Linköping Studies in Science and Technology Dissertation No. 1184
A Novel Route for Construction of Multipurpose Receptors through Chemical Modification of Glutathione Transferases
Johan Viljanen
Molecular Biotechnology Department of Physics, Chemistry and Biology Linköping University, SE-581 83 Linköping, Sweden Linköping 2008
© 2008 Johan Viljanen ISBN: 978-91-7393-893-8 ISSN: 0345-7524 Printed in Sweden by LiU-Tryck Linköping 2008
This thesis describes how the human Alpha class glutathione transferase (GST) A1-1 can be reprogrammed either to function as a multipurpose biosensor for detection of small molecule analytes, or as a handle providing for more efficient protein purification. A novel, user-friendly, and efficient method for site-specific introduction of functional groups into the active site of hGST A1-1 is the platform for these achievements. The designed thioester reagents are glutathione-based and they are able to label one single nucleophile (Y9) and leave the other 50 nucleophiles (in hGST A1-1) intact. The modification reaction was tested with five classes of GSTs (Alpha, Mu, Pi, Theta and Omega) and was found to be specific for the Alpha class isoenzymes. The reaction was further refined to target a single lysine residue, K216 in the hGST A1-1 mutant A216K, providing a stable amide bond between the protein and the labeling group. To further improve the labeling process, biotinylated reagents that could deliver the acyl group to Y9 (wt hGST A1-1) or K216 in the lysine mutant, while attached to streptavidin-coated agarose beads, were designed and synthesized. A focused library of eleven A216K/M208X mutants was made via random mutagenesis to provide an array of proteins with altered micro-environments in the hydrophobic binding site, where M208 is situated. Through the invented route for site-specific labeling, a fluorescent probe (coumarin) was introduced on K216 in all double mutants, with the purpose of developing a protein-based biosensor, akin to the olfactory system. The array of coumarin-labeled proteins responded differently to the addition of different analytes, and the responses were analyzed through pattern recognition of the fluorescence signals. The labeled proteins could also be site-specifically immobilized on a PEG-based biosensor chip via the single C112 on the surface of the protein, enabling development of surface-based biosensing systems. Also, a refined system for efficient detection and purification of GST-fusion proteins is presented. Through a screening process involving A216K and all produced A216K/M208X mutants, two candidates (A216K and A216K/M208F) were singled out as scaffolds for the next generation of fusion proteins. In addition to the features present in commercially available GST fusion constructs, the new mutants can be sitespecifically labeled with a fluorophore in bacterial lysates providing quick and sensitive monitoring of expression and purification. Furthermore, the proteins could be labeled with a unique aldehyde moiety providing for a novel protein purification scheme.
iii
iv
This thesis is based on five peer-reviewed scientific papers, and will be referred to in the text by their roman numerals (Papers I-V). They are enclosed at the end of the thesis.
I
Programmed Delivery of Novel Functional Groups to the Alpha Class Glutathione Transferases. Sofia Håkansson, Johan Viljanen and Kerstin S. Broo Biochemistry, 2003, 42, 10260-10268.
II
Combinatorial Chemical Reengineering of the Alpha Class Glutathione Transferases. Johan Viljanen, Lotta Tegler and Kerstin S. Broo Bioconjugate Chemistry, 2004, 15, 718-727.
III
Surface-Assisted Delivery of Fluorescent Groups to hGST A1-1 and a Lysine Mutant. Johan Viljanen, Lotta Tegler, Jenny Larsson and Kerstin S. Broo Bioconjugate Chemistry, 2006, 17, 429-437.
IV
A Multipurpose Receptor Composed of Promiscuous Proteins. Analyte Detection through Pattern Recognition. Johan Viljanen, Jenny Larsson, Andréas Larsson and Kerstin S. Broo Bioconjugate Chemistry, 2007, 18, 1935-1945.
V
Orthogonal Protein Purification – Expanding the Repertoire of GST Fusion Systems. Johan Viljanen, Jenny Larsson and Kerstin S. Broo Protein Expression & Purification, 2008, 57, 17-26.
v
I
I participated to a minor part in the planning process, performed 50 % of the experimental work and did a minor part of the writing.
II
I participated in the planning process, planned and performed a major part of the experiments, analyzed the data together with Lotta Tegler and contributed to parts of the writing.
III
I planned the project together with my supervisor, planned and performed a major part of the experiments, analyzed most of the data and contributed to parts of the writing.
IV
The basic ideas were planned together with my supervisor. I planned and performed all experimental work, except the cloning experiments and the surface chemistry that were performed in collaboration with Jenny Larsson and Andréas Larsson, respectively. I analyzed most of the data and wrote a major part of the manuscript.
V
I participated in the planning process, planned and performed a major part of the experimental work, analyzed most of the data and wrote a major part of the manuscript.
VI A Promiscuous Glutathione Transferase Transformed into a Selective Thiolester Hydrolase. Sofia Hederos, Lotta Tegler, Jonas Carlsson, Bengt Persson, Johan Viljanen and Kerstin S. Broo Organic & Biomolecular Chemistry, 2006, 4, 90-97.
vi
Denna avhandling handlar om hur skräddarsydda proteiner kan användas som en ny multifunktionell biosensor för detektion av giftiga molekyler, men också som nya handtag för effektivare proteinrening. Proteiner är kroppens maskiner och förekommer i praktiskt taget alla cellens processer. Idag är det möjligt att genom genetiska eller kemiska metoder manipulera naturliga proteiner för att ändra deras funktion eller skapa nya funktioner. Att kunna omprogrammera ett protein till att utföra andra uppgifter än dess naturliga har öppnat upp dörrar till att använda proteiner i nya, icke-naturliga sammanhang. Inom bioteknikindustrin är proteiner, konstruerade av människohänder, användbara verktyg. Biosensorer, biokatalysatorer och biomarkörer är några exempel på användningsområden för designade proteiner. I avhandlingen presenteras två alternativa vägar, A) och B), varigenom nya funktioner kan introduceras kovalent i det naturliga enzymet glutationtransferas. Mestadelen av arbetet har baserats på humant glutationtransferas A1-1 (hGST A1-1), och mutanter av den. Glutationtransferaser är en familj avgiftningsenzymer som finns hos alla däggdjur. De katalyserar konjugeringen av tripeptiden glutation till en mängd olika hydrofoba elektrofila molekyler (ofta giftiga molekyler för organismen), vilket gör att konjugatet blir mer vattenlösligt och kan sedermera lätt exporteras ut från cellen. Eftersom de har utvecklats att fungera som avgiftningsenzymer så har de en promiskuös bindningsficka för hydrofoba elektrofiler, tillsammans med en närliggande och specifik bindningsficka för glutation. Detta gör att de kan binda en mängd olika molekyler i den promiskuösa bindningsfickan och utföra många olika typer av enzymatiska reaktioner. Denna egenskap är ovanlig hos enzymer, då de allra flesta endast kan binda en specifik molekyl och utföra en specifik reaktion. Andra egenskaper hos glutationtransferaser är att de är stabila och de är lätta att producera och rena fram. Vidare är glutationtransferaser mycket välstuderade proteiner, och kristallstrukturer av många varianter finns tillgängliga. Sammantaget pekar det mesta mot att glutationtransferaser är nästintill optimala kandidater till att fungera som en plattform för utveckling av nya proteiner med nya egenskaper. A) Ett nytt användarvänligt och effektivt sätt att lägesspecifikt introducera funktionella molekyler med olika egenskaper i aktiva ytan hos proteinet har tagits fram. Genom att blanda designade och syntetiserade tioestrar av glutation med humana glutationtransferaser från Alfaklassen upptäcktes det att proteinerna blev modifierade av motsvarande acylgrupp på en specifik aminosyra i aktiva ytan, nämligen den katalytiskt viktiga tyrosin 9 (Paper I). Endast tyrosin 9 utav 51 befintliga nukleofila aminosyror i hela proteinet kunde genom den här metoden bli selektivt inmärkt med varierande acylgrupper, med utbyten som överskred 90 % efter en timmes inkubering (Paper II). Metoden har vidareutvecklats genom design och syntes av biotinylerade, glutationbaserade tioestrar, varigenom öveskottet av reagens effektivt kunde bindas till en agarosmatris immobiliserad med avidin, efter reaktionen. Mångsidigheten för reaktionen
vii
har även utvecklats ytterligare genom att införa mutationen A216K, vilken då också kunde märkas in lägesspecifikt och bilda stabila amidbindningar med en rad acylgrupper (Paper III). Detta har lett till att exempelvis en reportermolekyl (fluorescent prob) förts in i ett framtaget mutantbibliotek av proteinet där mikromiljön i aktiva ytan varierats, med avsikt att utveckla en proteinbaserad biosensor liknande våra lukt- och smaksystem (Paper IV). Även ett nytt potentiellt system för effektiv rening och detektion av önskade målproteiner har tagits fram genom den här metoden (Paper V). B) För att markant utöka möjligheterna att introducera nya funktionella molekyler i hGST-A1-1 gjordes försök till totalsyntes av proteinet. Genom att kemiskt syntetisera proteinet från scratch, ges nära gränslösa möjligheter att manipulera både själva proteinskelettet och aminosyrornas sidokedjor. Två olika designstrategier presenteras, varigenom syntetiska polypeptider kan konjugeras ihop stegvis med naturliga peptidbindningar, för att slutligen få fram ett helsyntetiskt, veckat och funktionellt protein.
viii
Acm Alloc ANT-NHS Boc CDNB DCM DIPCDI DIPEA DMF EDC ESI-MS FRET GS-Al GS-ANT GSB GSBd5 GSC-thioester GS-DNB GSH GST GS-thioester HOBt HPLC K216Cou MALDI-MS Msc NA NCL NHS OD PCR PEG PDEA SBzl SPCL SPPS TFA TFE UV
S-acetamidomethyl Allyloxycarbonyl Succinimidyl N-methylanthranilate t-butoxycarbonyl 1-chloro-2,4-dinitrobenzene Dichloromethane 1,3-diisopropylcarbodiimide Diisopropylethylamine N,N-dimethylformamide N-ethyl-N´-(3-dimethylaminopropyl)carbodiimide Electrospray ionization mass spectrometry Fluorescence resonance energy transfer Thioester of glutathione and p-formylbenzoic acid Thioester of glutathione and N-methylanthranilic acid Thioester of glutathione and benzoic acid Thioester of glutathione and the deuterated analogue of benzoic acid Thioester of glutathione, where the glycine residue was changed to a cystein residue Thioester of glutathione and 1-chloro-2,4-dinitrobenzene Glutathione Glutathione transferase Thioester of glutathione 1-hydroxybenzotriazole High performance liquid chromatography The conjugate between lysine 216 and coumarin Matrix-assisted laser desorption ionization mass spectrometry Methylsulfonylethyloxycarbonyl Neutravidin Native chemical ligation N-hydroxysuccinimide Optical density Polymerase chain reaction Polyethylene glycol 2-(2-pyridinyldithio)-ethylamine Benzylmercaptane Solid phase chemical ligation Solid phase peptide synthesis Trifluoroacetic acid Trifluoroethanol Ultraviolet
ix
The structure, three- and one-letter code of the 20 common amino acids.
H2 N
NH2 NH
O
O O-
NH2 H3 N
COO
-
Alanine, Ala, A
H3 N
COO
-
H3N
Arginine, Arg, R
COO
H3 N
COO-
H3 N
Asparagine, Asn, N
O SH
-
Aspartic acid, Asp, D
NH2
O
O-
H
COO
-
Cysteine, Cys, C
H3N
COO-
H3 N
Glycine, Gly, G
COO-
H3 N
Glutamine, Gln, Q
COO-
Glutamic acid, Glu, E
NH3 HN
H3 N
N
COO-
Histidine, His, H
H3N
COO-
H3 N
Isoleucine, Ile, I
COO-
Leucine, Leu, L
H3 N
COO-
Lysine, Lys, K
S
H3 N
COO-
Methionine, Met, M
H3N
OH
COON C + H2 H
COO-
Phenylalanine, Phe, F
H3 N
Proline, Pro, P
COO-
Serine, Ser, S
OH NH
OH H3 N
COO-
Threonine, Thr, T
H3 N
COO-
H3N
Tryptophan, Trp, W
COO-
Tyrosine, Tyr, Y
x
H3 N
COO-
Valine, Val, V
! "#$% #" ........................................................................................................................... 1 1.1. PROTEINS – THE FUNDAMENT OF LIFE......................................................................................... 1 1.2. WHY STUDY PROTEINS? ............................................................................................................... 3 & #'" (#' )"#* )"#*#+, #+,.................................................................................................. 5 #+, 2.1. PROTEIN ENGINEERING................................................................................................................ 5 2.2. CHEMICAL STRATEGIES IN PROTEIN DESIGN .............................................................................. 6 - ''"+"''"+ #. )%/" +0 .#/ )' * ) *00 1 '0 2 - ''"+"''"+ #. )%/" +0 0 .#/ )' * ) *00 1 '0 2 3............................................................................................................................................................ 9 3 3.1. GLUTATHIONE TRANSFERASES .................................................................................................... 9 3.2. STRUCTURE AND FUNCTION OF CYTOSOLIC GSTS ..................................................................... 9 3.3. A CANDIDATE FOR PROTEIN ENGINEERING (PAPER I) .............................................................. 11 3.3.1. Site-specific modification of Y9 in the Alpha class GSTs................................................. 12 3.3.2. Only the Alpha class GSTs became modified.................................................................... 14 3.3.3. The modification reaction was not unique for GSB ......................................................... 15 3.3.4. The GSH backbone is required for specificity................................................................... 16 3.3.5. Kinetics and stability .......................................................................................................... 16 3.4. EXPANDING THE REPERTOIRE OF ACYLATING REAGENTS (PAPER II) ..................................... 18 3.5. IMPROVEMENT OF THE MODIFICATION END-PRODUCT STABILITY .......................................... 20 3.5.1. Labeling of a single lysine residue in an hGST A1-1 mutant........................................... 20 3.5.2. Reaction mechanism .......................................................................................................... 22 3.6. DESIGN, SYNTHESIS AND FUNCTION OF NOVEL LABELING REAGENTS (PAPER III).................. 23 4 "#5'* (#0'"0#0 1 ' 53 4 "#5'* (#0'"0#0 1 ' 53 ..................................................................................... 27 4.1. THE CONCEPT OF BIOSENSORS................................................................................................... 27 4.2. PRINCIPLES IN THE DESIGN OF FLUORESCENT MOLECULAR SENSORS ..................................... 29 4.3. REENGINEERED GSTS AS BIOSENSORS ..................................................................................... 33 4.3.1. Signaling potential of the site-specifically labeled receptor family .................................. 35 4.4. AFFINITY STUDIES ...................................................................................................................... 38 4.5. ON THE MECHANISM OF BINDING OF ANALYTES BY THE PROTEINS ......................................... 40 4.6. IMMOBILIZATION OF A216KCOU ON A HYDROGEL SURFACE ..................................................... 42 4.7. CONCLUSIONS ............................................................................................................................ 45 6 '"+"'''$ "$ )' )'/ / **, /# **, /#$.'$ +0 $.'$ +00 0 .%0# 0 .%0#" #'"0 " #'"0 1 ' 53.......................................................................................................................................... 47 1 ' 53 5.1. CONSTRUCTION AND CHARACTERIZATION OF THE A216K/M208X MUTANTS ....................... 48 5.1.1 Specific activity and stability............................................................................................... 49 5.1.2. Labeling propensity and GSH affinity............................................................................... 49 5.2. SUMMARY ................................................................................................................................... 50 5.3. OUTLOOK ................................................................................................................................... 51 7 0%//, #. '0........................................................................................................... 53 7 0%//, #. '0 8 #* 0,")'00 #. 8 #* 0,")'00 #. h+0 !2 +0 !2! .................................................................................. 55 7.1. NATIVE CHEMICAL LIGATION ................................................................................................... 55 7.2. SYNTHETIC STRATEGY A (NCL) ............................................................................................... 57 7.2.1. Results................................................................................................................................. 59 7.3. SYNTHETIC STRATEGY B (SPCL) .............................................................................................. 62 9"#:*'$+'/'"0 ............................................................................................................. 65 '.''" '0 .................................................................................................................................. 67
! ! The processes that are the basis for life are taken for granted by many of us. Fundamental questions about how the building blocks of nature (atoms) can be combined to generate the larger building blocks (molecules) that in turn can generate the macromolecules of life (such as DNA, sugars, lipids and proteins) tend to be unanswerable and classified as miracles by many non-scientists. The miracle of life is that a myriad of chemical reactions in the cell are occurring simultaneously with great accuracy and at an astonishing speed. Understanding the molecular nature of life’s processes and the intricate networks of biomolecular interactions present in all living beings is the ultimate goal for scientists in the field of chemistry and biology. During the past 60 years, enormous efforts have been put in to understand the cellular machinery and the interplay between the different cellular constituents that makes life possible.
1.1. Proteins – the fundament of life The focus of this thesis has been a class of biological molecules called proteins. Proteins are the most versatile macromolecules. They control and assist in almost all the chemical, biological and physical processes that result in life. Proteins can function as receptor molecules, transport molecules, messenger molecules, molecular motors, defenders against microorganisms and other toxic compounds, and controllers of electron flow. However, of all the roles of proteins, probably the most important is catalysis. In the absence of catalysis, most reactions in biological systems would take place too slowly to provide products at an adequate pace for a metabolizing organism. The protein catalysts that serve this function in organisms are called enzymes.
1
Irrespective of the function of the protein, all proteins are made of basically 20 naturally occurring amino acids which are connected in long chains through peptide bonds (Figure 1.1).
AMINO ACID 1 R1 +
H 3N
R2
+ COO
DIPEPTIDE
AMINO ACID 2
+
-
H 3N
O COO
+
-
H 3N
H 2O
R2 NH
R1
COO -
PEPTIDE BOND
Figure 1.1. The amino acids are linked by peptide bonds to form polypeptide chains.
The amino acids have various sizes, shapes and functional groups. In order for a protein to function, it must adopt a certain three-dimensional structure that places the involved parts of the protein in the correct spatial arrangement, i.e. the protein must be folded. The information needed for the protein to fold into a functional threedimensional unit is contained within the one-dimensional amino acid sequence. The structural organization of a protein is usually described on four levels and may be defined as (Figure 1.2): (i) the primary structure, defined by the linear sequence of amino acids within a polypeptide; (ii) the secondary structure, which refers to the folding of local regions into motifs such as Į-helices and ȕ-sheets; (iii) the tertiary structure, the compact asymmetric structure resulting from packing of secondary structural elements and; (iv) the quaternary structure, defined by the assembly of subunits of proteins having more than one polypeptide chain.
PRIMARY STRUCTURE
SECONDARY STRUCTURE
TERTIARY STRUCTURE
Met-Ala-Glu-ArgLys-Lys-Ser-PheAsn-Pro-Ser-ThrTyr-Lys-Gly-……. Į-helix
ȕ-sheet
Figure 1.2. Structural levels in proteins.
