulate the incorporation of non-natural
amino acids into proteins in vivo have focused on the aaRS. In these investigations, such manipulation has focused on methionyl-tRNA synthetase (MetRS).
1.3 Methionyl-tRNA Synthetase
MetRS catalyzes the covalent attachment of methionine to its cognate tRNAs.
Homodimeric native methionyl-tRNA synthetase of E. coli is a Class 1 aaRS comprising
two identical subunits of 76 kDa molecular weight (105,106). The formation of the active
dimer is thought to be mediated by a C-terminal peptide of the native enzyme; removal of
9 approximately 120 amino acids from the C-terminus, either by mild trypsinolysis or
genetic engineering, eliminates the ability of the enzyme to dimerize, but produces a fully
active monomer with a single active site (105,107,108). The native dimer is also
observed in T. thermophilus (109), B. stearothermophilus (110), and in eukaryotic
enzymes, although yeast MetRS (both mitochondrial (1 1 1) and cytoplasmic (1 12,1 13))
lack the C-terminal peptide and are active in the monomeric form in vivo. Yeast and
human MetRS contain unique N-terminal extensions, not found in the prokaryotic
enzymes (99,1 14,1 15), that may be involved in import processes in the eukaryotic cell
(99).
Truncated forms of several bacterial MetRS (E coli, B. stearothermophilus, T.
thermophilus, M. tuberculosis) (96,108-1 10,1 16, 11 7) and both yeast (S. cerevisiae) and
human MetRS (117) have been cloned and expressed in E. coli, providing the opportunity
to study the enzymatic mechanisms of a variety of MetRS in vitro via site-directed
mutagenesis. Important structural and functional similarities among the MetRS of
varying origins exist; in fact, MetRS of both prokaryotic (including B.
stearothermophilus and T. thermophilus) and eukaryotic (yeast and human) origin are
capable of recognizing E. coli tRNA'^^' (117), despite the presence of only 21-26%
sequence homology (117). The sequences and structures that control the activation of
methionine and aminoacylation of tRNA*^" appear to be well-conserved and are generally
found in both prokaryotic and eukaryotic MetRS. Therefore, although the discussion
10 below features specific details for the E. coli enzyme, the general features and mechanism of action described are likely common to all MetRS.
The class I MetRS contains an active site consisting of a Rossmann nucleotide binding fold and nearby signature sequences HIGH and KMSKS, and catalyzes the attachment of methionine to the 2'-hydroxyl group of the terminal adenosine in tRNA'^^'.
The crystal structures of various MetRS confirm that the enzyme is organized into two subunits: the N-terminal domain which contains the active site, and the C-terminal domain which provides for recognition of tRNA""''. MetRS is a metalloenzyme, the E. coli MetRS containing one tightly bound zinc ion per protein chain (1 18-120), which is
thought to aid in the correct folding of the enzyme (121). No direct catalytic role for the
metal ion has been indicated, although maintenance of zinc in the enzyme appears to be
critical for activation and aminoacylation, presumably by correctly orienting the side
chains involved in the methionine binding pocket (1 16,120,122). The three dimensional
structures solved for the monomeric forms of both E. coli and T. thermophilus MetRS
(96,123-125) illustrate that the zinc-binding region is located in a connective peptide that
makes contacts with both the Rossmann fold and the KMSKS domains. The connective
peptide forms an isolated domain located over the catalytic crevice of the MetRS (124),
consistent with zinc's putative role in correctly orienting side chain residues in the active
site.
11 The solution of these crystal structures, coupled with site-directed mutagenesis, has provided insight into the possible mechanisms by which MetRS recognizes and activates methionine and then covalently attaches it to tRNA'^^ The orientation of bound methionine in the active site of E. coli MetRS, described in the recently published crystal structure for MetRS complexed with methionine (125), is shown in Figure 1.4. Residues highlighted in gray are those that line the pocket containing methionine (which can be distinguished by the yellow thioether side chain). The hydrophobic pocket enclosing the side chain of methionine comprises residues Alal2, Leu 13, Tyrl5, Trp253, Ala256,
Pro257, Tyr260, Ile297, His301, and Trp305. Tyrl5, Trp253, Pro257, and Tyr260 are in
nearest proximity to the side chain, while Trp305 closes the bottom of the hydrophobic
cavity (although it is not in direct contact with the terminal methyl group of methionine).
Comparison of the sequences of E. coli MetRS and other MetRS shows that
Tyrl5, Trp253, AIa256, Tyr260, and His301 are strictly conserved, while conservative
replacements are observed for Leu 13. The other residues are not strictly conserved
between species, although Trp305 is always aromatic. These observations are in good
agreement with site-directed mutagenesis studies indicating that mutation of Tyrl5,
His301, or Trp305 to Ala results in significant decreases in the rates of activation and
aminoacylation for the enzyme (126-128). Interestingly, mutation of Trp305 to Phe
produces an enzyme with full activity toward methionine, consistent with the
conservation of the aromaticity of this residue across species. Site-directed mutagenesis
12 also indicates important roles for Phel97 and Val298, as mutation of either of these residues to Ala also significantly decreases the rates of activation and aminoacylation of methionine by MetRS (116). Although these residues are not found directly in the
binding cavity, their mutation must result in a conformational change in the active site
that causes a loss of enzyme activity.
The sulfur atom of methionine is anchored in the site by accepting 2 hydrogen
o bonds (3.4A), one from the hydroxyl group of Tyr260 and the other from the amide N-H of Leu 13. His301 is hydrogen bonded to Tyr260 and shifts toward the methionine ligand
upon binding. Residues Asp52 and Arg233 have both been shown to be critical for
enzyme activity by site-directed mutagenesis (127). Consistent with the importance of
Asp52, the NHj moiety of methionine makes one hydrogen bond with the Asp52
carboxyl group (125). Arg 233, while not directly in contact with methionine, is linked to the aromatic pocket and to the zinc-binding domain via an extended hydrogen-bonded
network involving two different stretches of water molecules (125). The charged side
chain of Arg233 is therefore thought to be important in stabilizing the structure of the
enzyme in this region via hydrogen bonding.
Comparison of the crystal structure of MetRS complexed with methionine (125)
to that of the free enzyme (124), shows that a large number of conformational changes
occur in the enzyme upon binding of methionine. Tyrl5 and Asnl7 undergo 180°
rotations toward the ligand, Trp253 shifts by 6A, His301 shifts by 0.5 A, and Tyr260
13 shifts slightly. These shifts in position are mediated by concomitant unwinding of the helical peptide 294-304 with the winding of peptide 251-256 into an a-helix; this conformational flexibility at the active site may facilitate binding of methionine, analogues with structures and chemical functionality dissimilar to those of methionine.
The enzyme is also capable of binding homocysteine (Hey, Appendix A), the immediate precursor to methionine in the cell. Incorporation of Hey into tRNA and
protein is prevented by an efficient editing mechanism, which has been demonstrated to
operate in E. coli, yeast, and mammalian cells (129-132). The distinct feature of this editing mechanism is that the enzyme-bound Hcy-AMP undergoes intramolecular
cyclization to form homocysteine thiolactone (133). The same active site residues that are indicated by site-directed mutagenesis to be important for binding of methionine by
MetRS are also important in the discrimination between methionine and homocysteine in
editing (128). This suggests that E. coli MetRS may utilize a single active site that
partitions an amino acid substrate through either a synthetic pathway or an editing
pathway. Studies in which free thiols can be utilized by MetRS for thioester formation
with activated methionine (134) suggest that this may occur via binding of the thiol
sidechain of homocysteine at a subsite within the active site. Alternatively, a lower
affinity of the more polar sidechain of homocysteine for the hydrophobic pocket may
leave the sidechain available for cyclization (125). The lack of more general editing by
14 MetRS is likely to be important for the incorporation of a variety of methionine
analogues into proteins in vivo.
In addition to binding methionine and discriminating against homocysteine,
MetRS must also bind ATP and tRNA; residues indicated to play a role in these binding events have also been deduced on the basis of site-directed mutagenesis, affinity labeling studies, and three-dimensional structures. The binding of ATP involves residues near the methionine binding site, in particular His21 (a well-conserved residue in the HIGH signature sequence), Lysl42, Lys335 (in the KMSKS sequence), Tyr 359, and a
phylogenetically conserved Tyr residue at position 358 (123,135,136).
Recognition of the tRNA^"anticodon is indicated to involve two peptidic regions,
391-395 and 451-467, in the C-terminal part of MetRS, specifically Asn452, Trp461,
Arg395, and Asn391 (137-140). Not only are these residues are highly conserved among
MetRS, but they also move upon binding of methionine (125), perhaps indicating a
signaling mechanism between the N-terminal and C-terminal domains of the enzyme.
Affinity labeling experiments indicate that the 3' end acceptor arm of the tRNA binds
near the KMSKS signature sequence (141-143). Consistent with the fact that the 3' end
of the tRNA must react with the anhydride bond of the aminoacyladenylate formed from
methionine and ATP, the KMSKS signature sequence is near the ATP binding site (see
above). Taken together, these observations suggest a picture of tRNA"^" charging in
which ATP and methionine bind in the N-terminal portion of MetRS, with binding of the
15 tRNA ^ anticodon stabilized by the C-terminal portion of the enzyme so that the acceptor
arm of the tRNA is aligned with the N-terminal active site containing the enzyme-bound
methionyladenylate (Figure 1.5).
Once charged to tRNA^=', methionine is not just incorporated within a protein sequence, but also has the special role of acting as the universal initiator of translation.
With these two parallel roles for methionine, all living cells possess two distinct Met-
species that tRNA are aminoacylated with methionine by MetRS in vivo and in vitro;
tRNA^" (elongator) and tRNA*^" (initiatior) are both recognized by MetRS on the basis of their anticodon, CAU, and are aminoacylated with methionine. Once formed, Met-
tRNA^" is used during elongation, but Met-tRNA^" is formylated by methionyl-
tRNA^^"' formyltransferase (MTF). The formylation of the Met-tRNA'^'' ensures that it
will only be accepted at the ribosomal P site, and will not be recognized by the elongation
factor EF-Tu or accepted at the ribosomal A site (Figure 1.1). The most important
features permitting formylation of tRNA*^" (and not tRNA^") are the presence of 3 G-C
base pairs in the initiatior tRNA acceptor stem coupled with a CI A72 base mismatch
(144). The ability of the MetRS to recognize methionine analogues and use them
efficiently during both initiation and elongation is a key factor in our use of such
analogues for protein engineering in vivo. That methionine analogues can be efficiently
used in both initiation and elongation is suggested by the previously reported in vivo
incorporation of a several methionine analogues (27,36,42,48,50,145). Even a mutant
16 Lys-tRNA^^^ can be formylated by MTF and used during initiation (146), suggesting that a variety of methionine analogues may serve efficiently during initiation as well.
In summary, successful incorporation of methionine analogues into proteins in vivo requires 1) transport of the analogue into the cell, 2) activation of the analogue by
MetRS and aminoacylation of tRNA"^"' with the analogue, 3) lack of editing of the analogue by MetRS, and 4) the ability of the analogue to to be appropriately recognized by the elongation factors and ribosome during both initiation and elongation. Previous investigations of the incorporation of methionine analogues into proteins in vivo indicate
that a variety of methionine analogues of varying chemical function and side chain length
meet these requirements (39,40,42,48,145,147). These studies were the basis of the
initial investigations described here, which were aimed at expanding the scope of
methionine analogues that can be utilized by the translational apparatus for the
incorporation of methionine analogues into proteins in vivo.
1.4 Preliminary Investigations of Methionine Analogue Incorporation
While in theory any one of the 20 proteinogenic amino acids can be replaced by
non-natural amino acids, methionine analogues are of particular interest for investigation.
Due to methionine's special function as the initiator amino acid, the ability of a
methionine analogue to support protein synthesis indicates that the analogue must also
17 serve efficiently as the initiator amino acid. The presence of analogues at the N-terminus and at other specific sites along the protein chain not only provides routes for labeling of proteins, but may also permit the production of interesting block copolymer architectures.
Investigations of methionine are further motivated by the importance of methionine in mediating protein structure and protein-protein recognition processes (44,148,149); controlled incorporation of methionine analogues may permit purposeful manipulation of these phenomena.
Toward these ends, a series of methionine analogues 2 - 9 (46) (Appendix B), was
previously synthesized and investigated for incorporation into proteins in vivo. Previous
reports of in the vivo incorporation of selenomethionine (27,48), telluromethionine (42),
norleucine (36), trifluoromethionine (50), ethionine (42), and 5-nitrosohomocysteine
(145) indicate that the MetRS is able to charge at least a small range of analogues to the
tRNA"^"'. It therefore seemed possible that at least some of the analogues tested would be utilized by the protein biosynthesis machinery. Indeed, 2 (homoallylglycine) and 3
(homopropargylglycine) demonstrated translational activity in all stages of protein
biosynthesis, with extents of substitution of up to 98% (45,46). (The incorporation of 9
(norleucine) had been previously reported (42).) In fact, 3 supported protein biosynthesis
at levels similar to that of methionine, as 40 mg of DHFR could be isolated from 1-L
cultures supplemented with cither methionine or 3. Lower levels of protein synthesis
18 were observed for cultures supplemented with - 2. In contrast, 4 8 did not support protein synthesis in the absence of methionine in a conventional bacterial expression host.
Despite the known importance of the aaRS in controlling amino acid
incorporation into proteins in vivo, there has been a lack of data correlating the kinetics of analogue activation and aminoacylation by MetRS with incorporation into proteins in vivo. Therefore, one goal of this dissertation work has been to correlate the activation of methionine analogues by MetRS in vitro with their incorporation into proteins in vivo.
Additionally, methods to expand the range of methionine analogues that can be
incorporated into proteins have been investigated; expanding the types of analogues that
can be utilized by a bacterial host will permit the production of proteins with novel
chemical and physical properties. With the recent publication of the crystal structure of
MetRS with bound methionine (125), comparisons of analogue structures with the
known, bound structure of methionine may also provide important information regarding
the critical features of methionine analogues that control their activation by MetRS.
19 tRNA / fMet fMet Arg
UAC UAC - GCU 5' AUG CGA ,GCU > 5" AUG CGA mRNA iGCU (a) mRNA
P site A site P site A site
(b)
fMet ^ fMet
Ala 1 Arg Arg
^ ( GCU CGA UAC 5' GCU AUG CGA GCU 5' AUG CGA IGCU mRNA (c) mRNA P site A site P site A site
Figure 1 . 1 Schematic representation of the template nature of in vivo protein
synthesis, starting with initiation of translation by the binding of
fMet-tRNAfMet at the ribosomal P site, (a) An aminoacyl-tRNA
(Arg in this example) is delivered to the ribosomal A site and binds
through its anticodon to the mRNA. (b) Peptide bond formation is
catalyzed by peptidyl transferase, (c) The ribosome translocates
down the mRNA chain, and the next aminoacyl-tRNA (Ala in this
example) is delivered to the ribosome.
20 CONCEPTS FROM iTRUCTURAL BIOLOGY AND) PRIMARY SEQUENCE POLYMER SCIENCE OF AMINO ACIDS
I COMPLEMENTARY DNA SEQUENCE
Chemical synlhesis 1
'lllllMlllllinn Synthetic Gene
1- Bacieriai screening Plasmid and amplificaiion
2. Enzymalic ligation 10 yield muliimers; Isolation of target length DNA Enzyme
Target Length DNA
E. coll
Enzyme
Recombinant E. coli Plasmid
I NH2
Amino Acid COOH Polymer
Figure 1.2 Schematic of gene construction and protein synthesis. An oHgonucleotide
that encodes the target sequence with potentially desirable materials
properties is synthesized via standard solid phase methods and ligated
enzymatically into a bacterial cloning vector. The integrity of the insert is
verified via restriction analysis and standard sequencing methods. The
DNA is multimerized and ligated into a cloning vector, and the desired
multimer length is selected after amplification. The target length DNA is
ligated into a bacterial expression vector, which then contains the DNA
sequence encoding the artificial protein of interest under control of an
appropriate promoter. A bacterial host containing this vector is induced to
produce the artificial protein polymer, which is purified by appropriate
protein purification protocols.
21 Acceptor stem
5'
D loop c 3
XXX
Anticodon
Figure 1 .3 General structure of tRNA. The identity elements for
recognition by the appropriate aaRS include the anticodon site,
the acceptor stem, and the D loop.
22 Figure 1 .4 The active site of MetRS. The thioether of methionine is shown in
yellow. Nitrogen is shown in blue, and oxygen in red. The
backbone of the protein is shown in green.
23 Figure 1.5 The structure of MetRS. The residues forming the binding
pocket for methionine are shown in red. Residues that interact
with ATP are shown in yellow. Residues indicated by site-directed
mutagenesis to direct the binding of tRNA^^^ are shown in blue.
24 1.5 References
(1) Szwarc, M. Nature 1956, 178, 1 168-1 169.
(2) Faust, R. and Kennedy, J.P. Polym. Bull. 1986, 75, 3 1 7-323.
(3) Schrock, R.R. Acc. Chem. Res. 1990, 23, 158-165.
(4) Corradini, P. Macromol. Symp. 1995, 89, 1-11.
(5) Brintzinger, H.H., Fischer, D., Mulhaupt, R., Rieger, B. and Waymouth, R.M.
Angew. Chem. Int. Ed. Engl. 1995, 34, 1 143-1 170.
(6) Dias, E.L., SonBinh, T.N. and Grubbs, R.H. J. Am. Chem. Soc. 1997, / 19, 3887- 3897.
(7) Chiefari, J. et al. Macromolecules 1998, 31, 5559-5562.
(8) Ibba, M. and Soil, D. Science 1999, 286, 1893-1897.
(9) Krejchi, M.T., Atkins, E.D.T., Waddon, A.J., Fournier, M.J., Mason, T.L. and
Tirrell, D.A. Science 1994, 265, 1427-1432.
(10) Parkhe, A.D., Fournier, M.J., Mason, T.L. and Tirrell, D.A. Macromolecules 1993, 26(24), 6691-6693.
(11) McGrath, K.P., Fournier, M.J., Mason, T.L. and Tirrell, D.A. J. Am. Chem. Soc. 1992, 114, 727-733.
(12) Creel, H.S., Fournier, M.J., Mason, T.L. and Tirrell, D.A. Macromolecules 1991,
24, 1213-1214.
Fournier, M.J., Mason, T.L. and ( 1 3) Yu, S.M., Conticello, V., Zhang, G., Kayser, C.,
Tirrell, D.A. Nature 1997, 389, 187-190.
Science (14) Petka, W.A., Hardin, J.L., McGrath, K.P., Wirtz, D. and Tirrell, D.A. 1998,257,389-392.
25 .
Hecht, S.M., Alford, (15) B.L., Kuroda, Y. and Kitano, S. J. Biol. Chem. 1978, 253, 4517-4520.
(16) Noren, C.J., Anthony-Cahill, S.J., Griffith, M.C. and Schultz, P.G. Science 1989, 244, 182-188.
(17) Bain, J.D., Glabe, C.G., Dix, T.A. and Chamberlin, A.R. J. Am. Chem. Soc. 1989, 111, 8013-8014.
(18) Gallivan, J.P., Lester, H.A. and Dougherty, D.A. Chem. Biol. 1997, 4, 739-349.
(19) Hohsaka, T., Ashizuka, Y., Sasaki, H., Murakami, H. and Sisido, M. J. Am. Chem. Soc. 1999, 121, 12194-12195.
(20) Furter, R. Protein Sci. 1998, 7, 4 1 9-426.
(21) Ohno, S., Yokogawa, T., Fujii, I., Asahara, H., Inokuchi, H. and Nishikawa, K. J.
Biochem. 1998, 124, 1065-1068.
(22) Liu, D.R. and Schultz, P.G. Proc. Natl. Acad. Sci. USA 1999, 96, 4780-4785.
(23) Pastrnak, M., Magliery, T.J. and Schultz, P.G. Helv. Chim. Acta 2000, 83, 2277- 2286.
(24) Wang, L., Magliery, T.J., Liu, D.R. and Schultz, P.G. J. Ant. Chem. Soc. 2000, 722,5010-5011.
(25) Kowal, A.K., Kohrer, C. and RajBhandary, U.L. Proc. Natl. Acad. Sci. 2001, 98, 2268-2273.
(26) Wei, Y., Hendrickson, W.A., Crouch, R.J. and Satow, Y. Science 1990, 249, 1398-1405.
J. 1990, 1665- (27) Hendrickson, W.A., Horton, J.R. and LeMaster, D.M. EMBO 9, 1672.
845-85 1 (28) Hendrickson, W.A. J. Synchrotron Rad. 1999, 6,
26 .
Yoshikawa, (29) E., Fournier, M.J., Mason, T.L. and Tirrell, D.A. Macromolecules 1994,27, 5471-5475.
(30) Tang, Y., Ghirlanda, G., Vaidehi, N., Kua, J., Mainz, D.T., Goddard III, W.A., DeGrado, W.F. and Tirrell, D.A. Biochemistry 2001, 40, 2790-2796.
(3 Deming, T.J., 1 ) Fournier, M.J., Mason, T.L. and Tirrell, D.A. J. Macromol. Sci. PureAppl. Chem. 1997,^34,2143-2150.
(32) Reizer, J., Finley, K., Kakuda, D., MacLeod, C.L. and Reizer, A.a.S., M. H.
Protein Sci. 1993, 2, 20-30.
(33) Tarn, R. and Saier, M.H. Microbiol. Rev. 1993, 57, 320-346.
(34) Jakubowski, H. and Goldman, E. Microbiol. Rev. 1992, 56, 412-429.
(35) Tuve, T. and Williams, H. J. Am. Chem. Soc. 1957, 79, 5830-583 1
(36) Cowie, D.B., Cohen, G.N., Bolton, E.T. and Robichon-Szulmajster, H.D.
Biochim. Biophys. Acta 1959, 34, 39-46.
(37) Richmond, M.H. Bacteriol. Rev. 1962, 26, 398-420.
(38) Richmond, M.H. J. Mol. Biol. 1963, 6, 284-294.
(39) Hortin, G. and Boime, \. Methods. Enzymol. 1983, 96, 177-184.
(40) Wilson, M.J. and Hatfield, D.L. Biochim. Biophys. Acta 1984, 781, 205-215.
(41) Bogosian, G., Violand, B.N., Dorward-King, E.J., Workman, W.E., Jung, P.E.
and Kane, I.E. J. Biol. Chem. 1989, 264, 531-539.
(42) Budisa, N., Steipe, B., Demange, P., Eckerskorn, C, Kellermann, J. and Huber, R
Eur. J. Biochem. 1995, 230, 788-796.
C.B., (43) Apostol, L, Levine, J., Lippincott, J., Leach, J., Hess, E., Glascock,
Weickert, M.J. and Blackmore, R.J. J. Biol. Chem. 1997, 272, 28980-28988.
1 13-121. (44) Yuan, T. and Vogel, H.J. Protein Sci. 1999, 5,
27 .
(45) van Hest, J.C.M. and Tirrell, D.A. FEBS Lett. 1998, 428, 68-70.
(46) van Hest, J.C.M., Kiick, K.L. and Tirrell, D.A. J. Am. Chem. Soc. 2000, 722, 1282-1288.
(47) Kothakota, S., Dougherty, M.J., Fournier, M.J., Mason, T.L., Yoshikawa, E. and Tirrell, D.A. Macromol. Symp. 1995, 98, 573-583.
(48) Cowie, D.B. and Cohen, G.N. Biochim. Biophys. Acta 1957, 26, 252-261
(49) Dougherty, M.J., Kothakota, S., Mason, T.L., Tirrell, D.A. and Fournier, M.J. Macromolecules 1993, 26, 1779-1781.
(50) Duewel, H., Daub, E., Robinson, V. and Honek, J.F. Biochemistry 1997, 36, 3404-3416.
(5 1 ) Piperno, J.R. and Oxender, D.L. J. Biol. Chem. 1968, 243, 59 14-5920.
(52) Ingraham, J.L., Maaloe, O. and Neidhardt, F.C. Growth of the Bacterial Cell; Sinauer Associates Inc.: Boston, MA, 1983.
(53) Mendel, D., Ellman, J.A., Chang, Z., Veenstra, D.L., Kollman, P.A. and Schultz, P.G. Science 1992, 256, 1798:1802.
(54) Cornish, V.W., Mendel, D. and Schultz, P. Angew. Chem. Int. Ed. Engl. 1995, 34, 621-633.
(55) Aberg, A., Yaremchuk, A., Tukalo, M., Rasmussen, B. and Cusack, S.
Biochemistry 1997, 36, 3084-3094.
(56) Arnez, J.G. and Steitz, T.A. Biochemistry 1994, 33, 7560-7567.
(57) Arnez, J.G., Harris, D.C., Mitschler, A., Rees, B., Francklyn, C.S. and Moras, D. EMBOJ. 1995, 7^,4143-4155.
(58) Arnez, J.G. and Steitz, T.A. Biochemistry 1996, 35, 14725-14733.
28 (59) Arnez, J.G., Augustine, J.G., Moras, D. and Francklyn, C.S. Proc. Natl. Acad Sci USA 1997,94,7144-7149.
(60) Arnez, J.G., Dock-Bregcon, A.C. and Moras, D. J. Mol. Biol. 1999, 1449- 1459.
(6 Badger, J. 1 ) et al. Proc. Natl. Acad. Sci. USA 1988, 85, 3304-3308.
(62) Belrhali, H. et al. Science 1994, 263, 1432-1436.
(63) Biou, v., Yaremchuk, A., Tukalo, M. and Cusack, S. Science 1994, 263, 1404- 1410.
Briand, (64) C, Poterszman, A., Eiler, S., Webster, G., Thierry, J.C. and Moras, D. J. Mol. Biol. 2000, 299, 1051-1060.
(65) Brick, P. and Blow, D.M. J. Mol. Biol. 1987, 194, 287-297.
(66) Brick, P., Bhat, T.N. and Blow, D.M. J. Mol. Biol. 1989, 208, 83-98.
(67) Bullock, T.L., Sherlin, L.D. and Perona, J.J. Nature Struct. Biol. 2000, 7, 497- 504.
(68) Cavarelli, J., Delagoutte, B., Eriani, G., Gangloff, J. and Moras, D. EMBO J.
1998, 17, 5438-5448.
(69) Delarue, M., Poterszman, A., Nikonov, S., Garber, M., Moras, D. and Thierry,
J.C. EMBO J. 1994, 13, 3219-3229.
8- (70) Desogus, G., Todone, P., Brick, P. and Onesti, S. Biochemistry 2000, 39, 84 1 8425.
L., Thierry, J.C. and Moras, D. EMBO (7 1 ) Eiler, S., Dock-Bregeon, A.C., Moulinier,
J. 1999, 18, 6532-6541.
(72) Fourmy, D., Meinnel, T., Mechulam, Y. and Blanquet, S. J. Mol. Biol. 1993, 231, 1068-1077.
1078-1089. (73) Fourmy, D., Dardel, F. and Blanquet, S. J. Mol. Biol. 1993, 231,
29 . .
