UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The development of reconstituted translation system for peptidomimetic mRNA display synthesis Stojanovic, Vesna 2008

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2009_spring_stojanovic_vesna.pdf [ 3.87MB ]
Metadata
JSON: 24-1.0066852.json
JSON-LD: 24-1.0066852-ld.json
RDF/XML (Pretty): 24-1.0066852-rdf.xml
RDF/JSON: 24-1.0066852-rdf.json
Turtle: 24-1.0066852-turtle.txt
N-Triples: 24-1.0066852-rdf-ntriples.txt
Original Record: 24-1.0066852-source.json
Full Text
24-1.0066852-fulltext.txt
Citation
24-1.0066852.ris

Full Text

THE DEVELOPMENT OF RECONSTITUTED TRANSLATION SYSTEM FOR PEPTIDOMIMETIC mRNA DISPLAY SYNTHESIS  by Vesna Stojanovic B.Sc., University of Belgrade, 1994 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in The Faculty of Graduate Studies (Pharmaceutical Sciences) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) December 2008   Vesna Stojanovic, 2008  ABSTRACT The generation of high affinity, selective, and in vivo-stable peptide-based drugs is currently a major challenge in the field of drug development. Technologies exist that permit the generation of a vast diversity of chemical and conformational space and an example of such a technology is mRNA display, which utilizes protein translation machinery to produce a wide array of polypeptides starting from a combinatorial library of mRNA templates. The intention of this research was to bridge mRNA display to a reconstituted translation system using protein synthesis using recombinant elements (PURE) system for a new drug discovery platform. We hypothesized that it is possible to generate mRNA-peptidomimetic fusions using reconstituted translation system and chemo-enzymatically charged tRNAs, to incorporate unnatural amino acids into mRNA-peptidomimetic fusions. Upon demonstating that the reconstituted system was functional, we have synthesized hexapeptide fusion products containing four alanine residues and one biocytin residue. Fusions were assayed using urea-PAGE in the presence of streptavidin which allowed for unambiguous evaluation of the full-length fusion fraction. It was determined that overall more fusion product was generated with template that codes for biocytin early in the coding sequence, but that the percent of biocytin-containing product stays similar regardless of the biocytin place in the coding region. We have also found that the change in template untranslated region length does not improve incorporation of biocytin in dipeptide fusions within the tested range. Finally, after first unsuccessful attempts to make sarcosine hexapeptide fusions, we investigated the effect of magnesium ion concentration on the translation reaction. As a result of four series of experiments performed involving both alanine and sarcosine fusion synthesis  ii  in parallel, we concluded that an increase in magnesium concentration from 5 mM to 20 mM coincided with enabling of the reconstituted system in making hexapeptide fusions with sarcosine in a significantly high number of cases. This research work arises from the need to enable a new drug discovery tool that will allow both synthesis and affinity maturation of peptide-based compounds. It represents our pioneering efforts to develop a new technology and ultimately help bring to existence compounds of significant therapeutic value.  iii  TABLE OF CONTENTS Abstract .....................................................................................................................................ii Table of Contents.....................................................................................................................iv List of Tables............................................................................................................................vi List of Figures .........................................................................................................................vii Glossary..................................................................................................................................viii 1  Introduction......................................................................................................................... 1 1.1  The search for new drug candidates ..............................................................................1  1.2  E. coli translation machinery ........................................................................................3 1.2.1 Ribosome ..............................................................................................................4 1.2.2 Translation factors.................................................................................................5 1.2.3 tRNA ....................................................................................................................6 1.2.4 mRNA...................................................................................................................7  1.3  Molecular display technologies ....................................................................................8 1.3.1 mRNA display technology ..................................................................................11  2  1.4  Rationale ....................................................................................................................14  1.5  Hypothesis .................................................................................................................15  1.6  Specific aims..............................................................................................................15  Materials and Methods ..................................................................................................... 