2
QUATERNARY STRUCTURE
1.2. Why study proteins? There are many different aims that may motivate the study of proteins. The human gene pool has recently been surveyed in an international project, called the HUGOproject 1, 2. Collaboration between researchers around the world has lead to the conclusion that approximately 23 000 protein-coding genes exist in the human genome. The full set of proteins (the proteome) encoded by the human genome is more complex. The human proteome is larger than the genome, i.e. there are more proteins than genes. This is due to post-translational modifications and alternative splicing of genes 1. One of the biggest challenges for scientists is now to determine the structure and function of proteins. This would give better insights into where and when different proteins are active in human tissues and organs, and also a better understanding of how proteins interact with each other in complex networks. The over all aim in this research area is not only to understand the human biology, but also the course of events for complex diseases, such as cancer and degenerative diseases, which in turn is important for drug development and treatment. Understanding the mechanisms of how a synthesized, linear polypeptide, becomes a folded functional protein is another challenging task in protein science 3. A detailed understanding would give us the tools to predict the three-dimensional structure from the primary structure and to design novel protein structures and functions not provided by nature. Recently there has been a huge interest in diseases connected to protein misfolding, aggregation, and amyloid formation. These diseases include cystic fibrosis, cancer, Alzheimer´s disease, Parkinson´s disease, Type II diabetes, Amyotrophic lateral sclerosis (ALS), and Creutzfeldt-Jacob´s disease 4, 5. At present, a large amount of research is being carried out to understand the molecular basis underlying these diseases and for early detection of amyloid formation 6-9. Apart from the fundamental protein research which deals with the function of living organisms, there are other aspects of the utility of proteins; for example protein-based, or peptide-based, platforms for the development of novel enzymes, receptors or biosensors 10-14. The ability to manipulate proteins with chemical methods in order to alter the natural function or introduce new functions has provided a basis for a range of different applications 15, 16. Post-translational derivatization of naturally occurring proteins and enzymes with artificial functional molecules represents a promising approach for the development of novel tailor-made biomolecules 15, 17. This opens up several avenues to induce selective reactivity in a site-specific manner, and that can be applied to protein identification and manipulation in both an in vivo and an in vitro
3
context. Also, synthetic and semi-synthetic approaches have been attractive to create proteins and enzymes with novel functions 18. This thesis describes the development of multipurpose artificial receptors. The enzyme glutathione transferase has been used as a scaffold with the aim to alter the natural function of the protein by chemical methods. In Chapter 2, an overview of the field of protein design is given. My achievements are introduced in Chapter 3. First, our workhorse enzymes; the glutathione transferases are introduced. Thereafter, the story of the development of a novel route for site-specific introduction of chemical moieties into the protein scaffold is told. In Chapter 4 and Chapter 5, the applications of a multipurpose receptor are presented. A summary of the published work (Papers I-V) are found in Chapter 6. Finally, an unfulfilled opportunity for synthetic reconstruction of the enzyme is presented in Chapter 7.
4
& ; & ;
; Since the first site-directed mutagenesis experiments, approximately 30 years ago 19, it has been possible to manipulate proteins at a molecular level, both with geneticand chemical methods. The purposes of protein manipulations, usually referred as “protein design” are to achieve a fundamental understanding of protein structurefunction relationships, and also to alter their characteristics or introduce novel functions. This chapter gives an overview of both genetic and chemical strategies that have provided the protein designer with a greatly expanded palette from which to modulate the specificity and/or reactivity of existing enzymes and to create novel proteins with novel functions.
2.1. Protein engineering With the technology available today it is possible to clone and overexpress proteins in bacteria, yeast cells, insect cells, and mammalian cells. The knowledge of how to mutate genes has also made it possible to change the properties, such as catalysis, reaction specificity or stability 20, 21. Performing changes in the DNA sequence of a cloned gene, coding for a particular protein, with the purpose of modulating the characteristics of the protein or introducing novel functions, is called protein engineering. For example, redesigning the active site of an existing enzyme by genetic methods can change the catalytic activity of the enzyme to catalyze reactions not possible by the natural enzyme 10, 22. Engineered proteins are now useful reagents for applications in biotechnology and as environmental-friendly catalysts for example in detergents, in the agrochemical and food industry, and in pharmaceuticals and therapeutics 23-25. There are two general methods for protein engineering. One is known as directed evolution, and it is based on random alterations of the gene sequence. Diverse gene
5
libraries can be created by various random methods, such as error prone PCR or DNA shuffling. High-throughput selection or screening procedures, such as phage display, are thereafter applied to identify proteins with desired qualities, expressed from the gene libraries 26. An advantage of this method is that desired changes are often caused by mutations that no one would have expected. The difficulty of finding a sufficiently effective screening or selection system for a specific enzymatic activity may, on the other hand, be the biggest drawback of this method 27. The other strategy is known as rational design, in which precise changes in the amino acid sequence are preconceived based on a detailed knowledge of the protein structure, function and mechanism 28-30. Computationally based methods are most often applied to evaluate the structural and functional influences of the proposed mutations 31-33. Even though there is a large amount of reports on successful work based on rational design, the drawback is that a mutation often leads to unexpected results and have frequently been found to be deleterious with regards to stability and activity.
2.2. Chemical strategies in protein design In concert with the genetic techniques, chemical strategies have had a significant impact in the field of enzyme design. Taking advantage of the knowledge gained in protein engineering, the construction of new proteins and peptides with new functions, de novo or template based, is now a very active area of research 15, 34, 35. Chemical modification of proteins is an intriguing approach for the development of new enzymes and proteins with novel functions. Covalent modification methods allow an almost unlimited range of functionality to be site-specifically introduced into proteins. Most strategies for site-specific modification of proteins exploit the distinctive reactivity of the thiol group in cysteine in vitro 36-38. The thiol-reactive functional groups are primarily alkylating reagents, including iodoacetamides, maleimides and benzylic halides. Additional methods that permit site-specific modification of other residues than cysteine, and that permit specific, single-site modification in the presence of multiple cysteines, or under in vivo conditions are highly desirable. Synthetic and semi-synthetic approaches to generate novel functional proteins and enzymes have been developed, and thus, opened up possibilities to incorporate functionalities at any position in the protein sequence 39. Chemical protein synthesis also promises the unlimited variation of the covalent structure of a polypeptide chain with the objective of understanding the molecular basis of protein function.
6
This work constitutes a demonstration of how the methods of protein research can generate reengineered proteins with novel functions. Instead of focusing on fundamental protein science, effort has been put in taking advantage of the existing knowledge to generate proteins with functions not provided by nature.
7
8
- +0 +0 1 2 1 23 3 This chapter starts with an introduction to the glutathione transferases. The sitespecific incorporation of artificial groups into the protein scaffolds is then presented. The last part describes how an acylating reagent was further functionalized with an affinity tag.
3.1. Glutathione transferases The glutathione transferases (GSTs; EC 2.5.1.18) are a superfamily of multifunctional dimeric enzymes involved in the mechanism of cellular detoxication 40-44. Three major families of proteins that are widely distributed in nature exhibit glutathione transferase activity. Two of these, the cytosolic and mitochondrial GSTs, are soluble enzymes that are only distantly related 45. The third family comprises microsomal GST and is referred to as membrane-associated proteins 45. The soluble GSTs have since their discovery in 1961 proven to be present in oxygen metabolizing organisms 46. Based on amino acid sequence similarities, seven classes of cytosolic GSTs have been recognized in mammalian species, designated Alpha, Mu, Pi, Theta, Zeta, Sigma and Omega 47, and within each class, several isoenzymes may exist 45.
3.2. Structure and function of cytosolic GSTs Today, there are numerous crystal structures available of mammalian GSTs from all classes, both with and without ligands 41, 47, 48. The cytosolic GSTs exist as homo- or heterodimers, and they catalyse the conjugation reaction between the tripeptide glutathione (Ȗ-Glu-Cys-Gly, GSH) and a number of hydrophobic, electrophilic
9
compounds of both endogenous and xenobiotic origins 41 (Figure 3.1). The glutathione-conjugates are then further metabolized through the mercapturic acid pathway and eventually excreted; the glutathione-conjugates have greater solubility in water, facilitating their export from the cell 49. Cl
NO 2
NO 2
NO 2
NO 2 SH
O -
OOC
NH
NH NH 3 +
CDNB
O -
COO -
OOC
NH
NH NH 3
O GSH
S
+
COO -
O GS-DNB
Figure 3.1. The nucleophilic aromatic substitution reaction between GSH and the most frequently used electrophilic substrate 1-chloro-2,4-dinitrobenzene (CDNB).
Other functions of GSTs include inactivation of Į/ȕ-unsaturated carbonyl derivates, isomerase activities, peroxidase activities, transportation and clearance of oxidative stress products and modulation of cell proliferation and apoptosis signaling pathways 50 . In addition to their catalytic activities, GSTs also function as ligand binding proteins and thereby facilitate the intracellular storage of a variety of hydrophobic non-substrate compounds, including hormones, bile acids and metabolites 51-53. Because the GSTs have evolved to function as detoxication enzymes, each subunit has a specific and highly conserved glutathione binding site (G-site), and an adjacent but far less specific hydrophobic binding site for electrophilic substrates (H-site) 54, Figure 3.2. The substrate specificity of an enzyme is usually dominated by electrostatic interactions between the enzyme and the substrate. In the H-site of the Alpha class GSTs, there are very few hydrophilic residues that can contribute with electrostatic interactions 55. The residues contributing to the G-site are well-conserved across the GST classes, whereas those of the H-site are highly variable, and thus produce the varying hydrophobic substrate specificities of the different isoforms.
10
A
B H-site
Į9
Y9
G-site
Figure 3.2. (A) shows the dimeric structure of hGST A1-1 (PDB entry: 1GUH) 55 in complex with S-benzylglutathione (shown in stick representation). (B) shows one of the subunits, where tyrosine 9 (Y9) is represented in stick model, and a close-up of the binding pocket. The locations of the G-site and the H-site are highlighted with arrows. The flexible helix 9 (Į9) is also indicated in the figure.
Each subunit contains two domains: an N-terminal Į/ȕ-domain, where the G-site is located and a C-terminal Į-helical domain, where the H-site is located (Figure 3.2). Xray crystallographic and site-directed mutagenesis studies have shown that each subunit contains a conserved tyrosine (Alpha, Mu, Pi, Sigma), serine (Theta, Zeta) or cysteine (Omega) residue in the G-site, that are crucial in the normal detoxication reaction 50. These residues hydrogen bond to, and effectively deprotonates, GSH to form the nucleophilic thiolate anion (GS-) 41. The pKa value of the thiol group is lowered from approximately 9 in solution to 5.7-7.5 (depending on the isoenzyme) when bound to the active site 56-58 and thus facilitates the conjugation reaction. The modular features of these proteins, in combination with their stability, ease of purification 59, and the wealth of accumulated knowledge of structure-activity relationships 60, are all factors that point to the GSTs as ideal candidates in protein engineering efforts with the goal of obtaining novel function.
3.3. A candidate for protein engineering (Paper I) The different classes of GSTs display different characteristics and the present study is focused on the GSTs from the Alpha class, and in particular, the human isoenzyme hGST A1-1. A unique structural feature of the Alpha class GSTs is the C-terminal Įhelix (Į9) 61 (Figure 3.2). In the apo form of the protein, Į9 is either disordered or
11
structured, but conformationally highly mobile 62, 63. In contrast, hGST A1-1 in complex with S-benzylglutathione or ethacrynic acid has a well-formed amphipatic Įhelix (Į9) 55, 62. Glutathione alone can also induce helical formation in the C-terminus of hGST A1-1 63, 64. The hydrophobic microenvironment in the H-site of the Alpha class GSTs is partly due to Į9 that closes like a lid over the H-site when a substrate is bound, and thus protects the H-site from surrounding water 55. The dynamic nature of Į9 suggests that it can be mutated without deleterious effects on the overall structure of the protein. In fact, numerous modulations of the C-terminal region have been done to study the effects on ligand binding, detoxication activity and overall stability 58, 6567 . 3.3.1. Site-specific modification of Y9 in the Alpha class GSTs In a study with the purpose of obtaining mechanistic information in the construction of a novel enzyme, it was found that hGST A1-1 could be site-specifically acylated at the side chain of tyrosine 9 (Y9) upon incubation with GSB, a thioester of GSH and benzoic acid (Figure 3.3).
O SH
O -
OOC
NH
NH NH 3+
S
O
COO -
-
OOC
O
NH
NH NH 3
+
COO -
O
GSH
GSB
O hGST A1-1
O
GST
+ GSH
GS
O Y9
GSB
Figure 3.3. The structures of GSH and GSB, and the modification reaction of hGST A1-1 and GSB.
This novel route for site-specific incorporation of an artificial group into the Alpha class GST had not been reported previously and the reaction was therefore thoroughly investigated by UV spectroscopy, HPLC and MALDI-MS and the results were the starting point for the development of a multipurpose artificial receptor.
12
Upon incubation with an excess of GSB, it was found that wild-type hGST A1-1 became acylated at one single position (Y9) by the benzoyl moiety. UV spectroscopy measurements showed a decrease in the GSB concentration that was found to be equivalent to that of the protein concentration, calculated as a monomer. Detailed proteolysis analyses were performed to identify the site of modification. Peptide fragments of modified hGST A1-1 following proteolytic digestion, using either trypsin (cleaves on the carboxyl side of arginine and lysine) or Staphylococcus Aureus Protease V8 (S. aureus V8) (cleaves on the carboxyl side of glutamic acid and aspartic acid), was analyzed with MALDI-MS and compared to those obtained from unmodified hGST A1-1. From these experiments it turned out that the fragment corresponding to amino acids 7-13 (LHY9FNAR), using trypsin, or 2-17 (AEKPKLHY9FNARGRME) using S. aureus V8, had gained an increased mass equivalent to a benzoyl moiety.
Voy ager Spec #1=>BC[BP = 2053.7, 18917]
2053.72
90
AEKPKLHY9(Benzoyl)FNARGRME
80
% Intensity
70 60 50
1.9E+4
2053.7227
100
1551.1222
AEKPKLHY9FNARGRME
40 30
1948.6110 1948.61
2822.7981
20 10 0 1450
1554.1328
with S. aureus V8.
2056.7194
2414.1345 2013.3287 2282.1889
1860
Figure 3.4. MALDI-MS spectrum of the resulting peptide fragments of hGST A1-1 following incubation with GSB and proteolysis
2270
2680
2825.8332
3090
0 3500
Mass (m/z)
By comparing the two proteolytic fragments, Y9 was the only nucleophile that could form a sufficiently stable ester under neutral conditions. Inspection of the crystal structure of hGST A1-1 in complex with S -benzyl-GSH (PDB entry: 1GUH), Figure 3.5, where Y9 is in close proximity to the thioether bond, also pointed out Y9 to be the amino acid residue that became modified. α9
4.20 Å
Y9
Figure 3.5. A close-up of the crystal structure of hGST A1-1 (PDB entry: 1GUH) in a complex with S-benzylglutathione. The distance between the hydroxyl oxygen of Y9 and the ligand benzyl CH2 is 4.2 Å (© 2003 American Chemical Society).
13
3.3.2. Only the Alpha class GSTs became modified To explore if the modification reaction was unique to hGST A1-1, GSB was added to a panel of five classes of GSTs (Alpha, Mu, Omega, Pi, and Theta). All GSTs have a catalytically important tyrosine (Alpha, Mu, and Pi), serine (Theta) or cysteine (Omega) in the G-site in the equivalent position. The sequence identity across the classes is low but all the structures have a similar fold, with each subunit consisting of two distinct domains 46. However, there are some striking structural differences between the classes. Comparison of the Alpha, Mu and Pi structures reveals several major similarities, as well as crucial differences. The Alpha class has a C-terminal helix (Į9) that provides a smaller and more hydrophobic active site in comparison with the Mu and Pi classes that have more open sites 55. The Theta class GST shows only 7 % overall sequence identity with the Alpha, Mu, and Pi classes 68 and has a unique substrate specificity compared to all other classes 46. There are considerable structural variations between the Omega class and the other classes. In particular, the enzyme has an unusual proline-rich N-terminal extension not present in other classes 46
.
A set of ten isoenzymes (hGSTs A1-1, A2-2, A3-3, A4-4, M2-2, M4-4, O1-1, P1-1, mGST M5-5, rGST T2-2) and also the A4-4 mutant Y9F were used in the screening experiment. Proteolysis and MALDI-MS analysis was used to identify the fragments corresponding to Y9 in hGST A1-1. The four human isoenzymes from the Alpha class have a tyrosine residue at position 9, and the results showed that only this tyrosine residue in the Alpha class could be acylated by GSB. The hGST A4-4 mutant Y9F did not show any sign of modification. All together this provides evidence that the modification reaction is class-specific and that Y9 is the site of modification.
14
B. Mu
A. Alpha
D. Pi
C. Omega
E. Theta
Figure 3.6. Crystal structures of one of the subunits of five GST classes. The catalytically important residue corresponding to Y9 in hGST A1-1 is shown in stick representation. (A) human Alpha class (PDB entry: 1GUH), (B) human Mu class (PDB entry: 1HNA) 69 , (C) human Omega class (PDB entry: 1EEM) 70, (D) human Pi class (PDB entry: 1GSS) 71, (D) human Theta class (PDB entry: 1LJR) 72.
The class-specific modification might be partially explained by the depressed pKa values of Y9 in the Alpha class ranging from 6.7 (hGST A4-4) 73 to 9.2 (hGST A2-2) 74
, but on the other hand, the values are reported to be in this range for the Pi enzyme also 75. Binding preference is probably not the reason either, because similar compounds have been shown to bind to the Alpha, Mu, and Pi class with comparable affinities 76. Thus, the main reason behind the class-specificity appears to be that the thioester functionality is better oriented in the Alpha class to facilitate the nucleophilic attack of Y9 compared to the corresponding catalytic residues in the other classes. 3.3.3. The modification reaction was not unique for GSB
Two additional GS-thioesters (GS-ANT, synthesized from glutathione and Nmethylanthranilic acid, and the naturally occurring S-lactoylglutathione), Figure 3.7,
15
were also incubated together with the four isoenzymes from the Alpha class to investigate whether GSB was the only reagent able to modify Y9. MALDI-MS analyses showed that all isoenzymes became modified by GS-ANT whereas no trace of modification was detected for S -lactoylglutathione.
O
O GS
O
O
N
GS HN
GS-ANT
OH
S-lactoylglutathione
O
O HN ANT-NHS
Figure 3.7. Structures of GS-ANT, S-lactoylglutathione and ANT-NHS.
3.3.4. The GSH backbone is required for specificity The modification of hGST A1-1 is very specific in that only one out of 51 possible nucleophiles, one out of ten tyrosine residues, becomes covalently modified. The underlying reason for this specificity is the GSH backbone for which the affinity of the G-site is highly conserved throughout the GSTs. The GSH backbone is required for specificity since an activated ester, ANT-NHS (succinimidyl N-methylanthranilate), Figure 3.7, mainly acylated surface-exposed lysine residues and only to a minor extent Y9. In contrast, the corresponding GS-thioester, GS-ANT, only acylated Y9. 3.3.5. Kinetics and stability The rate of the modification reaction, and the stability of the formed Y9-benzoyl ester, was investigated by adding increasing concentrations of GSB to hGST A1-1 at different pH values. 100 µM GSB was shown to modify hGST A1-1 (5 µM) roughly quantitatively within 40 minutes, and the pH dependence showed a slight optimum at pH 7.5. The pKa of Y9 in hGST A1-1 is 8.1 57, but even though the concentration of reactive tyrosinate ion increases with increasing pH, the stability of both the thioester and the Y9 ester are decreased at higher pH values due to hydrolysis.