(74) Ilyin, V.A., Temple, B., Hu, M., Li, G.P., Yin, Y.H., Vachette, P. and Carter,
C.W. Protein Sci. 2000, 9, 2 1 8-23 1
(75) Logan, D.T., Mazauric, M.H., Kern, D. and Moras, D. EMBO J. 1995, 14, 4 1 56- 4167.
(76) Mosyak, L., Reshetnikova, L., Goldgur, Y., Delarue, M. and Safro, M.G. Nature Struct. Biol. 1995, 2, 537-547.
(77) Nureki, O. et al. Science 1995, 267, 1958-1965.
(78) Onesti, S., Miller, A.D. and Brick, P. Structure 1995, 3, 163-176.
(79) Onesti, S., Desogus, G., Brevet, A., Chen, J., Plateau, P., Blanquet, S. and Brick, P. Biochemistry 2000, 39, 12853-12861.
Perona, J.J., (80) Rould, M. A. and Steitz, T. A. Biochemistry 1993, 32, 8758-877 1
(8 1 ) Poterszman, A., Delarue, M., Thierry, J.C. and Moras, D. J. Mol. Biol. 1994, 244, 158-167.
(82) Qiu, X.Y., Janson, C.A., Blackburn, M.N., Chhohan, LK., Hibbs, M. and Abdel-
Meguid, S.S. Biochemistry 1999, 38, 12296-12304.
(83) Rath, V.L., Silvian, L.F., Beijer, B., Sproat, B.S. and Steitz, T.A. Structure Fold.
Des. 1998, 6, 439-449.
(84) Rees, B., Webster, G., Delarue, M., Boeglin, M. and Moras, D. J. Mol. Biol. 2000, 299, 1157-1164.
(85) Reshetnikova, L., Moor, N., Lavrik, O. and Vassylyev, D.G. J. Mol. Biol. 1999, 287, 555-568.
1 135-1 (86) Rould, M.A., Perona, J.J., Soil, D. and Steitz, T.A. Science 1989, 246, 142.
213-218. (87) Rould, M.A., Perona, J.J. and Steitz, T.A. Nature 1991, 352,
(88) Ruff, M. et al. Science 1991, 252, 1682-1689.
30 .
(89) Sankaranarayanan, R. et al. Cell 1999, 97, 37 1-38 1
(90) Sankaranarayanan, R., Dock-Bregeon, A.C., Rees, B., Bovee, M., Caillet, J., Romby, P., Francklyn, C.S. and Moras, D. Nature Struct. Biol. 2000, 7, 461-465.
Sauter, Lorber, E., (91) C, Cavarelli, J., Moras, D. and Giege, R. J. Mol. Biol. 2000, 299, 1313-1324.
Schmitt, E., (92) Panvert, M., Blanquet, S. and Mechulam, Y. EMBO J. 1998, 17, 6819-6826.
(93) Schmitt, E., Moulinier, L., Fujiwara, S., Imanaka, T., Thierry, J.C. and Moras, D.
EMBO J. 1998, 17, 5227-5237.
(94) Sherlin, L.D., Bullock, T.L., Newberry, K.J., Lipman, R.S.A., Hou, Y.M., Beijer,
B., Sproat, B.S. and Perona, J.J. J. Mol. Biol. 2000, 299, 431-446.
(95) Silvian, L.F., Wang, J.M. and Steitz, T.A. Science 1999, 285, 1074-1077.
(96) Sugiura, I. et al. Struct. Fold. Des. 2000, 8, 197-208.
(97) Eriani, G., Delarue, M., Poch, O., Gangloff, J. and Moras, D. Nature 1990, 347, 203-206.
(98) Rossmann, M.G., Moras, D. and Olsen, K.W. Nature 1974, 250, 194-199.
(99) Meinnel, T., Mechulam, Y. and Blanquet, S. tRNA: Structure, Biosynthesis, and Function; ASM Press: Washington, D. C, 1995.
(100) Cusack, S., Hartlein, M. and Leberman, R. Nucleic Acids Res. 1991, 19, 3489- 3498.
(101) Moras, D. Trends Biochem. Sci. 1992, 7 7, 1 59- 1 64.
(102) Edelmann, P. and Gallant, J.A. Cell 1977, 10, 131-137.
Mechanism; W. H. Freeman and Company: ( 1 03) Fersht, A. Enzyme Structure and New York, 1999.
31 .
(104) Fukai, S., Nuieki, O., Sekine, S.-I., Shimada, A., Tao, J., Vassylycv, D.G. and Yokoyama, S. Cell 2000, 103, 793-803.
Barker, D.G., (105) Ebel, J.-P., Jakes, R. and Bruton, C.J. Eur. J. Biochem. 1982, /27, 449-457.
Dardel, F., Fayat, (106) G. and Blanquet, S. J. Bacteriol. 1984, 160, 1 1 15-1 122.
(107) Cassio, D. and Waller, J.-P. Eur. J. Biochem. 1971, 20, 283-300.
(108) Mcllot, P., Mechulam, Y., LeCorre, D., Blanquet, S. and Fayat, G. J. Mol. Biol. 1989, 208, 429-443.
(109) Nureki, O., Muramatsu, T., Suzuki, K., Kohda, D., Matsuzawa, H., Ohta, T.,
Miyazawa, T. and Yokoyama, S. J. Biol. Chem. 1991, 266, 3268-3277.
(110) Mechulam, Y. et al. Nucleic Acids Res. 1991, 19, 3673-3681.
(111) Tzagoloff, A., Vambutas, A. and Akai, A. Eur. J. Biochem. 1989, / 79, 365-37 1
(112) Walter, P., Gangloff, J., Bonnet, J., Boulanger, Y., Ebel, J.-P. and Fasiolo, F.
Proc. Natl. Acad. Sci. USA 1983, 80, 2437-2441.
(113) Fasiolo, F., Gibson, B.W., Walter, P., Chatton, B., Biemann, K. and Boulanger,
Y. J. Biol. Chem. 1985, 260, 15571-15576.
(114) Walter, P., Weygand-Durasevic, I., Sanni, A., Ebel, J.-P. and Fasiolo, F. J. Biol. Chem. 1989, 264, 17126-17130.
(115) Lage, H. andDietel, M. Gene 1996, 178, 187-189.
(116) Schmitt, E., Panvert, M., Mechulam, Y. and Blanquet, S. Eur. J. Biochem. 1997, 246, 539-547.
M.-K., Yoon, J.H. and Kim, S. J. (1 17) Jo, Y.J., Lee, S.W., Jo, M.K., Lee, J.W., Kang, Biochem. Mol. Biol. 1999, 32, 547-553.
32 (118) Posorske, L.H., Cohn, M., Yanagisawa, N. and Auld, D.S. Biochim. Biophys. Acta 1979,576, 128-133.
(119) Mayaux, J.-F. and Blanquet, S. Biochemistry 1981, 20, 4647-4654.
Fourmy, D., (120) Meinnel, T., Mechulam, Y. and Blanquet, S. J. Mol. Biol. 1993, 231, 1068-1077.
(121) Mayaux, J.F., Kalogerakos, T., Brito, K.K. and Blanquet, S. Eur. J. Biochem. 1982, 728,41-46.
( 1 22) Nureki, O., Kohno, T., Sakamoto, K., Miyazawa, T. and Yokoyama, S. J. Biol. Chem. 1993, 268, 15368-15373.
( 1 23) Brunie, S., Zel wer, C. and Risler, J.-L. J. Mol. Biol. 1990, 2/6,411 -424.
( 1 24) Mechulam, Y., Schmitt, E., Maveyraud, L., Zelwer, C, Nureki, O., Yokoyama,
S., Konno, M. and Blanquet, S. J. Mol. Biol. 1999, 294, 1287-1297.
(125) Serre, L., Verdon, G., Choinowski, T., Hervouet, N., Risler, J.-L. and Zelwer, C.
J. Mol. Biol. 2001, 306, 863-876.
(126) Fourmy, D., Mechulam, Y., Brunie, S., Blanquet, S. and Fayat, G. FEBS Lett. 1991, 292, 259-263.
1 ( 27) Ghosh, G., Pelka, H., Schulman, L.H. and Brunie, S. Biochemistry 1991 , 30, 9569-9575.
Natl. ( 1 28) Kim, H. Y., Ghosh, G., Schulman, L.H., Brunie, S. and Jakubowski, H. Proc.
Acad. Sci. USA 1993, 90, 1 1553-1 1557.
(129) Jakubowski, H. Proc. Natl. Acad. Sci. USA 1990, 87, 4504-4508.
(130) Jakubowski, H. EMBO J. 1991, 10, 593-598.
237-240. (131) Jakubowski, H. and Goldman, E. FEBS Lett. 1993, 317,
1 1528-1 1535. (132) Gao, W., Goldman, E. and Jakubowski, H. Biochemistry 1994, 33,
33 (133) Jakubowski, H. and Fersht, A.R. Nucleic Acids Res. 1981, 9, 3105-3117.
( 1 34) Jakubowski, H. Biochemistry 1996, 35, 8252-8259.
(135) Mechulam, Y., Dardel, F., Le Corre, D., Blanquet, S. and Fayat, G. J. Mol Biol 1991,277,465-475.
Ghosh, G., (136) Brunie, S. and Schulman, L.H. J. Biol. Chem. 1991, 266, 17136- 17141.
( 1 Ghosh, 37) G., Pelka, H. and Schulman, L.D. Biochemistry 1990, 29, 2220-2225.
(138) Ghosh, G., Kim, H.Y., Demaret, J.-P., Brunie, S. and Schulman, L.H.
Biochemistry 1991, 30, 1 1767-1 1774.
(139) Meinnel, T., Mechulam, Y., LeCorre, D., Panvert, M., Blanquet, S. and Fayat, G. Proc. Natl. Acad. Sci. USA 1991, 88, 291-295.
(140) Kim, H.Y., Pelka, H., Brunie, S. and Schulman, L.H. Biochemistry 1993, 32, 10506-10511.
(141) Hountondji, C., Blanquet, S. and Lederer, F. Biochemistry 1985, 24, 1 175-1 180.
(142) Hountondji, C., Schmitter, J.M., Fukui, T., Tagaya, M. and Blanquet, S.
Biochemistry 1990, 29, 1 1266-1 1273.
(143) Meinnel, T. et al. Biochimie 1990, 72, 625-632.
(144) Schmitt, E., Guillon, J.M., Meinnel, T., Mechulam, Y., Dardel, F. and Blanquet,
S. Biochimie 1996, 78, 543-554.
(145) Jakubowski, H. J. Biol. Chem. 2000, 275, 21813-21816.
(146) Li, S., Kumar, N.V., Varshney, U. and RajBhandary, U.L. J. Biol. Chem. 1996, 277, 1022-1028.
2281-2286. (147) Skinner, C.G., Edelson, J. and Shive, W. J. Am. Chem. Soc. 1961, 83,
34 (148) Schenck, H.L., Dado, G.P. and Gellman, S.H. J. Am. Chem. Soc. 1996 / 18 12487-12494.
(149) Maier, K.L., Lenz, A.G., Beck-Speier, I. and Costabel, U. Methods Enzymol 1995, 257,455-461.
35 CHAPTER 2
IN VITRO ACTIVATION OF METHIONINE ANALOGUES BY METHIONYL-tRNA SYNTHETASE
2.1 Introduction and Objectives
Early investigations of the in vivo incorporation of methionine analogues (1-3)
demonstrated that a variety of methionine analogues can be incorporated into proteins by
the biosynthetic machinery. Initial investigations in our laboratories demonstrated that
the unsaturated analogues 2 and 3 replace methionine in proteins at levels of substitution
as high as 98% (2,3), while analogues 4 - 8 do not support protein synthesis in vivo in a
conventional bacterial host. Although protein biosynthesis is thought to be controlled
most stringently by the aaRS, there had been no systematic investigations of the role of
MetRS in controlling the incorporation of methionine analogues into proteins in vivo.
Therefore, the objectives of these investigations were to characterize the kinetics of
activation of a series of methionine analogues 2 - 13 (Appendix B) in vitro and to
correlate these results with the ability of these analogues to support protein synthesis in
the absence of added methionine in vivo.
36 2.2 Experimental Section
2.2.1 Analogue Synthesis
Each of the - analogues 2 7 and 11 (Figure 2.1) was prepared by alkylation of diethyl acetamidomalonate with the appropriate tosylate followed by decarboxylation and deprotection of the amine function, as previously described (2-4). Analogues 8, 9, 12, and 13 are available commercially (Sigma-Aldrich, St. Louis, MO). Analogue 10 was a
generous gift from Helen Blackwell and was prepared as described by Blackwell et. al
(5). Analogues 2 - 7 and 10 - 13 were used as the racemates, and all concentrations given are for the L-isomers of the analogues.
2.2.2 Determination of Translational Activity
Buffers and media were prepared according to standard protocols (6). Due to the
preference of MetRS for methionine over analogues, it is necessary that a bacterial
expression host that cannot synthesize its own methionine is used during protein
expression experiments aimed at the incorporation of methionine analogues. In addition,
the medium must be depleted of methionine and supplemented with the analogue of
and then interest. (The medium is supplemented with methionine to permit cell growth,
37 at the time of protein expression, the medium is depleted of methionine and supplemented only with the non-natural amino acid, as described below.)
The E. coli methionine auxotroph CAG 18491 (A', rph-1, metE3079.:TnlO) kindly provided by the Yale E. coli Genetic Stock Center, was transformed with plasmids pREP4 and pQE15 (Qiagen), to obtain the expression host CAG18491/pQE15/pREP4.
The plasmid pQE15 encodes the protein murine dihydrofolate reductase (DHFR) under
the control of a bacteriophage T5 promoter. The expression plasmid also encodes an N- terminal hexahistidine sequence that permits purification of the target protein by
immobilized metal chelate affinity chromatography. Furthermore, DHFR contains eight
methionine residues that can be replaced by methionine analogues, and its expression is easily monitored by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and
visualized by Coomassie blue staining.
2.2.2.1 Small Scale Protein Expression
M9AA medium (50 mL) supplemented with 1 mM MgS04, 0.2 wt% glucose, 1 mg/L thiamine chloride and the antibiotics ampicillin (200 mg/L) and kanamycin (35 mg/L) was inoculated with 2 mL of an overnight culture of CAG 18491/ pREP4/pQE15.
When the turbidity of the culture reached an optical density at 600 nm (ODeoo) of 0.8, a
culture. The cells were medium shift was performed to remove methionine from the cell
38 sedimented for 10 min at 3030 x at ^ 4°C, the supernatant was removed, and the cell pellet was washed twice with 20 of 1 x mL M9 salts. Cells were resuspended in 50 mL of the M9AA medium described above, without methionine. Test tubes containing 5 mL aliquots of the resulting culture were prepared, and were supplemented with 300 jiL 1
mg/mL either methionine or 2-13. A culture lacking methionine (or any analogue)
served as the negative control. Protein expression was induced by addition of isopropyl- p-D-thiogalactopyranoside (IPTG, Calbiochem) to a final concentration of 0.4 mM.
Cultures were grown for 4.5 hours, the OD^fx, was measured, and 1 mL of each of the
samples was sedimented. After the supernatant was decanted, the cell pellets were
resuspended in a volume of HjO (OD(,o„ x 100 |iL) to yield a normalized OD(,oo of 10.
Protein expression was monitored by SDS polyacrylamide gel electrophoresis (12%
acrylamide running gel, 12 mA, 14 h) and visualized by Coomassie blue staining.
2.2.2.2 Large Scale Protein Expression
Similar procedures were used for preparation and isolation of DHFR from media
supplemented with 1, 2, 3, or 9. M9AA medium (100 mL) supplemented with 1 niM
ampicillin (200 MgS04 0.2w% glucose, 1 mg/L thiamine chloride and the antibiotics
mg/L) and kanamycin (35 mg/L) was inoculated with E. coli strain
39 CAG18491/pQE15/pREP4 and grown overnight at 37°C. This culture was used to inoculate 900 mL M9AA medium supplemented as described. The cells were grown to an OD^oo of approximately 0.9 and the medium shift was performed as described for the small-scale experiments. Washes were conducted with 500 mL of 1 x M9 salts, and the
washed cell pellet was resuspended in 1 L of M9 + 19AA media (- methionine). An
aliquot (20 mL) of 1 mg/mL of the analogue was added, followed by IPTG (0.4 mM), and cultures were grown for 4.5 hours. The OD^tx, was measured, and 1 mL-samples
were sedimented and resuspended as described for the small-scale cultures. Protein
expression was monitored by SDS polyacrylamide gel electrophoresis. The 1-L culture
was sedimented (9800 x g, \0 minutes, 4°C), the supernatant removed, and the cell pellet
stored at -80°C overnight.
2.2.2.3 Protein Purification
The cell pellet was thawed for 30 min at 37°C, 30 mL of buffer (6 M guanidine-
HCl, 0.1 M NaH2P04, 0.01 M Tris, pH 8) was added, and the mixture was shaken at room
temperature for 1 hour. The cell lysate was then sonicated for 60 seconds in 1 second
bursts. The cell debris was sedimented (15,300 x g, 30 min, 4°C) and the supernatant was
subjected to immobilized metal affinity chromatography (Ni-NTA resin) under
denaturing conditions according to the procedure described by Qiagen (7). The
40 supernatant was loaded onto 10 mL of resin which was then washed with 50 niL of guanidine buffer followed by 25 mL of urea buffer (8 M urea, 0. 1 M NaH2P04 and 0.01
M Tris, pH 8). Similar urea buffers were used for three successive 25 mL washes at pH values of 6.3, 5.9 and respectively. 4.5, Target protein was obtained in washes at pH 5.9 and 4.5. These washes were combined and dialyzed (Spectra/Por membrane 1, MWCO =
6-8 kDa) against running distilled water for 4 days, followed by batchwise dialysis against doubly distilled water for one day. The dialysate was lyophilized to yield 35 - 40
mg of modified DHFR for cultures grown on medium supplemented with 3, similar to the
yield obtained for DHFR from cultures supplemented with methionine. Proteins
produced from cultures (1 L) grown on medium supplemented with 2 yielded 10 mg of
DHFR, while supplementation with 9 yielded 20 mg of protein. A control experiment in
2xYT medium afforded 60 mg of DHFR per liter culture.
2.2.3 MetRS Expression and Purification
The fully active, truncated form of the wild-type MetRS was purified from 24-
hour cultures of JMlOl cells carrying the plasmid pGG3 (8). (The pGG3 plasmid was
kindly donated by Professor Hieronim Jakubowski of UMDNJ-New Jersey Medical
School, Newark, New Jersey.) The plasmid pGG3 encodes MetRS under control of the
constitutive expression E. coli promoter metGpl (Genbank accession number X55791);
41 of the enzyme in the cellular host provided sufficient levels of enzyme for facile purification and use. The enzyme was purified by size exclusion chromotography as
previously described The cell (9). pellet from a 24-hour culture was resuspended in 360
of mL 20 mM Tris-Cl, 0.1 mM EDTA (pH 7.6) and stored at -SOT overnight. The
suspension was thawed ice on in a cold room (4°C) and 360 |iL of p-mercaptoethanol (p-
ME) was added. Streptomycin sulfate (12 g, 3% w/v) was added to the mixture and the
solution was stirred at 4°C for two hours. The cell lysate was clarified by centrifugation
X (25000 g, 45 minutes, 4°C). Ammonium sulfate (80 g, 35%) was added, the solution
was stirred at 4°C overnight, and the precipitate was removed from the solution by
centrifugation as above. Additional ammonium sulfate (95g, 70%) was added to the
supernate, the solution was stirred at 4°C for 4 hours, and the precipitate was collected by
centrifugation as above. The pellet was dissolved in 2 mL 10 mM phosphate buffer, 10
mM P-ME (pH 6.8), and 1 mL of this solution was loaded onto a Sephacryl S-100 HiPrep
size exclusion column and eluted isocratically with 10 mM phosphate, 10 mM P-ME
buffer at pH 6.8. The fractions containing MetRS (as determined by absorbance
measurements at 280 nm) were collected and concentrated in a centricon filter apparatus
(10,000 MWCO) via centrifugation. The purified enzyme was stored as 3-|iM solutions
MetRS, in 40% glycerol at -20°C until used for enzyme assays. (A truncated mutant
W305F, was also purified identically from overnight cultures of DH5aF' carrying the
42 plasmid PBSM547W305F. The plasmid pBSM547W305F was kindly donated by
Professor Yves Mechulam of the Ecole Polytechnique, Palaiseau Cedex, France.)
2-2.4 Activation of Methionine Analogues in Vitro
Activation of methionine analogues by wild-type MetRS was assayed via the
amino-acid-dependent ATP-PP, exchange reaction, also as previously described (9-11).
The assay, which measures the ^^P-radiolabeled ATP formed by the enzyme-catalyzed
exchange of "P-pyrophosphate (PP^) into ATP, was conducted in 150 |il of reaction
buffer (pH 7.6, 20 mM imidazole, 0.1 mM EDTA, 10 mM p-mercaptoethanol, 7 mM
MgCl2, 2 mM ATP, 0.1 mg/ml BSA, and 2 mM PPi(in the form of sodium pyrophosphate
Life (NEN Science Products, Inc.) with a specific activity of 0.2-0.5 TBq/mole)). (1
10'° Becquerel (Bq) = 1 disintegration per second (dps); 3.7 x Bq = 1 Curie (Ci).)
Assays to determine if the methionine analogues 2 - 13 support PPj exchange in the
presence of MetRS were conducted in solutions 75 nM in enzyme and 5 mM in the L-
isomer of the analogue with a reaction time of 20 minutes. Quantitative kinetic
parameters for analogues 5 and 11 were obtained with an enzyme concentration of 75 nM
and analogue concentrations of 100 |iM to 20 mM. Parameters for methionine were
obtained by using concentrations ranging from 10 |J.M to 1 mM. values for
43 methionine matched those previously reported (12), although the measured was somewhat lower than the literature value. Aliquots of 20 ^il were removed from the
reaction mixture at various time points and were quenched in 0.5 ml of a solution comprising 200 NaPP,, mM 7% w/v HCIO4, and 3% w/v activated charcoal. The charcoal was rinsed twice with 0.5 mL of a 10 mM NaPP., 0.5% HCIO4 solution and was then resuspended in 0.5 mL of this solution and counted via liquid scintillation methods.
Kinetic constants were calculated by nonlinear regression fit of the data to a Michaelis
Menten model.
2.2.5 Computational Modeling of Methionine Analogues
Single-point energy ab initio calculations (Hartree-Fock model, 6-3 IG* basis set)
were performed for methionine and for analogues 2, 3, 5, and 11 with fully extended side
chains. Electron density maps are shown as surfaces of electron density 0.08
electrons/au^ Isopotential plots are represented as surfaces where the energy of
interaction between the amino acid and a point positive charge is equal to -10 kcal/mole.
Calculations were performed by using the program MacSpartan (Wavefunction, Inc.,
Irvine, CA, USA).
44 2.3 Results and Discussion
2-3.1 Determination of Translational Activity of Methionine Analogues
bacterial A host strain (CAG18491/pQE15/pREP4) suitable for testing the
translational activity of methionine analogues 2 - 8 and 10 - 13 was prepared by
transformation of E. coli strain CAG 18491, a methionine auxotroph, with the repressor plasmid pREP4 and the expression plasmid pQE15. The expression plasmid pQEI5
encodes murine DHFR, which contains 8 methionine residues as possible sites of
substitution by non-natural amino acids. The translational activity of each analogue was
assessed on the basis of its capacity to support synthesis of DHFR in cultures of
CAG18491/pQE15/pREP4 that had been depleted of methionine.
The results of the iti vivo assays illustrated in Figure 2.1 corroborate our previous
reports that 2 and 3 serve effectively as methionine surrogates in bacterial protein
synthesis (2, 3)- Analysis of purified DHFR produced from cultures supplemented with
these analogues confirmed incorporation of 2 and 3 at levels up to 98%, also as
previously reported. In contrast, analogues 4 - 8 and 10 -13 do not support measurable
levels of protein synthesis in bacterial cultures depleted of methionine. It is highly
unlikely that recognition by EF-Tu, recognition by the ribosome, or transport into the cell
remarkably are the limiting factors for incorporation of the.se analogues. The ribo.some is
45 permissive toward amino acid analogues with widely varying chemical functionality, as has been demonstrated by the numerous analogues incorporated into proteins in in vitro translation experiments (13-20).
Transport of many analogues into the bacterial cell is indicated by a number of
literature reports. Analogue is 4 an antagonist for methionine, inhibiting the growth of E. coli cells (21) and replaces leucine 8 in human hemoglobin expressed in E. coli (22).
Although there is no similar evidence reported for the other analogues, the fact that trifluoromethionine and ethionine are incorporated into proteins expressed in E. coli
(1,23-25) suggests that neither the trifluoromethyl group nor the longer side chain will
inhibit transport of analogues 6 and 7 into E. coli cells.
2.3.2 MetRS Expression and Purification
The attachment of an amino acid to its cognate tRNA proceeds in two steps, as
below,
aaRS + aa+ATP [aaRS:aa~AMP] + PP,
[aaRS:aa~AMP] + tRNA'''' " aa-tRNA'"' + AMP + aaRS
where aaRS is the enzyme, aa is the amino acid, PP, is pyrophosphate, aaRS:aa~AMP is
the ami noacyl adenylate complexed with the enzyme, and aa-tRNA"" is the
46 aminoacyl-tRNA. Activation, the first step, involves the enzyme-catalyzed formation of an aminoacyl adenylate (aa~AMP) and can be studied by monitoring the rate of exchange
of radiolabeled pyrophosphate (-^^P-PP. ) into ATP (26,27). Aminoacylation, the second step, can be studied by monitoring the amount of radiolabeled amino acid attached to
tRNA in the presence of the enzyme. Because initial recognition of an amino acid by its
aaRS is perhaps the most critical step in the incorporation of non-natural amino acids into proteins in vivo, the in vitro activation of methionine analogues by MetRS has been
measured and compared to results of studies of in vivo incorporation.
Purified enzyme preparations are necessary to conduct the in vitro assays, and
SDS-PAGE analysis was used as an assessment of MetRS purity. The truncated form of
MetRS was expressed from 24-hour cultures and purified by size-exclusion methods.
The SDS-PAGE analysis of the isolated MetRS (both wild-type and mutant) is shown in
Figure 2.2. As indicated in the figure, both the wild-type and mutant forms of MetRS
were readily purified from cell lysates by size-exclusion chromatography.
2.3.3 Activation of Methionine Analogues by MetRS in Vitro
The relative rates of activation of methionine and methionine analogues 2 -13 by wild-type MetRS were estimated by the ATP-PP, exchange assay. The results of the in
vitro assays shown in Figure 2.3 are consistent with the in vivo results shown in
47 Figure 2.1, as the analogues that support the highest rates of PP. exchange also support protein synthesis in the absence of methionine. Methionine (1) is activated most efficiently by the enzyme, causing exchange of 9 nmoles PP, over the time course of the reaction. Analogues 2 and cause 3 exchange of PP. at rates similar to that of norlcucine
(9), while the remaining analogues 4, 6 - 8, and 12 - 13 cause exchange of PP, at levels n. higher than background (Figure 2.3, lane 14) under these experimental conditions.