16 2.1  Materials ....................................................................................................................16  2.2  Methods .....................................................................................................................19 2.2.1 Ribosome purification .........................................................................................19 2.2.2 Translation factor purification .............................................................................20 2.2.3 Preparation of 10-formyl-tetrahydrofolate ...........................................................22 2.2.4 tRNA synthesis ...................................................................................................23 2.2.5 tRNA aminoacylation..........................................................................................25 2.2.6 Preparation of NVOC-amino acid-pdCpA conjugates..........................................26 iv  2.2.7 Ligation of NVOC-amino acid-pdCpA conjugate to a respective tRNA(-CA)......32 2.2.8 Translation in vitro..............................................................................................32 3  Results ............................................................................................................................... 40 3.1  Assembling the system and initiation study ................................................................40  3.2  Elongation study.........................................................................................................46 3.2.1 Translation of 5GCC template.............................................................................46 3.2.2 Translation of CAC1 and CAC5 templates ..........................................................49 3.2.3 Translation of CAC1 and 1CAC templates ..........................................................56 3.2.4 Translation of templates CAC1 through 5............................................................59  3.3  Sarcosine incorporation study.....................................................................................61 3.3.1 Sarcosine incorporation under standard conditions ..............................................61 3.3.2 Effect of magnesium ion concentration increase on alanine and sarcosine incorporation.......................................................................................................63  4  Disscusions and Conclusion .............................................................................................. 83 4.1  Initiation study ...........................................................................................................83  4.2  Elongation study.........................................................................................................85 4.2.1 Translation of template 5GCC.............................................................................85 4.2.2 Translation of templates CAC1 and CAC5 ..........................................................85 4.2.3 Translation of templates CAC1 and 1CAC ..........................................................87 4.2.4 Translation of templates CAC1 through 5............................................................88  4.3  Sarcosine incorporation study.....................................................................................90  4.4  Conclusion .................................................................................................................92  4.5  Future experiments.....................................................................................................94  Bibliography............................................................................................................................ 97 Appendix ............................................................................................................................... 104  v  LIST OF TABLES Table 1.  Template sequences. ................................................................................................... 18  Table 2.  Yield estimate calculations of Ala-tRNAGGC and Sar-tRNAGGC titration experiment with CAC5 template at a high Mg(OAc)2 concentration......................... 67  Table 3.  One-way ANOVA analysis of total band volume per lane data for alanine translations (0 mM Mg(OAc)2 data point omitted)..................................................... 76  Table 4.  One-way ANOVA analysis of percent top band to total signal per lane data for alanine translations (0 mM Mg(OAc)2 data point omitted). ....................................... 78  Table 5.  One-way ANOVA analysis of percent total band volume per lane data for sarcosine translations (0 mM Mg(OAc)2 data point omitted). .................................... 80  Table 6.  One-way ANOVA analysis of percent top band to total signal per lane data for sarcosine translations (0 mM Mg(OAc)2 data point omitted). .................................... 82  Table 7.  The list of required purified components for the translation system. ....................... 104  Table 8.  Set of DNA primers used for E. coli tRNA library synthesis................................... 106  Table 9.  Standard conditions (composition) of PURE translation system.............................. 107  vi  LIST OF FIGURES Figure 1.  The structure of puromycin. ....................................................................................... 10  Figure 2.  Synthesis of the aminoacyl-pdCpA conjugate............................................................ 