16
The benzoyl ester formed at Y9 was stable in sodium phosphate buffer solution, pH 7.0, for at least 24 hours. A pulse and chase experiment using GSB and the deuterated analogue of GSB (GSBd5), showed that there is no exchange of bound benzoic acid, Figure 3.8. On the other hand, incubation with GSH for 10 minutes completely removed the ester, but an addition of S-methyl-GSH protected the benzoyl ester at Y9 significantly. Interestingly, the ANT-modified hGST A1-1 was significantly more stable towards GSH.
V
V
y
a
g
e
r
S
p
e
c
#
1
=
>
N
F
0
.
7
=
100
>
B
60
% Intensity
2050.14 2052.14
70
[
B
P
=
1
0
6
6
.
2
,
1
0
4
8
o
y
a
g
e
r
S
p
e
c
#
1
=
>
N
F
0
.
7
=
>
B
C
[
B
P
=
1
5
4
9
.
1
,
4
2
6
8
]
2056.06
2051.06
]
2050.06
80
90 80
C
90
2051.13
100
% Intensity
o
2057.06
70
2052.06
60 50 40 30 20
50
10
40
0 2040
30
2048
2056
20
2064
2072
2080
M as s (m /z)
2056
10 0 2040
2048
2056
2064
2072
2080
M as s (m /z)
Figure 3.8. MALDI-MS spectrum after a pulse and chase experiment where an excess of GSB was added to hGST A1-1 followed by a chase of an equal amount of the deuterated analogue GSBd5 after 5.5 h incubation. The reaction was then allowed to proceed overnight. The figure shows that there was no exchange of bound benzoic acid. The inset shows that GSB and GSBd5 react with hGST A1-1 in identical ways (© 2003 American Chemical Society). The potential for GSB and GS-ANT to deliver their acyl group to Y9 was also tested in E. coli lysates that were doped with hGST A1-1. GSB was able to modify Y9 only when a protease inhibitor cocktail was added to the reaction mixture, which was not necessary for GS-ANT. It might be that GS-ANT stabilized Į9 more than GSB, and protected it from proteolysis. The speed of the reactions were similar to that in buffer solution for both reagents, and by adding a protease inhibitor cocktail, the stability of the Y9 esters were also similar to that in buffer solution.
17
3.4. Expanding the repertoire of acylating reagents (Paper II) With the purpose of exploring the potential of the novel site-directed modification reaction, a combinatorial screening approach was set up. The GSTs are promiscuous proteins with a broad substrate specificity 40, and the range of GS-thioesters that could deliver their acyl groups to the proteins could potentially be expanded to include more reagents than GSB and GS-ANT. A panel of 17 thioesters of glutathione, with a variety of functionalities, were synthesized in parallel and used in screening experiments with the Alpha class GSTs, Figure 3.9. O O
HO
O
O
O HO
HO
HO
O
HO
O
O
OH 1. 3,5-dimethylbenzoic acid
2. Naphtalene-2-carboxylic acid
3. 5-carboxy-fluorescein
O
O
4. 7-methoxy-coumarin-3-carboxylic acid
O O
O
O
F
HO
HO
HO HO
O
O
O
N H
F 5. 3,4-difluorobenzoic acid
6. 4-acetamidobenzoic acid
7. 2-biphenylcarboxylic acid
O
8. 3-methylflavone-8-carboxylic acid
O
O
HO
O
HO
HO N
HO
F
F
O
F 9. 2,6-dimethylbenzoic acid
10. 3-(trifluoromethyl)-benzoic acid
11. 2-phenyl-4-quinolinecarboxylic acid
12. 4-benzoylbenzoic acid
O O
O HO
O
HO HO
S
HO
O 13. 2-thiophenecarboxylic acid
O
H
N 14. 3-(dimethylamino)-benzoic acid 15. p-formylbenzoic acid
HO 16. Phenylacetic acid 17. 6-phenylhexanoic acid
Figure 3.9. The structures of the carboxylic acids that were conjugated with GSH. GSB (not shown) was also included in the experiment.
The combined screening of both reagents and proteins was done in order to investigate whether any of the Alpha class isoenzymes would be better suited as a
18
scaffold for protein engineering experiments and also to obtain general information about reagent requirements. Out of 18 GS-thioesters investigated (GSB included), 14, or 78 %, were able to covalently modify Y9 of hGST A1-1. The modification reaction was also versatile with respect to the other Alpha class GSTs (hGST A2-2, A3-3 and A4-4), where 72 – 83 % of the reagents were able to acylate the proteins. The set of chemical groups that could be site-specifically introduced into the active site of the proteins now included fluorescent groups, a photochemical probe and an aldehyde moiety.
Table 3.1. Modification reactions of the human Alpha class GSTs. hGST GS-thioester GSB GS-1 GS-2 GS-3 GS-4 GS-5 GS-6 GS-7 GS-8 GS-9 GS-10 GS-11 GS-12 GS-13 GS-14 GS-15 GS-16 GS-17
A1-1 + + + + + + + + + + + + + +
A2-2 + + + + + + + + + + + + + + +
A3-3 + + + + + + + + + + + + + + +
A4-4 + + + + + + + + + + + + +
+ and – corresponds to modified or not modified Y9-fragment
Since the Alpha class GSTs have a relaxed hydrophobic substrate specificity that originates from their biological role as detoxication enzymes, it may not be that surprising that they accept the vide range of reagents. There does not seem to be a systematic way of predicting the ability of a GS-thioester to modify an Alpha class GST. The corresponding acids of the GS-thioesters vary in reactivity, size, hydrophobicity, substitution pattern, etc. No general trends were found, with the exception of GS-9, which is di-ortho-substituted, and did not react with any of the Alpha class GSTs. The likely explanation is that GS-9 is too sterically hindered, especially since the di-meta-substituted analogue, GS-1, is an exellent derivatization reagent for all four proteins. The extent of modification varied, which is not surprising since standardized conditions with respect to incubation time, temperature and reagent
19
concentration were used for all reactions. Not all reagents work for all Alpha class GSTs, and the main reason behind a successful modification depends on the combination of protein and reagent that in turn leads to an optimal orientation of the thioester functionality in close proximity to the reactive Y9 residue. The multiple set of acyl groups being able to site-specifically modify the proteins is due to the adjacent promiscuous H-site that accepts a large variety of groups. Even though the affinity of the acyl group for the H-site may be relatively low, one can envision that the acyl group hitches a ride with the glutathione backbone into the active site. Generally, GSH has an affinity for the G-site that is about an order of magnitude stronger than for a general substrate for the H-site 58, 77. Again, an example of how to utilize the fact that GSTs are promiscuous detoxication enzymes.
3.5. Improvement of the modification end-product stability The ester formed at the side chain of Y9 in the wild-type hGST A1-1 was not stable towards GSH, and the possibility of using Y9-modified proteins in purification or fusion protein experiments is thus limited since the intracellular concentration of GSH ranges from 0.1 to 10 mM. It was also found that by adding another GS-thioester reagent (GS-2) to benzoic acid-modified hGST A1-1 resulted in scrambling of the bound acyl group from benzoic acid to naphtoic acid 78. Thus, to allow this reaction to be more efficiently utilized, an effort was put into designing a more stable linkage between the protein and the acyl group.
3.5.1. Labeling of a single lysine residue in an hGST A1-1 mutant To circumvent these problems, Hederos et al 79, prepared three Lys-mutants of hGST A1-1 (Y9K, A216K and Y9F/A216K) to explore the possibility of generating a more stable amide bond in the modification reaction 79, Figure 3.10. The Y9K mutant was prepared to test whether the targeted tyrosine could be directly substituted by a lysine. Position 216 was chosen since an A216H mutant of hGST A1-1 has been shown to function as a novel hydrolytic enzyme, where an Y9-ester was found as a reaction intermediate 22, and an introduced lysine residue at position 216 could therefore be in spatial range for acylation. The double mutant Y9F/A216K was prepared to explore the role of the critical Y9 residue in a possible modification reaction.
20
Į9
K216
Y9
Figure 3.10. A close-up of the crystal structure of hGST A1-1 (PDB entry: 1GUH) in a complex with S-benzylglutathione. The targeted positions (Y9 and K216) are shown in stick representation, and helix 9 (Į9) is also indicated in the figure.
The panel of GS-thioesters (Figure 3.9) was incubated together with the three Lysmutants to test the modification reaction. Following incubation with the reagents and proteolytic digestion (S. aureus V8), the resulting peptide fragments were monitored by MALDI-MS. Out of 18 GS-thioesters tested, eight (44 %) were able to modify A216K, whereas Y9K was unaffected by the addition of the reagents. The double mutant Y9F/A216K did react with one reagent only, GSB. None of the other GSthioesters showed any trace of modification with this mutant. Hederos et al showed that the formed amide bond between A216K and the acyl groups is stable against 1 mM GSH for more than 24 hours and that the reaction can take place in an E. coli lysate. Again, the reaction was site-specific and only one nucleophile out of 52 present (one out of 25 lysine residues) was targeted. O O
NH 2 GS
R NH
R
A216K K216
K216
Figure 3.11. The site-specific modification reaction of A216K
21
Table 3.2. Modification reactions of A216K, Y9F/A216K and Y9K. hGST A1-1 GS-thioester
a
Y9F/A216Ka
A216K
Y9Kb
GSB + + GS-1 GS-2 GS-3 GS-4 + GS-5 + GS-6 + GS-7 GS-8 + GS-9 GS-10 GS-11 GS-12 GS-13 + GS-14 + GS-15 GS-16 + GS-17 a + and – corresponds to modified or not modified K216-fragment. b + and – corresponds to modified or not modified K9-fragment.
3.5.2. Reaction mechanism The reaction mechanism seemed to proceed via an Y9-ester whereupon the acyl group was transferred to K216. Interestingly, the formation of the Y9-ester does not seem to be an absolute requirement for acylation of K216 since the mutant Y9F/A216K becomes acylated at K216 by GSB. However, the rate of modification by GSB was slower for Y9F/A216K than for A216K, indicating that the hydroxyl group of Y9 speeded up the reaction rate. A suggestion is that either the intramolecular acyl transfer from Y9 to K216 is very fast, or Y9 aids as a general base on K216, lowering its pKa and thus making it possible for a direct nucleophilic attack on GSB. A third alternative is that Y9 is involved in the stabilization of the tetrahedral transition state formed upon a nucleophilic attack from K216. Figure 3.12 shows an example, where the modification reaction between A216K and GS-4 was followed by MALDI-MS. Time-spots were withdrawn from the reaction mixture for proteolysis and subsequent analysis.
22
Y9F/A216K + GS-4, 24 h
A
V
y
a
g
e
r
S
p
e
c
#
1
[
B
P
=
1
9
3
2
.
1
,
3
9
8
5
V
1 9 32 .10
% Intensity
60 1 12 3 .8 1
40 1 5 51 .93
30 20
99 4.7 7
10
1 9 34 .07
1508
60
1772
V
2036
90
% Intensity
80
y
a
g
e
r
S
p
e
c
#
1
[
B
P
=
9
7
0
.
3
,
2
1
2
1
3
30 20
D 90 80
Modified Y9 19 4 8.1 3 15 5 1.9 0
11 96 .71 10 10 .30 9 94 .33
10 0 980
10 14 .31
1
[
B
P
=
9
7
0
.
1
,
1
2
5
3
1
]
1 9 49 .68
1508
1772
2 1 51 .69
2036
2300
A216K + GS-4, 24 h
15 5 3.9 1
2036
y
a
g
e
r
S
p
e
c
#
1
=
>
N
F
0
.
7
[
B
P
=
1
3
2
5
.
8
,
3
0
0
9
]
Unmodified K216
Unmodified Y9
60 11 96 .77
50
15 5 0.9 7
-
40 1 32 7.8 3 10 08 .33
19 4 7.1 9
11 9 8.7 8
10
21 5 0.1 5
1772
o
Modified K216
70
20 19 5 0.1 6
1508
2 14 9 .6 9
1 32 5.8 3
30
1 32 7.7 7 11 9 8.7 2
1244
#
13 25 .48
V
Unmodified Y9
13 2 5.7 7
40
c
1 9 47 .71
1244
100
60 50
e
Unmodified Y9
]
15 49 .89
Modified K216
p
M as s (m /z)
1 0 08 .30
70
S
1 5 51 .57
40
0 980
2300
% Intensity
100
o
r
Modified Y9
Modified K216
50
A216K + GS-4, 30 min
Unmodified K216
e
Unmodified K216
M as s (m /z)
C
g
70
20 99 2 .1 2 11 23 .49 10
1244
a
30
1 12 5.8 2
0 980
y
80
1 54 9 .9 1
70 50
o
1 54 9 .5 6
100 90
80
A216K + GS-4, 5 min
B
]
Unmodified K216
90
% Intensity
o
Unmodified F9
100
0 980
2300
M as s (m /z)
1244
1508
1772
2036
2300
M as s (m /z)
Figure 3.12. (A) MALDI-MS of the resulting fragments of the mutant Y9F/A216K following incubation with GS-4 for 24 h, and proteolysis. (B-D): The action of both the Y9 fragment (aa 2-17) and the K216 fragment (aa 216-222) of A216K incubated with GS-4 could be followed semi-quantitatively with MALDI-MS. Early time points showed an increasing amount of Y9-intermediate that reached a maximum at around 10 minutes. At all time points thereafter, the amount of Y9-intermediate was tiny, whereas a fast increase in K216 modification was obtained.
3.6. Design, synthesis and function of novel labeling reagents (Paper III) The successful site-specific modification of hGST A1-1 and its mutant A216K with fluorescent molecules encouraged us to develop a biotinylated labeling reagent that could, either deliver the fluorescent acyl group to the folded protein scaffold whilst attached to an avidin-modified solid support (NA beads), or a system where the solid support could potentially be used to mop up excess and used up reagent following the modification step. Biotin was chosen because of the multitude of commercially available protein-friendly products that uses the highly specific and extraordinarily strong biotin-avidin interaction (Ka = 1015 M-1). If successful, this would allow for a site-specific attachment of a fluorescent group, yielding a pure, labeled protein in two steps, without further processing.
23
Because of the physiological importance of the GSTs, a lot of work has been put into investigating the substrate requirements, both for GSH
80-82
and also for the
50
electrophiles . Generally, alterations in the C-terminal glycyl moiety of the tripeptide are more accepted by the GSTs than alterations in the N-terminal Ȗ-Glu part
81, 82
.
Therefore, a tripeptide scaffold, Ȗ-Glu-Cys-Cys, where the glycine residue was altered to a cysteine residue was synthesized with orthogonal protecting groups on the cysteines. A thiol-reactive and water soluble biotin linker was then coupled to the Cterminal cystein residue, whereas the other cystein residue was coupled to a set of four different fluorescent probes in parallel (Figure 3.13) to produce the biotinylated thioesters (GSC-thioesters). A particular biotinylated GĺC thioester will henceforth be named GSC-Xbio, where X is the corresponding acyl group. In addition to the GSCthioesters, an Į-NH2 acylated GSB analogue was synthesized and a sepharose-linked GS-Cou was also produced to investigate the possibilities of alterations in the Nterminal part of the tripeptides in the protein modification reactions. OCH 3
A
Fmoc
O
H N
1. β-mercaptoethanol/DMF 2. EZ-link PEO Iodoacetyl biotin DIPEA, DMF
SPPS O
S S
O
O
O HN
S N H
O
H N
O O
O
S S
O OCH 3
O
O
O HN
O
S
O
H N
N H
1. 1% TFA/DCM 2. 7-methoxycoumarin3-carboxylic acid DIPCDI, HOBt, DIPEA DMF 3. TFA/TIS/H 2O
O
O
O
O
H N
S
O
O
O
N H
NH
S N H
H 3C
O
O O
O -
O
O
O
O
S H N
N H
NH 3 +
O
Figure 3.13.
O O-
O
H N
S
O
O
O
Solid-phase
N H
NH
S N H
GSC-Cou bio
O
synthesis of the biotinylated GSC-thioesters
B
OH
using orthogonal
O
O
protecting N
O
O
O
OH N H O
Fluorescein
O
O
Ant
DMACA
24
groups.
The outcome of adding the biotinylated GSC-thioesters to hGST A1-1 and A216K was investigated by MALDI-MS. We found that hGST A1-1 could be modified at the Y9 residue with two of the C-terminally altered GSC-thioesters (GSC-Antbio and GSC-Coubio) and the A216K mutant reacted with GSC-Coubio. Following completed modification reaction, NA beads were added to the reaction mixture as a scavenger to remove excess and used up reagent. To analyze the scavenging step we used reversed phase HPLC (Figure 3.14), and found that the treatment with NA beads quantitatively had removed residual reagent. In contrast to the modification experiments where the NA beads had been pre-incubated with the GSC-thioesters, allowing for surfaceassisted delivery of the acyl group, we found several advantages performing the modification reaction in solution with subsequent scavenging. For example, the rate of the modification reaction was significantly higher in solution and it was easier to control the extent of the reaction in this format. The modification could also be carried out both in an E. coli lysate containing expressed A216K and in a buffered solution containing 1 mM GSH. A Buffer
B NA beads
NA gel after wash
A216K eluate 1600000
1200000 800000 400000 0 370
410
450
490
Figure 3.14. (A) A photograph under an UV lamp, Ȝ = 337 nm, showing the scavenging step with NA beads. A216K was incubated with an excess of GSC-Coubio followed by incubation with the NA beads, and then eluted by centrifugation. (B) HPLC chromatograms, Ȝ = 355 nm, showing the labeling mixture before and after the scavenging step (© 2006 American Chemical Society).
None of the Į-NH2-derivatized reagents were able to modify hGST A1-1 and A216K. It was concluded that alterations in this part of the GSH backbone are not viable routes to user-friendly labeling reagents, in line with previous studies
25
. The Ȗ-
81, 83
glutamyl moiety has been shown to be the most important recognition site on the GSH molecule 82-84. Two of the biotinylated GSC-thioesters were thus successful in labeling hGST A1-1. One of these (GSC-Coubio) was also a reagent for the A216K mutant. Two types of residues are thus targeted, and this increases the flexibility of the system. The wildtype system may be used if reversible modification is desired since the ester bond formed can be quickly and gently broken simply by addition of GSH. The A216K system provides a more stable linkage that resists GSH and is therefore well suited for use in fusion proteins in purification systems (Chapter 5). The introduction of a fluorophore into the active site of A216K also provides possibilities to use A216KCou in biosensor applications (Chapter 4).
26
4 " 1 53 4 "
1 53 This chapter starts with an overview of the field of biosensors, followed by a summary of the basic principles of fluorescence, which is the sensing technique that has been used in this work. Finally, the development of a protein-based, multipurpose biosensor is described.
4.1. The concept of biosensors A biosensor is a device that allows for the detection and quantification of biomolecules in a single step: the direct spatial combination of biological recognition and transduction into a recordable signal
85
. According to the classification of a
biosensor, one should distinguish between a biosensor and a chemical sensor. Although chemical sensors may be used to monitor biological processes, as the in vivo pH or oxygen sensors
86, 87
, they incorporate a non-biological specificity-conferring
88
part or receptor . Biosensors possess a biological recognition element, typically, an oligonucleotide, peptide, enzyme, receptor-protein, antibody, organelle or whole cell, and the main purpose of the recognition system is to provide the sensor with a high degree of selectivity for the analyte to be measured 88.