Although analogues 5 and 11 effect very slow change of PP^, the rate of activation is apparently too low to support detectable levels of protein synthesis in vivo.
Kinetic parameters were obtained for methionine, 2, 3, 5, 9, and 11 as described
in the experimental section; results are shown in Table 2.1. Our measured value of K,„
for methionine matched previously reported values (12), although the value determined
for k,„, was slightly lower than that reported. Comparison of the k,.,„/K,„ values obtained
for methionine (0.54 s ' |liM ') and the above analogues show that the analogues are 500-
fold to 13825-fold poorer substrates for the enzyme, with catalytic efficiencies in
activation decreasing in the following order: Met >3>9>2>5>11. Even 3, which
supports the same protein yields as methionine in one-liter cultures, is 500-fold poorer a
substrate for the enzyme than methionine. That the loss of catalytic activity is manifest
both in increasing values of K,„ and decreasing values of k,,„ indicates that the analogues
are less efficient in both the binding and catalytic events of the enzyme.
48 Table 2. also 1 demonstrates that methionine analogues that are activated up to
2000-fold more slowly by MetRS than methionine can support protein biosynthesis in a conventional bacterial expression host in the absence of methionine. These results are comparable to those reported previously for the in vitro activation and in vivo incorporation of phenylalanine analogues (28-30); comparisons for other amino acids have been limited by a lack of in vitro activation data. The data suggest that non-natural amino acids can support protein synthesis in vivo even with surprisingly inefficient activation of the amino acid by its aaRS, and that under normal cellular conditions, activation of a natural amino acid by its aaRS is not the limiting factor in protein biosynthesis.
2.3.4 Determination of Protein Yields
The trend in protein yield shown in Table 2. 1 , coupled with the early observation
that the levels of protein synthesis in cultures supplemented with methionine or 3 were
significantly greater than those in cultures supplemented with 2 (2,3), led us to probe the
correlation of the catalytic efficiency of activation in vitro with protein yield in vivo.
This correlation was probed by comparing the kinetic constants for analogue activation
by MetRS with the yield of the target protein DHFR obtained from one-liter cultures of
the bacterial host CAG18491/pQE15/pREP4.
49 The relative kinetic constants for analogue activation and the corresponding protein yields are illustrated on the bar graph shown in Figure 2.4. Analogues with the highest values support the highest levels of protein synthesis; the protein yields scale remarkably well with the k,,/K, values, at least for the poorer substrates. Analogue
3 supports protein synthesis with yields equivalent to those obtained with methionine (35 mg/L), despite the fact that 3 is a 500-fold poorer substrate for MetRS than methionine.
Bacterial cultures supplemented with 9 (1050-fold lower k,,/KJ produce approximately
as 57% much DHFR as cultures supplemented with methionine, and cultures
supplemented with 2 (1850-fold lower kJKJ produce 28% of the control yield of protein. Bacterial cultures supplemented with 5 (4700-fold lower k,,/KJ or 11 (13825-
fold lower k,3/K^) do not support measurable levels of protein synthesis in this
expression host. These results indicate that the rate of methionine analogue activation in
vitro does indeed correlate with protein yields in vivo, and suggest that the kinetics of
activation can play a critical role in controlling the rate of protein synthesis in
methionine-depleted cultures supplemented with analogues that are poor substrates for
MetRS.
50 2.3.5 Computational Modeling of Methionine Analogues
Computational modeling of methionine analogues provides a basis for comparison of structural and electronic properties of the methionine analogues and
potential insight into the analogue features that are important for analogue activation by
MetRS. Comparisons of the structural and electronic properties between methionine and
the analogues may be useful for better understanding the activation of the analogues,
especially given that the crystal structure of MetRS with bound methionine has been
solved (31), and the side chain features important in binding are indicated.
Figure 2.5 compares the isopotential surfaces calculated for methionine and for
analogues 2, 3, 5, and 11. The representation of the extended side chain conformations is
consistent with that observed for methionine in the active site of MetRS (31). That 2
might serve as a substrate for the MetRS is not surprising, given the similar geometries
accessible to 1 and 2, the availability of 7i-electrons near the side-chain terminus of 2, and
importance the known translational activity of 9, the saturated analogue of 2. Indeed, the
that the n- of the thioether of methionine as a hydrogen-bond acceptor (31) suggests
chain of electrons of 2 may serve a similar function. The hydrophobicity of the side 9,
ability of to be activated by the highly coupled with its similar size, likely explains the 9
observed for 3 (i.e., hydrophobic active site of MetRS. The high translational activity
without loss of protein yield), was not near- quantitative replacement of methionine
51 anticipated, since the colinearity of the side - chain carbons 4 6 imposes on 3 a geometry substantially different than that of methionine. However, the electron density associated with the triple bond of 3 is positioned similarly to that of the thioether of the natural
substrate 1, despite the differences in side-chain geometry. Additionally, the isopotential surface of 3 (Figure 2.5) extends from the side chain in the same direction and at a similar distance as that for methionine whereas that of 2 does not, which may make 3 a better substrate for MetRS than 2. Furthermore, given the highly aromatic nature of the methionine binding pocket in MetRS, alkynyl C-H/ti contacts (32) and the polarizability of the unsaturated side chain may also play a significant role in recognition of 3 by the
* enzyme.
Figure 2.5 also compares the geometries of 1, 5, and 11. The latter two analogues
are not translationally active in a conventional bacterial expression host and are not
activated efficiently by MetRS in vitro. Although the geometries of 1 and 5 appear similar in the representation shown, the fixed planarity of the C4-C5 double bond may
preclude the side-chain conformation required for efficient recognition of 5 by MetRS.
The isopotential surface also extends only from the sides of the side chain and not in the
same position as that in methionine, which may reduce the ability of 5 to act as a good
substrate for MetRS. The linearity of the side chain of 11, coupled with its very different
geometry than 1 or 3, must preclude the analogue from achieving a side chain orientation
that permits efficient activation by MetRS. That 11 supports even low rates of PPj
52 exchange may be due to the presence of 7i-electrons in a position on the side chain similar to that of the thioether in methionine. These results suggest that the presence of tt
electrons and the polarizability of the triple bond play a significant role in the recognition of both 3 and 11 by MetRS, given that the analogues are activated by the enzyme despite their linear geometries.
2.4 Summary and Conclusions
Strategies to optimize the incorporation of non-natural amino acids into proteins in vivo are important for engineering protein materials. Toward these ends, a set of twelve methionine analogues was assayed for translational activity in E. coli and activation by MetRS in vitro. Results of the in vitro assays corroborate the in vivo
results, with analogues that support the highest level of PP, exchange in vitro also serving
most effectively as methionine surrogates in vivo. Our results indicate that methionine
analogues that are up to 2000-fold poorer substrates for MetRS in vitro are still able to
support protein synthesis in a conventional bacterial host. Furthermore, quantitative
assessment of the kinetics of activation indicates that the catalytic efficiency for a given
analogue also correlates well with in vivo yields of the test protein DHFR obtained from
cultures supplemented with the analogue, indicating a critical role for MetRS in
methinine analogue incorporation into proleins.
53 The electron density maps and isopotential surfaces for methionine and analogues
2, 3, 5, and 11 indicate the importance of side chain length and electron density in the activation of the analogues by the enzyme. Analogues 2, 3, and 9 all have side chains the same length as that of methionine and all support protein synthesis and PP, exchange
under the assay conditions. In contrast, none of the analogues 7, 8, 12, or 13 show translations activity or support PP^ exchange in these in vitro assays. The position of electron density along the side chains of 2 and 3 likely plays a role in their activation and
may permit relatively efficient activation of 3 despite the linearity of its side chain. The
importance of conformational flexibility was indicated by observations that the internal
alkene function of 4 and 5, and the internal alkyne function of 11, prevented their
incorporation into test protein under conditions employing a conventional bacterial host.
Additionally, the analogues 6 and 10 yielded no evidence of translational activity or PP,
exchange.
These results suggest that the success or failure of analogue incorporation in vivo
is indeed controlled by MetRS. Strategies to incorporate non-natural amino acids into
proteins should therefore include assessment of the aaRS activities of the bacterial host.
54 '
/
Table 2.1 Kinetic parameters for methionine analogues in the ATP-PP, exchange reaction and protein yields for bacterial cultures supplemented with the analogues.
Analogue (^iM) (s ') k,,/K, (s 'liM ') Protein Yield, mg/L
1 24.3 ± 2 13.3 ± 0.2 5.47 X 10 35
3 2415 ± 170 2.60 ± 0.3 1.08 X 10"^ 35
9 4120 ±900 2.15 ±0.6 5.22 X 10 ' 20
' 2 4555 ± 200 1.35 ±0.1 2.96 X 10 10
5 15675 ±250 1.82 ±0.6 1.16 X 10' 0
11 38650 ± 2000 1.51 ±0.5 3.91 X 10' 0
55 ipi IIP ^ map mv
- Met + Met 2 3 4 5 6
Figure 2.1 SDS-PAGE analysis of DHFR synthesis by E. coli strain CAG18491/pQE15/pREP4. Cultures were supplemented with nothing
(-Met) or with methionine or one of the analogues 2-13, as indicated.
56 (a) (b)
62 kDa
Figure 2.2 SDS-PAGE analysis of MetRS purification by size-exclusion
chromatography, (a) Wild-type MetRS (b) Mutant MetRS, W305F. The
molecular weight marker at 62 kDa is indicated; both MetRS enzymes have molecular weights of approximately 62.5 kDa. The proteins were
visualized by Coomassie Blue staining.
57 1 10'
Analogue
Figure 2.3 Activation of methionine and methionine analogues by MetRS. The
amount of PPj exchanged in 20 minutes, as measured in the ATP-PPj
exchange assay, is shown for methionine (1) and analogues 2 - 13. The
background (14) is given for a reaction mixture lacking enzyme and amino
acid.
58 0.0012
1 00%
3 9 2 5 11
Analogue
Figure 2.4 Comparison of kinetic parameters and protein yields for methionine
analogues. Kinetic parameters k^^/K^, normalized to that for methionine
(Table 2.1), are given for the analogues as indicated. The protein yield for
1 L cultures of CAG18491/pQE15/pREP4 supplemented with the
analogue is given as a percentage of the yield obtained for cultures supplemented with methionine.
59 Figure 2.5 Electron density maps (colored surfaces) and negative isopotential
surfaces (meshes) for methionine (a) and for analogues 2, 3, 5, and
11 (b - e, respectively). The electron density maps indicate electron-rich
(red) and electron-poor (blue) regions of each molecule. For simplicity,
the neutral amino acid form is shown; this avoids representations of the
highly extended isopotential surfaces of the carboxylate anion of the
zwitterion and facilitates comparison of side-chain electronic structure.
60 2.5 References
(1) Budisa, N., Steipe, B., Demange, P., Eckerskorn, C, Kellermann, J. and Huber, R. Eur. J. Biochem. 1995, 230, 788-796.
van Hest, (2) J.C.M. and Tirrell, D.A. FEES Lett. 1998, 428, 68-70.
(3) van Hest, J.C.M., Kiick, K.L. and Tirrell, D.A. J. Am. Chem. Soc. 2000, 722, 1282-1288.
(4) Coulter, A.W. and Talalay, P. J. Biol. Chem. 1968, 243, 3238-3247.
Blackwell, H.E. (5) and Grubbs, R.H. Angew. Chem. Int. Ed. Engl. 1998, 37, 3281- 3284.
(6) Sambrook, J., Fritsch, E.F. and Maniatis, T. Molecular Cloning: A Laboratory
Manual; Cold Spring Harbor Laboratory Press: Cold Spring Harbor, NY, 1989.
(7) The Qiagen Expressionist, Qiagen: Valencia, CA, 1998.
(8) Ghosh, G., Brunie, S. and Schulman, L.H. J. Biol. Chem. 1991, 266, 17136- 17141.
(9) Mellot, P., Mechulam, Y., LeCorre, D., Blanquet, S. and Fayat, G. J. Mol. Biol. 1989, 208, 429-443.
(10) Blanquet, S., Fayat, G. and Waller, J.-P. Eur. J. Biochem. 1974, 44, 343-351.
(11) Ghosh, G., Pelka, H. and Schulman, L.D. Biochemistry 1990, 29, 2220-2225.
(12) Kim, H.Y., Ghosh, G., Schulman, L.H., Brunie, S. and Jakubowski, H. Proc. Natl.
Acad. Sci. USA 1993, 90, 1 1553-1 1557.
(13) Bain, J.D., Glabe, C.G., Dix, T.A. and Chamberlin, A.R. J. Am. Chem. Soc. 1989, 7/7,8013-8014.
61 Noren, C.J., (14) Anthony-Cahill, S.J., Griffith, M.C. and Schultz, P.G. Science 1989, 244, 182-188.
(15) Cornish, V.W., Mendel, D. and Schultz, P. Angew. Chem. Int. Ed. Engl. 1995, 34, 621-633.
( 1 6) Nowak, M.W. et al. Science 1995, 268, 439-442.
(17) Saks, M.E., Sampson, J.R., Nowak, M.W., Kearney, P.C., Du, F., Abelson, J.N.,
Lester, H.A. and Dougherty, D.A. J. Biol. Chem. 1996, 277, 23169-23175.
(18) Gallivan, J.P., Lester, H.A. and Dougherty, D.A. Chem. Biol. 1997, 4, 739-349.
(19) Turcatti, G., Nemeth, K., Edgerton, M.D., Knowles, J., Vogel, H. and Chollet, A.
Receptors and Channels 1997, 5, 201-207.
(20) Hohsaka, T., Ashizuka, Y., Sasaki, H., Murakami, H. and Sisido, M. J. Am. Chem. Soc. 1999, 727, 12194-12195.
(21) Skinner, C.G., Edelson, J. and Shive, W. J. Am. Chem. Soc. 1961, 83, 2281-2286.
(22) Apostol, I., Levine, J., Lippincott, J., Leach, J., Hess, E., Glascock, C.B.,
Weickert, M.J. and Blackmore, R.J. J. Biol. Chem. 1997, 272, 28980-28988.
(23) Cowie, D.B., Cohen, G.N., Bolton, E.T. and Robichon-Szulmajster, H.D.
Biochim. Biophys. Acta 1959, 34, 39-46.
(24) Hendrickson, W.A., Horton, J.R. and LeMaster, D.M. EMBO J. 1990, 9, 1665- 1672.
(25) Duewel, H., Daub, E., Robinson, V. and Honek, J.F. Biochemistry 1997, 36, 3404-3416.
(26) Meinnel, T., Mechulam, Y. and Blanquet, S. tRNA: Structure, Biosynthesis, and Function; ASM Press: Washington, D. C, 1995.
and Company: New (27) Fersht, A. Enzyme Structure and Mechanism; W. H. Freeman York, 1999.
62 (28) Gabius, H.J., von der Haar, F. and Cramer, F. Biochemistry 1983, 22, 2331-2339,
(29) Kothakota, S., Mason, T.L., Tirrell, D.A. and Fournier, M.J. J. Am Chem Soc 1995, 7/7,536-537.
(30) Ibba, M., Kast, P. and Hennecke, H. Biochemistry 1994, 33,1 \ 07 -j \ \ 2.
(3 1 Serre, L., Verdon, G., Choinowski, ) T., Hervouet, N., Risler, J.-L. and Zelwer C J. Mol. Biol. 2001, 306, 863-876.
(32) Steiner, T., Starikov, E.B., Amado, A.M. and Teixeira-Dias, J.J.C. J. Chem. Soc
Perk. Tran.s. - 2 1995, 7, 1 32 1 1 326.
63 CHAPTER 3
ALTERATION OF THE METHIONYL-tRNA SYNTHETASE ACTIVITY OF A BACTERIAL EXPRESSION HOST
3.1 Introduction and Ohj ecfivps:
The incorporation of methionine analogues in vivo correlates well with the in vitro activation of the analogues by MetRS, as analogues that support high levels of PP, exchange in vitro also support protein synthesis in vivo. Two analogues, 5 and 11, which
support low levels of PP^ exchange in vitro, are unable to support protein biosynthesis in
a conventional bacterial host. Several other methionine analogues (4, 6-8, 10, 12-13)
were not utilized by the bacterial host during protein biosynthesis (and did not support measurable levels of PPj exchange in the in vitro assays described in Chapter 2). The
yields of protein obtained from cultures supplemented with a given analogue correlated
well with the catalytic efficiency of the analogue activation by MetRS. Based on these
results, we speculated that manipulation of the MetRS activity of the expression host
might enable additional methionine analogues to serve as substrates during bacterial
protein synthesis.
Two methods were used to investigate this hypothesis. First, overexpression of
the wild-type MetRS in a bacterial host was investigated via the introduction of plasmid
64 copies of the gene encoding MetRS. This strategy has not been employed previously lor incoiporating non-natural amino acids into proteins in vivo, but reports of the in vivo misacylation of tRNA substrates by overexpressed aaRS supported the viability of the approach (1-4). Overexpression of a mutant MetRS may also serve as an additional strategy for expanding the substrate range of the translation^ apparatus. Substitution of
Trp305 by Phe (W305F) produces a mutant MetRS with the same enzymatic rates of activation or aminoacylation as the wild-type MetRS (5,6). These results have
significance for the in vivo incorporation of methionine analogues, as the substitution of the bulky Trp with the smaller Phe in the active site of MetRS (7) may result in relaxed substrate specificity and permit incorporation of analogues not utilized by the wild-type
MetRS. Previous investigations demonstrating the utility of this approach for expanding
the substrate range of PheRS to include /?-Cl-Phe and /7-Br-Phe (8-11) suggest that a
similar strategy could be successful for MetRS.
The goals of these investigations were as follows. First, production of bacterial hosts capable of overexpressing wild-type or mutant MetRS was accomplished by insertion of the gene encoding the MetRS or W305F onto a multiple-copy expression
plasmid followed by transformation of the plasmid into an appropriate bacterial host.
The ability of analogues 2 - 13 to support protein biosynthesis in the modified bacterial
hosts was tested, and the effect of overexpression of MetRS on the yields of analogue-
containing protein produced by the bacterial host was studied. Differences in the abilities
65 of the wild-type and mutant MetRS to activate the methionine analogues were assessed by monitoring the in vitro ATP-PPj exchange activity of both enzymes.
3.2 Experimental Section
3.2.1 Genetic Manipulation of a Bacterial Expression Host
Recombinant DNA manipulations were carried out using standard, commerically
available reagents (from Sigma, Aldrich, Qiagen, New England Biolabs, Inc.) and standard protocols (12) unless otherwise noted. The expression plasmids pQE9 and pQE15 were obtained from Qiagen. The RGS-His antibody and anti-mouse IgG
horseradish peroxidase conjugate used for Western blotting procedures were obtained
from Qiagen and Amersham Life Sciences, respectively. The ECL Western blotting
detection reagents for developing Western blots were obtained from Amersham
Pharmacia Biotech.
Insertion of the gene encoding MetRS into the expression plasmid pQE15 was
accomplished using the genetic strategy outlined in Figure 3.1. Simply, a linker was
designed to permit the cohesive ends of the MetRS gene to be changed to Nhe I. The
resulting fragment was then ligated into the unique Nhe I site of the expression plasmid pQE9 or pQE15. Specific experimental protocols are delineated below.
66 3-2.1.1 Construction of Plasmid pUC19-Nhelink
Single-stranded oligonucleotides encoding the Nhe linker sequence shown in
Figure 3.2 were supplied by Genosys (The Woodlands, TX) and dissolved in water to yield a final concentration of 1 [ig/mL. The oligonucleotides encode a series of
restriction enzyme sites that permit the insertion of a DNA fragment at Kpn I and Sac I sites, with subsequent liberation of the gene fragment by digestion with Nhe I. The oligonucleotides were annealed and phosphorylated as follows. To a reaction mixture of
|iL of each 10 oligonucleotide was added 5 |a,L of 2M NaCl, 2 mL IM MgClj and 73 |xL water to yield a final volume of 100 |a.L. The solution was vortexed, incubated in boiling
water for 5 minutes, and then placed in a styrofoam cooler and allowed to cool to room temperature slowly overnight. The DNA was precipitated by adding 10 |aL of 3M
sodium acetate (pH 4.8), vortexing, adding 300 \iL of cold (-20°C) ethanol, and then
incubating the DNA at -80°C for 20 minutes. After centrifugation, the supernatant was removed and the DNA dried under vacuum for 5 minutes. The DNA was then
phosphorylated by redissolving the DNA pellet in 44.5 [iL water, adding 5 [iL of T4
polynucleotide kinase buffer (to bring the final reaction conditions to 70 mM Tris-HCl
(pH 7.6), 10 mM MgClj, 5 mM dithiothreitol), and adding 5 units of T4 polynucleotide
kinase. The reaction was incubated at 37°C for 1 hour and immediately mixed with 5.6
67 ^IL of lOx agarose gel loading buffer and isolated by agarose gel electrophoresis with a
2% 2w/v agarose gel (visualization with ethidium bromide).
The annealed, phosphorylated, double-stranded DNA was purified by excising the appropriate band (103 bp) from the agarose gel and extracting the DNA from the agarose via the Qiagen QIAquick spin gel extraction protocol (13). The cohesive ends of the
linker were designed as Aat II and Sph I, so that the linker could be added easily to the cloning vector pUC19, retaining the necessary unique restriction sites. The plasmid pUC19 (~1 )xg) was linearized by digestion for 4 hours at 3TC with 10 units each of Aat
II and I Sph enzymes in NEB buffer 4 (50 mM potassium acetate, 20 mM Tris-acetate, 10
mM magnesium acetate, 1 mM dithiothreitol, pH 7.9). All digestions throughout this
chapter were performed with a similar amount of DNA and at 37°C for four hours unless
otherwise noted. The linearized plasmid was not dephosphorylated, but was isolated by
agarose gel electrophoresis as above. The insert and linearized plasmid were ligated in a
molar ratio of 5: 1 (50: 10 ng), with 5 units of T4 DNA ligase in a total volume of 20 mL
T4 DNA ligase buffer (50 mM Tris-HCl (pH 7.8), 10 mM MgClj, 10 mM dithiothreitol, 1
mM ATP, and 25 }Xg/L BSA). The ligation was conducted at room temperature for 4
hours and then the ligation mixture was transformed into the E. coli cloning strain
DH5aF' according to the following procedure for making and transforming competent
68 cells. All ligation procedures were done identically as described here unless otherwise noted.
E. coli DH5aF cells were grown to an optical density (OD550) of approximately
0.5, placed on ice for 10 minutes, and collected by centrifugation at 2000 x g for 15 minutes at 4°C. The cells were resuspended in 2 mL of TFB 1 buffer ( 1 0 mM morpholino ethanesulfonic acid (MES) (pH 6.2), 100 mM RbCl2, 10 mM CaCl2.2H20, 50 mM
MnCl2-4H20) by gentle tilting of the tube. The volume was then adjusted to 16 mL with additional TFBl buffer. The suspension was left on ice in the cold room for 15 minutes and then the cells were pelleted by centrifugation at 2000 x g for 15 minutes at 4°C. The cells were resuspended in 2 mL TFB2 buffer (10 mM 3-(N-morpholino)propane sulfonic acid (MOPS), 75 mM CaCl2'2H20, 10 mM RbClj and 15% glycerol) and chilled on ice for 15 minutes before being dispensed into 200 |j.L aliquots which were stored at -80°C.
An aliquot of DH5aF' competent cells (50 }xL) was thawed on ice for 20 minutes.
To the cells, 5 \xL ligation reaction was added, and the mixture was incubated on ice for
45 minutes. The sample was heat-shocked at 42°C for approximately 90 seconds
followed by incubation on ice for 5 minutes. The sample was then diluted with 500 |liL
2xYT media and incubated on a rotary incubator for 1 hour at 37°C. Cells (200 \\L) were
spread on agar plates containing 200 |ig/mL ampicillin and were grown for 16 hours at
i>rc.
69 Eighteen of the approximately 400 single colonies from these plates were used to inoculate eighteen overnight cultures (5 mL of 2x YT). An aliquot ( 1 .5 mL) of the overnight culture was centrifuged to harvest the cells, and the DNA was isolated from the cells by the spin miniprep procedure described by Qiagen (14). Restriction analysis of the DNA was conducted by digesting the DNA at 37°C for 4 hours with Kpn I and Sac I
(10 units of each enzyme in NEB 1 buffer with BSA (10 mM bis tris propane-HCl, 10
mM MgCl2, 1 mM dithiothreitol, 100 M-g/mL BSA, pH 7.0)) to confirm the presence of the Nhelinker insert. Similar digestion with Nhe I for 4 hours at 37°C also confirmed these results (10 units Nhe I in NEB buffer 2 with BSA (50 mM NaCl, 10 mM Tris-HCl,
10 MgC12, mM 1 mM dithiothreitol, 100 ^g/mL BSA, pH 7.9)). Two samples were submitted for oligonucleotide sequencing analysis (Oligonucleotide Synthesis and
Sequencing Facility, California Institute of Technology, Pasadena, CA), which confirmed
the sequence (in one of the two samples) of the insert (103 bp) in the plasmid pUC19-
Nhelink.
3.2.1.2 Construction of Plasmid pUC19-W305F
The gene encoding MetRS W305F (2449 bp) was removed from plasmid
pBSM547W305F (a kind gift from Professor Yves Mechulam of the Ecole
Polytechnique, Palaiseau Cedex, France) by restriction enzyme digestion with Kpn I and
70 Sac I, employing 10 units of each enzyme in NEB 1 buffer with BSA (10 mM bis tris
propane-HCl, 10 mM MgCl,, 1 mM dithiothreitol, 100 |ig/mL BSA, pH 7.0). The
plasmid pUC19-Nhelink was linearized by similar digestion with these two enzymes and
dephosphorylated by adding calf intestinal alkaline phosphatase (CIP, 1 unit) directly to
the digestion mixture and incubating for an additional 30 minutes at 37°C. Both the
W305F gene and linearized pUC19-Nhelink plasmid were isolated and purified by
agarose gel electrophoresis gel (1.0% with visualization by ethidium bromide) as
described above. The purification of the W305F gene required separation of two
similarly sized fragments (2449 and 2862 bp), so the fragment was purified by gel
electrophoresis twice prior to its use in the ligation mixture. The W305F and linearized
pUC19-Nhelink were ligated, and the ligation mixture was transformed into DH5aF
using the procedures described above. Restriction digestion analysis (Nhe I) of the DNA
isolated from 9 of the resulting single transformants indicated the presence of insert
(2498 bp) in 7 out of 9 colonies tested. One sample was retained as pUC19-W305F and
the isolated DNA was stored at -20°C.