29  Figure 3.  Preparative scale HPLC purification chromatogram of NVOC-Sarcosine-pdCpA.... 30  Figure 4.  ESI-MS NVOC-Sarcosine-pdCpA analysis mass spectrum....................................... 31  Figure 5.  Translation of 1GCC template.................................................................................... 41  Figure 6.  Percent translation efficiency at various concentrations of ribosome. ....................... 45  Figure 7.  Translation of 5GCC template.................................................................................... 48  Figure 8.  Translation of CAC1 and CAC5 templates with a single aminoacyl-tRNA. ............. 51  Figure 9.  Translation of CAC1 and CAC5 templates. ............................................................... 53  Figure 10.  Titration of EF-Tu in CAC1 and CAC5 template translation..................................... 55  Figure 11.  3’UTR-independent biocytin incorporation. .............................................................. 58  Figure 12.  Codon position-dependent biocytin incorporation. .................................................... 60  Figure 13.  Undetectable incorporation of sarcosine under standard conditions. ......................... 62  Figure 14.  Titration of Mg(OAc)2 in CAC5 template translation. ............................................... 64  Figure 15.  Titration of Ala-tRNAGGC and Sar-tRNAGGC in CAC5 translation, at a high Mg(OAc)2 concentration. ........................................................................................... 66  Figure 16.  Mg(OAc)2 effects on CAC5 template translation....................................................... 73  Figure 17.  Magnesium effect on total fusion material production based on scintillation counts.......................................................................................................................... 74  Figure 18.  Magnesium effect on total band volume signal using alanyl-tRNA........................... 75  Figure 19.  Magnesium effect on percent top band to total signal using alanyl-tRNA. ............ 77  Figure 20.  Magnesium effect on total band volume signal using sarcosyl-tRNA. ...................... 79  Figure 21.  Magnesium effect on percent top band to total signal using sarcosyl-tRNA. ............ 81  Figure 22.  A selection of SDS-PAGE gel analysis of FPLC purified components of reconstituted system.................................................................................................. 105  vii  GLOSSARY A AARS BME C CPM DPM dNTP EF FPLC G HPLC IF IPTG IVTT ESI-MS mRNA MTF NRPS NTP NVOC PAGE PCR PURE RF RNA RRF RRL rRNA SD SDS-PAGE T TBE %TE 10-formyl-THF TLC tRNA U 3’-UTR  Adenine Aminoacyl-tRNA synthetase β-mercaptoethanol Cytosine Counts per minute Disintegrations per minute Deoxyribonucleotide triposphate Elongation factor Fast protein liquid chromatography Guanine High performance liquid chromatography Initiation factor Isopropyl-ß-D-thiogalactopyranoside In vitro transcription/translation Electrospray ionization mass spectrometry Messenger ribonucleic acid Methionyl-tRNA transferase Nonribosomal peptide synthetase Ribonucleotide triphosphate 6-Nitroveratryloxycarbonyl chloroformate Polyacrylamide gel electrophoresis Polymerase chain reaction Protein synthesis using recombinant elements Release factor Ribonucleic acid Ribosome recycling factor Rabbit reticulocyte lysate Ribosomal ribonucleic acid Shine-Delgarno Sodium dodecyl sulfate polyacrylamide gel electrophoresis Thymine Tris/Borate/EDTA Percent translation efficiency 10-formyl-tetrahydrofolate Thin layer chromatography Transfer ribonucleic acid Uracil 3’-untranslated region viii  1  INTRODUCTION  1.1  The search for new drug candidates  An ongoing interest exists in peptides and peptide-like molecules in the field of drug discovery as more than one half of all current drugs and an even higher percentage of drug targets have a peptide as their native ligand/recognition motif [1]. Standard ways of searching for new drug candidates involve screening collections of natural materials (fermentation broths, plant extracts) or more systematically, manufacturing and screening combinatorial libraries of compounds. Combinatorial libraries of peptides and peptide-like molecules can be generated biologically or synthetically, each having their inherent advantages and disadvantages [2-5]. Synthetic libraries are smaller in size (only up to 107 unique molecules), require complicated coding schemes and/or deconvolution steps, but can use diverse building blocks to generate stable, drug-like molecules [6, 7]. By contrast, the complexity of biologically generated libraries can be on the order of 1014 unique molecules and their encoded members are easily identified by DNA sequencing. As a major disadvantage, however, biological libraries are restricted to the twenty naturally occurring amino acids, thus making these compounds readily hydrolysable in vivo [8, 9]. Two main ways to generate peptides biologically are nonribosomal and ribosomal peptide synthesis. Nonribosomal peptide synthesis is carried out by large, multifunctional enzymes collectively called nonribosomal peptide synthetases (NRPSs) in a template-independent fashion [10]. NRPSs can synthesize peptides and peptide-like molecules out of most diverse building blocks such as D-, N-methyl, and beta-amino acids, as well as fatty acids, sugars and heterocyclic molecules. In addition, NRPS products tend to exist in circular structures, which  1  may be found in ribosomally produced peptides only after they undergo posttranslational