27
Antibody
DNA
Microorganism
Biological recognition system
Receptor Enzyme Interface Transducer type
Electrochemical Transducers e.g. - Amperometric - Potentiometric
Optical/Electronic Transducers e.g. - Surface Plasmon Resonance (SPR)
Optical Transducers e.g. - Absorbance - Fluorescence
Acoustic Transducers e.g. -Quartz Crystal Microbalance (QCM)
Figure 4.1. Schematic representation of biosensors.
In a biosensor, a biomolecular recognition event induces a change in a chemical or physical parameter (wavelength, frequency, temperature, pH, conductivity, refractive index, etc.), that is converted into a measurable signal by a transducer element. The signal transduction devices are generally divided into physical or electrochemical 85. Most biosensors are either enzyme-based or affinity-based, depending on whether the sensor signal originates from an enzymatic turnover of the analyte or from an affinity interaction between the analyte and the recognition element. Enzymes were historically the first molecular recognition elements included in biosensors, that were primarily developed for medical applications such as detection of glucose in blood 89, 90
. Although enzyme-based biosensors show significant promises for certain
environmental monitoring tasks, they also show several inherent limitations. The main 91
. Therefore, the fastest growing area in
limitation involves their lack of versatility
biosensors involves affinity-based biosensor techniques
92
. Numerous biosensing
techniques have been reported that allow researchers to better study kinetics, structure and solid/liquid interface phenomena associated with protein-ligand or protein-protein binding interactions 93-102. In recent years, there has been an increasing interest in the development of protein microarrays
103-106
. Assembling biosensors into microarrays
provides the possibility of simultaneous monitoring of many (thousands) biomolecular interactions on a small area (~1 cm2)
103, 107
. This allows for a high
throughput of analyses and at a reduced time and cost. In contrast to the successfully used DNA microarrays 108 (developed in the 1990s), microarrays of proteins are more
28
difficult to produce because of the complex nature of proteins compared to DNA. All together, tremendous efforts are being invested in the development of new biosensing techniques because of their applications in fields such as clinical/diagnostics, environmental
monitoring,
food
industry,
military/antiterrorism and nanotechnology
bioprocess
monitoring,
92, 109
.
The following part describes the achievements and prospects in the design and operation of a protein-based biosensor for which the transduction mechanism is based on fluorescence. The system involves the concepts of an affinity-based biosensor and is based on the coumarin-modified A216K mutant of hGST A1-1 (A216KCou).
4.2. Principles in the design of fluorescent molecular sensors Fluorescence is an ubiquitous luminescence phenomenon in which susceptible molecules (most often conjugated polycyclic aromatic molecules) emit light from electronically excited states created by either a physical (e.g. absorption of light), mechanical (friction), or chemical mechanism. When the molecules return to the ground state, the excess in energy is released as photons that are detected as fluorescence. The energy of the emission is typically less than that of absorption because of energy losses in the excited state through rotation and vibration of the molecule. Hence, fluorescence typically occurs at lower energy or longer wavelengths than the excitation wavelength. The shift in energy, and the resulting shift in wavelength maximum to longer wavelengths, is called the Stokes´ shift. In contrast to this conventional view of fluorescence, there is a relatively novel imaging technique in cell biology, called “Two-Photon Fluorescence Microscopy” (2PFM) 110. With this imaging technique, the sample is illuminated with wavelengths around twice the wavelength of the absorption peak of the fluorescent molecule used.
29
Excitation maximum )
Fluorescence emission (- - -)
Emission maximum
Fluorescence excitation (
Stokes´ shift
300
Wavelength of light (nm)
600
Figure 4.2. Definition of excitation, emission and Stokes´ shift.
The categories of molecules capable of undergoing electronic transitions that result in fluorescence are known as fluorescent probes, fluorochromes, or dyes. Probes that are covalently conjugated to a larger macromolecule (such as a nucleic acid, lipid, enzyme, or protein) are termed fluorophores. In general, fluorophores are divided into two broad classes, termed intrinsic and extrinsic. Intrinsic fluorophores, such as aromatic amino acids, porphyrins, and green fluorescent protein, are those that occur naturally
111
. Extrinsic fluorophores are synthetic dyes or modified biochemicals that
produce fluorescence with specific spectral properties. Several commonly used techniques within biological, biochemical, and biophysical sciences utilize the phenomenon of fluorescence, since it provides a simple, safe, and sensitive platform for analysis of conformation, folding and interaction. Many strategies based on fluorescence spectroscopy are available, including steady-state, time-resolved, and anisotropy measurements. In this study, three common parameters have been used in steady-state fluorescence experiments, and they are listed below. The simplest parameter is the fluorescence emission intensity. The intensity of fluorescence can be decreased by a vide variety of processes that are called quenching. A common example of quenching is observed with the collision of an excited state fluorophore and another (non-fluorescent) molecule (usually oxygen, halogens, amines and many electron-deficient organic molecules) in solution during the excited state lifetime, resulting in deactivation of the fluorophore and a return to the ground state. A second type of quenching mechanism, termed static quenching, arises from non-fluorescent complexes formed between the quencher and the
30
fluorophore that serve to limit the absorption by reducing the population of active, excitable molecules. For either static or collisional quenching to occur; the fluorophore and the quencher must be in direct contact. This method was used to measure the specific binding of analytes to our designed biosensor. A Wavelength shift of the emitted light for a fluorophore can occur in response to a change in its interaction with the molecular environment. The fluorescence wavelength is more red-shifted in a polar than in a hydrophobic environment. This occurs because the excited state has a larger dipole moment than the ground state, thus, after excitation, the surrounding solvent dipoles can relax around the excited state molecule, and this takes place more readily in polar solvents. Wavelength shift data was collected to get information of the micro-environmental surroundings of the fluorophore.
Fluorescence Resonance Energy Transfer (FRET) is a non-radiative process that should not be directly linked to the fluorescence process. Hence, this process is sometimes termed RET, i.e. the concept of fluorescence is excluded in the terminology because the process does not involve the appearance of a photon. FRET refers to a long-range dipole-dipole interaction between the electronic transition dipoles of the donor and the acceptor, and this phenomenon requires overlap of the absorption spectrum of the acceptor with the emission spectrum of the donor to occur. The distance at which 50 % of the donor emission energy is transferred to the acceptor is called the Förster distance (R0)
111
, and it is determined by the donor-acceptor pair
and is usually between 10-60 Å. Consequently, detectable FRET between two fluorophores, one (the donor) being initially excited and the other (the acceptor) being able to receive the energy of excitation and to transform it to its own emission, may be observed at separation distances of 10-60 Å and sometimes up to 100 Å, depending on the donor-acceptor pair. Hence, the method is very suitable for investigating conformational changes and quantifying distances within macromolecules
112, 113
. In
each monomer of A216K, a single tryptophan residue (W21) is positioned within 1015 Å (estimated from the crystal structure of wt hGST A1-1) from the introduced coumarin fluorophore at position 216, Figure 4.3.
31
W21 K216
13.6 Å
Figure 4.3. A close-up of the crystal structure of hGST A1-1 (PDB entry: 1GUH). A216 is replaced with K216. The distance between NH2 of the introduced K216 and W21 is 13.6 Å.
This, in combination with the spectral overlap between donor (W21) emission and acceptor (coumarin) absorption allowed for FRET experiments to be performed. This method was used to measure potential distance changes between W21 and coumarin upon analyte binding. Additionally, correction for the Inner Filter Effect is an important issue that was applied in the affinity studies discussed in part 4.4. A number of photophysical variables affect the accuracy of fluorescence measurements, and the most important in the context of equilibrium constant determinations is the absorption of the incident light or absorption of the emitted light. The relative importance of each process depends upon the optical densities (OD) of the sample at the excitation and emission wavelengths. If a significant part of the total absorption is due to the species, usually the ligand, whose concentration is varied in the course of the experiment; the absorption effect will distort the concentration dependence of the fluorescence, leading to an incorrect value for the equilibrium constant. Therefore, fluorescence intensities were corrected for the inner filter effect when the optical density exceeded OD = 0.05 111 at the wavelength of excitation and emission, using the following equation 111:
§ ODex + ODem · Fcorr = Fobs anti log¨ ¸ 2 © ¹
(4.1)
32
where Fobs and Fcorr are the experimental values and the corrected fluorescence intensities, respectively, and ODex and ODem are the optical densities of the sample at the wavelengths of excitation and emission, respectively.
4.3. Reengineered GSTs as Biosensors The introduced K216 is located in the C-terminal helix, in or near the H-site of the folded protein scaffold (Figure 3.10). The C-terminal helix (Į9) undergoes a conformational transition upon substrate binding
61-63, 114
, and we hypothesized that
K216Cou could report changes if the protein binds a molecule in the vicinity of the coumarin moiety. To test this hypothesis, A216KCou was used in pilot screening experiments together with GSH and a library of 12 hydrophobic molecules in a combinatorial fashion. The fluorescence spectra of 1 µM A216KCou with 0-500 µM of analyte in a buffered solution were recorded using Ȝexc = 280 nm (excitation of W21) and also Ȝexc = 355 nm (direct excitation of the coumarin fluorophore). It was found that, generally, addition of GSH or a hydrophobic molecule quenched the fluorescence of A216KCou to varying degrees, and small wavelength shifts were also observed. This was true both for the direct excitation of the coumarin fluorophore and excitation of W21 with concomitant resonance energy transfer to the coumarin. The ability of A216KCou to respond to the addition of different analytes opened up possibilities for biosensing. The human olfactory system has inspired researchers to develop an electronic nose based on relatively nonspecific sensor electrodes
115, 116
. The identification of a
particular compound is then based on mathematical interpretation of the signals from the array electrodes
117, 118
. Colorimetric sensor arrays for molecular recognition are
also common in biosensing applications
119-122
. We realized that this approach could
be extended to use proteins as receptors and that A216KCou is in fact exceptionally well suited for this type of experiment. The task was to design and create a multipurpose, protein-based affinity array that could sense and signal specific binding of different molecules where the readout could be interpreted through pattern recognition of fluorescence signals. Reports have shown that the contribution of the active-site M208 to the binding of the electrophilic substrate is of importance 123. It has also been shown that mutating M208 results in altered substrate preferences 123, 124. On the basis of this, a focused library of A216K/M208X mutants was made via random mutagenesis to provide an array of
33
proteins with altered micro-environments in the hydrophobic binding site (H-site). The idea was to deform the H-site to an appropriate degree, to obtain different substrate preferences. Too much would perhaps result in nonexistent binding in the Hsite and thus no analyte recognition or spoilage of the G-site with no labeling as a potential outcome. The drawback of too little deformation would be reduced diversity.
M208 K216
Į9
Figure 4.4. A close-up of the crystal structure of hGST A1-1 (PDB entry: 1GUH). A216 is replaced with K216. Positions K216 and M208 are shown in stick representation.
Eleven double mutants (A216K/M208L, F, S, Y, A, E, R, T, G, K and C) were isolated and purified, and all of the double mutants could be site-specifically labeled by GSC-Coubio at position 216 to form the K216Cou conjugates (Figure 3.11). A more detailed description of the characteristics of the double mutants is found in chapter 5. With the overall aim to develop a protein-based platform for biosensing, we wanted to (i) explore the signaling potential of the site-specifically labeled receptor family, (ii) investigate if the proteins were able to specifically bind a potential ligand in the active site and (iii) immobilize the labeled proteins on a hydrogel chip for surface-based biosensing applications, such as protein microarrays.
34
4.3.1. Signaling potential of the site-specifically labeled receptor family In order to explore the signaling potential of the receptor family and to find out what kind of information that could be extracted, a combinatorial screening experiment was set up with three analytes (n-valeric acid, fumaric acid monoethyl ester and litchocholic acid, Figure 4.5) and all 12 proteins (the double mutants and A216K) in a microplate format. These analytes were not known substrates for GSTs and two of them (n-valeric acid and fumaric acid monoethyl ester) had quite similar structures and sizes but were dissimilar compared to the third analyte (litchocholic acid). H
O OH
O
H
O OH
O
HO
H HO
O n-valeric acid
Fumaric acid monoethyl ester
H
H Lithocholic acid
Figure 4.5. Structures of n-valeric acid, fumaric acid monoethyl ester and lithocholic acid. The outcome of adding 200 µM of analyte to 1 µM (final concentrations) of each of the 12 labeled proteins was analyzed with respect to direct excitation of the coumarin probe at 355 nm and excitation of the tryptophan residue (W21) at 280 nm. The emission was monitored from 370 to 500 nm (the emission maximum (Ȝem,max) for coumarin was 408 nm without added analytes). Addition of GSH to GSTs can affect the binding of a second substrate in the H-site
58, 65, 67
, and therefore the equivalent
parameters were collected following addition of 200 µM GSH to each sample. In these experiments, both the changes in fluorescence intensities (ǻFCou, ǻFCou,FRET,
ǻFCou,GSH, ǻFCou,FRET,GSH) and the positions of the emission maxima (ǻȜem,max,Cou, ǻȜem,max,FRET, ǻȜem,max,Cou,GSH, ǻȜem,max,FRET,GSH) were documented. This resulted in a matrix of eight data points for each analyte and protein combination or 96 data points per analyte. The variation in the collected parameters between n-valeric acid and fumaric acid monoethyl ester was small, but detectable. This is not surprising since the two molecules are similar in size and structure. Adding lithocholic acid to the panel of proteins resulted in larger changes in fluorescence intensities and wavelength shifts, and the pattern of the collected parameters varied significantly compared to the results for any of the two smaller molecules, Figure 4.6. In order to simplify the
35
interpretation, the differences in changes between (i) fumaric acid monoethyl ester and n-valeric acid (Figure 4.7 A) and (ii) lithocholic acid and n-valeric acid (Figure 4.7 B) were calculated. These changes reported by a single protein or just a few proteins in an array would not amount to anything useful, but the combined effect from all 12 proteins was quite promising. For efficient use of this method, the data can be analyzed in a quantitative way, by for example PCA (Principal Components Analysis).
A
B
C
Figure 4.6. Graphic representation of the changes in fluorescence by adding (A) n-valeric acid, (B) fumaric acid monoethyl ester and (C) lithocholic acid to the panel of A216K/M208X mutants. The values on the y-axis correspond to the changes in the different parameters (in % compared to the proteins without analyte). A positive value means % quenching or % blue-shift.
36
A
Fumaric acid monoethyl ester – Valeric acid
B
Lithocholic acid – Valeric acid
Figure 4.7. Graphic representation of the changes in the eight measured fluorescence parameters between (upper) fumaric acid monoethyl ester and valeric acid and (lower) lithocholic acid and valeric acid. Valeric acid has thus been used as a normalizing substance. The values on the y-axis correspond to the absolute % values following subtraction of the changes in the parameters for valeric acid from the corresponding parameters for either of fumaric acid monoethyl ester or lithocholic acid. The mutants A216K/M208L and F were omitted in the investigation with lithocholic acid (© 2007 American Chemical Society).
37
4.4. Affinity studies The capacity of binding was investigated by measuring the dissociation constant (Kd) for a subset of analytes using conventional titration experiments and steady-state fluorescence spectroscopy. Using steady-state fluorescence as a tool to obtain information for a given protein-ligand interactions can in many cases be rather difficult and complex. Many parameters have to be taken into account to reveal correct information. In the present system, a protein was labeled with a fluorescent probe and the fluorescence signal (i.e. emission intensity and/or emission wavelength maxima) was expected to change upon analyte addition since the fluorophore was positioned near or in the active site in the folded protein. A proposed binding event was assumed to be bimolecular, i.e. one analyte per monomer. Each monomer of the dimeric enzyme hGST A1-1 is assumed to have an independently functional active site in the nucleophilic aromatic substitution reaction 125 , that is known as “all-of-the-sites” reactivity. This has also been suggested for GST A1-1 from rat 126. It has also been reported that both G-site- and H-site ligands bound noncooperatively to rGST-A1-1 with a stoichiometry of 2 molecules per enzyme molecule (1 molecule per subunit) 127. The binding of an analyte to a single site on a protein is described by the following equations: P + A ļ PA
(4.2)
At equilibrium, the dissociation constant Kd is defined by
Kd =
[P][A] , [PA]
(4.3)
where [P] is the concentration of free protein, [A] is the concentration of free analyte and [PA] is the concentration of the PA complex. High affinity is characterized by a low value of Kd. On the basis of previous studies 58, low micromolar to low millimolar affinities were expected. The Kd of GSH for wild-type hGST A1-1, measured by pre-equilibrium absorption experiments, was 190 ± 50 µM. The experiments were performed in black microplates with a clear bottom. Thus, information regarding the optical density, fluorescence intensity, and wavelength shift could be obtained in the same
38
experimental setup. During fluorescence spectroscopy measurements, the observed fluorescence Fobs is the weighted average of the fluorescence emitted from the free protein and from the protein bound to an analyte: Fobs =
Fbound [PA] + F free [P ]
(4.4)
[P]tot
where Fbound is the fluorescence of the protein bound to an analyte and Ffree is the fluorescence of the free protein. Equation (4.4) can be rearranged using (4.3) and (4.5):
[P ]tot = [P] + [PA]
(4.5)
where [PA] is the concentration of protein bound to an analyte and [P]tot is the total concentration of protein, giving the actual equation used for fitting to the experimental data:
Fobs =
Fbound ⋅ [ A] + F free ⋅ K d
(4.6)
[A] + K d
Here, Fobs is the observed fluorescence intensity, Fbound is the fluorescence of the protein bound to the analyte, Ffree is the fluorescence of free protein, and [A] is the concentration of free analyte. [A] can be expressed as:
[A] = − [P]tot + K d − [A]tot 2
§ [P ] + K d − [A]tot · + ¨ tot ¸ + K d ⋅ [A]tot 2 © ¹ 2
(4.7)
where [A]tot is the total concentration of analyte. The affinities (Kd-values) for GSH for seven tested mutants varied over a quite narrow range (72-179 µM). This was in line with our expectations because GSH is known to bind in the G-site and we had mutated the H-site. The affinity for 1-chloro-2,4dinitrobenzene (CDNB) varied between 601 µM for A216KCou/M208E to 6.8 mM for A216KCou/M208A. The larger variation for CDNB binding between the mutants was not surprising because position 208 is located in the H-site, where CDNB is known to bind. The nucleophilic aromatic substitution reaction between GSH and CDNB to form GS-DNB is catalyzed by wt hGST A1-1, and the proteins showed affinities for GS-DNB in between those of GSH and CDNB, respectively. S-hexyl-GSH, a wellknown inhibitor of the alpha class GSTs, showed the highest affinity, 2.5 µM for
39
A216KCou. An example of the affinity determination using steady-state fluorescence
Fluorescence intensity at 408 nm / a.u.
measurements is shown in Figure 4.8.
Fluorescence intensity / a.u.
50000
40000
30000
20000
3
45x10
40 35 30 25 20
0.0
0.2
0.4
0.6 0.8 [GSH]tot / M
1.0
1.2
-3
1.4x10
10000
0 380
400
420
440
460
480
500
Wavelength / nm
Figure 4.8. Emission spectra of A216KCou/M208F with increasing concentrations of GSH (excitation at 355 nm). The emission intensities at 408 nm were plotted against the total concentration of GSH and an equation that describes the dissociation of a bimolecular complex was fitted to the experimental data to obtain the Kd value, 72 µM (inset).