3.2.1.3 Construction of Expression Plasmids pOE9-W305F. pOE15-W305F. and pOE15-MRS
The gene encoding W305F was removed from the plasmid pUC19-W305F by
digestion with Nhe I (as described above) and isolated and purified by agarose gel
71 electrophoresis. The expression plasmids pQE9 and pQE15 (Qiagcn, Valencia, CA) were linearized, also by digestion with Nhe I, and similarly purified. The fragment W305F was ligated into each of these expression plasmids separately and the resulting ligation mixtures transformed into cloning strain DH5aF according to the procedures described above. After amplification and isolation of the plasmid DNA by miniprep procedures
(14), digestion with Nhe I was conducted to confirm the presence of the W305F insert
(2498 bp) in both of the resulting plasmids pQE9-W305F (5 of 9 colonies) and pQE15-
W305F (5 out of 10 colonies). Additional restriction analysis with enzymes Aat II and
Nhe\{\Q units each, NEB buffer 4) was conducted to confirm the presence of the gene encoding DHFR (920 bp) in the pQE15-W305F plasmid. The orientation of the W305F gene in pQE9-W305F and pQE15-W305F was deduced by restriction analysis with Hind
III (10 units, NEB buffer 2).
Transformation of pQE15-W305F into DHFaF' resulted, in one case, in genetic
recombination of the mutant gene with the chromosomal copy of the wild-type MetRS
gene, yielding plasmid pQE15-MRS. Along with the repressor plasmid pREP4, the
plasmids pQE9-W305F, pQE15-W305F, and pQE15-MRS were each transformed into
the expression strains CAG 18491 or B834(DE3) (Stratagene) (made competent by the
same procedures described for DH5aF') and plated onto agar plates containing 200
^ig/mL ampicillin and 35 |Xg/mL kanamycin. Isolation of single colonies yielded the
72 modified expression hosts CAG18491/pQE9-W305F/ pREP4, CAG 18491 /pQE 15-
W305F/pREP4, and CAG18491/pQE15-MRS/pREP4, along with hosts
B834(DE3)/pQE9-W305F/ pREP4, B834(DE3)/pQE15-W305F/pREP4, and
B834(DE3)/pQE15-MRS/pREP4. Cell stocks from overnight cultures were prepared and
stored at -80°C, and DNA stocks were stored at -20°C. Plasmid DNA from all cultures
used for protein expression experiments was sequenced to confirm that it encoded the
appropriate MetRS.
3.2.2 Protein Expression
3.2.2.1 Confirmation of MetRS and DHFR Expression
Single colonies of the bacterial expression hosts CAG18491/pQE15/pREP4 and either CAG18491/pQE15-W305F/pREP4 or CAG18491/pQE15-MRS/pREP4 were used to inoculate 5 mL 2xYT cultures containing 200 |J,g/mL ampicillin and 35 |ig/mL kanamycin. The cultures were grown to an ODgoo of 0.8 before inducing expression of
the target protein DHFR with 0.4 mM IPTG, and the cultures were permitted to grow for
an additional 4.5 hours. Samples (1 mL) of each of the cultures were sedimented and
resuspended in distilled H2O as previously described. The cell lysates from these experiments were analyzed by SDS-PAGE, and the proteins visualized by staining with
73 Coomassie blue. Similar experiments were conducted for bacterial host
CAG18491/pQE9-W305F/pREP4, except that overnight cultures were grown without
induction of protein expression.
3-2.2.2 Translational Activity of Methionine Analogues in Vivo
The translational activity of the methionine analogues 2 - 13 was tested in both
conventional bacterial hosts equipped with the plasmid pQE15 and modified bacterial
hosts carrying expression plasmids 1 pQE 5-MRS or pQE 1 5-W305F. The hosts were
produced by transformation with the appropriate expression plasmid and the repressor
plasmid pREP4 as described above. The small-scale expression experiments in minimal
media were conducted as described in Section 2.2.2, except that the supplementation
level of the analogues was either 20 mg/L, 60 mg/L, or 500 mg/L of the L-isomer. In
experiments to test analogue incorporation, analogue concentrations of 60 mg/L were
used. In experiments designed to test the yield of protein obtained from a conventional
versus a modified bacterial host, supplementation of 20 mg/L was employed. For large-
scale experiments designed to determine protein yields, and for producing proteins for
amino acid analysis and N-terminal sequencing, supplementation of 20 mg/L was
employed for 5, while 500 mg/L was employed for analogues 4, 7, 8, 11, and 12. Protein expression was monitored by SDS-PAGE, visualized either by Coomassie blue staining
74 or by Western blotting. The accumulation of the target protein DHFR was taken as
evidence for incorporation of the non-natural amino acid. Incorporation was confirmed
by purifying the protein by metal chelate affinity chromatography (15,16) as previously
described, and analyzing the purified protein by amino acid analysis and N-terminal
sequencing, and when possible, NMR spectroscopy.
3.2.2.3 Determination of Protein Yields
Western A blotting procedure designed to assess the relative yields of protein obtained from 5-mL cultures was conducted as previously described (17). Proteins were transferred from an acrylamide gel to a nitrocellulose membrane, and Western blots were
developed by treatment with a primary RGS-His antibody (20 |a.g/mL in 5% w/v dried milk blocking solution), followed by washing and treatment with a secondary anti-mouse
IgG conjugated to horseradish peroxidase (50 |Xg/mL in 5% w/v blocking solution).
Chemiluminescence detection of the labeled proteins was afforded by use of ECL
detection reagents (Amersham Pharmacia Biotech) and documented on film. Films were
exposed for varying times to ensure that band intensity was not saturated. Levels of
protein synthesis were estimated by the intensity of the band on the gel, as determined by
a Pharmacia Ultrascan XL laser densitometer and analysis by Pharmacia GelScan XL
software. Large scale yields were determined by purifying DHFR from 1 L cultures via
75 metal chelate affinity chiomalography under denaturing conditions, as previously described in Chapter 2(16).
^•^•3 Determination of Intracellular ConcentmUnn. nf MethioninP A Prll^ei-'i
The approximate intracellular concentrations of methionine and methionine analogues 8 or 9 during protein expression were determined by amino acid analysis of cell ly sates from 50-mL samples of CAG 18491 /pQE15-MRS/pRnP4 grown on medium
containing 1 AA and 9 supplemented with methionine, 8, or 9. Cultures of 250 mL were grown to an OD(^^^ of 0.7 on medium supplemented with methionine. After a medium
shift, the cells were resuspended in a medium containing 20 mg/L or 500 mg/L of methionine, or 8, 9. Protein expression was induced by the addition of IPTG (0.4 mM),
and 50 mL samples were harvested in duplicate the when cultures reached an OD of 1 .0
(approximately 2 hours). cell pellets The were washed once with 1 x M9 salts. Vice
amino acids were extracted as previously described (18), Harvested cells were
resuspended in 2.5 mL distilled H2O, and the pJl was adjusted to 10 with a few drops of
5N NII4OII. The samples were placed in a boiling water bath for 10 minutes and then
cooled to room temperature. The pH was adjusted to 2 with a few drops of 5N I ICl, and
the precipitate removed by cenlrifugation. The supernatant was submitted lor
quantitative amino acid analysis under conditions in which hydrolysis was not employed
76 Methionine, 8, and 9 were chosen for these preUminary analyses owing to the fact that these analogues can be detected directly by standard amino acid analysis.
3-2.4 Protein Characterization
3-2.4.1 Amino Acid Analysis and N-terminal Sequencin g
The composition of each protein was determined by amino acid analysis performed by the Analytical Chemistry and Peplide/DNA Synthesis Facility at Cornell
University, Ithaca, NY. N-terminal .sequencing was performed by the Protein/Peptidc
Microanalytical Laboratory at the California Institute of Technology, Pasadena, CA.
3.2.4.2 NMR Spectroscopy
Proton NMR spectra were recorded using a Varian Inova NMR spectrometer with
proton acquisition at 599.69 MHz. Samples were dis.solved in under slightly acidic
conditions and spectra were recorded at 25°C overnight. A simple presaturation pulse
was used for water suppression. Unambiguous assignment of certain resonances for analogue 5 was confirmed using ID TOCSY experiments.
77 ID TOCSY spectra were recorded on a Varian Inova NMR spectrometer with proton acquisition at 599.69 MHz. A ID TOCSY pulse sequence with gradient-based coherence path rejection and DIPSI-2 mixing sequence (10 kHz bandwidth) was used to establish chemical shift correlations This (19). pulse sequence is designed to determine the identity of protons in a single spin system (in this case, amino acid) using selective irradiation of a single resonance followed by observation of protons to which
magnetization is transferred. Selective irradiation of the resonance at 5.35 ppm
(unambiguously assigned to olefin protons of 5) was achieved with a 23.67 ms c-SNOB pulse with a 60 Hz bandwidth (20). The .selectivity of the pulse is demonstrated in a separate, simple ID experiment in which the selective pulse is applied alone; no other
resonances are observed in the spectrum under these conditions.
3.2.5 Activation of Methionine Analogues bv MetRS in Vitro
The activation of methionine analogues 2 - 13 was measured as described in
Chapter 2, but in this case, included measurement of the activity for the analogues with
the mutant MetRS W305F as well. The kinetic parameters for the analogues determined
as described in Section 2.2.4 were determined with W305F using the same concentrations
as previously described. Assay conditions for analogues that were not analyzed
78 previously are given below; measurements were conducted with both the wild-type and mutant MetRS.
For analogue 8, enzyme concentrations of 60 nM and analogue concentrations of
2.5 mM to 35 mM were used in the ATP-PP, exchange assay. Analogues 4 and 12 are such poor substrates for the enzyme that only k,,/K, values could be determined accurately. For 12, enzyme concentrations of 100-150 nM and analogue concentrations of 5 mM to 50 mM were used. The lower solubility of 4 precluded the use of these high analogue concentrations, so the k,yK,„ value was estimated by calculating the slope of the plot of velocity vs. substrate concentration at low substrate concentration. Enzyme concentrations of 150 nM and analogue concentrations of 0.6 mM to 6.6 mM were used.
Measurable velocities could not be obtained for analogue 7, owing to the very low solubility of this analogue at the pH of the enzyme assay reaction buffer.
Additionally, ATP-PPj exchange assays were conducted on the whole cell lysates from both conventional bacterial expression hosts (CAG18491/pQE15/pREP4 and
B834(DE3)/pQE15/pREP4) and modified bacterial hosts (CAG 18491 /pQE 15-
MRS/pREP4 and B834(DE3)/pQE15-MRS/pREP4) to confirm that the addition of the
MetRS gene to the expression plasmid pQE15 results in an increased level of enzyme
activity in the bacterial host. These assays were conducted in a manner similar to that
used for analysis of the methionine analogues. M9AA medium was inoculated with the
appropriate bacterial host and the culture was grown at 37°C for 8 hours. An aliquot
79 (3.75 mL) of each culture was centrifuged at 12000 x to ^ pellet the cells. The cell pellet
was washed twice with 1 mL of 1 x M9 salts to remove residual amino acids from the cell
pellet. The cell pellets were then resuspended in a volume {OD,^, x 100 |iL) of ATP-PP,
exchange assay buffer sufficient to normalize the OD,oo of the samples to 37.5.
The resuspended cells were lysed by a single freeze-thaw cycle, vortexed, and 30
^iL of the cell lysate was used directly in assays with a total volume of 75 ^iL and a
saturating methionine concentration of 750 [iM. A control assay in which methilonine
was excluded was also analyzed for each cell lysate to yield a background level of ATP-
PPj exchange that was subtracted from the velocities observed for the cell lysates
supplemented with methionine. This procedure corrected for any ATP-PP. exchange that
might be catalyzed by the other 19 aaRS and cognate amino acids present in the cell
lysate. Using a saturating concentration of methionine permits determination of the
maximal velocity, which yields direct comparison of the enzyme concentrations in each
cell lysate, assuming a constant value of between the MetRS of the different cultures.
Aliquots (20 |lL) were quenched at various time points. Aliquots from the assay mixtures
of the conventional bacterial host cell lysate were quenched at 1.5, 4.5, and 9.0 minutes,
while those from the modified bacterial host cell lysate were quenched at 60, 90, and 180
seconds. The charcoal was washed twice with the wash buffer, and counted via liquid
scintillation methods. Plots of the moles PPj exchanged vs. time permitted the calculation
80 of the maximal velocity for each sample. Maximal velocities are reported as moles of PP,
exchanged per second.
3.2.6 Characterization of GFP Fluorescenr.p.
The effect of incorporation of methionine analogues on protein function was
investigated by monitoring the fluorescence of the green fluorescent protein (GFP)
(2 1 produced ,22) from bacterial cultures supplemented with various methionine analogues. The data were kindly provided by Pin Wang at the California Institute of
Technology, Pasadena, CA, and the procedures were as follows. The 714 bp fragment encoding GFP (Clonetech, Palo Alto, CA) was amplified via PCR methods to yield a
gene fragment with Bam UVHind III cohesive ends. This gene fragment was introduced into the multiple cloning site of the expression vector pQE31 (Qiagen, Valencia, CA) to yield the plasmid pQE31-GFP (23). This plasmid was transformed along with pREP4 into the methionine auxotroph CAG 18491. The expression plasmid encodes GFP under the control of a bacteriophage T5 promoter, and encodes GFP with an N-terminal
hexahistidine sequence for facile purification by metal chelate affinity chromatography.
The GFP contains 6 methionine residues, all potential sites for substitution with methionine analogues. Protein expression experiments were conducted as described for
the DHFR expression, except that 50 mL cultures of CAG18491/pQE31-GFP/pREP4
81 were grown on M9 medium supplemented with 500 mg/L of tiie methionine analogues 2,
3, or 9 (23).
The proteins were purified by metal chelate affinity chromatography under nati ve
conditions (24). The cell pellet was resuspended in 2 mL of sonication buffer (50 mM
NaH2P04, 300 NaCl, mM 10 mM imidazole, pH 8.0). After lysis with lysozyme and
sonication on ice, the cell debris was pelleted and the supernatc loaded onto a nickel spin column equilibrated with 0.6 mL of the sonication buffer. The protein was washed with
2 X 0.6 mL wash buffer (50 mM NaH^PO^, 300 mM NaCl, 20 mM imidazole, pH 8.0)
and eluted with 2 x 0.2 mL elution buffer (50 mM NaH2P04, 300 mM NaCl, 250 mM
imidazole, pH 8.0). The protein was stored at 4°C in the elution buffer. The undiluted
sample was excited at 396 nm and the intensity of the fluorescence was monitored at 506
nm using a Photon Technology International fluorescence spectrophotometer.
Fluorescence spectra were normalized for protein concentration, which was determined
by monitoring the absorption maximum of the sample at 280 nm (23).
82 3.3 Results and Discussion
3.3.1 Genetic Manipulation of a Bacterial Expression Host
The construction of a plasmid that permits overexpression of MetRS in a bacterial
expression host was achieved according to the genetic strategy outlined in Figure 3.1.
The gene encoding a mutant MetRS was removed from plasmid pBSM547W305F (5) by
treatment with restriction enzymes Sac I and Kpn I. The Sac UKpn I fragment (2449 bp)
was ligated into the cloning vector pUC19-Nhelink (which was constructed to permit the
cohesive ends of the W305F gene to be changed to Nhe I), to yield plasmid pUC19-
W305F. A schematic of the linker inserted into pUC19 to form pUC19-Nhelink is shown
in Figure 3.2, and its sequence is given in Appendix D. Restriction digestion of the
pUC19-W305F with Nhe I confirmed the presence of the W305F fragment (2498 bp)
(Figure 3.3). The W305F gene with Nhe I cohesive ends (Appendix E) was ligated into
the unique Nhe I site of either the plasmid pQE15 or pQE9 to yield the plasmids pQE15-
W305F and pQE9-W305F. Restriction analysis of each of the plasmids with Nhe I
confirmed the presence of the gene encoding W305F (2498 bp, Figures 3.3 and 3.4).
confirmed Restriction digestion of the pQE15-W305F plasmid with Aat II and Nhe I also
plasmid (Figure 3.4). The the presence of the gene encoding DHFR (920 bp) in the
was confirmed by restriction orientation of the MetRS genes in the expression plasmids
83 analysis with Hind III, as shown in Figures 3.3 and 3.4, and allowed determination of the
plasmid identity; there are no likely consequences on the expression of the MctRS based
on orientation. Both orientations of W305F were observed in the plasmid pQE9-W305F
(Figure 3.3), while only the orientation shown in Figure 3.2 (corresponding to orientation
2 in the PQE9-W305F) was observed for pQE15-W305F (Figure 3.4). Transformation of
the PQE15-W305F plasmid into a recA positive cell strain resulted in genetic
recombination of the mutant MetRS gene with the chromosomal copy of the wild-type
MetRS gene, yielding plasmid pQE15-MRS. The plasmid DNA from all cell cultures
used for protein expression experiments was sequenced to confirm its identity.
These plasmids were used, along with the repressor plasmid pREP4, to transform
the E. coli methionine auxotrophic strains CAG 18491 or B834(DE3). Ovcrexpression of
MetRS and DHFR was confirmed by SDS-PAGE analysis of cell lysates of 8-hour
cultures that had been induced to express DHFR for 4 hours. The presence of the band
indicating expression of MetRS is clearly visible along with the band corresponding to
DHFR in cultures of CAG18491/pQEI5-MRS/pREP4, as shown in Figure 3.5. Similar
results confirming the expression of W3()5F were observed for overnight cultures of
CAG18491/pQE15-W305F/pREP4 and CAG18491/pQE9-W3()5F/pREP4 (no target
protein expression could be monitored for the hosts containing pQE9, as this plasmid
contains only a multiple cloning site).
84 The data illustrated in Figure 3.6, obtained from ATP-PPj exchange assays of cell lysates of CAG18491/pQE15-MRS/pREP4 and B834(DE3)/pQE15-MRS/pREP4,
confirm the results in Figure 3.5 and indicate that the modified bacterial host
CAG18491/pQE15-MRS/pREP4 exhibits a V^,, value for methionine activation
approximately 50-fold higher than that observed for the control host
CAG18491/pQE15/pREP4 (Figure 3.6a). A 30-fold increase was observed for the
modified host B834(DE3)/pQE15-MRS/pREP4 in comparison to the control host
B834(DE3)/pQE15/pREP4 (Figure 3.6b). These values are consistent with the reported
copy number of the pQE15 plasmid (approximately 50) (16).
3.3.2 Translational Activity of Methionine Analogues in a Modified Bacterial Expression Host
Bacterial hosts capable of overexpressing MetRS or W305F were produced by
transforming the E. coli methionine auxotrophs CAG 18491 or B834(DE3) with the
repressor plasmid pREP4 and an expression plasmid that encodes truncated forms of
either the wild-type MetRS or a mutant MetRS W305F, as described in the previous
analogues. section. These hosts were used to test the translational activity of methionine
were identical; results The results obtained for both CAG 18491 and B834(DE3) cultures
for CAG 18491 are shown and discussed below.
85 Methionine analogues 2 - 13 were tested for translational activity in both conventional and modified bacterial expression hosts. Cultures of
CAG18491/pQE15/pREP4 or CAG18491/pQE15-MRS/pREP4 in M9AA media were grown to an optical density of 0.90, and the cells were sedimented by centrifugation.
Cells were washed three times with M9 salts and resuspended to an optical density of
0.90 in M9 test media containing 19 amino acids plus 1) neither methionine nor analogue
(negative control); 2) methionine (60 mg/L, positive control); or 3) an analogue of interest (60 mg/L). Expression of DHFR was induced by addition of 0.4 mM isopropyl
P-D-thiogalactopyranoside (IPTG), and protein synthesis was monitored after 4.5 hours.
Expression of DHFR was monitored by SDS-polyacrylamide gel electrophoresis (SDS-
PAGE); accumulation of target protein was taken as evidence for translational activity of the methionine analogue. The results from these experiments are summarized in Figure
3.7.
The target protein was not observed in the negative control culture of
CAG18491/pQE15/pREP4, or in cultures of CAG 18491 /pQE15/pREP4 or
CAG18491/pQE15-MRS/pREP4 supplemented with 4, 6, 7, 8, 10, 12, or 13. In contrast,
DHFR was detected in both bacterial host cultures supplemented with methionine (1), 2,
3, or 9, as indicated by the appearance of a protein band at the position expected for
DHFR in SDS-PAGE.
86 For the negative control cultures and for cultures supplemented with 5 and 11,
however, the behavior of the bacterial hosts differed, as shown in Figure 3.7. The target
protein DHFR was not detected in the CAG18491/pQE15/pREP4 cultures supplemented
with 5 or 11, while strong induction of DHFR was observed for CAG 18491 /pQE 15-
MRS/pREP4 under the same conditions. Even the unsupplemented control culture of
CAG18491/pQE15-MRS/pREP4 shows evidence of DHFR synthesis, suggesting that
introduction of pQE15-MRS does indeed increase the rate of activation of methionine in
the modified host. Results identical to those obtained for CAG 18491 /pQE 15-
MRS/pREP4 were observed for bacterial cultures that were equipped with the mutant
MetRS W305F (CAG18491/pQE15-W305F/pREP4).
The yields of DHFR obtained from 1-L cultures of CAG 18491 /pQE 15-
MRS/pREP4 supplemented with 20 mg/L of 5 were approximately 8 mg/L. Consistent
with the fact that 11 is a poorer substrate for MetRS than 5, 60 mg/L supplementation by
11 was required to observe measurable accumulation of target protein. Because
supplementation by 11 at a concentration of 500 mg/L improved the yield of protein in 5-
mL cultures (Figure 3.8), the yield for 1-L cultures was determined at this level of supplementation and was found to be 35 mg/L. Purified DHFR containing 5 or 11 was analyzed by amino acid analysis, N-terminal sequencing, and, when possible, NMR
spectroscopy (see Section 3.3.6).
87 ^•^•3 Protein Yields from MoHifieH Bacterid] Fvpression Ho.t.
Overexpression of MetRS in a bacterial host has expanded the number of
methionine analogues that can be incorporated into proteins in vivo. Coupled with the
results from Chapter 2 (which indicate that the activation of methionine analogues by
MetRS controls the yields of target protein obtained from a conventional bacterial host),
these results suggested that protein yields obtained from bacterial cultures supplemented
with methionine analogues might be improved by increasing the MetRS activity of the
bacterial host. To test this hypothesis, we compared the yields of protein prepared in the
conventional bacterial expression host, CAG18491/pQE15/pREP4, to those obtained
from a modified host, CAG18491/pQE15-MRS/pREP4.
Protein synthesis was monitored for 5-ml cultures of these hosts supplemented
with 20 mg/L of methionine or analogues 2, 3, or 9; Western blot analyses of protein
synthesis are shown in Figure 3.9. Although very low levels of protein synthesis were
observed for negative control cultures of CAG18491/pQE15-MRS/pREP4, amino acid
analyses, N-terminal sequencing, and NMR analyses of proteins produced in cultures of
this host supplemented with 5 (a poorer substrate for MetRS than 2, 3, or 9) still show 90-
96% replacement of methionine by 5 (25). Thus, the level of protein synthesis shown in
Figure 3.9 results from the incorporation of the analogue and is not likely due to
incorporation of residual methionine; these results were confirmed by amino acid
88 analysis. For cultures supplemented with methionine or 3, the modified host,
CAG18491/pQE15-MRS/pREP4, does not exhibit higher levels of protein synthesis than
the conventional host CAG18491/pQE15/pREP4. Analysis by laser densitometry
confirms these results and reveals approximately equal accumulation of target protein for
both strains; identical results have been obtained for yields of purified DHFR obtained
from large-scale expressions. Activation of the analogue by MetRS therefore does not
appear to limit protein synthesis in cultures supplemented with 3.
For cultures supplemented with 2 or 9, however, the modified bacterial host exhibits significantly increased levels of protein synthesis in comparison with the conventional host (Figure 3.9). Laser densitometry analysis indicates that the level of
protein synthesis in the modified host is increased approximately 1 .5-fold over that in the conventional host for cultures supplemented with 2, and approximately 1 .4-fold for cultures supplemented with 9. Activation of these analogues by MetRS appears to limit
protein synthesis in the conventional host, such that increasing the MetRS activity of the
host is sufficient to restore high levels of protein synthesis. Indeed, as shown in Figure
3.10, the yield of DHFR obtained from large-scale cultures of CAG 18491 /pQE 15-
MRS/pREP4 supplemented with 2 or 9 is increased to approximately 35 mg/L (as
compared to yields of 10 and 20 mg/L, respectively, from cultures of
CAG18491/pQE15/pREP4, (Table 2.1)). The results indicate that simple overexpression
of MetRS can improve protein yields for cultures supplemented with methionine
89 analogues that are poor substrates for MetRS, and may provide an attractive general
method for efficient production of chemically novel protein materials in VIVO.
3-3-4 Incorporation of Additional Methionine An.lnpnes into Protein. i„ u;.,^
Overexpression of MetRS has been indicated not only to expand the range of
methionine analogues that can be incorporated into proteins in vivo, but also to improve
the yields of protein obtained from cultures of conventional bacterial hosts grown in
media supplemented with poor substrates of MetRS. The improvement in incorporation
and protein yield most likely results from the increase in the rate of analogue activation
afforded by overexpression of the enzyme in the bacterial host in cultures supplemented
with mg/L of 20 the analogue. For 11, it was observed that raising the supplementation
to at least mg/L in 250 cultures of the modified bacterial host also improved protein yield
(Figure 3.8), which suggested that the concentration of methionine analogues in the cell
could be increased by raising their concentration in the medium. Therefore, the
importance of the level of amino acid supplementation on the incorporation of analogues
into proteins in vivo was probed.
The expression of the target protein DHFR was monitored by SDS-PAGE
analysis for both conventional (CAG18491/pQE15/pREP4) and modified bacterial
expression hosts (CAG18491/pQE15-MRS/pREP4) with supplementation at 500 mg/L of
90 8, 10, and 4, 6, 7, 12, 13 (3.8 to 4.3 mM). These analogues were chosen owing to their
inability to support protein biosynthesis in our previous experiments. The results from
the SDS-PAGE analysis of these cultures are shown in Figure 3.11. Negative control
cultures did not show any target protein synthesis, while both positive control cultures
supported protein synthesis in vivo. For the cultures of the conventional bacterial host
(Figure 3. 1 (a)), none of the 1 above analogues supported protein synthesis in vivo. For
the modified bacterial host CAG18491/pQE15-MRS/pREP4 (Figure 3.1 1 (b)), however,
the accumulation of target protein was observed in cultures supplemented with 500 mg/L
of analogues 4, 7, 8, or 12. Increasing the level of amino acid in the medium, coupled
with overexpression of the MetRS, is required for these methionine analogues to support
protein biosynthesis; their activation appears to be the rate-limiting step in protein
biosynthesis. Identical studies were conducted with bacterial cultures of
CAG18491/pQE15-W305F/pREP4 to ascertain if overexpression of the mutant MetRS would rescue the translational activity of any additional analogues. No additional analogues were utilized by the bacterial host equipped with extra copies of the mutant
MetRS.
Proteins produced by CAG18491/pQE15-MRS/pREP4 cultures (50 mL)
supplemented with 500 mg/L of the methionine analogues 4, 7, 8, and 12 were purified
and analyzed by amino acid analysis and N-terminal sequencing to confirm the
incorporation of these methionine analogues into the target protein DHFR (Section 3.3.6).
91 Characterization of the ATP-PP, exchange activity supported by the analogues was also
revisited (Section 3.3.7).