modifications [11]. As a result of their composition and topology, NRPS peptides have high affinity for their target, are highly selective and relatively resistant to proteases, thus bearing some desirable drug-like characteristics [12], but so far all reported attempts to generate novel NRPS products by genetically manipulating NRPSs remained unrewarded [13, 14]. In contrast to nonribosomal peptide synthesis, ribosomal peptide and protein synthesis is a method abundantly utilized in all living organisms. This process is based on transcription of genetic information encoded in DNA into sequences of mRNA, and further translation of that information into polypeptide sequences. The fact that this biosynthetic method requires a template that contains encoded information of the product of synthesis provides for an advantage over other methods, as the template molecule sequence can easily be randomized causing the subsequent, controlled randomization of polypeptide product. Furthermore, if the coding template molecule stays bound to its polypeptide product, the link between the polypeptide and template gives rise to a new entity, an mRNA display molecule. In mRNA display both product and template can be easily manipulated (e.g., sequenced or amplified) using basic molecular biology techniques, and this ease and speed in manipulation is not the case with any of the other methods for randomized polypeptide libraries generation (NRPS peptide libraries or synthetic polypeptide libraries). In summary, based on the drug-like qualities of some NRPS products, as well as the nature and availability of ribosomal synthetic machinery, the idea emerged to attempt generation of new, synthetic NRPS-like products. The new material would be generated (1) by using an inexpensive and well studied, E. coli translation machinery, and (2) in a format that allows for quick amplification, sequencing and screening against a selected target (i.e., in an mRNA  2  display library format). Both of these premises form the basis of this research project. In the following sections I will give a brief introduction overview of both E. coli translation machinery and mRNA display technology. 1.2  E. coli translation machinery  Bacterial translation is a complex, five-stage process that starts with (1) activation of amino acids, continues with (2) initiation, (3) elongation, (4) termination and release (considered one step), and ends with (5) folding and posttranslational processing. The main components of the translation machinery are: the ribosome, the protein factors that regulate the process, the transfer ribonucleic acid (tRNA) molecules with their respective amino acid residues, and the messenger ribonucleic acid (mRNA) template. The first stage of translation involves charging tRNA molecules with their respective amino acids, while the last stage of translation usually involves enzymatic glycosylation, phosphorylation, methylation, isoprenylation or acetylation or a removal of one or more amino acids, usually from the Nterminus from the nascent product. Both of these stages take place away from the ribosome, in the cytosol surrounding the ribosome [11]. The first stage of translation, charging of tRNA molecules, is conducted by the enzymatic action of twenty different aminoacyl-tRNA synthetases (AARS), which charge their respective natural amino acids onto their cognate tRNA molecules with relatively high specificity, using the energy of ATP and requiring Mg+2. This step (charging of tRNA molecules) represents the first of three proofreading mechanisms of protein synthesis [10]. The other two proofreading steps are functions of the ribosome and EF-Tu (one of the translation factors), and they arise later in translation. These three steps together represent an important quality control feature of ribosomal translation. Since this feature’s role is to  3  ensure that only naturally corresponding elements are met in ribosomal translation, in our technology development where we attempt to introduce unnatural amino acids into ribosomal synthesis, this feature will need to be mitigated or its elements altogether removed. This concept will be discussed in later sections about peptidomimetic mRNA display technology. While the first and last stages of translation take place in the cytosol surrounding the ribosome, the three core steps of translation take place in the ribosome and are orchestrated by nine proteins – translation factors: initiation factors (IF1, IF2, and IF3), elongation factors (EF-G, EF-Tu, and EF-Ts), and three of four termination factors (RF1 or RF2, RF3, and RRF). In addition, there are other numerous factors and energy sources necessary for the system to work, namely, methionyl-tRNA transformylase (MTF) and 10-formyl-5,6,7,8tetrahydrofolic acid (10-formyl-THF), nucleoside triphosphates (NTPs), creatine phosphate, creatine kinase, myokinase, nucleoside-diphosphate kinase, pyrophosphatase, mono- and divalent cations (Mg+2 being the most dominant one), etc. [15]. The three core steps of the translation process, initiation, elongation, and termination/ release, will be analyzed in the following sections. 1.2.1 Ribosome The ribosome is a large, 2.3-MDa ribonucleoprotein complex that translates genetic information into proteins [16]. The bacterial 70S ribosome is comprised of two subunits: a small (30S) and a large (50S) subunit. Each subunit is composed of ribosomal RNA (rRNA) and ribosomal proteins. In E. coli, the 30S subunit includes a single 1542-nucleotide base rRNA (16S) and 21 ribosomal proteins, whereas the 50S subunit consists of two molecules of rRNA, one comprised of 115 nucleotide bases and another comprised of 2904 nucleotide bases (5S and 23S, respectively) and 33 ribosomal proteins.  