4.5. On the mechanism of binding of analytes by the proteins The most common result when a ligand was added to the panel of labeled proteins was quenching of the fluorescence of coumarin. This was true for both direct excitation of the coumarin probe at 355 nm and for excitation of W21 at 280 or 295 nm. There was a small variation in the degree of quenching of the coumarin fluorescence between the labeled A216K/M208X mutants, probably resulting from differing microenvironments in the binding site leading to different orientations of K216Cou and the analyte. In many instances a fluorophore can be quenched both by collisions (dynamic quenching) and by complex formation (static quenching) with the same quencher
111
. Static and dynamic quenching can be distinguished by their
differing dependence on temperature, and is preferably carried out by fluorescence lifetime measurements
111
. Another method to distinguish static and dynamic
quenching is by examination of the absorption spectra of the fluorophore. Complex formation frequently results in perturbation of the absorption spectrum of the fluorophore, whereas no changes in the absorption spectrum occur in dynamic
40
quenching
111
. For this setup, where steady-state fluorescence was used; it was not
possible to reveal information whether the apparent quenching effect was predominantly dynamic, static, or a combination of the both. No observable changes in the absorption spectrum of coumarin were detected upon analyte addition, and the binding curves were not affected by differing temperatures. For some of the analytes, the effect was the opposite i.e. an increase of the coumarin fluorescence. This was particularly true for S-hexyl-GSH, which presented a binding curve where the fluorescence intensity was increased with increasing concentration of S-hexyl-GSH (Figure 4.9). The reason for the intensity increase is not clear at a photophysical level, but one suggestion is that S-hexyl-GSH lacks functionalities capable of quenching the coumarin probe and potentially shields the coumarin probe from intramolecular quenching moieties.
Normalized fluorescence at 408 nm
1.0 0.8 0.6 0.4 0.2 0.0 -8
10
-7
10
-6
-5
10
10
-4
10
-3
10
[analyte]free / M
Figure 4.9. Binding curves obtained by plotting normalized fluorescence intensity at 408 nm versus the concentrations of free analytes, [analyte]free. The concentrations were calculated from the Kd values. A216KCou with S-hexyl-GSH (+, Kd = 2.5 µM), GSH (ǻ, Kd = 108 µM), GS-DNB (Ƈ, Kd = 347 µM), CDNB (ź, Kd = 1.3 mM) (© 2007 American Chemical Society).
One suggestion is that a combination of mechanisms gives rise to the different fluorescence effects. The occurrence of quenching or a “light-up” effect depends upon the mechanism, which in turn depends upon the chemical properties of the individual molecules. In the presented system, the coumarin probe is positioned in the flexible Cterminal helix, which is presumed to undergo a conformational transition upon substrate binding. Because of this flexibility, and that the chemical properties differ between the analytes, it is not surprising that the apparent fluorescence effects differ between the analytes.
41
The interpretation of the apparent mechanism of energy transfer from W21 to the coumarin probe upon analyte addition was somewhat complex because competing mechanisms between quenching and energy transfer were observed. Also, the influence of the inner filter effect added an obstructing parameter that complicated the data analysis. Because of the mentioned complications, and also because of the short distance between W21 and the coumarin fluorophore, it was not possible to observe any distance changes upon analyte addition. The tryptophan emission alone was not observable when the coumarin fluorophore was attached to K216, i.e. the efficiency of energy transfer was close to 100%. According to the Förster theory
111
, the transfer
efficiency is strongly dependent on the distance when the donor-acceptor distance is close to R0. The Förster distance for tryptophan-coumarin derivates was not found in the literature but 20-25 Å should be a good estimate 128. Hence, in the current system, only large distance increases upon analyte addition would be detectable. In this work, “FRET” should be interpreted as a combined parameter which could be utilized within the pattern recognition analysis for extracting as much information as possible from the experimental array.
4.6. Immobilization of A216KCou on a hydrogel surface Studies of biomolecular interactions in general, provides two fundamentally different approaches. Either all participating molecules are free in a solution or one of the interaction partners is immobilized on a surface. Most biosensors are based on biomolecules bound to a surface, which is often part of the transducer element. The general advantages of surface immobilization are that it allows convenient washing of the solid support without loss of the capture molecule, and also allows for surface regeneration which reduces sample consumption and processing time. When immobilizing fragile biomolecules (e.g. proteins), it has been shown that by coating the surface with a PEG-based hydrogel matrix, problems of denaturation and nonspecific binding associated with adsorption are dramatically reduced
129-131
.
Additionally, the PEG-based hydrogel promotes native interactions in a solution-like environment, and provides a 3D architecture that offers the potential of higher loading capacities and thus an increased response. The hydrogel surface used in this work was a patterned PEG-COOH matrix where carboxyl groups provided handles for sitedirected, covalent attachment of biomolecules. A detailed description of the preparation of the array surface is given in Larsson et al. 132.
42
An advantage with human GST A1-1 is that the protein has one single, and reduced, cysteine residue (C112) per monomer that lies at the surface of the protein, adjacent to the binding site. It has been reported that chemical modification of C112 has no effect on the catalytic turnover of hGST A1-1 133. This residue is therefore ideally located to serve as an orthogonal reactive handle for either, probes that can function as a “FRETpair” together with the coumarin fluorophore, or specific surface immobilization. Leaving the C112 in A216KCou intact, thus provided for site-specific immobilization of the protein on a PDEA activated surface. As a complement, an NHS-activated surface was also used for attaching the protein via nucleophilic amino acids (Figure 4.10). O
Cl H+ N Cl
N C N
O
-
HO N +
NH O
O
EDC
NHS
N O C HN
OH
A216KCou
NH 2
O
a)
HN
O
A216KCou
O
O N O
S
H 2N
b)
S
N
O
PDEA
HN S S N
Figure 4.10. Reaction scheme for activation of surface carboxyl groups with N-hydroxysuccinimide (NHS), mediated by N-ethyl-N´-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC) and subsequent immobilization of A216KCou (a) via nucleophilic amino acid residues on the protein or (b) via Cys112 subsequent to activation by 2-(2-pyridinyldithio)ethylamine (PDEA).
43
A216KCou
SH
O HN S S
A216KCou
The immobilization was examined by IR-spectroscopy, proteolytic digestion of the protein whilst attached to the surface, and fluorescence microscopy. Indeed, it was possible to immobilize A216KCou using either of the methods described (Figure 4.11). The higher intensity of the emitted light from the spots when A216KCou was immobilized on the NHS-activated surface indicated higher protein concentration. A possible reason can be that the protein contains many NHS-reactive residues and thus, due to the relative slow molecular tumbling of the protein, enables more successful reactions per unit of time, compared to the single-point attachment via C112. These experiments were an initial attempt to explore the potential of using A216KCou, or mutants of it, in surface-based biosensing systems. OH
OH
B
A HO
O C
S S
HO
NH
A216KCou
O HOOC
HOOC COOH
COOH
OH
OH
HOOC
HOOC OH
A216KCou
A216KCou
HN
OH
O HN C
A216KCou
S
S
H N O
HO
HO OH
OH
Figure 4.11. Schematic illustration of the immobilization of A216KCou on a PEGbased biosensor chip and fluorescence microscopy. By use of masks, arrays of matrix spots were grafted onto a formed thiol self-assembled monolayer (SAM). (A) immobilization via NHS-reactive groups on the protein and (B) site-specific immobilization on a PDEA-activated PEG-matrix via Cys112. At the bottom are the pictures from fluorescence microscopy following immobilization.
44
4.7. Conclusions The naturally occurring enzyme hGST A1-1 has in a three step iterative procedure been turned into a multipurpose protein-based biosensor, using protein engineering and chemical modification. This was possible by taking advance of the benefits of the GSTs, i.e. promiscuous binding sites, high resistance to mutations and a site-specific attachment point for surface immobilization in combination with our novel sitespecific route that targets a single lysine. This work constitutes proof of principle, demonstrating that the ideas of the electronic nose and tongue could be transferred into the protein world.
45
46
6 ' ; +0 1 53 1 53 In parallel with the work on paper IV, a more thorough study of the A216K/M208X mutants was performed with the aim of finding a candidate that could improve on GST-based fusion systems. A screening protocol was set up, where several parameters were measured for all individual mutants, such as expression level, specific activity, thermal stability, and rate of chemical modifications both in an E. coli lysate directly after expression and in buffered solution (purified protein).
Improvement of the technologies for protein production and purification are always welcomed in the maturing field of proteomics. The GST gene fusion system is a well established system for expression, purification and detection of target proteins and peptides
134-138
. A vector encoding for a GST fusion tag with an integrated protease
site allows for convenient expression of the target protein, and subsequent purification by glutathione resins. The GST carrier can then be cleaved from the fusion protein by digesting with a site-specific protease such as thrombin or blood coagulation factor Xa. This chapter describes the development of an alternative method based on protein design for (i) improving expression yields, (ii) effective detection and (iii) purification of GST fusion proteins. By using our unique protein labeling kit with which a lysine residue can be covalently linked to an orthogonal group not present in other proteins, both in E. coli lysates and buffered solutions, we envisioned that the produced A216K/M208X protein library could expand the properties of a GST-fusion system. Introduction of a fluorescent probe in an E. coli lysate would provide a novel and sensitive colorimetric detection assay for monitoring the level of expressed protein, and introduction of, for example,
47
an aldehyde moiety would provide for alternative purification schemes. A selection protocol was carried out to find candidates from the A216K-mutants that subsequently could be used in protein purification schemes using GSTs as fusion partners. Good candidates should be easily expressed and purified in high yields, be thermally stable, be able to bind GSH and GSH-derivatives, be purified by standard GST-GSH protocols, and catalyze the standard CDNB reaction.
5.1. Construction and characterization of the A216K/M208X mutants The vector containing the hGSTA1-1 mutant A216K gene was used as a template for random replacement of M208 by site-directed mutagenesis. Twenty clones were isolated and sequenced to produce eleven double mutants (A216K/M208S, A, Y, F, G, T, L, R, C, E, and K). In these semi high-throughput preparations, the cells were lysed through freeze-thawing. Following purification, and determination of the expression level, the A216K/M208X mutants were characterized with respect to their (i) specific activity towards CDNB, (ii) thermal stability, (iii) affinity towards GSH, and (iv) their propensity to react with four different labeling reagents (GSB, GS-Cou, GSC-Coubio, and GS-Al) in both E. coli lysates and buffer solutions.
O
O
-
OOC
O S
O NH
GSC-Coubio
O H N
NH 3+
CH 3
O
COO -
O NH
S
O
O
NH
O
NH
S N H
O
O
O S
O -
OOC
NH NH 3+
O
GSB
O NH
COO
-
O
O
O CHO
GS-Cou
GS-Al
Figure 5.1. The structures of the synthesized glutathione-based thioesters used in this study.
48
5.1.1 Specific activity and stability All of the produced A216K/M208X mutants showed catalytic activity towards CDNB, and demonstrated that the protein tolerated the double mutations. A216K had a 3-fold lower specific activity compared to the human wild-type enzyme, but surprisingly, two of the double mutants (A216K/M208F and A216K/M208C) were actually better catalysts than the single mutant A216K. Activity measurements were also utilized to obtain information of the stabilities of the proteins, and demonstrated that the most of the proteins could be stored at pH 7, and room temperature, for 4 days while retaining more than 60 % activity (Table 5.1).
Table 5.1: Yields, specific activities and stabilities of the produced double mutants. Mutant
A216K A216K/M208S A216K/M208A A216K/M208Y A216K/M208F A216K/M208G A216K/M208T A216K/M208L A216K/M208R A216K/M208C A216K/M208E A216K/M208K
Yield
a
Specific activity
b
Stability
In % using A216K as a reference
Towards CDNB + GSH (µmol*min-1*mg-1)
Remaining CDNB activity over 4 days (%)
100 26 59 72 26 77 103 99 23 74 38 37
27.4 ± 3.5 8.1 ± 0.3 8.0 ± 0.8 9.9 ± 0.9 35.6 ± 2.4 3.1 ± 0.2 12.6 ± 3.2 21.2 ± 0.2 2.6 ± 0.2 30.7 ± 1.7 19.1 ± 0.8 2.6 ± 0.1
91 93 56 68 71 87 61 83 119 6 65 58
(a) The yield of pure A216K from a one-liter culture was 18.8 mg under these circumstances. (b) The mean values of two ore more measurements
5.1.2. Labeling propensity and GSH affinity The labeling reactions were tested towards all mutants in E. coli lysates and buffer solution, using four different labeling reagents. Time-points were withdrawn from the labeling mixture for proteolysis, and subsequently analyzed by MALDI-MS in a semiquantitative fashion by dividing the m/z peak resulting from the modified fragment containing K216 by that of the respective unmodified + modified fragment. The outcomes of the site specific-labeling reactions are summarized in Table 5.2. The affinities for GSH to the coumarin-labeled mutants ranged between 72 µM – 179 µM, as measured by fluorescence spectroscopy (chapter 4).
49
Table 5.2: Modification reactions of the produced double mutants. a
a
a
a
Mutant
GSB % reacted after 4 h (E. coli lysate)
GS-Cou % reacted after 8 min (E. coli lysate)
GSC-Coubio % reacted after 5 h (pure proteins)
GS-Al % reacted after 1 h (E. coli lysate)
A216K A216K/M208S A216K/M208A A216K/M208Y A216K/M208F A216K/M208G A216K/M208T A216K/M208L A216K/M208R A216K/M208C A216K/M208E A216K/M208K
73 74 58 28 100 53 60 30 n.m. 36 50 36
49 100 100 56 63 61 85 100 37 100 n.m. n.m.
97 100 100 27 100 94 79 32 85 80 100 86
71 63 n.m. n.m. 65 n.m. 47 22 n.m. n.m. 54 n.m.
(a) Semi-quantitative calculations from MALDI-MS.
5.2. Summary It turned out that the parent protein A216K and the double mutant A216K/M208F performed best when summarizing all tested parameters (Table 5.1 and 5.2). They could be quickly and quantitatively labeled with a coumarin probe in E. coli lysates. As shown in Paper III for coumarin-labeled A216K, the emission intensity from fluorescence measurements is dependent on the protein concentration (Figure 5.2) and the technique can thus be used to monitor the level of expressed protein. These mutants were also the best catalysts for the conjugation reaction between GSH and CDNB, and they could be stored in room temperature for several days with only a small loss of activity. Their affinities towards GSH were in micromolar range, and they could easily be purified using immobilized GSH on sepharose matrix. Finally, these candidates could be labeled with an aldehyde moiety, providing alternative orthogonal purification schemes using, for instance, immobilized hydrazide derivatives as capture molecules.
50
800000
1000000
Unknown
Fluorescence intensity / a.u.
RFU
600000
800000
10 µM (known) 9 µM
400000
200000
600000
0
400000
0
5 µM
0,4
0,8
1,2
[A216K] (µM)
200000
1 µM (known) 0 370
410
450
490
Wavelength / nm
Figure 5.2. Fluorescence spectra (Ȝexc = 360 nm) resulting from incubating two known (1 and 10 µM) and two unknown (5 and 9 µM) concentrations of A216K with GSC-Coubio (100 µM) for 5 h with subsequent treatment with NA beads. The inset shows the intensity of emission at 407 nm plotted versus concentration of protein. The two unknown samples are highlighted with arrows (© 2006 American Chemical Society). As can be seen in Tables 5.1 and 5.2, there are possibilities to choose mutant candidates other than A216K and A216K/M208F, depending on the desired performance. A drawback with A216K/M208F was the relative poor yield. For instance, A216K/M208T could be expressed in higher yields than A216K/M208F, but in turn, it had lower specific activity, stability, affinity for GSH, and it showed slower modification rates. If the focus instead would be on modification rates, A216K/M208S would be an interesting candidate.
5.3. Outlook The results in this chapter show that, based on rational design, using site-directed mutagenesis and subsequent site-specific labeling, it can be possible to improve existing GST gene fusion systems. It would be very interesting to fuse the gene of A216K or A216K/M208F to a fusion partner of interest and evaluate the outcome of expression, labeling, and purification. If successful, there are several parameters that can be fine-tuned. It should be mentioned that a major part of the experiments were done in a combinatorial fashion to prove our ideas, and even though they were done
51
carefully, many steps from cloning to purification could be optimized utilizing alternative methods and reagents.
52
7 0 ; 7 0 ; Paper I •
A novel method of site-specific introduction of non-coded groups in the Alpha class GSTs was invented.
•
One single tyrosine residue (Y9) out of 51 possible nuclephiles (in hGST A11) was targeted for labeling.
•
The yield of the modification reaction was close to 100 % in less than one hour under optimal conditions.
•
It was possible to modify hGST A1-1 in E. coli lysates
Paper II •
A set of 17 GS-thioesters with alternating acyl groups were synthesized and screened against the human Alpha class GSTs to explore the reagent requirements in the modification reaction.
•
The modification reaction was versatile with respect to all human Alpha class GSTs (A1-1, A2-2, A3-3 and A4-4) and the range of accepted GS-thioesters varied from 72 to 83 %.
•
The set now included fluorescent probes and a unique handle (an aldehyde) for chemical derivatization.
Paper III •
To further improve the ease of handling of the glutathione-based labeling reagents, biotinylated reagents (GSC-Xbio) were designed and synthesized.
•
Two of the biotinylated GSC-thioesters could deliver the acyl group to Y9 in wild-type hGST A1-1 or K216 in the A216K mutant of hGST A1-1, creating a stable amide bond.
•
The reaction was quantitative, fast, and could be performed directly in an E.
coli lysate after expression. •
Streptavidin-coated agarose beads could then be used to mop up excess and used up reagent following completed reaction, and it was also possible to modify the proteins while the reagents were attached to the beads.
53
Paper IV •
The ideas of using advanced pattern recognition were applied to construct a protein-based multipurpose biosensor, i.e. an attempt to mimic the function of the olfactory system.
•
The hydrophobic binding site (H-site) of the hGST A1-1 mutant A216K was scrambled through mutations at position 208 to alter the micro-environment, and a library of 11 A216K/M208X mutants were isolated.
•
All double mutants could be site-specifically modified at K216 by a fluorescent probe.
•
Detection of analyte binding occurred through pattern recognition of fluorescence signals.
•
The analytes bound in a specific manner in the binding pocket and the affinities of a subset of analytes for the labeled proteins were measured. The affinities varied between 2.5 µM to 6.8 mM.
•
The labeled proteins could be site-specifically immobilized on a PEG-based biosensor chip via residue C112 on the protein surface.
Paper V •
Improvements on protein purification schemes that are using GSTs as fusion partners are presented.
•
Several parameters for the produced A216K/M208X library were tested, such as expression yield, thermal stability, specific activity, and modification propensity.
•
The specific introduction of a fluorescent probe in E. coli lysates directly after expression provided a quick and sensitive monitoring of expression yields, and labeling with an aldehyde moiety provided a specific handle, not present in other proteins expressed in E. coli, for orthogonal protein purification.
•
Two of the candidates from the protein library performed well in all tested parameters and could possibly improve on existing GST fusion systems.
54
8 ; ; +0 !2 +0 !2! Since the early-1990s, chemical synthesis has emerged as a powerful technique to generate proteins with novel properties 139-142. The total synthesis of protein constructs enables complete control over design and incorporation of non-coded elements at any position in the protein. This is not possible by recombinant DNA-based methods that are subject to the limitations of ribosomal synthesis of the polypeptide chain. However, considerable progress has been made in expanding the number and nature of genetically encoded amino acids in E. coli, yeast and mammalian cells in the recent years
143
. By means of a unique codon and corresponding tRNA/aminoacyl-tRNA
synthetase pair, Schultz and co-workers have incorporated more than 30 unnatural amino acids into proteins. The chemical synthetic strategies are however considered as simpler and more straightforward, and promise the unlimited variation of the covalent structure of a polypeptide chain
144
. This chapter describes an unfulfilled
opportunity to synthetically reconstruct hGST A1-1, by applying two different designed synthetic strategies.