3-3-5 Intracellular Concentrati ons of Methionine Analn^npg
The intracellular concentrations of methionine analogues 8 and 9 were
preliminarily investigated to determine to what extent the intracellular concentration of
the analogues is increased upon the increase in supplementation. Cultures of
CAG18491/pQE15-MRS/pREP4 (250 mL), supplemented with 20 mg/L or 500 mg/L of
methionine, or 8, 9, were grown as previously described. Samples (50 mL) were
collected in duplicate after two hours of protein expression, and the concentration of free
amino acid in the cell lysate was determined by amino acid analysis as described in the
experimental section. The choice of 8 and 9 as analogues in these preliminary
investigations was based on the ease of their quantitative determination under standard
amino acid analysis conditions. As shown in Table 3.1, the concentrations of the
analogues in the medium increased by a factor of 5 to 10 upon increasing the
supplementation from 20 mg/L to 500 mg/L. The concentrations of the analogues in cultures supplemented with 4.3 mM of 8 or 3.8 mM of 9 were approximately 650 |iM and
200 |J,M, respectively. These values are higher than the reported intracellular
concentration of methionine in cellular hosts grown on medium lacking any amino acids
92 (- 10 ^IM) but are (18), consistent with the concentration of methionine observed in
cultures supplemented with 500 mg/L (3.4 mM) methionine (350 ^iM, Table 3.1). This
indicates that their transport is not inhibited relative to that of methionine and may
suggest that the analogues are transported into the cell by a similar transport mechanism,
3.3.6 Protein Characteri/atinn
3.3.6.1 Amino Acid Analvsis an d N-termin?^l Sequencing
Increasing MetRS activity by overexpression of MetRS in a bacterial host is necessary and sufficient to observe translational activity of 5 and 11 under convenient conditions in vivo. Further increasing the rate of analogue activation by concomitantly elevating the concentration of analogue in the medium rescues the translational activity of
4, 7, 8, and 12. Amino acid analysis of proteins produced from cultures supplemented
with 500 mg/L of these analogues was conducted to confirm replacement of methionine.
As demonstrated in Table 3.2, amino acid analysis shows a decrease in methionine
content from the expected value of 3.8 mol%. It is impossible to detect 4, 5, 7, 11, or 12
directly by amino acid analysis, owing to their instability under the analysis conditions.
If, however, depletion of methionine is assumed to result from replacement by the
analogue, the observed analysis corresponds to overall extents of incorporation for the
93 analogues ranging from 60% to 98%, as shown in Table 3.3. The amino acid profile
given in Table 3.2 also demonstrates that even at high levels of analogue
supplementation, the experimentally observed amino acid content is within experimental
error (±10% of the expected value) of the theoretical values for DHFR, with the
exception of the decrement in methionine content.
Retention of the N-tcrminal (initiator) methionine in DHFR is expected on the
basis of the identity of the penultimate amino acid (26), so N-terminal sequencing
provided an additional means of assessing the extent of replacement of methionine by the
analogues. Because the analogues are not degraded under the analysis conditions, they
can be detected directly, and this analysis provides direct evidence for the presence of the
analogues in the target protein. Comparison of chromatograms of the N-terminal
residues of DHFR and those for DHFR containing methionine analogues (Figures 3. 12,
3.13, 3.14, 3.15, 3.16, and 3.17) demonstrates that the methionine that normally occupies
the initiator position of DHFR is replaced with the analogues. The signal corresponding
to methionine elutes at approximately 13.0 minutes, while those corresponding to the
analogue are listed in the appropriate figure caption. The large peaks that elute at
approximately 14.7 minutes correspond to piperidylphenylthiourea (pptu), a product of
the analysis resulting from the buffer, and the small peak at approximately 18.5 minutes
corresponds to diethylphthalate (diet), an internal standard. These results clearly indicate
the incorporation of a variety of methionine analogues at the initiator site of DHFR.
94 Integration of the peak areas corresponding to methionine and the analogues indicates up
to quantitative replacement of methionine by the analogues, with results given for each
analogue in Table 3.3. These results are consistent with the amino acid analyses and suggest that analogues 5, II, 8, 12, 4, and 7 are increasingly poor substrates for MetRS.
3.3.6.2 NMR Spectroscopy
A direct assessment of the extent of incorporation of 5 into DHFR is provided by
NMR spectroscopy. (Similar results could not be obtained for the other analogues due to the lack of resonances distinct from those of DHFR.) Comparisons of the 600 MHz proton NMR spectra (Figure 3.18) of DHFR, 5, and DHFR-5 indicate the appearance, in the DHFR-5 spectrum (Figure 3.18c), of the vinylene protons of 5 at 65.35 (6-CH) and
55.60-5.70 (y-CH). The resonances at 55.35 and 55.70 occur at the same chemical shift values as in free 5 and are clearly due to incorporation of 5 into DHFR. That the
resonance at 5.60 ppm arises from the y-CH vinylene proton of 5 is suggested by the fact
that the integrated intensity of the resonance at 55.35 equals the sum of the integrations of
the resonances at 55.60 and 55.70.
This assignment is confirmed by ID TOCSY (Total Correlation Spectroscopy)
experiments that indicate that the protons at both 55.60 and 55.70 are members of the
95 same spin system (and therefore the same amino acid) as those at 65.35 (The TOCSY
spectra are shown in Appendix C). More importantly, the ID TOCSY experiments also
show that the protons at 65.35 (and therefore those at 65.60 and 65.70) are associated with the spin system of the entire side chain of 5: ('H-NMR (D,0): 6 1 .6 (d, 3H, J = 6.4
Hz, CH3-CH=CH-CH,), 2.50 (t, 2H, 7 = 5.8, 13 Hz, CH3-CH=CH-CH,), 3.70 (t, IH, 7
5.8 Hz, H,N-CH-COOH), 5.30-5.43 (m, lH,y, = 14.7, 7 = 6.4 Hz, CH^-CH^CH-CH,),
5.70 (m, lH,y3= 14.7,7= 13Hz, CH3-CH=CH=CH2) ppm)
Selective irradiation of the resonance at 65.35 followed by immediate observat ion shows only the resonance at 65.35. Observation after a mixing time of 20 ms, however, shows the protons at 65.60 and 65.70, indicating that those protons are members of the
same spin system (and therefore the same amino acid residue) as those corresponding to the resonance at 65.35. The protons of the side chain P and e carbons are also observed
after 20 ms mixing time, at chemical shift values characteristic of the free amino acid (2.5 ppm (p-CH^) and 1.6 ppm (e-CH^)).
Integration of the spectrum suggests that 5 of the 8 methionine positions occupied
by 5 are represented by the resonance at 65.60; these protons must reside in a
magnetically-distinct environment from the protons at 65.70. These results
unequivocally demonstrate the translational activity of 5 in the host strain outfitted with
96 elevated MetRS activity. Integration of the NMR speetrum indieates 90 ± 6%
replacement of methionine.
3-3.7 Activation of Methionine. Analogues hy MptRc; ;^ y/^^^
Activation of methionine analogues 4, 7, 8, and 12 by MetRS was assayed by the
ATP-PP. exchange assay as previously described. Previous characterization of the exchange supported by these analogues (Figure 2.3) indicated that none of them supported measurable PP. exchange at 20 minutes at an analogue concentration of 5 mM.
In these experiments, analogues at concentrations of 15 mM were incubated with the enzyme for 3.5 hours prior to quenching of a 20 aliquot of the assay mixture. As indicated in Figure 3.19, each of the analogues supports levels of PP, exchange greater than the background under these experimental conditions, confirming the role of MclRS
in the incorporation of these analogues into proteins in vivo.
Quantitative assessment of the catalytic efficiency by which MetRS activates
these analogues was achieved by determining the kinetic parameters of these analogues
as previously described. The analogues are increasingly poor substrates for the enzyme
as compared to analogues 5 and 11 (Table 3.4), with k,,„,/K,„ values up to 340,000-fold
lower than those observed for methionine. As discussed for 5 and 1 1 (Chapter 2), the
loss of catalytic activity manifest both in increasing values of K„, and decreasing values
97 of k,.. indicates that these analogues are less efficient than methionine in both the binding
and catalytic events of the enzyme. These results also confirm that, under normal
cellular conditions (i.e., normal expression levels of MetRS and lower amino acid
supplementation), activation of analogues 4, 7, 8, and 12 appears to be the rate limiting
step in protein biosynthesis and prevents the accumulation of measurable amounts of
target protein.
Increasing the MetRS activity of the cellular host and increasing the concentration of analogue in the medium increases the rate of activation of analogues 4, 7, 8, and 12 sufficiently to rescue their translational activity in vivo. Results in Figure 2.4 indicate that the catalytic efficiency of analogue activation can be reduced approximately 2000- fold relative to that of methionine (to a k,,/K, ' value of approximately 3.0 x 10"' s jiM ')
prior to the loss of the ability of an analogue to support protein synthesis under a standard set of experimental conditions. The observation of translational activity for 4, 7, 8, and
in cellular 12 hosts in which MetRS activity is increased 50-fold and intracellular
analogue concentration is likely increased 5-10-fold is consistent with these previous
observations, as the rates of activation are effectively raised 250-500 fold under these conditions. Even for the poorest substrate measured for MetRS (12, which has a KJK^
10'^ value of 1.6 x s'' |LlM"'), the rate of activation would be effectively increased above
the kga/K^ threshold of 3.0 x 10"* s ' |iM ' as a result of the increase in activity and
98 supplementation, which would rescue the ability of the analogue to support protein
synthesis in vivo.
The importance of side chain length in facilitating activation of an analogue by
MetRS is again illustrated, as 7, 8, and 12 all have side chains of different length than
methionine and are much less efficiently activated by MetRS. Although 4 has the same
side chain length as methionine, the conformational restriction placed on the side chain
by the c/5-conformation of the double bond imposes on 4 a geometry that likely does not bind well at the active site of MetRS. The presence of 7i-electrons in the side chains of 4 and 12 may also play a role in their activation by MetRS. The lack of observed translational activity for 10 likely results both from the fact that the side chain is longer than that of methionine, and because the electron density is located at the y-position of the side chain instead of near the 5-position. The shorter side-chain length and linear
geometry of 13 may restrict the accessibility of the n electrons of the triple bond and limit
binding of this analogue at the MetRS active site. It is surprising, however, that the translational activity of 6 is not rescued under any of the in vivo assay conditions reported
here, given the translational activity of both 9 and trifluoromethionine. Perhaps the
hydrogen bonding of the thioether sulfur of trifluoromethionine in the active site of
MetRS permits accommodation of the electronegative fluorine groups, while the lack of
those interactions for the side chain of 6 precludes activation of the analogue by MetRS.
99 Taken together with the studies in Chapter 2, these results corroborate the importance
both of side chain length and of electron density near the 6-position of the analogue side
chain in activation of the analogues by MetRS.
Quantitative assessment of the kinetic parameters for activation of the analogues
by W305F was also conducted. Consistent with the in vivo results, the mutant MetRS
W305F is only a factor of 4 more efficient in activation than the wild-type MetRS, as
delineated in Table 3.5. Nevertheless, this small improvement in catalytic efficiency may
point to the potential for altering the MetRS activity through directed evolution as a
means to further expand the substrate range of the translational apparatus.
3.3.8 Characterization of GFP Fluorescence
The effect on protein function by replacement of methionine by methionine
analogues was preliminarily investigated by monitoring the fluorescence of green
fluorescent protein (GFP) produced from bacterial cultures supplemented with various
methionine analogues. Bacterial cultures CAG18491/pQE31-GFP/pREP4 were obtained
by transformation of the E. coli methionine auxotroph CAG 18491 with the expression plasmid pQE31-GFP and the repressor plasmid pREP4. The expression plasmid pQE31-
GFP encodes GFP with an N-terminal hexahistidine sequence under the control of a
bacteriophage T5 promoter. Cultures (50 mL) of the bacterial host were grown in M9
100 medium lacking methionine and supplemented with analogues 2, 3, or 9. The GFP was
purified by metal chelate affinity chromatography under native conditions, and the
fluorescence measured from the elution buffer solutions is shown in Figure 3.20.
As can be seen in Figure 3.20, GFP produced from cultures supplemented with
500 mg/L 2, 3, or 9 has an emission spectrum that is similar to that observed for GFP
produced from cultures supplemented with the same level of methionine. While the emission maximum is observed at 506 in all nm of the spectra, it is obvious from the normalized spectra shown in Figure 3.20 that the intensity of the emission peak is significantly reduced in the GFP containing the methionine analogues 2, 3, or 9. This suggests that the replacement of methionine does not directly alter the chemical nature of the chromophore, but causes changes in protein structure that reduce the fluorescence of the protein.
Regardless, these results demonstrate that, even at high levels of substitution, the
incorporation of non-natural amino acids does not necessarily result in a complete loss of
structure or function. There have been a variety of literature reports in which biological
activity of proteins is retained after substitution by amino acid analogues (27-32); in fact,
selenomethionine supports cell growth (28). Certainly the retention or loss of protein
function upon incorporation of non-natural amino acids must be assessed on a case-by-
case basis. The effect of analogue incorporation on protein function will depend on the
natural amino acid being replaced, the chemical nature of the non-natural amino acid, and
101 the number, positions, and importance of the natural amino acid in the structure and
function of the target protein.
3-4 Summary and Conclii.sion<;
The incorporation of 4, 5, 7, 8, 11, and 12 into proteins in vivo constitutes the first example of broadening the amino acid substrate range of the E. coli translational apparatus via overproduction of wild-type MetRS in a bacterial host. (Overexpression of a mutant MetRS W305F provided little additional advantage.) Under an appropriate set of experimental conditions employing overexpression, analogues 5 and 11, which are not incorporated by a conventional bacterial host, replace methionine at levels exceeding
96%. Increasing the analogue supplementation to 500 mg/L in cultures of the modified bacterial host permits incorporation of a larger set of methionine analogues (4, 7, 8, and
12) with replacement levels up to 92%. The poorest of these substrates is approximately
340,000-fold poorer a substrate for MetRS during activation than methionine, which
suggests that, under appropriate experimental conditions, much poorer substrates for
MetRS than previously imagined can be utilized by the protein biosynthesis machinery.
As a result of the versatile chemistry of unsaturated groups (33-39), the incorporation of
the unsaturated analogues 4, 5, 11, and 12 may provide new opportunities for
macromolecular synthesis through protein engineering. More generally, these results
102 indicate that this simple strategy- overexpression of wild-type - aaRS may be used to
modify proteins by incorporation of non-natural amino acids that are poor substrates for
aaRS and that would be essentially inactive in conventional expression hosts.
Our results also demonstrate that increasing the MctRS activity of a bacterial expression host improves the yields of proteins containing methionine analogues that are poor substrates for MctRS. Overexpression of aaRS ,nay therefore also be a general strategy for improving yields of proteins containing other non-natural amino acids, as well as proteins rich in particular natural amino acids. Manipulation of the aaRS activities of a bacterial host therefore has enormous potential for broadening the scope of protein engineering by permitting production of natural and artificial proteins with novel chemical and physical properties.
103 Table 3.1 Intracellular amino acid concentrations as determined by amino acid analysis.
Analogue Level of supplementation Intracellular concentration, \iM*
mg/L (jxM)
Met un 35 ±5
Met 20 (135) 58 ±7
Met 500 (3400) 350 ± 40
8 20 (170) 78 ±8
8 500 (4300) 600 ± 50
9 20 (150) 48 ± 10
9 500 (3800) 220 ± 20
Average values from duplicate analyses are reported.
104 Table 3.2 Amino acid analysis results for DHFR containing methionine analogues.
Amino acid Theoretical mole % DHFR-Met DHFR-Ana*
Ala 2.9 3.6 3 6 Cys
Asx 7.1 7.6 1 6 Glx 11.9 12.3 12 1
Phe 4.3 4.4 4.6
Gly 7.6 8.5 8.5
His 3.8 3.9 3.8
He 6.7 5.6 5.9
Lys 8.0 8.1 8.2
Leu 9.5 9.6 9.8
Met 3.8 3.7 0.2
Pro 6 2 6.6 Arg 5.7 6.0 6.1
Ser 7.6 7.8 8.0
Thr 2.9 3.3 3.1
Val 6.7 6.3 6.4
Trp 1.9 N/A N/A
Tvr 2.9 2.9 2.9
* Ana = analogue. Average values from duplicate analyses of each of 6 DHFR-Ana samples are reported. (DHFR was produced from cultures supplemented with 4, 5, 7, 8, 11, and 12.)
105 Table 3.3 Extent of replacement of methionine by methionine analogues,
Extent of Replacement, %
Analogue Amino Acid Analysis N-terminal Sequencing
5 92 ±3 97 ±2
11 98 ±2 97 ±2 8 92 ±3 87 ±2 12 84 ±5 84 ±2 4 76 ±5 79 ±5 7 60 ±5 56 ±5
106 Table 3.4 Kinetic parameters for activation of methionine analogues by MetRS.
Analogue k "cat' Relative
Met 24.3 13.3 0.547 1 3 2415 2.61 10"^ 1.08 X 1/500 9 4120 2.15 5.22 X 10' 1/1050
2 4555 ' 1.35 2.96 X 10 1/1850
5 15675 1.82 1.16 X 10' 1/4700
11 38560 1.51 3.91 X 10-^ 1/14000
8 14380 0.17 1.18 X 10-' 1/45600
4 N/A 10-^ N/A 4.31 X 1/127200 12 N/A N/A 1.58 X 10"^ 1/342000 7 ND* ND ND ND
* Not determined
107 Comparison of catalytic efficiencies for the activation of methioni analogues by MclRS and W305F.
W305F
' Analogue ' k /K s uM ' i\ci alive ' k,„,/K,.„ s laM Relative
Met 0 ^47 1 1.06 1
10"* 3 1 08 X 1 /DUU 4.18 X 10 1/260
^ 9 5 22 x1 0 1 /I n^M 1.84 X 10 1/580
2 2.96 X 10 ' 1/1850 1.27 X I0-' 1/850
5 " 1.16 X 10" 1/4700 4.35 X 10 1 /24()0
11 10' 3.91 X 1/14000 1.47 10" X I /7200
8 10' ' 1.18 X 1/45600 3.74 X 10 1/28400
4 10' ' 4.31 X 1/127200 2.42 X 10 1/43800
** 12 1.58 X ' lO 1 /342000 1.83 X 10 1/58000
108 Nhel 3348
Figure 3. 1 Genetic strategy for increasing the MetRS activity in the bacterial host.
(a) The plasmid pUC19 is linearized with restriction digestion by Aat II
and Sac I, and the linker insert with these cohesive ends is ligated into
the plasmid to yield pUC19-Nhe. (b) The pUC19-Nhe is linearized by
digestion with Kpn I and Sac I, and the W305F gene fragment with the
same cohesive ends is ligated into the plasmid to yield pUC19-W305F.
(c) The W305F gene is removed from pUC19-W305F by digestion with
Nhe I and ligated into the unique Nhe I site of pQE15 to yield pQE15- W305F.
109 Aat II compatible Nhe I Sac I Kpn I Nhe I Sph I
Figure 3.2 Schematic of Nhe linker designed to change the identity of the cohesive ends of the gene encoding W305F into Nhe I. The W305F fragment can
be removed from its original plasmid pBSM547W305F by restriction
digestion with Kpn I and Sac I and inserted in between those sites in this
linker. The W305F fragment can then be removed from the linker with
restriction digestion with Nhe I and ligated into the unique Nhe I site in the plasmids pQE9 and pQE15.
110 4000 bp W305F
1500 bp
ExpecteH
Lanes 1,10: MW markers
Lane 2: Bluescript plasmid Kpn l/Sac I 2850, 2450 Lane 3: W305F Nhel 2498 Lane 4: pUC19-W305F Nhel 2498, 2194 Lane 5: pUC19-W305F Hind III 1559, 3133 Lanes 6,7 PQE9-W305F Nhel 3440, 2498 Lane 8: PQE9-W305F Hind III 4271, 1667 Orientation 1
Lane 9 PQE9-W305F Hind III 4867, 1071 Orientation 2
Figure 3.3 Restriction digestion results for the construction of plasmid pQE9-W305F.
Ill 4000 bp —
- W305F - 2498 bp 1500 bp —
1000 bp —
DHFR - 920 bp
Plasmid Digest
Lane 1: MW markers
Lane 2; PQE15 uncut Lane 3 pQE15/pREP4 Nhel 4004 Lane 4: pQE15/pREP4 Hind III 4004
Lane 5: pQE15-W305F/pREP4 uncut Lane 6: pQEI5-W305F/pREP4 Nhel 4004, 2498 Lane 7: pQE15-W305F/pREP4 Hind III 5430, 3740, 1072 Lane 8: pQE15/pREP4 Aat WNhe I 3740, Lane 9: 3084, 920 pQE15-W305F/pREP4 Aat WNhe I 3740, 3084, 2498, 920 Lane 10: W305F Nhel 2498
Figure 3.4 Restriction digestion results for plasmid pQE15-W305F.
112 62
llll^^
mm- mm^ 25 kDa - DHFR
(a) (b)
Figure 3.5 SDS-PAGE analysis of the expression of DHFR and MctRS in
CAG18491/pREP4 bacterial hosts equipped with (a) pQE15 or (b) PQE15-MRS.
113 1
1.2 10-''
(a)
11 1 10
8 10 12 (0
Q. ® 6 10-^2 o
4 10-12
2 10 12
iO 6 10
(b)
5 10 11
4 10-11
CO of Q. 3 10 11 CO 0) o
2 10 11
11 1 10
r-— PQEI5 MRS PQE15 MRS Met +Met Met +Met +Met, corrected
Figure 3.6 Activation rates of methionine by whole cell lysates, (a)
CAG18491/pREP4 and (b) B834(DE3)/pREP4, equipped with the expression plasmids pQE15 or pQE15-MRS (MRS).
114 Figure 3.7 SDS-PAGE analysis of DHFR expression in bacterial cultures
supplemented with methionine analogues as indicated, (a) The
conventional bacterial host CAG18491/pQE15/pREP4. (b) The modified bacterial host CAG18491/pQE15-MRS/pREP4.
115 (a) (b) (c)
^m^i iHlPi ^Biiii
Figure 3.8 Western blot analysis of DHFR yield from modified bacterial host
CAG18491/pQE15-MRS/pREP4 cultures supplemented with 11 (L- isomer) at concentrations of (a) 60 mg/L, (b) 250 mg/L, and (c) 500 mg/L.
116 1
I 1 I 1 I ^ , PQE15 MRS PQE15 MRS pQE15 MRS pQE15 MRS
Western blot analysis of DHFR synthesis by bacterial expression hosts CAG18491/pQE15/pREP4 (pQE15) and CAG18491/pQE15-MRS/pREP4 (MRS). Bacterial cultures were supplemented with methionine, 2, 3, or 9.
117 Figure 3.10 Yields of purified DHFR obtained from 1 -liter cultures supplemented with methionine, 2, 3,or 9. The solid bars represent yields from
CAG18491/pQE15/pREP4 cultures, while the striped bars represent yields
from CAG18491/pQE15-MRS/pREP4 cultures.
118 Methionine - +
(b) (a) (b) (a) (b) (a) (b)
8
(a) (b) (a) (b) (a) (b)
10 12 13
Figure 3. 1 1 SDS-PAGE analysis of DHFR synthesis in bacterial cultures
supplemented with methionine analogues at 500 mg/L. Cultures were
supplemented with methionine, 4, 6, 7, 8, 10, 12, and 13 (the analogues
that did not support protein synthesis at lower levels of supplementation),
as indicated, (a) CAG18491/pQE15/pREP4. (b) CAG18491/pQE15- MRS/pREP4.
119 (a) Met PP'"
w p diet AJ
(b)
pptu
60 10.0 14.0 18.0 Retention time, minutes
Figure 3. 1 2 N-tcrminal sequencing results indicating occupancy of the
initiator site in DHFR produced in bacterial cultures
supplemented with 4, which elutes at 14.8 minutes. N-lerminal
residues are shown for (a) DHFR-Met, (b) the free amino
acid 4, and (c) DHFR-4, as determined by Edman degradation.
120 pptu (a) Met
D Q A P AJ
(b)
diet
(c) pptu
, —' -I.I I . 6.0 10.0 14.0 18.0
Retention time, minutes
Figure 3.13 N-terminal sequencing results indicating occupancy of the
initiator site in DHFR produced in bacterial cultures
supplemented with 5, which elutes at 16.0 minutes. N-terminal
residues are shown for (a) DHFR-Met, (b) the free amino
acid 5, and (c) DHFR-5, as determined by Edman degradation.
121 (a) Met PP^"
D Q ^ diet Ljll
(b) pptu
w diet
10.0 14.0 18.0 20.0
Retention time, minutes
Figure 3.14 N-terminal sequencing results indicating occupancy of the
initiator site in DHFR produced in bacterial cultures
supplemented with 7, which elutes at 20.8 minutes. N-terminal
residues are shown for (a) DHFR-Met, (b) the free amino
acid 7, and (c) DHFR-7, as determined by Edman degradation.
122 (a) Met pptu
D Q A W diet aJ
(b) 8 pptu
diet
__A-_
(c) 8 pptu
Met diet
60 10.0 14.0 18.0 Retention time, minutes
Figure 3.15 N-terminal sequencing results indicating occupancy of the
initiator site in DHFR produced in bacterial cultures
supplemented with 8, which elutes at 14.2 minutes. N-terminal
residues are shown for (a) DHFR-Met, (b) the free amino
acid 8, and (c) DHFR-8, as determined by Edman degradation.
123 (a) pptu Met
p Q A p W diet AJ
(b)
11 pptu
diet
(C)
11
pptu Met diet
—a— -. ,/ I /I. A . 6.0 10.0 14.0 18.0 Retention time, minutes
Figure 3.16 N-terminal sequencing results indicating occupancy of the
initiator site in DHFR produced in bacterial cultures
supplemented with 11, which elutes at 1 1.7 minutes. N-terminal
residues are shown for (a) DHFR-Met, (b) the free amino
acid 11, and (c) DHFR-11, as determined by Edman degradation.
124 (a) Me pptu
W diet
(b) 12 pptu
diet li A
(c) 12
pptu Met
diet
6.0 10.0 14.0 18.0 Retention time, minutes
Figure 3.17 N-terminal sequencing results indicating occupancy of the
initiator site in DHFR produced in bacterial cultures
supplemented with 12, which elutes at 11.6 minutes. N-terminal
residues are shown for (a) DHFR-Met, (b) the free amino
acid 12, and (c) DHFR-12, as determined by Edman degradation.
125 (b)
(d)
^.iV ^> \ Ajv A/ '
2 6.8 6.4 6.0 5.6 5.70 5.60 5.50 5.40 5.30
Chemical shift, ppm
18 1 H NMR spectra (599.69 MHz) of DHFR (a), 5 (b), and
DHFR-5 (c + d). Samples were dissolved at concentrations of
approximately 10 mg/mL in D2O containing 2% [D2lformic acid
and spectra were recorded at 25 °C overnight.
126 2.5 10'
7 8 10 12 13 Bkgd
Analogue
Figure 3. 19 Aclivation of methionine analogues by MelRS. Assay reactions were
supplemented with (he methionine analogues indicated, at a concentration of 15 mM. The reaction was quenched at 3.5 hours. The background is given for a reaction mixture lacking both analogue and enzyme.