4  The bacterial ribosome has three distinct sites that are able to bind tRNA: aminoacyl (A) site, peptidyl (P) site and exit (E) site. While both ribosome subunits contribute in constituting A and P sites, the E site is primarily situated within the 50S subunit [17]. Each one of the ribosome sites plays an important role in steps of translation, as outlined in the next section. 1.2.2 Translation factors Initiation is the second step of bacterial translation and involves the interaction of the 30S subunit with the Shine-Dalgarno sequence on mRNA. That sequence in its several variations is a common attribute of all bacterial mRNA templates, and it is complementary to the 3′ end of the smaller subunit rRNA (16S) according to Watson-Crick base pairing, where adenine (A) forms a base pair with uracil (U) and guanine (G) with cytosine (C). [18]. The process also involves all three initiation factors, IF1, IF2, and IF3. The function of IF1 is to bind to the A site to prevent tRNA binding during initiation and to increase the affinity of IF2 for the ribosome. IF2 is a GTPase that preferentially binds to initiator tRNA acylated with formyl-methionine and helps to bring the aminoacyl-tRNA to the P site of the ribosome [19]. IF3 is known to bind strongly to the 30S subunit to block the E site from tRNA binding and to prevent its association with the 50S subunit prior the joining of all elements required for initiation. It also helps in selection of initiator tRNA by destabilizing the binding of other tRNAs onto the P site of the ribosome [20]. Additionally, IF3 has been found to dissociate deacylated tRNA molecules from the 30S subunit in the last step of translation termination [21]. The end of the initiation finds aminoacylated initiator tRNA in the P site of the ribosome, while leaving an empty A site to mark the beginning of elongation. Elongation is the third translation phase in which aminoacyl-tRNA is brought into the A site as a ternary complex with EF-Tu and GTP. Correct template codon and tRNA anticodon interaction stabilizes tRNA binding to the A site, then triggers GTP hydrolysis by EF-Tu and 5  tRNA movement to the P site. Consecutive peptide bond formation leaves the ribosome with a deacylated tRNA in the P site and peptidyl tRNA in the A site. Next translocation of the tRNAs and mRNA is facilitated by EF-G, which is also a GTPase, and the final result is a ribosome with a deacylated tRNA in the E site, peptidyl tRNA in the P site, and an empty A site, ready for the next round of elongation [18, 22, 23]. The EF-Tu–GTP complex regeneration involves the action of EF-Ts and GTP. The elongation cycle is repeated until the entire coding sequence of the mRNA template is translated and one of the three stop (termination) codons appears in the A site, whereupon translation is terminated [24]. The termination and release, the fourth phase of translation, starts with the recognition of the stop codon and involves one of the two release factors, RF1 or RF2 [25]. While both factors recognize stop codon UAA, RF1 recognizes also UAG, and RF2 recognizes UGA. Hydrolysis of peptidyl tRNA by RF1 or 2 is required for binding GTP to RF3 that takes place on the ribosome. RF3 has a high affinity for the ribosome, which in turn stimulates dissociation of RF1 or 2 [26]. The hydrolysis of GTP is required for subsequent dissociation of RF3. The structure of RRF [28] is very similar to tRNA, supporting the notion that it mimics a tRNA in its binding to the A site of the ribosome [29]. After release of the peptide chain, the ribosome is left with mRNA and a deacylated tRNA in the P site. This complex is disassembled to prepare the ribosome for the next round of protein synthesis with the combined action of RRF and EF-G, while using the energy of GTP [27]. 1.2.3 tRNA During translation, twenty natural amino acid residues are carried to the ribosome attached to their cognate tRNA. The function of tRNAs as adaptor molecules in translation was first proposed by Francis Crick in 1958 [30, 31], and was confirmed experimentally in the 1960s [32]. Despite being relatively small (on average 76 nucleotides long), the tRNA structure is 6  rich in stereo-chemical information and exhibits specific folding for selective RNA–protein interactions, ensuring relatively high fidelity of the translation process (an error rate of translation is 1 in 104 incorporations under normal growth conditions) [33, 34]. It has been identified that structurally and functionally tRNA has four “arms” – an acceptor stem or amino acid arm (3’ end that always ends in –CCA attaches to the aminoacyl residue), an anticodon loop or arm (complementary to an mRNA codon, and specific to the respective amino acid), and D and T arms. The T arm is responsible for binding to the ribosome, while both D and T are important for overall folding of the molecule. 