7.1. Native Chemical Ligation Native chemical ligation (NCL) is the most practical and robust method for the chemoselective reaction of unprotected peptides. Native chemical ligation has become established as an effective method for the synthesis of large polypeptides that fold efficiently to form active protein molecules
18, 145-147
. So far, more than 300
biologically active proteins from more than 20 different families have been successfully prepared by total chemical synthesis with this method
144
. In the NCL
procedure, a peptide containing a C-terminal thioester reacts with another peptide
55
containing an N-terminal cysteine residue in aqueous solution at neutral pH, and this results in the formation of a native peptide bond between the two peptide segments 139 (Figure 7.1).
Figure 7.1. Native chemical ligation (NCL). Unprotected peptide segments are reacted by means of reversible thiol/thioester exchange to give a thioester-linked reaction product, followed by an irreversible intramolecular rearrangement forming a polypeptide product that is linked by a native amide bond.
Coupling long peptides by this technique is in many cases nearly quantitative and provides synthetic access to proteins otherwise impossible to make, and in particular, it provides almost inexhaustible opportunities to decorate the protein with artificial groups at any position
147
. The designed Synthetic strategy A is based on NCL.
However, most chemical protein syntheses by NCL have involved the joining together of just two unprotected peptide segments
139, 148-152
. Sequential NCL in solution of
three or more peptide segments involves additional chemical manipulations directed by the need for temporary protection of the middle segments, each of which carries chemoselective reactive functionalities
153
. The native chemical ligation method has
been developed to apply the principles of polymer-supported organic synthesis, establishing access to much larger polypeptide chains by performing sequential solidphase ligations of unprotected peptide segments
154
. Solid-phase chemical ligation
(SPCL) has all the advantages of traditional solid-phase organic chemistry. Specifically, the SPCL method allows for rapid purification of the intermediate polymer support-bound ligation products and facilitates the changes in solvent and other conditions necessary for different steps of the synthetic cycle. In addition, potential solubility problems for the intermediate product polypeptides will be reduced because of their attachment to a polymeric support and this also reduces the possibilities to form aggregates
155
. The designed Synthetic strategy B is based on
SPCL.
56
7.2. Synthetic Strategy A (NCL) The 221 residue amino acid sequence of hGST A1-1 (without the N-terminal methionine residue) was divided into four polypeptide segments (amino acids 1-45, 46-114, 115-167 and 168-221) (Figure 7.2). The chosen cleavage sites excluded residues known to be important for structure and function of hGST A1-1.
A.
B.
C.
D.
1
AEKPKLHYFN 11ARGRMESTRW 21LLAAAGVEFE 31EKFIKSAEDL
41
DKLR(NĺD)-SBzl
46
(DĺC(Msc))GYLM 51FQQVPMVEID 61GMKLVQTRAI
81
YGKDIKERAL 91IDMYIEGIAD 101LGEMILLLPV 111C(Acm)PPE-SBzl
71
LNYIASKYNL
115
(EĺC(Acm))KDAKL 121ALIKEKIKNR 131YFPAFEKVLK 141SHGQDYLVGN
151
KLSRADIHLV
168
(EĺC)LD 171SSLISSFPLL 181KALKTRISNL 191PTVKKFLQPG 201SPRKPPMDEK
211
SLEEARKIFR 221F
161
ELLYYVE-SBzl
Figure 7.2. The four synthesized polypeptide segments.
For segment A, the C-terminal asparagine was changed to aspartic acid. The aspartic acid (segment A) and the glutamic acid (segments B and C) were coupled to the resin via their ȕ-carboxy group and Ȗ-carboxy group, respectively. A resin yielding a Cterminal amide upon resin detachment was used for segments A and D. For segment A, this means that the introduced aspartic acid will again become an asparagine upon resin detachment. Segments B and C were synthesized on a resin yielding a Cterminal acid upon resin detachment. The N-terminal aspartic acid (segment B) and glutamic acid (segments C and D) were changed to cysteine residues. The introduced cysteine residues in segments B and C were orthogonally protected with a methylsulfonylethyloxycarbonyl (Msc) and S-acetamidomethyl (Acm) group, respectively. Also, the cysteine residue 111 (segment B) was protected with the Acm group. The thioester at the C-terminus of segments A, B, and C, were made on the
57
resin following completed peptide synthesis and orthogonal deprotection of the Įcarboxy groups. The designed scheme for the total synthesis of hGST A1-1 was based on sequential consecutive thioester-mediated native chemical ligations (Figure 7.3), followed by alkylation of the cysteine residues at positions 46, 115, and 168 with bromoacetic acid to form non-coded amino acid side chains, similar to glutamic acid (Figure 7.4).
D
C
!"#$ % ! &'$
D
C
B
( (
) ) )
* !"#$ % ! ) &'$
C
B
) ) )
D
A )
!"#$ % !
A
B
)
C
D
) ) )
+,+#$ -+.$"
Figure 7.3. Scheme of strategy A for total synthesis of hGST A1-1 by NCL.
/
A
Figure 7.4. Alkylation of the cysteine residues at positions 46, 115 and 168 with bromoacetic acid forming non-coded amino acid side chains similar to glutamic acid.
)
B
/
A
58
/
D
)
C
) ) )
B ) ) )
C
D
7.2.1. Results Four polypeptides were successfully synthesized using Fmoc-based solid-phase peptide synthesis (SPPS), and purified with HPLC. The resulting peptide masses from MALDI-MS analysis (Figure 7.5) were in good agreement compared to the theoretical masses.
V
o
90
a
g
e
r
S
p
e
c
#
1
[
B
P
=
5
3
0
8
.
1
,
5
4
4
6
]
V
5446.0
*
A
90 80
70 60 50
5178.42
40 30
g
e
r
S
p
e
c
#
1
=
>
N
F
0
.
7
[
B
P
=
8
1
8
5
.
4
,
1
3
7
8
4
]
8184.13
1.4E+4
*
B
70 60 50 40
10
0 1000
2800
4600
6400
8200
0 10000
0 6000
6800
M as s (m /z)
V
o
y
a
g
e
r
S
p
e
c
#
1
=
>
N
F
0
.
7
[
B
P
=
6
3
8
4
.
8
,
2
7
2
3
7
]
80 % Intensity
6285.44
50 40 30 20
y
a
g
e
r
S
p
e
c
#
1
=
>
N
F
0
.
7
[
B
P
=
6
1
5
9
.
7
,
3
9200
5
0
9
0 10000
]
6160.94
3508.5
*
D
70 60 50 40
2879.01
30
3193.09
6218.58
20
10 0 1999.0
o
100 90
70 60
8400
V
2.7E+4
*
C
7600 M as s (m /z)
6385.42
100
% Intensity
a
20
10
80
y
30
20
90
o
100
% Intensity
% Intensity
80
y
5308.36
100
10 3599.4
5199.8
6800.2
8400.6
0 10001.0
M as s (m /z)
0 1000
2800
4600
6400
8200
0 10000
M as s (m /z)
Figure 7.5. MALDI-MS spectra of the four synthesized polypeptides. (A) corresponds to segment A (aa 1-45), calculated m/z: 5306 D, (B) to segment B (aa 46-114), calculated m/z: 8182 D, (C) to segment C (aa 115-167), calculated m/z: 6387 D and (D) to segment D (aa 168-221), calculated m/z: 6160 D.
The ligation reaction between segments C and D was monitored by analytical HPLC (Figure 7.6), and it was essentially complete after 48 h. The ligated product was purified with HPLC and the product mass was identified with MALDI-MS (Figure 7.7). Unfortunately, the oxidized D+D could not be satisfyingly separated from the product.
59
Ligated C+D
C
PhSH
D t = 48 h t=2h
15
10
20
25
30
minutes
Figure 7.6. Native chemical ligation between segments C and D monitored by analytical HPLC (UV profiles at 220 nm are shown). The reaction was performed in 6M GuHCl in 0.1M NaPi, pH 7.1, with 2% (v/v) thiophenol (PhSH). Segment C was added in three aliquots (t=0, t=5h, and t=24h, total molar ratio of 1.5:1 between C and D). In the upper trace (t=48h), PhSH has been extracted away.
V
100
Unreacted D
o
y
a
g
e
r
S
p
e
c
#
1
=
>
N
F
0
.
7
=
>
N
F
0
.
7
[
B
P
=
1
2
4
1
3
.
1
,
4
3
1
9
]
Ligated product C+D
12412.19
4318.8
90
% Intensity
80 70 60 50 40
Oxidized D+D
12314.55
6208.03 Ligated product m/2+
6159.67
30 20 10 0 5000
7000
9000
11000
13000
0 15000
M as s (m /z)
Figure 7.7. MALDI-MS spectrum of the purified product after ligation between segments C and D.
The next step was to selectively remove the Acm group of the cysteine residue at the N-terminus of the ligated product (C + D). Despite several tested deprotection mixtures, this was found to be rather difficult, and yielded at most only 20 – 30 % Acm-deprotected product, determined from semi-quantitative analysis by electrospray ionization mass spectrometry (ESI-MS) (Figure 7.8). The reason for this was probably that the polypeptide formed aggregates that resulted in a buried cysteine residue.
60
+17 +16
+15
+18 +14
*
*
+13
*
*
*
+12
* *
+11
* m/z
Figure 7.8. ESI-MS spectrum following Acm deprotection of the cysteine residue at the N-terminus of the C+D segment, showing multiple protonation states of the polypeptide. The peaks highlighted with asterisks correspond to Acm-deprotected polypeptide and the peaks with outlined protonation states correspond to the polypeptide with the Acm group.
A small scale test to couple the C + D polypeptide to B was carried out. The ligation reaction was followed by HPLC and a peak was observed that corresponded to the ligation product between the two polypeptides, as determined by MALDI-MS (data not shown). However, at this stage it became hard to interpret the results, again, probably because of aggregation of the polypeptides that resulted in broad peaks, both in the HPLC and the MALDI-MS experiments. The conditions during the Acm removal, the ligation reactions, and the HPLC runs could be improved by including 10-20 % trifluoroethanol (TFE) that is known to interrupt formation of aggregates in concentrations above 10 %.
61
7.3. Synthetic strategy B (SPCL) This strategy was developed to be able to carry out the consecutive peptide ligations of unprotected peptide segments, whilst the C-terminal peptide was attached to a water compatible polymeric support. Unfortunately, this strategy was not applied in the lab due to instrumental problems at that time, but is presented here as an opportunity for total synthesis of hGST A1-1, providing possible improvements compared to the synthetic strategy A. A short summary of the synthetic strategy B: •
Fmoc-Lys(Alloc)-OH (a lysine residue that is orthogonally protected at the side chain with an Alloc group) is attached to the solid support.
•
The amino group is activated with bromoacetic acid following Fmoc removal, and the first amino acid of segment D (Fmoc-Phe-OH) is then coupled to the
ȕ-carbon via its carboxy group. The rest of segment D is then synthesized by SPPS. •
Following Alloc removal, levulinic acid is coupled to the side chain of the Cterminal Lys residue, providing an orthogonal ketone handle.
•
The synthesized peptide is cleaved from the solid support and purified with HPLC.
•
A water-compatible solid support is reacted with N-Boc-aminooxyacetic acid, the Boc group is removed, and the purified segment D is coupled to the solid support via its keto group by oxime-forming chemoselective reaction.
•
Segment C, B, and A is then ligated consecutively by SPCL with orthogonal deprotection of the Cys-protecting groups in between.
•
Following alkylation of the three inserted Cys residues (changed from Glu or Asp to Cys) with bromoacetic acid and removal of the Acm group of the intrinsic Cys 112, the synthesized polypeptide is cleaved from the solid support by aqueous base, purified, and eventually folded in buffer soulution.
62
Fmoc NH
NH 2
+
CO 2 H
NH(Alloc) 1. Fmoc deprotection 2. Coupling of Bromoacetic acid O HN
Br
NH
O
NH(Alloc)
Coupling of Fmoc-Phe-OH O
O
Fmoc HN
HN
O
NH
O
NH(Alloc)
1. Fmoc deprotection 2. SPPS O Cys(Acm)
HN
O HN
O
NH
O 1. 2. 3. 4.
NH(Alloc)
Alloc deprotection Coupling of Levulinic acid TFA cleavage HPLC purification
O Cys(Acm)
HN
O O
HN
O
NH 2
O
+
O
H 2N O
NH
NH O
1. Coupling with N-Boc-aminooxyacetic acid to the resin 2. Boc deprotection 3. Oxime ligation
O Cys(Acm)
HN
O HN
O
NH 2
O 1. 2. 3. 4. 5. 6. 7.
O
NH
N NH O Acm deprotection SPCL O Alkylation w. bromoacetic acid Acm deprotection of cys 112 Cleavage from resin with aqueous base HPLC purification Folding
Figure 7.9. Reaction scheme of the total synthesis of hGST A1-1 by SPCL.
63
64
När jag för tio år sedan flyttade från Alingsås till Linköping för att börja studera kemi trodde jag nog inte att jag en dag skulle bli doktor. Men intresset för ämnet bara ökade med åren, mycket tack vare den höga kvalitén på kemiutbildningen i Linköping. Nu har jag (snart) nått fram till doktorstiteln, och det är inte utan all fantastisk hjälp och support som många människor har bidragit med. Såklart finns det en massa personer jag skulle vilja tacka för att på ett eller annat sätt bidragit till att denna avhandling blivit verklighet… …först och främst min fantastiska handledare Kerstin Broo! Jag är otroligt lycklig för att jag fick genomföra mina forskarstudier med dig som handledare. Du har varit ett grymt stöd och pushat på i både med- och motgång, alltid funnits tillgänglig när det behövts och gett mycket positiv feedback. Du har gett mig stora möjligheter att utvecklas som självständig forskare som jag kommer att ha stor nytta av i framtiden. Ärligt talat har jag svårt att se hur en handledare skulle kunna vara bättre än vad Kerstin varit. …ett stort tack till min biträdande (eller hur man nu ska säga) handledare Nalle Jonsson! När Kerstin flyttade till Göteborg så har du klivit in och ”tagit hand om mig” på många sätt. Även fast du är en upptagen man så har du alltid tagit dig tid för diskussioner och hjälpt till när det behövts. Du har varit suverän! …jag vill tacka Lars Baltzer, som var min handledare under examensarbetet och som bidrog till att jag fick möjlighet att påbörja mina forskarstudier. Du väckte mitt intresse för forskning i gränsområdet mellan organisk kemi och biokemi, och du har bidragit till att jag kommit dit jag står idag. …under mitt examensarbete hade jag förmånen att arbeta tillsammans med Gunnar Dolphin. Du var en suverän handledare och mentor! Jag lärde mig enormt mycket kemi under halvåret vi samarbetade! …under mina första år som doktorand hade jag mycket trevliga GST-samarbeten med Sofia Hederos. Tack för att du tog dig tid och introducerade mig i alla moment och lade grunden till min egen forskning. Jag vill även tacka Lotta Tegler, Beatrice Carlsson, Sofia Carlsson och Jenny Larsson Viljanen, för mycket trevliga samarbeten genom åren. …tack Andréas Larsson för gott samarbete med yt-projektet! Om vi skulle haft mer tid kunde det blivit nåt stort av det… …tack Uno Carlsson, Per Hammarström och Martin Karlsson på biokemi för att ni alltid ställer upp och diskuterar mina frågeställningar av mer biokemisk karaktär. …utan vår suveräna administratör Susanne Andersson skulle inte mycket fungera på avdelningen. Ett jättetack för att du alltid ställer upp och svarar på mer eller mindre dumma frågor och fixar allt! …många personer har kommit och gått i vår ”espresso-korridor” under åren. Det är få förunnat att få jobba med så duktiga och trevliga människor runtomkring sig. Tack
65
alla i det gamla gardet för trevliga fikastunder och diverse påhitt efter arbetstid: Karin, Helena, Viji, Jonas N, Jonas W N, Johan R, Alina, Tess, Jesus, Martin, Laila, Gunnar H, Kathrine, Cissi L, Kaisong, Yi, Andreas, Klas och alla exjobbare, projektanställda och stipendiater. ...jag vill även tacka alla gamla doktorandkollegor på ”riktig kemi” för att ni förgyllt min tid som doktorand: Markus, Johan O, Anders, P-O, Thorsten, Fredrik W och alla andra. …ett stort tack till alla hjältar som fortfarande är kvar på kemiavdelningen och kämpar på med era projekt. Tillsammans med er har jag haft förträffligt roligt under arbetstid, på konferenser och på galna kemifester. Jag kommer sakna er! Patrik, Leffe, Janosch, Anngelica, Cissi, Karin A, Karin S, Patricia, Ina, Daniel S, Satish, Sofie, Fredrik, Andreas, Alma, Timmy, Roger, Lan, Bäck, Veronica, Daniel K, Robban och alla stipendiater. …tack all övrig personal på kemiavdelningen: Rita, CGA, POK, NOP, Helena, Kristina, Hasse, Roger, Stefan, Ingemar, Peter K, Peter N, Annika, Gunilla, Lars, Misha, Bosse, Krusell, Maria S, Lars-Göran, Magda med flera, för att ni bidragit till det goda arbetsklimatet och alltid varit hjälpsamma. …tack whiskyklubben för alla trevliga sammankomster där vi tillsammans har smakat och betygsatt livets vatten. Jag ser fram emot att göra en del gästspel framöver… …jag är väldigt glad att jag lärt känna så många underbara människor under tiden som student i Linköping. Nu är ni spridda över landet och vi träffas ju alltför sällan, men när vi väl gör det så har vi grymt roligt. Jag är säker på att det ges många fler tillfällen framöver! Ett speciellt tack till V75-klubben…vad vore en lördag utan bongen? Snart smäller det! Tack ALLIHOP, ni vet vilka ni är! …livet handlar ju inte bara om forskning, även om det blivit en herrans massa av den varan de senaste åren. Jag vill tacka mina vänner utanför universitetsvärlden: Tomas, Tobbe, Olof och Daniel, det finns inga bättre vänner än ni! När vi träffas, dock alltför sällan, så mår jag riktigt bra! Nu hoppas jag det blir mycket oftare. Mina barndomsvänner Tobbe W, Niclas och Peno, vi har många minnen ihop och nu är det dags för en tjôtkväll igen. Makkara trallen alle! …jag hade inte stått här idag utan min familj. Jag är otroligt stolt över att vuxit upp med mina underbara bröder och deras familjer: Tony, Tommy, Timo och Janne. Tack för allt ni lärt mig och att ni alltid finns där! Jag har fått den bästa uppväxt man kan tänka sig, tack vare mina alltid stöttande föräldrar, mamma Marja-Liisa och pappa Jorma. Ni vet nog fortfarande inte riktigt vad jag gör, men nu kan jag i alla fall säga att jag är färdig! Ni är bäst allihop! …sist men inte minst, min fru Jenny, du har fått leva med en tankspridd och grubblande forskare. Du har många gånger fått agera bollplank för mina idéer och lyckats få mig att behålla lugn och harmoni genom den här resan. Jag älskar dig! Mot nya utmaningar….Johan Viljanen, Linköping, April 2008
66
(1)
Lander, E. S., Linton, L. M., Birren, B., Nusbaum, C., Zody, M. C., Baldwin, J., Devon, K., Dewar, K., Doyle, M., FitzHugh, W., Funke, R., Gage, D., Harris, K., Heaford, A., Howland, J., Kann, L., Lehoczky, J., LeVine, R., McEwan, P., McKernan, K., Meldrim, J., Mesirov, J. P., Miranda, C., Morris, W., Naylor, J., Raymond, C., Rosetti, M., Santos, R., Sheridan, A., Sougnez, C., Stange-Thomann, N., Stojanovic, N., Subramanian, A., Wyman, D., Rogers, J., Sulston, J., Ainscough, R., Beck, S., Bentley, D., Burton, J., Clee, C., Carter, N., Coulson, A., Deadman, R., Deloukas, P., Dunham, A., Dunham, I., Durbin, R., French, L., Grafham, D., Gregory, S., Hubbard, T., Humphray, S., Hunt, A., Jones, M., Lloyd, C., McMurray, A., Matthews, L., Mercer, S., Milne, S., Mullikin, J. C., Mungall, A., Plumb, R., Ross, M., Shownkeen, R., Sims, S., Waterston, R. H., Wilson, R. K., Hillier, L. W., McPherson, J. D., Marra, M. A., Mardis, E. R., Fulton, L. A., Chinwalla, A. T., Pepin, K. H., Gish, W. R., Chissoe, S. L., Wendl, M. C., Delehaunty, K. D., Miner, T. L., Delehaunty, A., Kramer, J. B., Cook, L. L., Fulton, R. S., Johnson, D. L., Minx, P. J., Clifton, S. W., Hawkins, T., Branscomb, E., Predki, P., Richardson, P., Wenning, S., Slezak, T., Doggett, N., Cheng, J. F., Olsen, A., Lucas, S., Elkin, C., Uberbacher, E., Frazier, M., et al. (2001) Initial sequencing and analysis of the human genome. Nature 409, 860-921.