127 Wavelength, nm
ure 3.20 Fluorescence spectra for GFP produced from CAG 18491 /pQE31-
GFP/pREP4 cultures supplemented with methionine, 2, 3, or 9. The
spectra were collected at room temperature with an excitation maximum of 396 nm and an emission maximum of 506 nm.
128 3.5 References
(1) Swanson Hoben P., , Sumner-Smith, M., Uemura, H., Watson, L. and Soil ' D Science 1988, 242, 1548-1551.
(2) Sherman, J.M., Rogers, M.J. and Soli, D. Nucleci Acids. Res. 1992, 20, 2847- 2852, '
(3) Varshney, U. and RajBhandary, U.L. J. Bacterial. 1992, 174, 7819-7826.
Li, S., Kumar, (4) N.V., Varshney, U. and RajBhandary, U.L. J. Biol. Chem 1996 277, 1022-1028.
Fourmy, D., (5) Mechulam, Y., Brunie, S., Blanquet, S. and Fayat, G FEBS Lett 1991, 292, 259-263.
(6) Kim, H.Y., Ghosh, G., Schulman, L.H., Brunie, S. and Jakubowski, H. Proc Natl Acad. Sci. USA 1993, 90, 1 1553-1 1557.
(7) Kast, P. and Hennecke, H. J. Mol. Biol. 1991, 222, 99-124.
(8) Ibba, M., Kast, P. and Hennecke, H. Biochemistry 1994, 33, 7107-71 12.
Ibba, (9) M. and Hennecke, H. FEBS Lett. 1995, 364, 273-275.
(10) Sharma, N., Furter, R., Kast, P. and Tirrell, D.A. FEBS Lett. 2000, 467, 37 40.
Sambrook, J., (11) Fritsch, E.F. and Maniatis, T. Molecular Cloning: A Laboratory Manual; Cold Spring Harbor Laboratory Press: Cold Spring Harbor, NY, 1989,
1 ( 2) QIAquick® Spin Handbook; Qiagen: Valencia, CA, 2000, p. 23.
1 ( 3) Qiaprep® Miniprep Handbook; Qiagen: Valencia, CA, 1 999.
1 ( 4) Ni-NTA Spin Handbook; Qiagen: Valencia, CA, 1 998, p. 1 6.
(15) The Qiaexpressionist; Qiagen: Valencia, CA, 1998.
129 . . . .
( 1 6) Kiick, K.L. and Tirrell, D. A. Tetrahedron 2000, 56, 9487-9493.
Raunio, (17) R. and Rosenqvist, H. Acta Chem. Scand. 1970, 24, ll-il-llAA.
(18) Uhrin, D. and Barlow, P.N. J. Magn. Reson. 1997, 126, 248-255.
(19) Kupce, E., Boyd, J. and Campbell, I.D. J. Magn. Reson. Ser. B. 1995, W6, 300-
(20) Inouye, S. and Tsuji, F.I. FEBS Lett. 1994, 341, 277-280.
(21) Tsien, R.Y. Anna. Rev. Biochem. 1998, 67, 509-544.
(22) Wang, P. 2001, unpublished results.
(23) Ni-NTA Spin Handbook; Qiagen: Valencia, CA, 1998, p. 18.
(24) Kiick, K.L., van Hest, J.C.M. and Tirrell, D.A. Angew. Chem. Int. Ed Engl 2000 59,2148-2152.
(25) Pipemo, J.R. and Oxender, D.L. J. Biol. Chem. 1968, 243, 5914-5920.
(26) Hirel, P.H., Schmitter, J.M., Dessen, P., Fayat, G. and Blanquet, S. Proc. Natl. Acad. Sci. USA 1989, 86, 8247-825 1
(27) Cowie, D.B. and Cohen, G.N. Biochim. Biophys. Acta 1957, 26, 252-261
(28) Tuve, T. and Williams, H. J. Am. Chem. Soc. 1957, 79, 5830-5831.
(29) Koide, H. et al. Proc. Natl. Acad. Sci. USA 1988, 85, 6237-6241
(30) Koide, H. et al. Biochemistry 1994, 33, 7470-7476.
(31) Lau, E.Y. and Gerig, J.T. Biophys. J. 1997, 73, 1579-1592.
(32) Yuan, T. and Vogel, H.J. Protein Sci. 1999, 5, 1 13-121.
(33) Schoenberg, A. and Heck, R.F. J. Org. Chem. 1974, 39, 3327-333 1
130 '''' ' "---^ ^-^-^ S 1^9r' Press:
^''^ ^-^-^^ Car«/,.,... "J";/;. f /.„.v«r/... in Organic Johnf Synthesis-'^>'"^^^5/5, Wiley and Sons: New York, 1995.
(36) Blackwell, H.E. and Grubbs, R.H. Ange.. Chem. Inu Ed. Engl 1998, 37, 3281-
(37) Clark, T.D., Kobayashi, K. and Ghadiri, M.R. Chem. Eur. J. 1999. 5, 782-792.
(38) Amatore, C. and Jutand, A. J. Organomet. Chem. 1999, 576, 255-277.
(39) Furstner, A., Guth, O., Rumbo, A. and Seidel, G. J. Am. Chem. Soc 1999 • J21 11108-11113.
131 CHAPTER 4
IN VIVO INCORPORATION OF AZIDO AMINO ACIDS
4-1 Introduction and Ohj prfivpc
In the previous chapters it has been demonstrated that the translational apparatus
of E. coli is remarkably permissive toward methionine analogues under appropriate
experimental conditions. In our efforts to further expand the repertoire of selective and
mild chemistries that can be used to engineer proteins in vivo, we turned our attention to
recently published work indicating that azidosugars can be utilized by the polysaccharide
synthesis machinery of the cell (1). Such permissiveness has allowed the engineering of
cell surfaces with sialic acid residues bearing azide and other functionality (1-4); the
azide can be subsequently derivatized under very mild aqueous conditions at high yields
(1). The incorporation of azidoamino acids into proteins in vivo may therefore be a feasible and desirable approach for engineering proteins that can be selectively modified under native conditions.
The chemical modification strategy for proteins decorated with azide functionality
is shown in Figure 4. 1 . The triarylphosphine reagent can selectively deliver a variety of
functional moieties (although the FLAG peptide is shown in the figure) via formation of
an aza ylide intermediate that rearranges to form an amide bond (1). Based on the
132 potential of this chemistry for selective chemical modification of proteins, two azido
ammo acids, 14 and 15 (Appendix B and Figure 4.2), were investigated as potential
methionine surrogates vi.o and in activation assays in .itro. Electron density maps
and isopotential surfaces for these analogues were generated. Combining the results
from in vitro and in vivo studies with models of the electronic and steric features of these
side chains may provide additional insight into the primary factors controlling the
activation of methionine analogues by MetRS. Chemical modification of the azido
residues in the target protein DHFR was also investigated.
4.2 Experimental Section
4.2.1 Computational Methods
The modeling of the electron density maps and isopotential surfaces of analogues
14 and 15 was conducted as described in Section 2.2.5.
4.2.2 Analogue Synthesis
Analogues 14 and 15 were kindly provided by Eliana Saxon, University of
California Berkeley, Berkeley, CA, and were synthesized by modification of serine and
133 homoserine, respectively (5-7). They were charaeterized by NMR spectroscopy and mass
spectrometry and used as provided in assays measuring thetr translational activity ,„ v,va
and their activation by MetRS in vitro.
4-2.3 Activation of Azidoamino Acids hy MetRS in Vitm
The activation of analogues 14 and 15 was measured by monitoring the ATP-PP,
exchange supported by the analogues, as described in Section 2.2.4. Assays designed to
measure the amount of PP. exchanged in a 20-minute period were conducted using 75 nM
MetRS and an analogue concentration of 5 mM. An additional assay to measure the
amount of PP^ exchange at 3.5 hours in a reaction mixture that was 75 nM in enzyme and
15 mM in analogue was also conducted for analogue 14. Quantitative determination of
the kinetic parameters for the activation of the analogue 15 by MetRS was also conducted
as previously described. Enzyme concentrations of 50 nM and analogue concentrations
of 200 \lM to 10 mM were used in these assays.
4.2.4 Translational Activity of Azidoamino Acids in Vivo
Experiments to measure the ability of the analogues to support protein synthesis
in vivo were conducted as previously described (Section 2.2.2). Small-scale (5 mL)
134 expressions were conducted for both analogues at analogue concentrations of 20 mg/L and 500 mg/L. Large-scale (1 L) expressions were conducted for analogue 15 with an analogue concentration of 400 mg/L. Proteins were purified via metal chelate affinity chromatography and analyzed by amino acid analysis and N-terminal sequencing as described in Section 3.2.4. Fragments of DHFR-15 were generated by trypsin digestion and were analyzed by matrix-assisted laser desorption ionization mass spectrometry
(MALDI-MS) employing a-cyano-4-hydroxycinnamic acid as the matrix. The digestion and MALDI-MS analysis were conducted by Gary Hathaway at the Protein/Peptide
Microanalytical Laboratory at the California Institute of Technology, Pasadena, CA.
4.2.5 GPP Fluorescence
A medium-scale (50 mL) expression of GFP (which contains 6 methionine residues) from bacterial cultures CAG18491/pQE31-GFP/pREP 4 was conducted as a
preliminary assessment of the impact of the incorporation of 15 on the activity of proteins. The expression protocols, purification procedures, and fluorescence
measurements were identical to the procedures described in Chapter 3. The fluorescence
spectra of the proteins were measured directly as described in Section 3.2.6 and
normalized for protein concentration, which in this case was determined by the Lowry method.
135 MQdification of Azide \i^.\^ y.. yj, ^ gn
Protein (DHFR) containing 15 in place of methionine (DHFR-15) was purified
via metal chelate affinity chromatography under denaturing conditions, dialy/ed, and
lyophili/ed, as described in Chapter 3 (8). This protein was also analyxed by amino acid
analysis and N-terminal sequencing to confirm the incorporation of 15. The procedures
described below were conducted by liliana Saxon at the University of California
Berkeley, Berkeley, CA. I)nFIM5 and DHFR-Met (10 ^M) were incubated (separately)
with the triarylphosphine-FLAG reagent, shown in Figure 4.1, (250 ^M) in PBS buffer
(137 mM NaCl, 2.7 mM KCl, 4.3 mM Na,IIPO,.7Il,0, 1.4 mM K11,P0„ pll 7.4) at room
temperature for up to hours. 6 The selectivity of the Staudingcr ligation was assessed by
Western blot analysis with either a primary RGS-Mis antibody or a primary FLAG
antibody. Western blots were developed as previously described. After incubation with
the appropriate primary antibody, the blots were washed and treated with a secondary
anti-mouse IgG conjugated to HRP and were developed with appropriate
chcmilumincscence reagents (7).
136 4.3 Results and Discussion
4.3.1 Computational Modeling of Analogues
The computational studies on azido amino acids are aimed not only at the design
of an azido methionine analogue that can replace methionine in vivo, but also at
establishing additional evidence for the important structural and electronic features that
make an analogue a good substrate for MetRS. The electron density maps and
isopotential surfaces for the azido methionine analogues 14 and 15 were calculated and
compared with those of methionine, as shown in Figure 4.2.
Our previous studies suggested that the presence of electron density at or near the
5-position of the analogue is one important feature for activation of a methionine
analogue by MetRS, as no single side-chain geometry appeared to be required for
analogue activation by the enzyme (9-12). These observations are consistent with crystal
structure studies demonstrating the importance of hydrogen bonding interactions between
the sulfur at the 5-position of the side chain of methionine and the hydroxyl group of
Tyr260 of MetRS (13). It is apparent from the representations in Figure 4.2 that 15 not
the 5- only is a good structural analogue for methionine, but also has electron density at
methionine. position of the side chain with an isopotential surface very similar to that of
methionine and has electron Analogue 14, on the other hand, is less similar structurally to
137 dcnsi.y U.ca.ed at the rposilion and on the opposite s.dc of ,he cx.cndecl side chain tin
•hal of methionine. Based on these resuhs and by considering the orientation of
metlHonine in the bniding poekel of MetRS (Figure 1 .4), we hypothesized thai 15. and
not 14, should serve well as a methionine surrogate in vitro and in vivo.
"^•3-2 Activation of Azidoamin^ /\r„k hy MMiSjnJ^
The results of ATP-PP. exchange assays for 14 and 15 support the above
hypothesis and indicate that only 15, and not 14, is able to support PP, exchange in vitro,
as illustrated in Figure 4.3. The in vitro results for all of the methionine analogues are
given to facilitate comparison between analogues. As shown in the figure, 15 is indicated
to support exchange at a rate similar to that observed for 3, one of the most efficiently
activated methionine analogues investigated to date. (Analogue 14 did not support PP,
exchange at levels above background even in a.ssays that were 15 mM in 14 and were
quenched after 3.5 hours.) The quantitative analysis of the kinetic parameters for 15 is
consistent with the data in Figure 4.3 and is summarized in Table 4. 1 . A k^^K,,, value of
10"'' 1.42 x s"' ' iiM was measured for activation of 15 by MetRS in vitro, indicating that it
may be a slightly better analogue for activation by MetRS than 3, which has a k,„/K,„
10"' ' '. value for activation of 1 . 16 x s ^iM The side chains of both 3 and 15 have
138 elecron densi.y near ,he 8-position of the amino acid, although .5 shares a greater
conformational similarity with methionine that may ailow it to be activated slightly more
efficiently by MetRS than 3.
^•^•3 Translational Anivity of Azido.nminn a.;h. yi^.^
Whether or not these In vitro observations correlate with the translational activity
of and in vivo 14 15 was tested in a standard small-scale (5-mL) protein expression assay
The ability of the analogues to support protein synthesis in bacterial cultures
supplemented with the analogue and depleted of methionine was monitored by SDS-
PAGE analysis. Both conventional (CAG18491/pQE15/pREP4) and modified
(CAG18491/pQE15-MRS/pREP4) bacterial expression hosts were investigated. The
results in Figure 4.4, shown for the conventional expression host
CAG18491/pQE15/pREP4, are representative of both hosts and demonstrate that only
analogue 15 supports protein synthesis in vivo. Analogue 14 did not support protein
biosynthesis even in a bacterial host equipped with extra copies of the wild-type or
mutant MetRS.
Consistent with the relatively high catalytic efficiency with which MetRS
activates this analogue (Table 4. 1), the synthesis of target protein is observed in cultures
of a conventional bacterial host supplemented with 15; overexpression of MetRS is not
139 required. Also consistent with the relatively high k.,/K.,. for this analogue (similar to the
k,„/K„ for 3, which supported protein yields identical to those for methionine), the yields
of protein obtained from 5- mL cultures supplemented with 20 mg/L 15 are similar to
those for cultures supplemented with 20 mg/L methionine, as shown in Figure 4.5.
Increasing the concentration of amino acid in the medium to 500 mg/L does not increase
the protein yield for cultures supplemented either with 15 or methionine, indicating that protein biosynthesis is not limited by the rate of activation of either of these amino acids under the in vivo assay conditions. Yields from large-scale expressions ( 1 L) corroborate these observations, with the yield of DHFR obtained from 1 -L cultures of
CAG18491/pQE15/pREP4 supplemented with 15 matching those obtained from cultures supplemented with methionine or 3 (35 mg/L).
4.3.4 Protein Characterization
Protein obtained from these expressions was purified and analyzed for amino acid
content and for the identity of the N-terminal amino acid. The results of the amino acid
analysis are listed in Table 4.2, and show that other than the decrement in methionine
content, the profile of amino acids is consistent with that expected for DHFR. As 15 is
unstable under the analysis conditions, the decrease in methionine content is taken as
indirect evidence of replacement of methionine by 15. Based on the decrease of
140 methionine content from an expected value of 3.8% to an observed value of 0.2%, the
amino acid analysis indicates 95 ± 2% replacement of methionme.
As DHFR retains the N-termmal methionine " (14), direct measure of the
incorporation of the analogue could be obtained from N-termmal sequencmg, shown in
Figure 4.6. The signal corresponding to methionine elutes at approximately 13.0
minutes while that corresponding to 15 elutes at 1 1.5 minutes. The large peaks that elute
at approximately 14.7 minutes correspond to piperidylphenylthiourea (pptu), a product of
the analysis resulting from the buffer, and the small peak at approximately 18.5 minutes
corresponds to diethylphthalate (diet), an internal standard. As demonstrated in the
figure, the N-terminal methionine of DHFR (Figure 4.6a) is nearly completely replaced
by 15 (Figure 4.6b) in DHFR produced from cultures supplemented with 15 (Figure
4.6c). Integration of the peak areas indicates 97 ± 2% replacement of methionine.
Peptide fragments from DHFR-15 were generated by digestion with trypsin and
analyzed by MALDI-MS. Results for all observed fragments in which methionine is encoded are delineated in Table 4.3, and a representative mass spectrum for the fragment with the encoded sequence IMQEFESDTFFPEIDLGK is shown in Figure 4.7. The methionine-containing fragment should exhibit a mass of 2146.4, although no such peak
is apparent in this data. Instead, a fragment with a mass of 2141.1 is observed, which is
consistent with the replacement of the single methionine in the fragment by 15. The
complete absence of a peak at the mass expected for the methionine-containing fragment
141 suggests essentially quantitative replacement of me.h.onine by 15. Th,s is further
corroborated by tryptic digest /MALDI-MS analysis of DHFR-Met. in which the
obtained for three additional tryptic fragments of DHFR-15 (Table 4.3). Again,
fragments are observed only at masses consistent with complete replacement of
methionine by 15. and not at masses indicating retention of methtonme. These results
suggest quantitative replacement of methionine by 15 for 6 different posmons encoding
methionine, consistent with both amino acid analyses and N-termi„al sequencing results.
4.3.5 GFP Fluorp,«;rpnrp
The impact of this level of azidoamino acid incorporation on protein function was
preliminarily investigated by monitoring the fluorescence observed in samples of GFP
produced from bacterial cultures supplemented with 15. The GFP contains 6 methionmes
and has an absorbance maximum at 396 nm and an emission maximum at 506 nm (15).
GFP produced from a culture supplemented with 500 mg/L of 15 exhibits an emission
spectrum similar to that observed for GFP produced from cultures supplemented with the
same level of methionine (Figure 4.7). While the emission maximum appears at 506 nm
in both spectra, it is obvious from the normalized spectra shown in Figure 4.8 that the
intensity of the emission peak is significantly reduced for the GFP containing the azide
142 analogue. The fac, ,ha, flucescence i.s obscvec, de^nsna.cs ,ha, ,hc <,„a„,i,a,ivc,
.nuh,,si.e rep,ace,.e„, of „,e,hioni„e by ,5 docs no, .esuU i„ a co.nplce loss o,' p,o,ei„
sTuCurc or func.ion. The selecCve „.odifica,io„ or covalen, eap.ure of HiologicaHy
active macromoleculcs therefore .ecn,.s a feasible application of the ,„corporation of
azidoamino acids into proteins ,„ Wv„. As discnsscd in Chapter 3, it is difficult to make