1.2.4 mRNA The template for bacterial translation is a molecule of mRNA. Bacterial mRNA always contains a Shine-Delgarno (SD) sequence in its 5’ sequence, typically 4-5 nucleotide bases long and 5-8 nucleotide bases upstream from the start codon. The SD sequence serves as a recognition site for the bacterial ribosome, more specifically for the 3’ end of the smaller subunit RNA (16S), ensuring correct AUG codon positioning for initiation [18, 35, 36]. No specific sequence has been defined upstream of the SD region, but it is usually U-, A-, or U and A-rich, which are sequences that minimize the secondary structure formation [37-39]. Other pairings between mRNA and the 16S rRNA upstream from the SD sequence proposed in an attempt to explain its stimulatory effect on translation [40] were ruled out due to lack of a change when complementary 16S rRNA sequences were mutated [41]. On the 3’ end of mRNA, many bacterial mRNA molecules contain an 80-250 nucleotide long sequence of adenosine bases, important as a recognition site for mRNA processing enzymes [42]. In between these 5’ and 3’ untranslated regions (UTRs), the mRNA contains the coding sequence for a protein or a peptide that routinely starts with AUG, coding for formylmethionine, and ends with one of the termination codons, UAA, UAG or UGA. Additionally, 7  in bacterial templates the termination codon of one gene often overlaps the start codon of the next gene (e.g., UGAUG), facilitating re-initiation of translation [43]. In summary, ribosomal translation is performed and controlled via joint actions of many components of the ribosomal machinery, and it is very much like translation from one language to another: its output (polypeptide) is made to correspond to the input (mRNA template) according to the translation key outlined by the universal genetic code, which shows the matching of each amino acid (the unit of polypeptide) with its corresponding nucleotide codon (the unit of mRNA coding region). Since this research is centered on enabling the E. coli translation machinery to generate peptide-like products in an mRNA display library format, in the next sections I will elaborate on what constitutes the second aspect of this research – molecular display technologies, and more specifically, mRNA display technology. 1.3  Molecular display technologies  Molecular display refers to the presentation (or the display) of molecules – most commonly proteins or peptides, on a macromolecular scaffold (e.g., on the surface of a virus, a cell, a ribosome, or an mRNA molecule). [44] The sequence information of the displayed molecule is encoded either in or on the entity displaying it, thereby establishing an association between the displayed peptide that results in a phenotype and the encoded sequence of the displayed molecule partitioned in such a way as to provide its unique genotype. The vast potential of molecular display technology was first explored in 1985 by G.P. Smith in his experiments with phage display [2]. He coupled encapsulated DNA (genotype) with displayed protein (phenotype), termed it "fusion phage" and described affinity purification of  8  “virions bearing a target determinant (antigen) from a 108-fold excess of phage not bearing the determinant, using minute amounts of antibody” [45]. Bacterial display was the first cell surface display system, whereby proteins are displayed on the surface of E. coli. This introduced the advantages of high transformation efficiency, ease of manipulation and facile quantitative screening by fluorescence-activated cell sorting (FACS) [46-48]. Yeast display offered all the convenience of bacterial display with the added benefit of eukaryotic protein folding pathways and a codon usage very similar to mammalian cells, therefore becoming most suitable for the display of mammalian extracellular proteins [49, 50]. Lastly, ribosome and mRNA display systems have emerged as cell-free protein engineering technologies by means of in vitro transcription/translation methods (IVTT) [4, 51, 52]. There are two main strategies in the development of IVTT. The first is based on the use of crude cell extracts from E. coli, rabbit reticulocytes, or wheat germ [53, 54]. This technique has been more extensively used due to its simplicity and availability, despite inherent problems such as the rapid depletion of energy charges independent of peptide bond formation and partial degradation of protein product or template by an abundance of proteases and nucleases, respectively [55, 56]. These problems have been overcome by the second strategy, first introduced in 1997 by Weissbach’s group and established in 2001 by Shimizu et al. under the name of protein synthesis using recombinant elements (PURE) [15, 57]. PURE entails the use of purified histidine-tagged protein components of the translation system that includes translation factors, aminoacyl-tRNA synthetases (AARSs), methionyl-tRNA synthetase (MTF), T7  9  RNA polymerase, as well as purified ribosomes, tRNAs, amino acids, NTPs, and a DNA template for the polypeptide or protein of interest. Using either one of the two IVTT strategies, nascent protein or a polypeptide in an mRNA display format becomes covalently linked in situ through the C-terminal end to its encoding mRNA through a puromycin residue, a tyrosyl-tRNA mimic (Figure 1). All templates for mRNA display synthesis are built to bear a puromycin moiety on their 3’ end. In contrast to mRNA display, ribosome display is dependent on the stability of the ternary complex that consists of polypeptide, mRNA template, and the ribosome [58, 59]. The ribosome display ternary complex is quite stable, but non-covalent; there is an extreme size difference between the complex and the displayed product that is likely to cause problems in the affinity maturation phase. In mRNA display, the covalent mRNA-protein fusion allows for more stringent selection methods (e.g., conducted at elevated temperatures) while the size disproportion between the complex and the product is much smaller [58, 60].  ← Resembles adenosine of a tRNA  ← Resembles a tyrosine residue  Figure 1. The structure of puromycin. The structure of puromycin mimics the 3‘ end of tyrosyl-tRNA, with a part of its molecule resembling the adenosine of the tRNA and a part resembling tyrosine. Instead of the tyrosyltRNA ester linkage, puromycin has a much more stable amide linkage (encircled). 