(2)
Venter, J. C., Adams, M. D., Myers, E. W., Li, P. W., Mural, R. J., Sutton, G. G., Smith, H. O., Yandell, M., Evans, C. A., Holt, R. A., Gocayne, J. D., Amanatides, P., Ballew, R. M., Huson, D. H., Wortman, J. R., Zhang, Q., Kodira, C. D., Zheng, X. H., Chen, L., Skupski, M., Subramanian, G., Thomas, P. D., Zhang, J., Gabor Miklos, G. L., Nelson, C., Broder, S., Clark, A. G., Nadeau, J., McKusick, V. A., Zinder, N., Levine, A. J., Roberts, R. J., Simon, M., Slayman, C., Hunkapiller, M., Bolanos, R., Delcher, A., Dew, I., Fasulo, D., Flanigan, M., Florea, L., Halpern, A., Hannenhalli, S., Kravitz, S., Levy, S., Mobarry, C., Reinert, K., Remington, K., Abu-Threideh, J., Beasley, E., Biddick, K., Bonazzi, V., Brandon, R., Cargill, M., Chandramouliswaran, I., Charlab, R., Chaturvedi, K., Deng, Z., Di Francesco, V., Dunn, P., Eilbeck, K., Evangelista, C., Gabrielian, A. E., Gan, W., Ge, W., Gong, F., Gu, Z., Guan, P., Heiman, T. J., Higgins, M. E., Ji, R. R., Ke, Z., Ketchum, K. A., Lai, Z., Lei, Y., Li, Z., Li, J., Liang, Y., Lin, X., Lu, F., Merkulov, G. V., Milshina, N., Moore, H. M., Naik, A. K., Narayan, V. A., Neelam, B., Nusskern, D., Rusch, D. B., Salzberg, S., Shao, W., Shue, B., Sun, J., Wang, Z., Wang, A., Wang, X., Wang, J., Wei, M., Wides, R., Xiao, C., Yan, C., et al. (2001) The sequence of the human genome. Science 291, 1304-1351.
67
(3)
Daggett, V., and Fersht, A. (2003) The present view of the mechanism of protein folding. Nat Rev Mol Cell Biol 4, 497-502.
(4)
Dobson, C. M. (2003) Protein folding and misfolding. Nature 426, 884-890.
(5)
Chiti, F., and Dobson, C. M. (2006) Protein misfolding, functional amyloid, and human disease. Annu Rev Biochem 75, 333-366.
(6)
Cohen, F. E., and Kelly, J. W. (2003) Therapeutic approaches to proteinmisfolding diseases. Nature 426, 905-909.
(7)
Muchowski, P. J., and Wacker, J. L. (2005) Modulation of neurodegeneration by molecular chaperones. Nat Rev Neurosci 6, 11-22.
(8)
Chaudhuri, T. K., and Paul, S. (2006) Protein-misfolding diseases and chaperone-based therapeutic approaches. Febs J 273, 1331-1349.
(9)
Nilsson, K. P., Aslund, A., Berg, I., Nystrom, S., Konradsson, P., Herland, A., Inganas, O., Stabo-Eeg, F., Lindgren, M., Westermark, G. T., Lannfelt, L., Nilsson, L. N., and Hammarstrom, P. (2007) Imaging distinct conformational states of amyloid-beta fibrils in Alzheimer's disease using novel luminescent probes. ACS Chem Biol 2, 553-560.
(10)
Toscano, M. D., Woycechowsky, K. J., and Hilvert, D. (2007) Minimalist active-site redesign: teaching old enzymes new tricks. Angew Chem Int Ed Engl 46, 3212-3236.
(11)
Cooper, W. J., and Waters, M. L. (2005) Molecular recognition with designed peptides and proteins. Curr Opin Chem Biol 9, 627-631.
(12)
Martin, L., and Vita, C. (2000) Engineering novel bioactive mini-proteins from small size natural and de novo designed scaffolds. Curr Protein Pept Sci 1, 403-430.
(13)
Skerra, A. (2000) Engineered protein scaffolds for molecular recognition. J Mol Recognit 13, 167-187.
(14)
Hamachi, I., Watanabe, J. I., Eboshi, R., Hiraoka, T., and Shinkai, S. (2000) Incorporation of artificial receptors into a protein/peptide surface: a strategy for on/off type of switching of semisynthetic enzymes. Biopolymers 55, 459468.
(15)
Qi, D., Tann, C. M., Haring, D., and Distefano, M. D. (2001) Generation of new enzymes via covalent modification of existing proteins. Chem Rev 101, 3081-3111.
(16)
Davis, B. G. (2003) Chemical modification of biocatalysts. Curr Opin Biotechnol 14, 379-386.
(17)
Zhang, Z., Smith, B. A., Wang, L., Brock, A., Cho, C., and Schultz, P. G. (2003) A new strategy for the site-specific modification of proteins in vivo. Biochemistry 42, 6735-6746.
68
(18)
Wilken, J., and Kent, S. B. (1998) Chemical protein synthesis. Curr Opin Biotechnol 9, 412-426.
(19)
Hutchison, C. A., 3rd, Phillips, S., Edgell, M. H., Gillam, S., Jahnke, P., and Smith, M. (1978) Mutagenesis at a specific position in a DNA sequence. J Biol Chem 253, 6551-6560.
(20)
Hult, K., and Berglund, P. (2003) Engineered enzymes for improved organic synthesis. Curr Opin Biotechnol 14, 395-400.
(21)
Van den Burg, B., Vriend, G., Veltman, O. R., Venema, G., and Eijsink, V. G. (1998) Engineering an enzyme to resist boiling. Proc Natl Acad Sci U S A 95, 2056-2060.
(22)
Hederos, S., Broo, K. S., Jakobsson, E., Kleywegt, G. J., Mannervik, B., and Baltzer, L. (2004) Incorporation of a single His residue by rational design enables thiol-ester hydrolysis by human glutathione transferase A1-1. Proc Natl Acad Sci U S A 101, 13163-13167.
(23)
Schmid, A., Dordick, J. S., Hauer, B., Kiener, A., Wubbolts, M., and Witholt, B. (2001) Industrial biocatalysis today and tomorrow. Nature 409, 258-268.
(24)
Zaks, A. (2001) Industrial biocatalysis. Curr Opin Chem Biol 5, 130-136.
(25)
Houde, A., Kademi, A., and Leblanc, D. (2004) Lipases and their industrial applications: an overview. Appl Biochem Biotechnol 118, 155-170.
(26)
Aharoni, A., Amitai, G., Bernath, K., Magdassi, S., and Tawfik, D. S. (2005) High-throughput screening of enzyme libraries: thiolactonases evolved by fluorescence-activated sorting of single cells in emulsion compartments. Chem Biol 12, 1281-1289.
(27)
Chen, R. (2001) Enzyme engineering: rational redesign versus directed evolution. Trends Biotechnol 19, 13-14.
(28)
Chen, R. (1999) A general strategy for enzyme engineering. Trends Biotechnol 17, 344-345.
(29)
Penning, T. M., and Jez, J. M. (2001) Enzyme redesign. Chem Rev 101, 30273046.
(30)
Park, H. S., Nam, S. H., Lee, J. K., Yoon, C. N., Mannervik, B., Benkovic, S. J., and Kim, H. S. (2006) Design and evolution of new catalytic activity with an existing protein scaffold. Science 311, 535-538.
(31)
Hellinga, H. W., and Richards, F. M. (1991) Construction of new ligand binding sites in proteins of known structure. I. Computer-aided modeling of sites with pre-defined geometry. J Mol Biol 222, 763-785.
(32)
Hellinga, H. W. (1998) Computational protein engineering. Nat Struct Biol 5, 525-527.
69
(33)
Looger, L. L., Dwyer, M. A., Smith, J. J., and Hellinga, H. W. (2003) Computational design of receptor and sensor proteins with novel functions. Nature 423, 185-190.
(34)
Baltzer, L., Nilsson, H., and Nilsson, J. (2001) De novo design of proteins-what are the rules? Chem Rev 101, 3153-3163.
(35)
Ghirlanda, G., Lear, J. D., Ogihara, N. L., Eisenberg, D., and DeGrado, W. F. (2002) A hierarchic approach to the design of hexameric helical barrels. J Mol Biol 319, 243-253.
(36)
Birkett, D. J., Dwek, R. A., Radda, G. K., Richards, R. E., and Salmon, A. G. (1971) Probes for the conformational transitions of phosphorylase b. Effect of ligands studied by proton relaxation enhancement, fluorescence and chemical reactivities. Eur J Biochem 20, 494-508.
(37)
Carlsson, U., Aasa, R., Henderson, L. E., Jonsson, B. H., and Lindskog, S. (1975) Paramagnetic and fluorescent probes attached to "buried" sulfhydryl groups in human carbonic anhydrases. Application to inhibitor binding, denaturation and refolding. Eur J Biochem 52, 25-36.
(38)
Hammarstrom, P., Owenius, R., Martensson, L. G., Carlsson, U., and Lindgren, M. (2001) High-resolution probing of local conformational changes in proteins by the use of multiple labeling: unfolding and self-assembly of human carbonic anhydrase II monitored by spin, fluorescent, and chemical reactivity probes. Biophys J 80, 2867-2885.
(39)
Clayton, D., Shapovalov, G., Maurer, J. A., Dougherty, D. A., Lester, H. A., and Kochendoerfer, G. G. (2004) Total chemical synthesis and electrophysiological characterization of mechanosensitive channels from Escherichia coli and Mycobacterium tuberculosis. Proc Natl Acad Sci U S A 101, 4764-4769.
(40)
Mannervik, B., and Danielson, U. H. (1988) Glutathione transferases-structure and catalytic activity. CRC Crit Rev Biochem 23, 283-337.
(41)
Armstrong, R. N. (1997) Structure, catalytic mechanism, and evolution of the glutathione transferases. Chem Res Toxicol 10, 2-18.
(42)
Mannervik, B. (1996) Evolution of glutathione transferases and related enzymes for the protection of cells against electrophiles. Biochem Soc Trans 24, 878-880.
(43)
Coles, B. F., and Kadlubar, F. F. (2005) Human alpha class glutathione Stransferases: genetic polymorphism, expression, and susceptibility to disease. Methods Enzymol 401, 9-42.
(44)
Liang, F. Q., Alssadi, R., Morehead, P., Awasthi, Y. C., and Godley, B. F. (2005) Enhanced expression of glutathione-S-transferase A1-1 protects against oxidative stress in human retinal pigment epithelial cells. Exp Eye Res 80, 113-119.
70
(45)
Hayes, J. D., Flanagan, J. U., and Jowsey, I. R. (2005) Glutathione transferases. Annu Rev Pharmacol Toxicol 45, 51-88.
(46)
Sheehan, D., Meade, G., Foley, V. M., and Dowd, C. A. (2001) Structure, function and evolution of glutathione transferases: implications for classification of non-mammalian members of an ancient enzyme superfamily. Biochem J 360, 1-16.
(47)
Mannervik, B., Board, P. G., Hayes, J. D., Listowsky, I., and Pearson, W. R. (2005) Nomenclature for mammalian soluble glutathione transferases. Methods Enzymol 401, 1-8.
(48)
Tellez-Sanz, R., Cesareo, E., Nuccetelli, M., Aguilera, A. M., Baron, C., Parker, L. J., Adams, J. J., Morton, C. J., Lo Bello, M., Parker, M. W., and Garcia-Fuentes, L. (2006) Calorimetric and structural studies of the nitric oxide carrier S-nitrosoglutathione bound to human glutathione transferase P11. Protein Sci 15, 1093-1105.
(49)
Habig, W. H., Pabst, M. J., and Jakoby, W. B. (1974) Glutathione Stransferases. The first enzymatic step in mercapturic acid formation. J Biol Chem 249, 7130-7139.
(50)
Hayes, J. D., and Pulford, D. J. (1995) The glutathione S-transferase supergene family: regulation of GST and the contribution of the isoenzymes to cancer chemoprotection and drug resistance. Crit Rev Biochem Mol Biol 30, 445-600.
(51)
Sayed, Y., Hornby, J. A., Lopez, M., and Dirr, H. (2002) Thermodynamics of the ligandin function of human class Alpha glutathione transferase A1-1: energetics of organic anion ligand binding. Biochem J 363, 341-346.
(52)
Axarli, I. A., Rigden, D. J., and Labrou, N. E. (2004) Characterization of the ligandin site of maize glutathione S-transferase I. Biochem J 382, 885-893.
(53)
Singh, S. V., Leal, T., and Awasthi, Y. C. (1988) Inhibition of human glutathione S-transferases by bile acids. Toxicol Appl Pharmacol 95, 248-254.
(54)
Mannervik, B., Alin, P., Guthenberg, C., Jensson, H., Tahir, M. K., Warholm, M., and Jornvall, H. (1985) Identification of three classes of cytosolic glutathione transferase common to several mammalian species: correlation between structural data and enzymatic properties. Proc Natl Acad Sci U S A 82, 7202-7206.
(55)
Sinning, I., Kleywegt, G. J., Cowan, S. W., Reinemer, P., Dirr, H. W., Huber, R., Gilliland, G. L., Armstrong, R. N., Ji, X., Board, P. G., and et al. (1993) Structure determination and refinement of human alpha class glutathione transferase A1-1, and a comparison with the Mu and Pi class enzymes. J Mol Biol 232, 192-212.
(56)
Graminski, G. F., Kubo, Y., and Armstrong, R. N. (1989) Spectroscopic and kinetic evidence for the thiolate anion of glutathione at the active site of glutathione S-transferase. Biochemistry 28, 3562-3568.
71
(57)
Dietze, E. C., Ibarra, C., Dabrowski, M. J., Bird, A., and Atkins, W. M. (1996) Rational modulation of the catalytic activity of A1-1 glutathione S-transferase: evidence for incorporation of an on-face (pi.HO-Ar) hydrogen bond at tyrosine-9. Biochemistry 35, 11938-11944.
(58)
Gustafsson, A., Etahadieh, M., Jemth, P., and Mannervik, B. (1999) The Cterminal region of human glutathione transferase A1-1 affects the rate of glutathione binding and the ionization of the active-site Tyr9. Biochemistry 38, 16268-16275.
(59)
Lim, K., Ho, J. X., Keeling, K., Gilliland, G. L., Ji, X., Ruker, F., and Carter, D. C. (1994) Three-dimensional structure of Schistosoma japonicum glutathione S-transferase fused with a six-amino acid conserved neutralizing epitope of gp41 from HIV. Protein Sci 3, 2233-2244.
(60)
Ketterer, B. (2001) A bird's eye view of the glutathione transferase field. Chem Biol Interact 138, 27-42.
(61)
Nieslanik, B. S., Ibarra, C., and Atkins, W. M. (2001) The C-terminus of glutathione S-transferase A1-1 is required for entropically-driven ligand binding. Biochemistry 40, 3536-3543.
(62)
Cameron, A. D., Sinning, I., L'Hermite, G., Olin, B., Board, P. G., Mannervik, B., and Jones, T. A. (1995) Structural analysis of human alpha-class glutathione transferase A1-1 in the apo-form and in complexes with ethacrynic acid and its glutathione conjugate. Structure 3, 717-727.
(63)
Grahn, E., Novotny, M., Jakobsson, E., Gustafsson, A., Grehn, L., Olin, B., Madsen, D., Wahlberg, M., Mannervik, B., and Kleywegt, G. J. (2006) New crystal structures of human glutathione transferase A1-1 shed light on glutathione binding and the conformation of the C-terminal helix. Acta Crystallogr D Biol Crystallogr 62, 197-207.
(64)
Zhan, Y., and Rule, G. S. (2004) Glutathione induces helical formation in the carboxy terminus of human glutathione transferase A1-1. Biochemistry 43, 7244-7254.
(65)
Dirr, H. W., and Wallace, L. A. (1999) Role of the C-terminal helix 9 in the stability and ligandin function of class alpha glutathione transferase A1-1. Biochemistry 38, 15631-15640.
(66)
Mosebi, S., Sayed, Y., Burke, J., and Dirr, H. W. (2003) Residue 219 impacts on the dynamics of the C-terminal region in glutathione transferase A1-1: implications for stability and catalytic and ligandin functions. Biochemistry 42, 15326-15332.
(67)
Nilsson, L. O., Edalat, M., Pettersson, P. L., and Mannervik, B. (2002) Aromatic residues in the C-terminal region of glutathione transferase A1-1 influence rate-determining steps in the catalytic mechanism. Biochim Biophys Acta 1597, 157-163.
72
(68)
Pemble, S. E., and Taylor, J. B. (1992) An evolutionary perspective on glutathione transferases inferred from class-theta glutathione transferase cDNA sequences. Biochem J 287 (Pt 3), 957-963.
(69)
Raghunathan, S., Chandross, R. J., Kretsinger, R. H., Allison, T. J., Penington, C. J., and Rule, G. S. (1994) Crystal structure of human class mu glutathione transferase GSTM2-2. Effects of lattice packing on conformational heterogeneity. J Mol Biol 238, 815-832.
(70)
Board, P. G., Coggan, M., Chelvanayagam, G., Easteal, S., Jermiin, L. S., Schulte, G. K., Danley, D. E., Hoth, L. R., Griffor, M. C., Kamath, A. V., Rosner, M. H., Chrunyk, B. A., Perregaux, D. E., Gabel, C. A., Geoghegan, K. F., and Pandit, J. (2000) Identification, characterization, and crystal structure of the Omega class glutathione transferases. J Biol Chem 275, 24798-24806.