generalizations regarding the impact of analogue n.corporat.on on protein structure an function, as such an impact must be assessed on an individual basis 4.3.6 .Modification of ih. DHFR-I.S hv staiidin per I i^,.,:^. The ability of DHFR-15 to be selectively modified by the Slaudinger ligation (shown in Figure 4. 1) has also been preliminarily investigated. DUFR produced from bacterial cultures of CAG18491/pQE15/pREP4 supplemented with 500 mg/L of 15 was purified under denaturing conditions, dialyzed, and lyophili/ed. Either DHFR-15 or DHFR-Met was incubated with a triarylphosphine-FLAG reagent in PBS buffer for up to hours in order 6 to selectively react the modified DHFR with the triarylphosphine-FLAG (7). Figure gives 4.9 Western blot results for some of these preliminary experiments. The Western blot in Figure 4.9a was developed with an RGS-His antibody, and bands with a molecular weight consistent with that of DHFR are observed in lanes 2-5, demonstrating that both DHFR-15 and DHFR-Met are recognized by the RGS-His 143 antibody, as expected. The incorporation of 15 at the N-terminal position (near the hexahistidine sequence) does not appear to compromise the binding of the antibody. Only the DHFR-15 that is incubated with the triarylphosphine-FLAG shows additional bands at higher molecular weights, however, consistent with the modification of multiple sites of DHFR-15. The Western blot in Figure 4.9b, developed with a FLAG antibody, exhibits bands only in lane 2. Coupled with the results in Figure 4.9(a), these data demonstrate that only DHFR-15 is labeled with the FLAG peptide; neither DHFR-15 or DHFR-Met incubated in buffer, nor DHFR-Met incubated with the triarylphosphine- FLAG are detected in the Western analysis employing the FLAG antibody. These results unequivocally demonstrate the selectivity of the triarylphosphine reagent for the azidohomoalanine in DHFR-15. The presence of multiple bands in lane 2 of Figures 4.9(a) and 4.9(b), however, indicates that the 8 positions containing 15 are not quantitatively modified. Since our characterization results indicate essentially quantitative replacment of methionine by 15, the lack of quantitative modification does not arise from the presence of methionine in the protein. Incomplete reaction of the azide groups with the triarylphosphine reagent may arise from insufficient length of reaction, a lack of accessibility of the azide groups to the triarylphosphine-FLAG, partial hydrolysis of the aza ylide intermediate in the context of the protein, or partial reduction of the azide groups to amine groups during protein biosynthesis and purification. 144 Reduction during protein biosynthesis and purification is not indicated by N- terminal sequencing, as a single peak with an area corresponding to 97% replacement is observed in the chromatogram of the N-terminal residue of DHFR-15 (Figure 4.6). There is, however, indication of the reduction in the MALDI-MS analysis of tryptic fragments of DHFR-15. Reduction of the azide to the amine results in a loss of 26 mass units, and in the mass spectrum, a small peak corresponding to such a mass loss is observed for each fragment containmg 15. A representative mass spectrum is given in Figure 4.10. Based on the N-terminal sequencing and mass spectral data, the degree of conversion to amine prior to chemical modification is not sufficient to account for the incomplete reaction of the azide functionality with the triarylphosphine-FLAG. Methods to quantify the degree of reduction are currently being investigated, along with methods to improve the extent of chemical modification. 4.4 Summarv and Conclusions These studies constitute the first example of the replacement of methionine residues by azidoamino acids in vivo. The capacity of azidoamino acids 14 and 15 to support protein synthesis in vivo correlates well with their activation by MetRS in vitro, as 14, which does not support measurable levels of PPj exchange, also does not support protein biosynthesis. In contrast, 15, which is the most efficiently activated analogue 145 investigated to date, supports prote.n synthes.s at yields similar to those supported by methionine, with essentially quantitative levels of replaeement. These results provide additional evidence indicating the critical role of MetRS in the incorporation of methionine analogues into proteins m vivo, and yield additional information confirming the important side-chain features controlling activation of methionine analogues by MetRS. That and 15, not 14, is activated efficiently by MetRS indicates that the presence of electron density near the 8-position of a methionine analogue is an important feature for efficient analogue activation. Furthermore, the incorporation of azidoamino acids into proteins provides an additional and novel method for the selective modification of proteins. The selectivity of the reaction between triarylphosphine reagents and the azidoamino acid residues in a protein has been demonstrated for the first time. That a protein containing an azidoamino acid might maintain its function has been indicated by experiments demonstrating that GFP produced from bacterial cultures supplemented with 15 retains a significant level of fluorescence. The ligation of phosphines and azides may therefore permit the modification of biologically active macromolecules or the trapping of transiently associate intermediates within the cellular environment. Investigations are underway to demonstrate quantitative modification of azidoamino acids incorporated into proteins in VIVO. 146 1 ' Table 4. Comparison of kinetic parameters for a.idoamino acds 14 and 15 whh those for methionine analogues 3 and 9. Analogue Relative Met 5.47 X 10 1 14 no activity n/a 15 1.42 X 10-' 1/390 3 10'' 1.16X 1/500 9 5.22 X lO-'' 1/1050 147 Table 4.2 Amino acid analysis results for DHFR-Met and DHFR-15. Amino acid Theoretical mole % DHFR-Met DHFR-15 Ala 2.9 3.6 3.8 Cys — Asx 7.1 7.6 8.1 Glx 11.9 12.3 12.6 Phe 4.3 4.4 4.3 Gly 7.6 8.5 8.4 His 3.8 3.9 3.9 He 6.7 5.6 6.2 Lys 8.0 8.1 8.5 Leu 9.5 9.6 9.4 Met 3.8 3.7 0.2 Pro 6.2 6.4 6.7 Arg 5.7 6.0 6.5 Ser 7.6 7.8 8.6 Thr 2.9 3.3 3.2 Val 6.7 6.3 6.7 Trp 1.9 N/A N/A Tyr 2.9 2.9 2.7 148 Table 4.3 Mass spectrometric analysis of tryptic fragments of DHFR-15. Fragment sequence # Methionines Mass Methionine 15 Observed QNLVIMGR 1 1 930.1 925.0 925.6 IMQEFESDTFFPEIDLGK 1 2146.4 2141.3 2141.1 VDMVWIVGGSSVYQEA MNQPGHLR 2 2674.0 2663.8 2663.4 GSHHHHHHGSGIMVRP LNSIVAVSQNMGIGK 2 3295.8 3285.6 3284.8 149 Electrophilic trap - N. HOaC-KDDDDKYD H OCH3 KDDDDKYD""^ © G N— Protein *V O Ph FLAG peptide Aza-ylide CH3O0 Amide bond H2O Protein H Protein KDDDDKYD' KDDDDKYD"^^ Figure 4. The Staudinger ligation 1 for the chemical modification of azido amino acids with a triarylphosphine-FLAG reagent. 150 (b) (c) Electron density maps (colored surfaces) and negative isopotential surfaces (meshes) for (a) methionine, (b) 14, and (c) 15. The electron density maps indicate electron-rich (red) and electron-poor (blue) regions of each molecule. 151 Figure 4.3 Activation of methionine and methionine analogues by MetRS. The amount of PP, exchanged in 20 minutes is shown for methionine (1) and methionine - analogues 2 15. The full set of analogues is given to facilitate comparison between analogues 2 - 13 (Chapter 2) and 14 and 15. The background (16) is given for a reaction mixture lacking both enzyme and amino acid. 152 Met +Met 14 15 Figure 4.4 SDS-PAGE analysis of the translational activity for methionine analogues 14 and 15. The SDS-PAGE analysis was conducted on whole cell lysates of CAG18491/pQE15/pREP4 cultures. 153 (a) (b) €i' mm 20mg/L 500mg/L 20 mg/L 500 mg/L Figure 4.5 Western blot analysis of the expression of DHFR in bacterial cultures supplemented with the indicated concentration of (a) methionine or (b) 15, 154 (a) pptii Met 4.0 8.0 12.0 16.0 20.0 Retention time, minutes Figure 4.6 N-terminal sequencing results indicating occupancy of the initiator site in DHFR produced in bacterial cultures supplemented with 15, which elutes at 1 1.5 minutes. N-terminal residues are shown for (a) DHFR-Met, (b) the free amino acid 15, and (c) DHFR- 15, as determined by Edman degradation. 155 100 2141.1 Fragment sequence: IMQEFESDTFFPEIDLGK - 2 1 46 4 IZQEFESDTFFPEIDLGK - 2141.3 c(0 50 2146.4 Mass (m/z) Figure 4.7 Mass spectrometric analysis of a tryptic fragment of DHFR-15. A fragment containing methionine (M), with the sequence shown, would exhibit a mass of 2 146.4. Substitution of methionine with 15 (Z) yields a fragment with an expected mass of 2141.3. The relative ratio of methionine-containing fragment vs. 15-containing fragment shown here is observed for three additional fragments. 156 500000 |— n I I I 400 450 500 550 600 Wavelength, nm Figure 4.8 Fluorescence spectra for GFP produced from cultures of bacterial host CAG18491/pQE31-GFP/pREP4 supplemented with methionine or 15. Spectra were collected at room temperature in 50 mM phosphate, 250 mM imidazole buffer at pH 8.0 and were normalized for protein concentration. The spectra were collected at room temperature with an excitation maximum of 396 nm and an emission maximum of 506 nm. 157 Figure 4.9 Western blot analysis of DHFR produced from cultures of bacterial host CAG18491/pQE15/pREP4 supplemented with methionine or 15 (7). (a) Western blot analysis with an RGS-His antibody. 1) Molecular weight markers (19.2, 24.7, 36.4 kDa), 2) DHFR-15 incubated with triarylphosphine-FLAG at room temperature in PBS buffer, pH 7.4, 3) DHFR-15 incubated in PBS buffer, 4) DHFR-Met incubated with the triarylphosphine-FLAG, 5) DHFR-Met incubated in PBS buffer, (b)Western blot analysis with a FLAG-antibody. (1-5) are as designated for (a). 158 100 2141.1 c B S 50 2115.1 Mass (m/z) Figure 4.10 Mass spectrometric analysis of a tryptic fragment of DHFR-15. The peak with a mass of 2141.1 corresponds to a fragment containing 15 (as shown in Figure 4.7). A loss of 26 mass units to yield the peak with a mass of 21 15. 1 is consistent with the reduction of the azide (15) to an amine. 159 4.5 Reference.^ ( 1 Saxon, E. and Bertozzi, ) C.R. Science 2000, 257, 2007-20 1 0. Mahal, (2) L.K. and Bertozzi, C.R. Chem. Biol. 1997, 4, 415-422. (3) Mahal, L.K., Yarema, K.J. and Bertozzi, C.R. Science 1997, 276, 1 125-1 128. (4) Yarema, K.J., Mahal, L.K., Bruehl, R.E., Rodriguez, E.C. and Bertozzi C R J Biol. Chem. 1998, 273, 31 168-31 179. (5) Arnold, L.D., May, R.G. and Vederas, J.C. J. Am. Chem Soc. 1988, 110, 2237- Mangold, (6) J.B., Mischke, M.R. and Lavelle, J.M. Mat. Res. 1989, 216, 27-33. Saxon, (7) E. 2001, unpublished results. (8) Ni-NTA Spin Handbook; Qiagen: Valencia, CA, 1998, p. 16. (9) van Hest, J.C.M. and Tirrell, D.A. FEBS Lett. 1998, 428, 68-70. (10) van Hest, J.C.M., Kiick, K.L. and Tirrell, D.A. J. Am Chem Soc 2000 722 1282-1288. (11) Kiick, K.L., van Hest, J.C.M. and Tirrell, D.A. Angew. Chem. Int. Ed Engl 2000 i9, 2148-2152. (12) Kiick, K.L. and Tirrell, D.A. Tetrahedron 2000, 56, 9487-9493. (13) Serre, L., Verdon, G., Choinowski, T., Hervouet, N., Risler, J.-L. and Zelwer, C. J. Mol. Biol 2001, 306, 863-876. Hirel, P.H., (14) Schmitter, J.M., Dessen, P., Fayat, G. and Blanquet, S. Proc. Natl. Acad. Sci. USA 1989, 86, 8247-8251. ( 1 5) Tsien, R. Y. Annu. Rev. Biochem. 1998, 67, 509-544. 160 CHAPTER 5 FUTURE RESEARCH DIRECTIONS Our studies have demonstrated that overexpression of MetRS in a bacterial host not only increases the number of methionine analogues that can be utilized during protein biosynthesis, but also increases the yields of proteins containing mclh.onme analogues. Efficient incorporation of a variety of unsaturated analogues in place of methionine could be exploited for selective chemical modification via ruthenium-catalyzed (1,2), palladium-catalyzed (3-5), tungsten-catalyzed (6), and other chemistries (7,8). The incorporation of azide functionality permits selective modification with triarylphosphine reagents and may present opportunities for other selective chemical transformations. Furthermore, proteins with high extents of replacement of methionine by unsaturated and azido amino acids may also retain their function, as indicated by preliminary experiments in which GFP containing methionine analogues exhibits significant, albeit reduced, fluorescence. Additional characterization of the modified GFPs (and other proteins, such as DHFR) by circular dichroism and NMR spectroscopies may provide evidence indicating the degree of perturbation of protein folding resulting from the incorporation of methionine analogues. Although these initial studies of modified GFPs indicate that protein fluorescence is reduced, they suggest that protein structure and function might be retained upon analogue incorporation. Incorporation of novel non-natural amino acids 161 v,a man,pula,ion of the aaRS activities of a bacterial host therefore has enormous potenfal for permitting production of natural and artificial proteins with novel chemical and physical properties. Strategies for manipulating the aaRS activities of a bacterial host include overexpression of wHd-type aaRS, site-directed mutagenesis of aaRS, or introduction of heterologous synthetases in the cell (9-11). Overexpression of wild-type MetRS has been demonstrated in our laboratories to permit the incorporation of a variety of methionine analogues, and it is likely that overexpression of other wild-type aaRS will provide similar results. Indeed, overexpression of wild-type LeuRS in a cellular host has rescued the translational activity of hexafluoroleucine, an analogue not utilized by the conventional translational machinery (12). Therefore it seems likely that expandmg this simple strategy - overexpression of - wild-type aaRS to include other aaRS should have similar positive results. Overexpression of a mutant PheRS prepared via site-directed mutagenesis (13) has permitted the in vivo incorporation of /7-Br-Phe, an analogue not normally utilized during protein biosynthesis (14). While overexpression of a mutant MetRS W305F has not afforded advantages over overexpression of wild-type MetRS, the mutant enzyme exhibits slightly higher rates for activation of methionine analogues than the wild-type MetRS. Therefore, mutation of aaRS activities of the cell via directed evolution methods (15,16) may be useful for producing aaRS capable of activating non-natural amino acids 162 not activated by wild-type aaRS. The demonstrated prom.scuity of the wUd-type MetRS, coupled with the increased activity of the mutant MetRS, suggests the potential of evolving MetRS activity toward methionine analogues such as those bearing Huorinated. conjugated, or surfactant-like groups. In addition to this evolutionary-based strategy for developing new aaRS activity, a rational approach to aaRS design may also be possible through the use of computational methods. Computational methods developed by Goddard and coworkers (17-19) have yielded free energies of binding for phenylalanine analogue/PheRS pairs that correlate well with in vitro activation and in vivo incorporation. The rational design of methionine analogue/MetRS pairs is a logical extension of this computational work, given the ample kinetic data now available for the activation of methionine analogues. In the long term, these studies would be useful for understanding the basic interactions that govern binding of an amino acid by an aaRS, without the need for synthesis of new analoguiles or crystallization of an aa/aaRS pair. The prospects are therefore promising for utilizing these approaches to design new aa/aaRS pairs for the purposeful production of chemically and physically novel biopolymers. 163 1 5.2 References Clark, (1) T.D., Kobayashi, K. and Ghadiri, M.R. Chem. Eur. J. 1999, 5, 782-792. Blackwell, (2) H.E. and Grubbs, R.H. Angew. Chem. Int. Ed. Engl. 1998, 37, 3281- 3284. (3) Amatore, C. and Jutand, A. /. Organomet. Chem. 1999, 575, 255-277. Tsuji, (4) J. Palladium Reagents and Catalysts: Innovations in Organic Synthesis; John Wiley and Sons: New York, 1995. (5) Schoenberg, A. and Heck, R.F. J. Org. Chem. 1974, 39, 3327-333 (6) Furstner, A., Guth, O., Rumbo, A. and Seidel, G. J. Am. Chem. Soc 1999 121 11108-11113. (7) Trost, B.M. and Fleming, I. Comprehensive Organic Synthesis; Pergamon Press- Oxford, 1991. (8) Rutjes, F.P.J.T., Wolf, L.B. and Schoemaker, H.E. J. Chem. Soc, Perkin Trans. I 2000,4197-4212. (9) Furter, R. Protein Sci. 1998, 7, 419-426. (10) Ohno, S., Yokogawa, T., Fujii, I., Asahara, H., Inokuchi, H. and Nishikawa, K. J. Biochem. 1998, 124, 1065-1068. (11) Liu, D.R. and Schultz, P.G. Proc. Natl. Acad. Sci. USA 1999, 96, 4780-4785. (12) Tang, Y. and Tirrell, D.A. 2001, unpublished results. (13) Ibba, M. and Hennecke, H. FEBS Lett. 1995, 364, 273-275. (14) Sharma, N., Furter, R., Kast, P. and Tirrell, D.A. FEBS Lett. 2000, 467, 37-40. (15) Arnold, F.H. and Moore, J.C. Adv. Biochem. Eng. Biotech. 1997, 58, 1-14. 164 (16) Arnold, F.H. Chem. Eng. Sci. 1996, 51, 5091-51 02, (17) Vaidehi, N., Jain, A. and Goddard III, W.A. /. Phys. Chem. 1996, 100, 10508- (18) Lim K.-T., Brunett, S., lotov, M., McClurg, R.B., Vaidehi, N., Dasgupta, S " Taylor, S. and Goddard III, W.A. J. Comput. Chem. 1997, 18, 501-521 (19) Vaidehi, N. and Goddard III, W.A. J. Phys. Chem. A 2000, 104, 2375-2383. 165 APPENDIX A LIST OF AMINO ACIDS ALIPHATIC Glycine (Gly G) Alanine (Ala A) Val ine (Val V) Leucine (Leu L) Isoleucine (I!e I) OH OH AROMATIC Phenylalanine (Phe F) Tyrosine (Tyr Y) Tryptophan (Trp W) HoN CYCLIC Proline (Pro P) BASIC Lysine (Lys K) Arginine (Arg R) I listidine (His H) OH Ha OH ^^'V^OH N^NH N NH2 ACIDIC Aspartic Acid (Asp D) Glutamic Acid (Glu E) Asparagine (Asn N) Glutamine (Gin Q) H 'y^oi OH ^^^OH ^0 OH NH2 O-'^OH O'^NHa HYDROXYL OR SULFUR CONTAINING Serine (SerS) Cysteine (Cys C) Threonine (Thr T) Methionine (Met M) Homocysteine (Hey) OH H. OH H2N^ ^OH SH ^OH SH Precursor to methionine 166 APPENDIX B LIST OF METHIONINE ANALOGUES Methionine (1) Homoallylglycine (2) Homopropargylglycine (3) OH 1^ cis-Crotylglycine (4) trans-Crotylglycine (5) 6,6,6 - Trifluoronorleucine (6) CF- 2-Aminoheptanoic acid (7) Norvaline (8) Norleucine (9) H OH 0-Allylserine (10) 2-Butynylglycine (11) Allylglycine (12) HaN^ ^OH H21 Propargylglycine (13) Azidoalanine (14) Azidohomoalanine (15) h.nX,h H2N OH OH N 167 APPENDIX C TOCSY SPECTRA FOR DHFR-5 (a) ppm (b) I ppa (a) Selective irradiation at 5.35 ppm, no mixing time (b) Selective irradiation at 5.35 ppm, 20 ms mixing time 168 APPHNDIX D DNA SEQUENCE OF NHELINKER Mcr I Gdi II Eag I Eae Sac I I BsiE HaiA I I Msp I Hae III Hpa II Mcr I ScrF T Gdi II Nci I Eag I Sau96 Eae I Nla III I Hae III Msp I Bspl286 Sph I I Sau96 I Aha II C3D6 I NspC I Ban II Nla Aba I IV Nla Mae II Msp I IV Nsp7524 I Bspl286 I Alu I Aat II Hpa Nsp I II B3P12Q Nhe I Rina I T Ban I Hpa II Alu I Ban Ssp I Rma I II Asp7iq BsmA I BsiE I Eco57 T I II Ben T I Mill Sph I BsmA I NspC I Mae II Nsp7524 Aha II HinP T Aat II Nhe I Nla III Msp I Alu I Nsp I Hpa II Rma I Hha__I Hae III xba I Rma I II I II M Ml I II GGCCGGACGTCTCTAGAGCTAGCGCATO 109 CCGGCCTGCAGAGATCTCGATCGO I I* II llh I I 81 92 99 83 93 102 83 97 104 86 98 105 86 102 87 104 89 104 104 169 APPENDIX E DNA SEQUENCE OF METRS NHE I CASSETTE HgiA I Hae III Gdi II Ea3 I BstU I Pnu4H Msp I I Bspl286 I Hae III I I II -I II .im'^'^V^^^'i^^^AAAA^ III I II I M 1 8 1R oc i-i IJ'' llll • • .1 2 11 19 .1° 64 ^2 30 41 49 13 22 33 42 53 4 22 33 42 53 13 25 33 42 55 18 25 43 1 20 25 25 Mbo II Mae II S^^ Pie I Eae I BstU I Mae III Rsa _ i ^ ^ I MOJ^ I Hph I FsD I 106 ' • i27 lis 'ill I.J I II 127 136 152 ^J^9 III 139 155 193 159 193 159 194 196 Hae III OCT Gdi II ^^^^ ^ Pnu4H I EcoR II Mae III BstU I HgiA I BstN I Sau96 I HinP I e t , 300 A 203' I • lllll^ .1 In' 11; I43 257' ].. -l.^ I U ' 273 281 291 206 229 298 247 260 273 286 206 214 229 298 247 286 298 214 234 248 214 249 250 251 Msp I Hpa II Fnu4H I ScrF I Bbv I Nci I Dde I Ben I Esp I Sec I Pie I Nla III Alu I Alu I Hph I Hinf I SfaK I Mse I Alu I I III I I I III II II CATGCTGAAAGCTCAGCAGCTTOGTATCACCCCGGAGC^ 4OO GTACGACrrrcGAGTCGTCGAACCATAGTGGCXXX^ • . 1 . . . I III . 1^, HIM- I • I I 301 310 318 327 355 364 387 395 311 330 355 312 331 316 331 316 331 332 332 Hga I Mbo II Tag I Ear I Bsr I Mse I 11 I I I GACAACTATCACTCGACGCACAGCGAAGAGAACCGCOUTIT^^ 500 CTGTTGATAGTGAGCTGCGTGTCGCTTCTCTTGGC^^ • . . . • • • . I I . I I I 413 425 436 483 415 425 Continued next page 170 " APPENDIX E Continued Sau3A Mbo I MsP I - Dpn I BaaM II Alw I ^ Sau96 I Map I Alu IT Mbo I Hpa II £iaL-U Dpn I ma III Rsr tt „ . ScrP I N<:r>R TT ft 1 ... t * ^ EsnM^ T .... ... " |03 512 525 536 «f 576 512 526 537 577 504 512 '^38 557 577 512 S77 514 578 515 578 580 581 581 BstU I Taq I 581 Hha I "^"P ^ "SP I Hha I Fnu4H I nnn T Hpa II "'^lO III II I Hinf I Mbo II • ' 609 Li I i II • . . 662 I I 612 II] 670 ^83 ^6 612 671 613 "639 NspB II Fnu4H I HgiE II c- -.X -r Sau96 I ' Nla IV T ^ ^ ^ Ava II Hae III ' r r" "'-< ' rrf ff" r- 800 • • s: • !" • I,. 704 738 799 738 800 740 744 745 ScrF I Nci I Msp I Hpa II Ben I I HinP I Aha II J EcoR V BstU I D-fcti T BstU I SsE_l Mae II Hga I [ I I I I- 815 824' L ' I 59 870 879 890 827 860 827 860 863 663 863 863 863 Fnu4H I BstU I Taq I Pie I HinP I Alu I Mbo II Rsa I Nla III Hinf I Hha 1 HinP III Bsr I Hint I Alu I • • • • lo7-- I22 • ill- iL ' ies ' L ' 937 953 970 994 938 957 977 939 Hae III Ftiu4H I Nla III Hae I Nla III Bbv I htne I Tthlll II TTCATCCXTTAAAGATATT^^ I^qq AAGTAGCCATTTCTATAACAAATGAAGGTCn03GACAAGACCGGA^^ . . . I- |. I III- I 1040 1048 1059 1079 1086 1041 1059 1094 Continued next page 171 APPENDIX E Continued Nla IV ScrF I Ban I EcoR II Fnu4H I BstN I I Hph HinP I BstU I pflM I Mae III Hha I BsmA^I Mse I AlwN I Hinf I Hga I Mae III ^ TACACTGCCACTTCCCGCGTrrCTACA^ ^^00 I • I • ' ' • • • • - I I I I I I .1 1103 ' 1116 1132 1149 1156 1168 1178 1194 1108 1116 1136 1155 1137 1161 1139 1161 1139 1161 Sau3A ScrF I HinP I Mbo I Hha I Dpn I Mbo II Mse I BstN I BstU I I Taq ScrF I Hpa I Mse I Mbo II Cla I EcoR 11 HinC II Hpa I Ear I EcoR V BstN I asm I EcoR V HinC II. II II I II I I I I I II III CTACACTGCGAAACTCTCrraXGCATTGATGATATCGATC^^ I300 GATGTGACGCTrTGAGAGAAGCXX:GTAACTACTATACXrrAGAGTTGGACCTr^ II • -11 I III • • • II I I II -III 1216 1232 1246 1270 1277 1295 1217 1235 1246 1283 1295 1221 1236 1246 1283 1296 1222 1238 1250 1284 1300 1222 1238 1300 1238 1300 NspB II Haeu illTT^ Fnu4H I ™® 1 Bsr I Rsa I SfaN I II ' I I I II CTGGCCTCCCGTAATGCGGGCTTTATCAACAAGCGTTTTGA 14 00 GACCGGAGGGCATTACGCCCGAAATAGTTGTTCCXAAAACTtXCGCACGACCff • • • . . . . . III- I I I I I 1302 1360 1378 1394 1303 1397 1305 1398 HinP I BstU I Hha I Sau3A I HinP I Hae II Mbo I Eco57 I Hph I Hha I Nla III Dpn I Taq I II I II II I i CTGAAGTGATTGGTGAAGCGTGGGAAAGCCGTGAATTTGGTAAAGCCXTI^ 1500 GACTTCACTAACCACTTCGCACCCTrTCGGCACTTAAACCATTI^^ • • • I I • • • M I- II I -I 1401 1412 1450 1459 1472 1491 1450 1462 1472 1451 1463 1472 1463 SfaN I Sec I BstU I Dsa I Fnu4H I Pst I Nla IV Hae III BspM I SfaN I BstU I Eco57 I II II I I M I I I GGCTCCGTGGGTGGTGGCGAAACAGGAAGGCCGCGATGCOSACCT^ 1600 CCGAGGCACCCACCACCGCTTTGTCCTICCGGCGCTACGGCTGGACGTCCXOT • • • • I I • I • I II- II I I I 1501 1529 1542 1566 1580 1598 1505 1530 1544 1505 1532 1535 Fok I Sec I Rsa I ScrF I Sau96 I Msp I EcoR II Fnu4H I SdO-I Hpa II Mnl I BstN I Bbv I Dra III CfrlO I Bsm I HgiE II NspB II Hae III Hinf I I I II I I III II II I AAGCCGGTACTGCCGAAACny^CGAGCCTKXy^GAAGCATT^^ 1700^^^^ TTCGGCCATGACGGCTITGACTGGCTCGCACGTtriTCOT • • • • II I- III- 1-1 II- I l-l -I 1603 1637 1658 1679 1688 1699 1604 1642 1659 1681 1691 1604 1659 1681 1699 1607 1659 1687 1659 1663 Continued next page 172 APPENDIX E Continued HinP I BstU I HinP I Taq I Fnu4H I Msp I I BSFM I Bbv I Hpa II HaeJ^rlII .S^^ sfaN I Mnl I ^^^^ ^^^^0 gsr I wm t I 1800 ^^25 * ' 1735 -il I I II 1. I75O II I 1712 1727 1738 ^''81 1790 1799 1712 1728 1782 1792 1784 1793 1784 1793 1787 1796 1789 Hinf I 1790 Sau3A I HinP I NspB II Mbo I HinP I Sau96 I Dpn I Hha I Hha I Hae III ^ II Alw I HDh T . ^ HinC II II BstU I II II. 