10  Another advantage of both IVTT generated display systems is their potential for affinity maturation through recursive mutagenesis, in which selected molecules are mutated after each round of selection. Hanes et al. applied recursive evolution in a ribosome display selection using standard Taq DNA polymerase (1 mutation in every 20,000 bases) to amplify the library after each round of selection [61]. More extensive diversification beyond the capabilities of Taq can be performed by employing methods such as error-prone PCR [62], mutagenic dNTP analogs [63], or DNA shuffling [64], allowing for generation and later investigation of even greater sequence space. Finally, both ribosome and mRNA display systems are particularly suitable for introducing chemical complexity in a library format by incorporating novel chemical functionalities since they are template based and manufactured in vitro [8], permitting a considerable amount of control over the created product. Therefore, the two display systems enable generation of the largest peptidic libraries yet achievable (up to 1014 unique molecules), since the library size is not limited by cellular transformation efficiencies that are prerequisites for establishing cell or viral surface display libraries [58]. 1.3.1 mRNA display technology The idea of using IVTT to obtain mRNA display libraries of compounds with NRPS product-like attributes has been proposed by several groups so far [8, 65-68]. Unnatural chemical functionalities can be incorporated into an mRNA display library co-translationally or posttranslationally by chemical modification [69, 70]. The former requires redesigning of the genetic code by assigning codons de novo to unnatural amino acids [67], as well as engaging different strategies for charging tRNA molecules. It has been demonstrated so far that it is possible to manipulate the genetic code as well as incorporate large numbers of diverse side chain molecules into a polymer using the 11  ribosomal machinery [71, 72]. The most common approach to expanding the genetic code in order to incorporate an unnatural amino acid into a polypeptide chain is nonsense suppression – a process for introducing a chemically charged nonsense suppressor tRNA to the translation reaction while the template contains a nonsense codon in its coding sequence [71, 73]. The nonsense suppressor tRNA is usually a recombinant tRNA molecule that codes for the amber/nonsense (stop) codon, UAG, but it can be made to code for either of the other two stop codons. Nonsense suppression has also been achieved in vivo by successfully transforming cells to synthesize suppressor tRNA and the corresponding mutant AARS [74, 75]. Nonsense suppression has been widely used to demonstrate the ribosome’s ability to incorporate different unnatural elements into natural proteins as well as to study the protein structure and function [76]. This approach however has its limitations, mainly due to the fact that there are only three stop codons, so if one stop codon is used as a termination signal, only two can be used to code for unnatural amino acids during one translation reaction. When nonsense suppression is used in conjunction with PURE methodology, it is possible to omit the release factor specific for the stop codon used in the coding region (RF1 or RF2) so that the mutated protein can be obtained at high yield and purity [15]. The reason for this improvement is that the respective release factor, if present, recognizes the stop codon and prematurely terminates protein synthesis. This introduced competition between translation factors and suppressor tRNAs is an additional limitation to the method [67]. In order to overcome the nonsense suppression limitations, the same approach has been extended to rarely used sense codons by “frameshifting” acylated tRNAs or by expanding the genetic code beyond the three-membered codons by developing four and five-membered  12  codons, as well as three-residue codons with novel base pairs [77-81]. However, overall efficiencies of successful frameshift and other kinds of suppression stayed frequently below 50%, especially when using the first IVTT strategy, making the synthesis of high quality chemically complex pools of encoded peptidomimetics still very difficult [80-82]. One of the reasons for the low efficiency of the strategies described so far was suspected to be primarily the competition from the natural amino acids reinforced by the proofreading mechanisms of AARSs, EF-Tu, and the ribosome. It was demonstrated that those limitations in incorporating unnatural amino acids could be largely overcome by excluding the factors leading to competition in translation, primarily natural amino acids, tRNAs, AARSs, and release factors [83]. For example, by omitting amino acids and AARSs and adding three different, chemically charged tRNAs to a PURE reconstituted translation system, Forster et al. succeeded in simultaneously reassigning three sense codons to unnatural amino acids. This successful reassignment of sense codons to unnatural amino acids sets a new alternative to nonsense suppression in expanding the genetic code, referred to as the sense suppression. Sense suppression has also been demonstrated using unmodified AARS to enzymatically charge tRNAs with unnatural elements [84] or modified pairs of AARS/tRNAs [85], done by Szostak’s research group, or by chemically transforming the natural amino acids after they have been enzymatically charged onto tRNA, as published by Fahnestock and Rich in 1971 [86], and Merryman and Green in 2004 [66]. Fahnestock and Rich performed chemical deaminination of Phe-tRNAPhe to α-hydroxy-Phe-tRNAPhe that was then used for the ribosomal synthesis of a Phe-polyester [86]. Using a similar approach, Merryman and Green were the first to chemically methylate the α-amino groups of the 20 canonical amino acids while attached to their cognate tRNA, and demonstrate translation of N-methyl peptides [66].  