(71)
Reinemer, P., Dirr, H. W., Ladenstein, R., Huber, R., Lo Bello, M., Federici, G., and Parker, M. W. (1992) Three-dimensional structure of class pi glutathione S-transferase from human placenta in complex with Shexylglutathione at 2.8 A resolution. J Mol Biol 227, 214-226.
(72)
Rossjohn, J., McKinstry, W. J., Oakley, A. J., Verger, D., Flanagan, J., Chelvanayagam, G., Tan, K. L., Board, P. G., and Parker, M. W. (1998) Human theta class glutathione transferase: the crystal structure reveals a sulfate-binding pocket within a buried active site. Structure 6, 309-322.
(73)
Hubatsch, I., and Mannervik, B. (2001) A highly acidic tyrosine 9 and a normally titrating tyrosine 212 contribute to the catalytic mechanism of human glutathione transferase A4-4. Biochem Biophys Res Commun 280, 878-882.
(74)
Pettersson, P. L., Johansson, A. S., and Mannervik, B. (2002) Transmutation of human glutathione transferase A2-2 with peroxidase activity into an efficient steroid isomerase. J Biol Chem 277, 30019-30022.
(75)
Karshikoff, A., Reinemer, P., Huber, R., and Ladenstein, R. (1993) Electrostatic evidence for the activation of the glutathione thiol by Tyr7 in piclass glutathione transferases. Eur J Biochem 215, 663-670.
(76)
Lyttle, M. H., Hocker, M. D., Hui, H. C., Caldwell, C. G., Aaron, D. T., Engqvist-Goldstein, A., Flatgaard, J. E., and Bauer, K. E. (1994) Isozymespecific glutathione-S-transferase inhibitors: design and synthesis. J Med Chem 37, 189-194.
(77)
Ladner, J. E., Parsons, J. F., Rife, C. L., Gilliland, G. L., and Armstrong, R. N. (2004) Parallel evolutionary pathways for glutathione transferases: structure and mechanism of the mitochondrial class kappa enzyme rGSTK1-1. Biochemistry 43, 352-361.
(78)
Viljanen, J., Tegler, L., and Broo, K. S. (2004) Combinatorial chemical reengineering of the alpha class glutathione transferases. Bioconjug Chem 15, 718-727.
73
(79)
Hederos, S., Karlsson, B., Tegler, L., and Broo, K. S. (2005) Ligand-directed labeling of a single lysine residue in hGST A1-1 mutants. Bioconjug Chem 16, 1009-1018.
(80)
Adang, A. E., Duindam, A. J., Brussee, J., Mulder, G. J., and van der Gen, A. (1988) Synthesis and nucleophilic reactivity of a series of glutathione analogues, modified at the gamma-glutamyl moiety. Biochem J 255, 715-720.
(81)
Adang, A. E., Meyer, D. J., Brussee, J., Van der Gen, A., Ketterer, B., and Mulder, G. J. (1989) Interaction of rat glutathione S-transferases 7-7 and 8-8 with gamma-glutamyl- or glycyl-modified glutathione analogues. Biochem J 264, 759-764.
(82)
Adang, A. E., Brussee, J., van der Gen, A., and Mulder, G. J. (1990) The glutathione-binding site in glutathione S-transferases. Investigation of the cysteinyl, glycyl and gamma-glutamyl domains. Biochem J 269, 47-54.
(83)
Adang, A. E., Brussee, J., Meyer, D. J., Coles, B., Ketterer, B., van der Gen, A., and Mulder, G. J. (1988) Substrate specificity of rat liver glutathione Stransferase isoenzymes for a series of glutathione analogues, modified at the gamma-glutamyl moiety. Biochem J 255, 721-724.
(84)
Sugimoto, M., Kuhlenkamp, J., Ookhtens, M., Aw, T. Y., Reeve, J., Jr., and Kaplowitz, N. (1985) Gamma-glutamylcysteine: a substrate for glutathione Stransferases. Biochem Pharmacol 34, 3643-3647.
(85)
Scheller, F. W., Wollenberger, U., Warsinke, A., and Lisdat, F. (2001) Research and development in biosensors. Curr Opin Biotechnol 12, 35-40.
(86)
Coon, R. L., Lai, N. C., and Kampine, J. P. (1976) Evaluation of a dualfunction pH and PCO2 in vivo sensor. J Appl Physiol 40, 625-629.
(87)
Frost, M. C., Rudich, S. M., Zhang, H., Maraschio, M. A., and Meyerhoff, M. E. (2002) In vivo biocompatibility and analytical performance of intravascular amperometric oxygen sensors prepared with improved nitric oxide-releasing silicone rubber coating. Anal Chem 74, 5942-5947.
(88)
Thevenot, D. R., Toth, K., Durst, R. A., and Wilson, G. S. (2001) Electrochemical biosensors: recommended definitions and classification. Biosensors & Bioelectronics 16, 121-131.
(89)
Clark, L. C., Jr., and Lyons, C. (1962) Electrode systems for continuous monitoring in cardiovascular surgery. Ann N Y Acad Sci 102, 29-45.
(90)
Updike, S. J., and Hicks, G. P. (1967) The enzyme electrode. Nature 214, 986988.
(91)
Rogers, K. R. (2006) Recent advances in biosensor techniques for environmental monitoring. Anal Chim Acta 568, 222-231.
(92)
Rogers, K. R. (2000) Principles of affinity-based biosensors. Molecular Biotechnology 14, 109-129.
74
(93)
Kronick, M. N., and Little, W. A. (1975) A new immunoassay based on fluorescence excitation by internal reflection spectroscopy. J Immunol Methods 8, 235-240.
(94)
Giaever, I. (1973) The antibody-antigen reaction: a visual observation. J Immunol 110, 1424-1426.
(95)
Marshall, A., and Hodgson, J. (1998) DNA chips: an array of possibilities. Nat Biotechnol 16, 27-31.
(96)
Tromberg, B. J., Sepaniak, M. J., Alarie, J. P., Vo-Dinh, T., and Santella, R. M. (1988) Development of antibody-based fiber-optic sensors for detection of a benzo[a]pyrene metabolite. Anal Chem 60, 1901-1908.
(97)
Nakanishi, K., Muguruma, H., and Karube, I. (1996) A novel method of immobilizing antibodies on a quartz crystal microbalance using plasmapolymerized films for immunosensors. Anal Chem 68, 1695-1700.
(98)
Changeux, J. P., Devillers-Thiery, A., and Chemouilli, P. (1984) Acetylcholine receptor: an allosteric protein. Science 225, 1335-1345.
(99)
Jordan, C. E., Frutos, A. G., Thiel, A. J., and Corn, R. M. (1997) Surface plasmon resonance imaging measurements of DNA hybridization adsorption and streptavidin/DNA multilayer formation at chemically modified gold surfaces. Analytical Chemistry 69, 4939-4947.
(100) Jonsson, U., Fagerstam, L., Ivarsson, B., Johnsson, B., Karlsson, R., Lundh, K., Lofas, S., Persson, B., Roos, H., Ronnberg, I., Sjolander, S., Stenberg, E., Stahlberg, R., Urbaniczky, C., Ostlin, H., and Malmqvist, M. (1991) RealTime Biospecific Interaction Analysis Using Surface-Plasmon Resonance and a Sensor Chip Technology. Biotechniques 11, 620-&. (101) de Lorimier, R. M., Smith, J. J., Dwyer, M. A., Looger, L. L., Sali, K. M., Paavola, C. D., Rizk, S. S., Sadigov, S., Conrad, D. W., Loew, L., and Hellinga, H. W. (2002) Construction of a fluorescent biosensor family. Protein Sci 11, 2655-2675. (102) Renard, M., Belkadi, L., Hugo, N., England, P., Altschuh, D., and Bedouelle, H. (2002) Knowledge-based design of reagentless fluorescent biosensors from recombinant antibodies. J Mol Biol 318, 429-442. (103) Templin, M. F., Stoll, D., Schwenk, J. M., Potz, O., Kramer, S., and Joos, T. O. (2003) Protein microarrays: promising tools for proteomic research. Proteomics 3, 2155-2166. (104) Zhu, H., and Snyder, M. (2001) Protein arrays and microarrays. Curr Opin Chem Biol 5, 40-45. (105) Bertone, P., and Snyder, M. (2005) Advances in functional protein microarray technology. Febs J 272, 5400-5411.
75
(106) Klenkar, G., Valiokas, R., Lundstrom, I., Tinazli, A., Tampe, R., Piehler, J., and Liedberg, B. (2006) Piezo dispensed microarray of multivalent chelating thiols for dissecting complex protein-protein interactions. Anal Chem 78, 3643-3650. (107) MacBeath, G., and Schreiber, S. L. (2000) Printing proteins as microarrays for high-throughput function determination. Science 289, 1760-1763. (108) Harrington, C. A., Rosenow, C., and Retief, J. (2000) Monitoring gene expression using DNA microarrays. Curr Opin Microbiol 3, 285-291. (109) Altschuh, D., Oncul, S., and Demchenko, A. P. (2006) Fluorescence sensing of intermolecular interactions and development of direct molecular biosensors. J Mol Recognit 19, 459-477. (110) Imanishi, Y., Lodowski, K. H., and Koutalos, Y. (2007) Two-photon microscopy: shedding light on the chemistry of vision. Biochemistry 46, 96749684. (111) Lakowicz, J. R. (1999) Principles of fluorescence spectroscopy, 2nd ed., Kluwer Academic/Plenum, New York. (112) Lakowicz, J. R., Gryczynski, I., Wiczk, W., Laczko, G., Prendergast, F. C., and Johnson, M. L. (1990) Conformational distributions of melittin in water/methanol mixtures from frequency-domain measurements of nonradiative energy transfer. Biophys Chem 36, 99-115. (113) Villebeck, L., Moparthi, S. B., Lindgren, M., Hammarstrom, P., and Jonsson, B. H. (2007) Domain-specific chaperone-induced expansion is required for beta-actin folding: a comparison of beta-actin conformations upon interactions with GroEL and tail-less complex polypeptide 1 ring complex (TRiC). Biochemistry 46, 12639-12647. (114) Nieslanik, B. S., Dabrowski, M. J., Lyon, R. P., and Atkins, W. M. (1999) Stopped-flow kinetic analysis of the ligand-induced coil-helix transition in glutathione S-transferase A1-1: evidence for a persistent denatured state. Biochemistry 38, 6971-6980. (115) Hines, E. L., Llobet, E., and Gardner, J. W. (1999) Electronic noses: a review of signal processing techniques. Iee Proceedings-Circuits Devices and Systems 146, 297-310. (116) Gardner, J. W., Shin, H. W., and Hines, E. L. (2000) An electronic nose system to diagnose illness. Sensors and Actuators B-Chemical 70, 19-24. (117) Stetter, J. R., Findlay, M. W., Schroeder, K. M., Yue, C., and Penrose, W. R. (1993) Quality Classification of Grain Using a Sensor Array and PatternRecognition. Analytica Chimica Acta 284, 1-11. (118) Eklov, T., and Lundstrom, I. (1999) Distributed sensor system for quantification of individual components in a multiple gas mixture. Analytical Chemistry 71, 3544-3550.
76
(119) Suslick, K. S., Rakow, N. A., and Sen, A. (2004) Colorimetric sensor arrays for molecular recognition. Tetrahedron 60, 11133-11138. (120) Rakow, N. A., Sen, A., Janzen, M. C., Ponder, J. B., and Suslick, K. S. (2005) Molecular recognition and discrimination of amines with a colorimetric array. Angewandte Chemie-International Edition 44, 4528-4532. (121) Buryak, A., and Severin, K. (2005) A chemosensor array for the colorimetric identification of 20 natural amino acids. Journal of the American Chemical Society 127, 3700-3701. (122) Janzen, M. C., Ponder, J. B., Bailey, D. P., Ingison, C. K., and Suslick, K. S. (2006) Colorimetric sensor Arrays for volatile organic compounds. Analytical Chemistry 78, 3591-3600. (123) Widersten, M., Bjornestedt, R., and Mannervik, B. (1994) Contribution of amino acid residue 208 in the hydrophobic binding site to the catalytic mechanism of human glutathione transferase A1-1. Biochemistry 33, 1171711723. (124) Hansson, L. O., Widersten, M., and Mannervik, B. (1999) An approach to optimizing the active site in a glutathione transferase by evolution in vitro. Biochem J 344 Pt 1, 93-100. (125) Lien, S., Gustafsson, A., Andersson, A. K., and Mannervik, B. (2001) Human glutathione transferase A1-1 demonstrates both half-of-the-sites and all-of-thesites reactivity. J Biol Chem 276, 35599-35605. (126) Danielson, U. H., and Mannervik, B. (1985) Kinetic independence of the subunits of cytosolic glutathione transferase from the rat. Biochem J 231, 263267. (127) Jakobson, I., Warholm, M., and Mannervik, B. (1979) The binding of substrates and a product of the enzymatic reaction to glutathione S-transferase A. J Biol Chem 254, 7085-7089. (128) Wu, P., and Brand, L. (1994) Resonance energy transfer: methods and applications. Anal Biochem 218, 1-13. (129) Mori, Y., Nagaoka, S., Takiuchi, H., Kikuchi, T., Noguchi, N., Tanzawa, H., and Noishiki, Y. (1982) A new antithrombogenic material with long polyethyleneoxide chains. Trans Am Soc Artif Intern Organs 28, 459-463. (130) Lee, J. H., Jeong, B. J., and Lee, H. B. (1997) Plasma protein adsorption and platelet adhesion onto comb-like PEO gradient surfaces. J Biomed Mater Res 34, 105-114. (131) Zhang, M., Desai, T., and Ferrari, M. (1998) Proteins and cells on PEG immobilized silicon surfaces. Biomaterials 19, 953-960. (132) Larsson, A., Du, C. X., and Liedberg, B. (2007) UV-Patterned Poly(ethylene glycol) Matrix for Microarray Applications. Biomacromolecules.
77
(133) Lyon, R. P., and Atkins, W. M. (2002) Kinetic characterization of native and cysteine 112-modified glutathione S-transferase A1-1: reassessment of nonsubstrate ligand binding. Biochemistry 41, 10920-10927. (134) Smith, D. B., and Johnson, K. S. (1988) Single-Step Purification of Polypeptides Expressed in Escherichia-Coli as Fusions with Glutathione STransferase. Gene 67, 31-40. (135) Toye, B., Zhong, G. M., Peeling, R., and Brunham, R. C. (1990) Immunological Characterization of a Cloned Fragment Containing the Species-Specific Epitope from the Major Outer-Membrane Protein of Chlamydia-Trachomatis. Infection and Immunity 58, 3909-3913. (136) Fikrig, E., Barthold, S. W., Kantor, F. S., and Flavell, R. A. (1990) Protection of Mice against the Lyme-Disease Agent by Immunizing with Recombinant Ospa. Science 250, 553-556. (137) Wu, X., and Oppermann, U. (2003) High-level expression and rapid purification of rare-codon genes from hyperthermophilic archaea by the GST gene fusion system. J Chromatogr B Analyt Technol Biomed Life Sci 786, 177-185. (138) Scheich, C., Sievert, V., and Bussow, K. (2003) An automated method for high-throughput protein purification applied to a comparison of His-tag and GST-tag affinity chromatography. Bmc Biotechnology 3, -. (139) Dawson, P. E., Muir, T. W., Clark-Lewis, I., and Kent, S. B. (1994) Synthesis of proteins by native chemical ligation. Science 266, 776-779. (140) Kochendoerfer, G. G., and Kent, S. B. (1999) Chemical protein synthesis. Curr Opin Chem Biol 3, 665-671. (141) Kochendoerfer, G. G., Salom, D., Lear, J. D., Wilk-Orescan, R., Kent, S. B., and DeGrado, W. F. (1999) Total chemical synthesis of the integral membrane protein influenza A virus M2: role of its C-terminal domain in tetramer assembly. Biochemistry 38, 11905-11913. (142) Camarero, J. A., and Mitchell, A. R. (2005) Synthesis of proteins by native chemical ligation using Fmoc-based chemistry. Protein and Peptide Letters 12, 723-728. (143) Wang, L., Xie, J., and Schultz, P. G. (2006) Expanding the genetic code. Annu Rev Biophys Biomol Struct 35, 225-249. (144) Dawson, P. E., and Kent, S. B. H. (2000) Synthesis of native proteins by chemical ligation. Annual Review of Biochemistry 69, 923-960. (145) Schneider, J., and Kent, S. B. (1988) Enzymatic activity of a synthetic 99 residue protein corresponding to the putative HIV-1 protease. Cell 54, 363368.
78
(146) Canne, L. E., Ferredamare, A. R., Burley, S. K., and Kent, S. B. H. (1995) Total Chemical Synthesis of a Unique Transcription Factor-Related Protein Cmyc-Max. Journal of the American Chemical Society 117, 2998-3007. (147) Kochendoerfer, G. G., Chen, S. Y., Mao, F., Cressman, S., Traviglia, S., Shao, H., Hunter, C. L., Low, D. W., Cagle, E. N., Carnevali, M., Gueriguian, V., Keogh, P. J., Porter, H., Stratton, S. M., Wiedeke, M. C., Wilken, J., Tang, J., Levy, J. J., Miranda, L. P., Crnogorac, M. M., Kalbag, S., Botti, P., SchindlerHorvat, J., Savatski, L., Adamson, J. W., Kung, A., Kent, S. B., and Bradburne, J. A. (2003) Design and chemical synthesis of a homogeneous polymer-modified erythropoiesis protein. Science 299, 884-887. (148) Liu, C. F., and Tam, J. P. (1994) Peptide segment ligation strategy without use of protecting groups. Proc Natl Acad Sci U S A 91, 6584-6588. (149) Muir, T. W., Williams, M. J., Ginsberg, M. H., and Kent, S. B. (1994) Design and chemical synthesis of a neoprotein structural model for the cytoplasmic domain of a multisubunit cell-surface receptor: integrin alpha IIb beta 3 (platelet GPIIb-IIIa). Biochemistry 33, 7701-7708. (150) Hackeng, T. M., Mounier, C. M., Bon, C., Dawson, P. E., Griffin, J. H., and Kent, S. B. (1997) Total chemical synthesis of enzymatically active human type II secretory phospholipase A2. Proc Natl Acad Sci U S A 94, 7845-7850. (151) Wilken, J., Hoover, D., Thompson, D. A., Barlow, P. N., McSparron, H., Picard, L., Wlodawer, A., Lubkowski, J., and Kent, S. B. (1999) Total chemical synthesis and high-resolution crystal structure of the potent anti-HIV protein AOP-RANTES. Chem Biol 6, 43-51. (152) Bang, D., Chopra, N., and Kent, S. B. (2004) Total chemical synthesis of crambin. J Am Chem Soc 126, 1377-1383. (153) Baca, M., Muir, T. W., Schnolzer, M., and Kent, S. B. H. (1995) Chemical Ligation of Cysteine-Containing Peptides - Synthesis of a 22-Kda Tethered Dimer of Hiv-1 Protease. Journal of the American Chemical Society 117, 1881-1887. (154) Canne, L. E., Botti, P., Simon, R. J., Chen, Y. J., Dennis, E. A., and Kent, S. B. H. (1999) Chemical protein synthesis by solid phase ligation of unprotected peptide segments. Journal of the American Chemical Society 121, 8720-8727. (155) Kent, S. B. (1988) Chemical synthesis of peptides and proteins. Annu Rev Biochem 57, 957-989.
79