'HTG'rr 1900 77 w A i^oTCTCAAACAA I I I r * I I II' 1859 1868 1863 1873 1867 1874 1822 1867 1874 Sau3A I Mbo I Dpn I Alw I Nla IV BstY I BamH T I Sau3A I HinP I ^ Mbo I T • • 1918 I92; ' * * I 934 943 ^^8 ' * \UJ 1918 1933 1994 ^lllg 1^84 1921 1 Qi-i ^^^5 ''''1921 '^'^ 1985 lilt 1985 1987 1987 1987 1987 1988 1988 1988 1988 ScrF I Sau96 I Ava II NlaMl IV Msp I Scr^^"'^ CfrlO I ^ ^^'^ "^"^ TTT^*"^ ^ Nla III F^u4H I EcoR II Hha I ^"j' ^ fiSJlA^ Eco57 I Hph I Bbv I BstN I ^ I ACTAACCAGCAGTGTGGTAATACCACay^TTGGACCGT^^ * ' ' * I I II *| I -I I -I II I I 2010 2032 2040 2050 2064 2074 2081 2090 2016 2032 2041 2050 2071 2081 2090 2032 2084 2035 2085 2035 2085 2087 2087 2090 Continued next page 173 APPENDIX E Continued SfaN I Msp I ScrF I Msp I scrF I SfaN I Nci I Fnu4H Hpa II I I Ben I Biyv CfrlO I Msd I 1 Dde I Fok I Hoa IT ^^^^ I Esp I Hpa II Ben T u u T ^ ^ ILJAJA.JA T T ' ^aq 2200 II • . I I "^YY'^'^'r'""^'^^'^'^*^^^'^^CGAAATA'IT • 91 7Q A, I I I I I 2122 2119 2126 2119 2127 2119 2127 2120 2123 ' Bsr ' ' 1^ I SfaN l^^ Bsr I p^^^ 2290 Nla IV •I 2391 2391 Nla IV Knn T Ban I Asp718 Msp I Hpa II ScrF I Nci I Msp I Ben I Hae III Sau96 I Gdi II Sau96 I Eag I Nla IV Eae I ^ Bspl286 I Msp Fnu4H T 3^ I Mae Alpl ' ^ Y^^' ?g?^^TCtI^^^ 2500 AGC 2503 TCG 174 APPENDIX F DNA AND AMINO ACID SEQUENCE OF METRS 1/1 61/21 31/11 91/31 ATG ACT CAA GTC GCG AAG AAA ATT CTG GTG ACG TGC GCA CTG CCG GGC TCA ATC CAC CTC GGC TAC GCT AAC CAT ATG CTG GAG CAC ATC CAG CGT TAG GAG CGA GCT GAT GTC TGG GTC Met thr gin val ala lys lys ile leu val thr cys ala leu pro gly ser ile his leu gly tyr ala asn his met leu glu his ile gin arg tyr gin arg ala asp val trp val 121/41 151/51 181/61 211/71 ATG GGC GGC CAC GAG GTC AAC TTC ATC TGC GCC GAC GAT GCC CAC GGT ACA ATC ATG CTG AAA GCT CAG CAG CCG CTT GGT ATC ACC CCG GAG CAG ATG ATG AGT CAG GAG ATT GGC GAA met arg gly his glu val asn phe ile cys ala asp asp ala his gly thr pro ile met leu lys ala gin gin leu gly ile thr pro glu gin met met ser gin glu ile gly glu 241/81 271/91 301/101 331/111 CAT CAG ACT GAT TTC GCA GGC TTT AAC ATC AGC TAT GAC AAC TAT CAC TCG ACG CAC AGC GAA GAG AAC GGC CAG TTG TCA GAA CTT ATC TAC TCT CGC CTG AAA GAA AAC GGT TTT ATT his gin thr asp phe ala gly phe asn ile ser tyr asp asn tyr his ser thr his ser glu glu asn arg gin leu ser glu leu ile tyr ser arg leu lys glu asn gly phe ile 361/121 391/131 421/141 451/151 AAA AAC CGC ACC ATC TCT CAG CTG TAC GAT CCG GAA AAA GGC ATG TTC CTG CCG GAC CGT TTT GTG AAA GGC ACC TGC CCG AAA TGT AAA TCC CCG GAT CAA TAC GGC GAT AAC TGC GAA lys asn arg thr ile ser gin leu tyr asp pro glu lys gly met phe leu pro asp arg phe val lys gly thr cys pro lys cys lys ser pro asp gin tyr gly asp asn cys glu Continued next page 175 APPENDIX F Continued 481/161 541/181 511/171 571/191 GTC TGC GGC GCG ACC TAG AGO CCG ACT GAA CTG ATC GAG CCG TCT GGC GOT ACG CCG GTA AAA TCG GTG GTT ATG CGT GAT TCT GAA CAC TTC TCT TTC AGO GAA TTC TTT GAT CTG CCC val cys gly ala thr tyr ser pro thr glu leu ile glu pro lys ser gly ala thr pro val ser val val met arg asp ser glu his phe ser phe ser glu phe phe asp leu pro 601/201 631/211 661/221 691/231 ATG TTG CAG GCA TGG ACC CGC AGC GGT GCG TTG CAG GAG CAG GTG ATG CAG GAG TGG TTT GAA GCA AAT AAA TCT GGC CTG CAA CAG TGG GAT ATC CCT TAG TTC GGT TCC CGC GAC GCC met leu gin ala trp thr arg ser gly ala leu gin glu gin val ala asn met gin glu trp phe glu lys ser gly leu gin gin trp asp ile pro tyr phe gly ser arg asp ala ^^I'^^l 751/251 TTT GAA ATT CCG AAC GCG CCG GGC AAA TAT TTC TAC GTC TGG CTG GAC CCA rrr »^ AGC ??C Ga5 GM AGC Sa phe glu lie pro asn ala pro gly lys tyr phe tyr val trp leu asp ala pro 71 " sil III Z ' 841/281 871/291 901/301 931/311 TAC TGG AAG AAA GAC TCC ACC GCC GAG CTG TAC CAC TTC ATC GGT AAA GAT ATT GTT TAC TTC CAC AGC CTG TTC TGG CCT GCC ATG CTG GAA GGC AGC AAC TTC CGC AAG CCG TCC AAC tyr trp lys lys asp ser thr ala glu leu tyr his phe ile gly lys asp ile val tyr phe his ser leu phe trp pro ala met leu glu gly ser asn phe arg lys pro ser asn 961/321 991/331 1021/341 1051/351 CTG TTT GTT CAT GGC TAT GTG ACG GTG AAC GGC GCA AAG ATG TCC AAG TCT CGC GGC ACC TTT ATT AAA GCC AGC ACC TGG CTG AAT CAT TTT GAC GCA GAC AGC CTG CGT TAC TAC TAC leu phe val his gly tyr val thr val asn gly ala lys met ser lys ser arg gly thr phe ile lys ala ser thr trp leu asn his phe asp ala asp ser leu arg tyr tyr tyr Continued next page 176 APPENDIX F Continued 1081/361 1141/381 1111/371 o5? - - «c III III lit Ill ^^cc 0„ ... AAT GCG GGC TTT ^^T AAC CTG GCC TCC CGT thr ala lys leu ser ser arg ile aso ;iqn = val gin arg val asn ala P^e asp ill val ^ ^ asn ala gly phe leu ala ser arg 1201/401 1261/421 1231/411 -C'^-'CTG GOT G.C CCG ?AC Z IS tf, C.G TTG GAA TTT GGT AAA ^""^ ^CG TGG GAA AGC CGT ^- ti: iTs li: iTr ::i i\i m it glu phe gly Tys ^rp glu ser arg 1321/441 1351/451 1381/461 1411/471 GCC GTG CGC GAA ATC ATG GCG CTG GCT GAT CTG GCT AAC CGC TAT GTC GAT CAG GCT CCG TGG GTG GTG GAA GCG AAA CAG GAA GGC CGC GAT GCC ATT TGC TCA ATG GAC CTG CAG GCA ala val arg glu ile met ala leu ala asp leu ala asn arg tyr val asp glu gin ala pro trp val val ala lys gin glu gly arg asp ala asp ile cys ser met leu gin ala 1441/481 1471/491 1501/501 1531/511 GGC ATC AAC CTG TTC CGC GTG CTG ATG ACT TAC CTG AAG CCG GTA CTG CCG AAA CTG ACC GAG CGT GCA GAA GCA TTC CTC AAT ACG GAA CTG ACC TGG GAT GGT ATC CAG CAA CCG CTG gly ile asn leu phe arg val leu met thr tyr leu lys pro val leu pro lys leu thr glu arg ala glu ala phe leu asn thr glu leu thr trp asp gly ile gin gin pro leu 1561/521 1591/531 1621/541 CTG GGC CAC AAA GTG AAT CCG TTC AAG GCG CTG TAT AAC CGC ATC GAT ATG AGG CAG GTT GAA GCA CTG GTG GAA GCC TCT AAA TGA leu gly his lys val asn pro phe lys ala leu tyr asn arg ile asp met arg gin val glu ala leu val glu ala ser lys OPA 177 BIBLIOGRAPHY ^ 1997, J6, 3084-3094. Amatore, C. and Jutand, A. J. Organomet. Chem. 1999, 576, 255-277. Apostol, I., Levine, J., Lippincott, J., Leach, J., Hess, E., Glascock, C.B., Weickert M J and Blackmore, R.J. J. Biol. Chem. 1997, 272, 28980-28988. Arnez, J.G. and Steitz, T.A. Biochemistry 1996, 35, 14725-14733 Arnez, J.G., Augustine, J.G., Moras, D. and Francklyn, C.S. Proc. Natl. Acad. Sci USA 1997,9^7144-7149. Arnez, J.G., Dock-Bregeon, A.C. and Moras, D. /. Mol. Biol. 1999, 286, 1449-1459. Arnez, J.G., Harris, D.C., Mitschler, A., Rees, B., Francklyn, C.S. and Moras, D EMBO J. 1995, 74,4143-4155. Arnold, F.H. and Moore, J.C. Adv. Biochem. Eng. Biotech. 1997, 58, 1-14. Arnold, F.H. Chem. Eng. Sci. 1996, 57, 5091-5102. Arnold, L.D., May, R.G. and Vederas, J.C. J. Am. Chem. Soc. 1988, 110, 2237-2241. Badger, J. et al. Proc. Natl. Acad. Sci. USA 1988, 85, 3304-3308. Bain, J.D., Glabe, C.G., Dix, T.A. and Chamberlin, A.R. J. Am. Chem. Soc. 1989, ///, 8013-8014. Barker, D.G., Ebel, J.-P., Jakes, R. and Bruton, C.J. Eur. J. Biochem. 1982, 727, 449-457 Belrhali, H. et al. Science 1994, 263, 1432-1436. Biou, v., Yaremchuk, A., Tukalo, M. and Cusack, S. Science 1994, 263, 1404-1410. 78 Blackwell, H.E. and Grubbs, R.H. Angew. Chem. Int. Ed. Engl. 1998, 37, 3281-3284. Blanquet, S., Fayat, G. and Waller, J.-P. Eur. J. Biochem. 1974, 44, 343-351. Bogosian, G., Violand, B.N., Dorward-Kmg, E.J., Workman, W.E., Jung, P.E. and Kane J.F. J. Biol. Chem. 1989, 264, 531-539. Briand, C, Poterszman, A., Eiler, S., Webster, G., Thierry, J.C. and Moras, D J Mol Biol. 2000,299, 1051-1060. Brick, P. and Blow, D.M. J. Mol. Biol. 1987, 194, 287-297. Brick, P., Bhat, T.N. and Blow, D.M. J. Mol. Biol. 1989, 208, 83-98. Brintzinger, H.H., Fischer, D., Mulhaupt, R., Rieger, B. and Waymouth, R.M. Angew.Chem. Int. Ed. Engl. 1995, 34, 1 143-1 170. Brunie, S., Zelwer, C. and Risler, J.-L. J. Mol. Biol 1990, 2/6, 41 1-424. Budisa, N., Steipe, B., Demange, P., Eckerskorn, C, Kellermann, J. and Huber, R. Eur. J. Biochem. 1995, 230, 788-796. Bullock, T.L., Sherlin, L.D. and Perona, J.J. Nature Struct. Biol. 2000, 7, 497-504. Cassio, D. and Waller, J.-P. Eur. J. Biochem. 1971, 20, 283-300. Cavarelli, J., Delagoutte, B., Eriani, G., Gangloff, J. and Moras, D. EMBO J. 1998, 17, 5438-5448. Chiefari, J. et al. Macromolecules 1998, 31, 5559-5562. Clark, T.D., Kobayashi, K. and Ghadiri, M.R. Chem. Eur. J. 1999, 5, 782-792. Cornish, V.W., Mendel, D. and Schultz, P. Angew. Chem. Int. Ed. Engl. 1995, 34, 621- 633. Corradini, P. Macromol. Symp. 1995, 59, 1-1 1. 179 Coulter, A.W. and Talalay, P. J. Biol. Chem. 1968, 243, 3238-3247. Cowie, D.B. and Cohen, G.N. Biochim. Biophys. Acta 1957, 26, 252-261, Cowie, D.B Cohen, G.N., Bolton, E.T. and Robichon-Szulma.ster, H.D. Bioc,u„ Biophys. Acta 1959, 34, 39-46. Creel, H.S., Fournier, M.J., Mason, T.L. and Tirrell, D. A. Macromolecules 1991 24 1213-1214. ' Cusack, S., Hartlein, M. and Leberman, R. Nucleic Acids Res. 1991, 79, 3489-3498. Dardel, F., Fayat, G. and Blanquet, S. J. Bacterial. 1984, 160, 1115-1 122. Delarue, M., Poterszman, A., Nikonov, S., Garber, M., Moras, D. and Thierry J C EMBOJ. 1994,75,3219-3229. Deming, T.J., Fournier, M.J., Mason, T.L. and T.rrell, D.A. J. Macromol Sci. Pure Annl Chem. 1997, AJ< 2143-2150. Desogus, G., Todone, F., Brick, P. and Onesti, S. Biochemistry 2000, 39, 8418-8425. Dias, E.L., SonBinh, T.N. and Grubbs, R.H. /. Am. Chem. Soc. 1997, 779, 3887-3897. Dougherty, M.J., Kothakota, S., Mason, T.L., Tirrell, D.A. and Fournier, M.J. Macromolecules 1993, 26, 1779-1781. Duewel, H., Daub, E., Robinson, V. and Honek, J.F. Biochemistry 1997, 36, 3404-3416. Edelmann, P. and Gallant, J. A. Cell 1977, 10, 131-137. Filer, S., Dock-Bregeon, A.C., Moulinier, L., Thierry, J.C. and Moras, D. EMBO J. 1999, 18, 6532-6541. Eriani, G., Delarue, M., Poch, O., Gangloff, J. and Moras, D. Nature 1990, 347, 203-206. 180 Fasiolo, F., Gibson, B.W., Walter, P., Chatton, B., Biemann, K. and Boulanger f Y J• , C/zem. 1985,260, 15571-15576. • Faust, R. and Kennedy, J.P. Polym. Bull. 1986, 75, 317-323. Fersht, A. Enzyme Structure and Mechanism; W. H. Freeman and Company New York 1999. Fourmy, D., Dardel, F. and Blanquet, S. J. Mol. Biol. 1993, 231, 1078-1 089, Fourmy, D., Mechulam, Y., Brunie, S., Blanquet, S. and Fayat, G. FEBS Lett 1991 292 259-263. Fourmy, D., Meinnel, T., Mechulam, Y. and Blanquet, S. J. Mol. Biol. 1993, 231 1068- 1077. Fukai, S., Nureki, O., Sekine, S.-I., Shimada, A., Tao, J., Vassylyev, D.G. and Yokoyama, S. Cell 2000, 103, 793-803. Furstner, A., Guth, O., Rumbo, A. and Seidel, G. J. Am. Chem. Soc. 1999 108- , 727, II 11113. Furter, R. Protein Sci. 1998, 7, 419-426. Gabius, H.J., von der Haar, F. and Cramer, F. Biochemistry 1983, 22, 2331-2339. Gallivan, J.P., Lester, H.A. and Dougherty, D.A. Chem. Biol. 1997, 4, 739-349. Gao, W., Goldman, E. and Jakubowski, H. Biochemistry 1994, 33, 1 1528-1 1535. Ghosh, G., Brunie, S. and Schulman, L.H. J. Biol. Chem. 1991, 266, 17136-17141. Ghosh, G., Kim, H.Y., Demaret, J. -P., Brunie, S. and Schulman, L.H. Biochemistry 1991, 30, 1 1767-1 1774. Ghosh, G., Pelka, H. and Schulman, L.D. Biochemistry 1990, 29, 2220-2225. Ghosh, G., Pelka, H., Schulman, L.H. and Brunie, S. Biochemistry 1991, 30, 9569-9575. 181 . 1 Hech,, S^M^ Airord, B.L., Kuroda, Y. and Ki,a„„, S. / Biol. CHem. „78, 25., 45,7. Hendrickson, W.A. / Synchroimn Rud 1999, 6, 845-85 Hendrickson, W.A.. Horlon, J.R. and LeMas(er, D.M. UMBO J. 1990, 9, 1665-1672. ^'--'^ m;':;8"7:s°2r' ' " ^- - Hortin, G. and Boimc, I. Methods. Enzymol. 1983, 96, 777-784. Hountondji, C, Blanquet, S. and Lcdcrer, F. Biochemistry 1985, 24^ 1 175-1 180. Hounlondji, C, Schmitler, J.M., Fukui, T., Tagaya, M. and Blanquet, S. Biochemistry 1990,29, II 266- II 273. Ibba, M. and Henncckc, H. FEBS Lett. 1995, 364, m-TTS. Ibba, M. and Soil, D. Science 1999, 2(^(5, 1893-1897. Ibba, M., Kast, P. and Hennecke, H. Biochemistry 1994, ii, 7107-71 12. Ilyin, V.A., Temple, B., IIu, M., Li, G.P., Yin, Y.H., Vachctle, P. and Carter, C.W. Protein Sci. 2000, 9, 2 1 8-23 1 Ingraham, J.L., Maaloe, O. and Neidhardt, F.C. Growth of the Bacterial Cell; Sinaucr Associates Inc.; Boston, MA, 1983. Inouyc, S. and Tsuji, F.I. FEBS Lett. 1994, 341, 277-280. Jakubowski, H. and Fersht, A.R. Nucleic Acids Res. 1981, 9, 3105-31 17. Jakubowski, II. and Goldman, E. FEBS Lett. 1993, 3J7, 237-240. 182 Jakubowski, H. and Goldman, E. Microbiol. Rev. 1992, 56, 412-429. Jakubowski, H. Biochemistry 1996, 35, 8252-8259 Jakubowski, H. EMBO J. 1991, 10, 593-598. Jakubowski, H. J. Biol. Chem. 2000, 275, 21813-21816. Jakubowski, H. Proc. Natl. Acad. Sci. USA 1990, 87, 4504-4508. Jo, Y.J., Lee, S.W., Jo, M.K., Lee, J.W., Kang, M.-K., Yoon, J.H. and Kim, S. J. Biochem. Mol. Biol. 1999, 32, 547-553. Kast, P. and Hennecke, H. J. Mol. Biol. 1991, 222, 99-124. Kiick, K.L. and Tirrell, D.A. Tetrahedron 2000, 56, 9487-9493. Kiick, K.L., van Hest, J.C.M. and Tirrell, D.A. Angew. Chem. Int. Ed. Engl. 2000 39 2148-2152. Kim, H.Y., Ghosh, G., Schulman, L.H., Brunie, S. and Jakubowski, H. Proc. Natl. Acad Sci. USA 1993, 90, 11553-11557. Kim, H.Y., Pelka, H., Brunie, S. and Schulman, L.H. Biochemistry 1993, 32, 10506- 10511. Koide, H. et al. Biochemistry 1994, 33, 7470-7476. Koide, H. et al. Proc. Natl. Acad. Sci. USA 1988, 85, 6237-6241. Kothakota, S., Dougherty, M.J., Foumier, M.J., Mason, T.L., Yoshikawa, E. and Tirrell, D.A. Macromol. Symp. 1995, 98, 573-583. Kothakota, S., Mason, T.L., Tirrell, D.A. and Fournier, M.J. J. Am. Chem. Soc. 1995, 117, 536-537. 183 Kowal, A.K., Kohrer, C. and RajBhandary, U.L. Proc. Natl. Acad. Sc. 2001, 98, 2268- Krejch. M.T., Atkins, E.D.T., Waddon, A.J., Fournier, M.J., Mason, T.L. and Tirrell D.A. Science 1994, 265, 1427-1432. Kupce, E., Boyd, J. and Campbell, I.D. J. Magn. Reson. Ser. B. 1995, 106, 300-303. Lage, H. and Dietel, M. Gene 1996, 178, 187-189. Lau, E.Y. and Gerig, J.T. Biophys. J. 1997, 73, 1579-1592. Li, S., Kumar, N.V., Varshney, U. and RajBhandary, U.L. J. Biol. Chem 1996 277,1022-1028. Lim, K.-T., Brunett, S., lotov, M., McClurg, R.B., Vaidehi, N., Dasgupta, S., Taylor, S. and Goddard III, W.A. / Comput. Chem. 1997, 18, 501-521. Liu, D.R. and Schultz, P.G. Proc. Natl. Acad. Sci. USA 1999, 96, 4780-4785. Logan, D.T., Mazauric, M.H., Kern, D. and Moras, D. EMBO J. 1995, 14, 4156-4167. Mahal, L.K. and Bertozzi, C.R. Chem. Biol. 1997, 4, 415-422. Mahal, L.K., Yarema, K.J. and Bertozzi, C.R. Science 1997, 276, 1 125-1 128. Maier, K.L., Lenz, A.G., Beck-Speier, I. and Costabel, U. Methods Enzymol. 1995, 257,455-461. Mangold, J.B., Mischke, M.R. and Lavelle, J.M. Mut. Res. 1989, 276, 27-33. Mayaux, J.-F. and Blanquet, S. Biochemistry 1981, 20, 4647-4654. Mayaux, J.F., Kalogerakos, T., Brito, K.K. and Blanquet, S. Eur. J. Biochem. 1982, 728, 41-46. McGrath, K.P., Fournier, M.J., Mason, T.L. and Tirrell, D.A. J. Am. Chem. Soc. 1992, 77^, 727-733. 184 1 Mechulam, Y. et al. Nucleic Acids Res. 1991, 19, 3673-368 /, 4o5-475. ' Mechulam, Y., Schmitt, E., Maveyraud, L., Zelwer, C, Nureki, O., Yokoyama S Konno, M. and Blanquet, S. / Mol. Biol. 1999, 294, 1287-1297. Meinnel, T. et al. Biochimie 1990, 72, 625-632. Meinnel, T., Mechulam, Y. and Blanquet, S. tRNA: Structure, Bwsynthesis, and Function; ASM Press: Washington, D. C, 1995. Meinnel, T., Mechulam, Y., LeCorre, D., Panvert, M., Blanquet, S. and Fayat, G Proc Natl. Acad. Sci. USA 1991, 88, 291-295. Mellot, P., Mechulam, Y., LeCorre, D., Blanquet, S. and Fayat, G. J. Mol Bwl 1989 208, 429-443. Mendel, D., Ellman, J. A., Chang, Z., Veenstra, D.L., Kollman, P. A. and Schullz, P.G. Science 1992, 256, 1 798- 1 802. Moras, D. Trends Biochem. Sci. 1992, 17, 159-164. Mosyak, L., Reshctnikova, L., Goldgur, Y., Delarue, M. and Safro, M.G. Nature Struct Biol. 1995, 2, 537-547. Ni-NTA Spin Handbook; Qiagen: Valencia, CA, 1998. Noren, C.J., Anthony-Cahill, S.J., Griffith, M.C. and Schultz, P.G. Science 1989, 2^^182-188. Nowak, M.W. et al. Science 1995, 268, 439-442. Nureki, O. et al. Science 1995, 267, 1958-1965. 185 . Nureki, O., Murarnatsu, T., Suzuki, K., Kohda, D., Matsuzawa, H., Ohta, T., M.yazawa T. and Yokoyama, S. J. Biol. Chem. 1991, 266, 3268-3277. Ohno, S Yokogawa, T., Fujii, I., Asahara, H., Inokuch., H. and Nishikawa K " J Biochem. 1998, 124, 1065-1068. ' Onesti, S., Desogus, G., Brevet, A., Chen, J., Plateau, P., Blanquet, S. and Brick P Biochemistry 2000, 1 39, 2853- 1 286 1 Onesti, S., Miller, A.D. and Brick, P. Structure 1995, 3, 163-176. Parkhe, A.D., Fournier, M.J., Mason, T.L. and Tirrell, D.A. Macromolecules 1993 26f24j, 6691-6693. Pastrnak, M., Magliery, T.J. and Schultz, P.G. Helv. Chim. Acta 2000, 83, 2277-2286. Perona, J.J., Rould, M.A. and Steitz, T.A. Biochemistry 1993, 32, 8758-8771. Petka, W.A., Hardin, J.L., McGrath, K.P., Wirtz, D. and Tirrell, D.A. Science 1998 281 389-392. Piperno, J.R. and Oxender, D.L. J. Biol. Chem. 1968, 243, 5914-5920. Posorske, L.H., Cohn, M., Yanagisawa, N. and Auld, D.S. Biochim. Biophys. Acta 1979 576, 128-133. Poterszman, A., Delarue, M., Thierry, J.C. and Moras, D. /. Mol. Biol. 1994, 244, 158- 167. Qiaprep® Miniprep Handbook; Qiagen: Valencia, CA, 1999. QIAquick® Spin Handbook; Qiagen: Valencia, CA, 2000. Qiu, X.Y., Janson, C.A., Blackburn, M.N., Chhohan, I.K., Hibbs, M. and Abdel-Mcguid, S.S. Biochemistry 1999, 38, 12296-12304. 186 Raunio, R. and Rosenqvist, H. Acta Chem. Scand. 1970, 24, 2737 2744. Rees, B Webster, G., Delarue, M., Boeglin, M. and Moras, D. J. Mol. B.ol. 2000, 299 1 164. Reshetnikova, L., Moor, N., Lavrik, O. and Vassylyev, D.G. / Mol. Biol. 1999 287 555. 568. Richmond, M.H. Bacteriol Rev. 1962, 26, 398-420. Richmond, M.H. J. Mol. Biol. 1963, 6, 284-294. Rossmann, M.G., Moras, D. and Olsen, K.W. Nature 191A, 250, 194-199. Rould, M.A., Perona, J.J. and Steitz, T.A. Nature 1991, 352, 213-218. Rould, M.A., Perona, J.J., Soil, D. and Steitz, T.A. Science 1989, 246, 1 135-1 142. Ruff, M. et al. Science 1991, 252, 1682-1689. Rutjes, F.P.J.T., Wolf, L.B. and Schoemaker, H.E. J. Chem. Soc, Perkin Trans. 1 2000 4197-4212. Saks, M.E., Sampson, J.R., Nowak, M.W., Kearney, P.C., Du, F., Abelson, J.N., Lester, H.A. and Dougherty, D.A. J. Biol. Chem. 1996, 277, 23169-23175. Sambrook, J., Fritsch, E.F. and Maniatis, T. Molecular Cloning: A Laboratory Manual; Cold Spring Harbor Laboratory Press: Cold Spring Harbor, NY, 1989. Sankaranarayanan, R. et al. Cell 1999, 97, 371-381. 187 ' Sankaranarayanan, R Dock-Bregeon, A.C., Rees. B., Bovee, M., Caille,, J., Ron,by P Francklyn. C.S. and Moras, D. Nmure Struc, Biol. 2000. 7, 46 1 -465. Sauter. C.^^E., Cavarelli. ,.. Moras, D. and Giege, R. / Mol. Bio,. 2000, 299, Saxon, E. and Bertozzi, C.R. Science 2000, 287, 2007-2010. Schenck, H.L., Dado, G.P. and Gellman, S.H. J. Am. Chem. Soc. 1996 118 12487 12494. ' ' Schmitt, E., Guillon, J.M., Meinnel, T., Mechulam, Y., Dardel, F. and Blanquet S Biochimie 1996, 78, 543-554. Schmitt, E., Moulinier, L., Fujiwara, S., Imanaka, T., Thierry, J.C. and Moras, D EMBO / 1998, 77,5227-5237. Schmitt, E., Panvert, M., Blanquet, S. and Mechulam, Y. EMBO J. 1998, 17, 6819-6826. Schmitt, E., Panvert, M., Mechulam, Y. and Blanquet, S. Eur. J. Biochem 1997 246 ' 539-547. ' Schoenberg, A. and Heck, R.F. J. Org. Chem. 1974, 39, 3327-3331. Schrock, R.R. Acc. Chem. Res. 1990, 23, 158-165. Serre, L., Verdon, G., Choinowski, T., Hervouet, N., Risler, J.-L. and Zelwer, C. J. Mol. Biol. 2001, 306, 863-876. Sharma, N., Furter, R., Kast, P. and Tirrell, D.A. FEBS Lett. 2000, 467, 37-40. Sherlin, L.D., Bullock, T.L., Newberry, K.J., Lipman, R.S.A., Hou, Y.M., Beijer, B., Sproat, B.S. and Perona, J.J. / Mol. Biol. 2000, 299, 431-446. Sherman, J.M., Rogers, M.J. and Soli, D. Nucleci Acids. Res. 1992, 20, 2847-2852. Silvian, L.F., Wang, J.M. and Steitz, T.A. Science 1999, 285, 1074-1077. 88 . Skinner, C.G., Edelson, J. and Shive. W. J. Am. Chen,. Soc. 1961, 83. 2281-2286. Steiner, T., S.ankov^E3., Amado. A.M. and Tei.eira-Dias, J.J.C. J. Chen,. Soc. Perk Trans. 2 1995, 7, 1321-1326. Sugiura, I. et al. Struct. Fold. Des. 2000, 8, 197-208. Swanson, R., Hoben P., , Sumner-Smith, M., Uemura, H., Watson, L. and Soli D Science 1988, 242, 1 548- 1 55 1 Szwarc, M. Nature 1956, 178, 1 168- 11 69. Tarn, R. and Saier, M.H. Microbiol. Rev. 1993, 57, 320-346 Tang, Y., Ghirlanda, G., Vaidehi, N., Kua, J., Mainz, D.T., Goddard III, W.A., DeGrado, W.F. and Tirrell, D.A. Biochemistry 2001, 40, 2790-2796. The Qiaexpressionist; Qiagen: Valencia, CA, 1998. Trost, B.M. and Fleming, I. Comprehensive Organic Synthesis; Pergamon Press- Oxford 1991. Tsien, R.Y. Annu. Rev. Biochem. 1998, 67, 509-544. Tsuji, J. Palladium Reagents and Catalysts: Innovations in Organic Synthesis; John Wiley and Sons: New York, 1995. Turcatti, G., Nemeth, K., Edgerton, M.D., Knowles, J., Vogel, H. and Chollet, A. Receptors and Channels 1997, 5, 201-207. Tuve, T. and Williams, H. J. Am. Chem. Soc. 1957, 79, 5830-5831. Tzagoloff, A., Vambutas, A. and Akai, A. Eur. J. Biochem. 1989, 179, 365-371. Uhrin, D. and Barlow, P.N. J. Magn. Reson. 1997, 126, 248-255. Vaidehi, N. and Goddard III, W.A. J. Phys. Chem. A 2000, 104, 2375-2383. 89 Vaidehi, N., Jain, A. and Goddard III, W.A. / Phys. Chem. 1996, 100, 10508-10517. van Hest, J.C.M. and Tirrell, D.A. FEBS Lett. 1998, 428, 68-70. van Hest, J.C.M., Knck, K.L. and Tirrell, D.A. J. Am. Chan. Soc. 2000, 122, 1282-1288 Varshney, U. and RajBhandary, U.L. J. Bacteriol. 1992, 174, 7819-7826. Walter, P., Gangloff, J., Bonnet, J., Boulanger, Y., Ebei, J.-P. and Fasi<,lo, F Proc Natl Acad. Sci. USA 1983, 80, 2437-2441. Walter, P., Weygand-Durasevic, I., Sanni, A., Ebel, J.-P. and Fasiolo, F. ./ Biol Chem 1989,26^17126-17130. Wang, L., Magliery, T.J., Liu, D.R. and Schultz, P.O. / Am. Chem. Soc 2000 122 5010-5011. Wei, Y., Hendrickson, W.A., Crouch, R.J. and Satow, Y. Science 1990, 249, 1398-1405. Wilson, M.J. and Hatfield, D.L. Biochim. Biophys. Acta 1984, 781, 205-215. Yarema, K.J., Mahal, L.K., Bruehl, R.E., Rodriguez, B.C. and Bertozzi, C.R. J. Biol. Chem. 1998,275,31168-31179. Yoshikawa, E., Fournicr, M.J., Mason, T.L. and Tirrell, D.A. Macromolecules 1994, 27, 5471-5475. Yu, S.M., Conticello, V., Zhang, G., Kayser, C., Fournier, M.J., Mason, T.L. and Tirrell, D.A. Nature 1997, 389, 187-190. Yuan, T. and Vogel, H.J. Protein Sci. 1999, 1 13-121. 190 I