13  Finally, the latest reported advances by Suga’s group involved the introduction of a new, ribozyme-based system for tRNA acylation called flexizyme, an AARS-like RNA molecule, that involves chemo-enzymatic ligation of a series of microhelix or full-length tRNA to respective unnatural amino acid [87]. Even though in vitro methods of Merryman and Green, Suga and Szostak have successfully incorporated some N-methylated amino acids into short peptides, the unnatural amino acid incorporation into peptides appears biased. This bias mainly depends on aminoacylation efficiency, which is unequal across the range of tested amino acids in all said studies [68, 88, 89]. This implied that if any of the proposed methods for tRNA charging are used in preparing adaptor molecules for mRNA display library synthesis, the libraries’ complexity and composition will inevitably depend on the bias introduced prior to their creation, leaving the library selection process evaluators guessing whether the best candidate even entered the process. 1.4  Rationale  The original intent of this study was to further explore the experimental approach to sense suppression first introduced by Frankel et al. in 2003. [8]. They used chemically charged tRNAs to reassign a Val and an Ala codon to N-methyl-Phe and demonstrated the synthesis of protease resistant poly-N-methyl-Phe peptides. In addition, Tan et al. used sense codon reassignment and chemically charged tRNAs to test the translation of several different backbone analogues of Phe and Ala and found that the N-methyl and α-hydroxy analogues of these amino acids are efficiently incorporated at a single codon in a tripeptide [90]. Both of these studies utilized a modification of Hecht’s method to chemo-enzymatically charge tRNAs [91] [92]. This method, unlike other listed methods for tRNA charging, has  14  not been tested sufficiently to show whether it would introduce a bias in incorporation efficiency of unnatural amino acid into mRNA display. Preliminary tests performed in the Frankel laboratory have shown little or no difference in mRNA display incorporation efficiency among 14 different N-methyl and cyclic amino acids (unpublished results). In my thesis research, the chemo-enzymatic method for tRNA charging and an in-house prepared reconstituted translation system [15] were used to (1) demonstrate site specific incorporation of a natural (alanine) and selected unnatural amino acids (sarcosine and biocytin) into mRNA display, as well as to (2) search for the most favorable conditions for selected unnatural amino acid incorporation. Once confirmed functional, this system is intended to provide a fundamental tool for synthesis of peptidomimetic libraries in an mRNA display format. Future research will help demonstrate whether obtained libraries possess drug-like qualities (e.g., proteolytic resistance), as well as use them as pools of ligand candidates against selected targets. 1.5  Hypothesis  It is possible to generate mRNA-peptidomimetic fusions using reconstituted translation system and chemo-enzymatically charged tRNAs for unnatural amino acid incorporation into mRNA-peptidomimetic fusions. 1.6  Specific aims  (1) Generate components of PURE system (reconstituted system for translaton in vitro). (2) Perform translations in vitro using the PURE system to incorporate natural and unnatural amino acids into a single template mRNA display fusion, and find the most favorable conditions for full-length fusion synthesis.  15  2  MATERIALS AND METHODS  2.1  Materials  Reagents were obtained from the following sources: Applied Biosystems/Ambion (Austin, Texas, USA) RNASecure® Amersham Biosciences (Piscataway, New Jersey, USA) Oligo(dT)-cellulose type 7. Bachem Americas, Inc. (Torrance, California, USA) N-methyl glutamic acid; homoproline; thioproline; t-Bu-glycine. Dr. Jack Szostak at Massachusetts General Hospital (Boston, Massachusetts, USA) All clones for protein components of reconstituted system (IF1, IF2, IF3, EF-Tu, EF-Ts, EFG, MTF, all AARS, T4 RNA ligase). Dr. Richard Fahlman, University of Alberta (Edmonton, Alberta, Canada) T7 RNA polymerase clone. E. coli Genetic Stock Center, Yale University, (New Haven, Connecticut, USA) A19 (CGSC#5997) E. coli bacterial cells for ribosome preparation. Fisher Scientific Ltd. (Vancouver, British Columbia, Canada) Acetonitrile; dichloromethane; ethanol; glacial acetic acid; glycerol; hydrochloric acid; methanol; tetrahydrofurane. Fluka and Riedel-de Haën, Sigma-Aldrich Laborchemikalien GmbH (Seelze, Germany) Biocytin. Integrated DNA Technologies (Toronto, Ontario, Canada)  16  M13 reverse and M13 forward primers, as well as custom made primers for Ala-tRNA construction and –CA tRNA library construction (Table 8, APPENDIX). National Diagnostics (Atlanta, Georgia, USA) SequaGel® and ProtoGel® reagents. New England Biolabs (Ipswich, Massachusetts, USA) All restriction digest enzymes and their respective buffers; T4 RNA ligase buffer. Roche Applied Science (Laval, Quebec, Canada) PWO SuperYield DNA polymerase. Sigma Chemical Co. (St. Louis, Missouri, USA) initiator tRNA and total tRNA from E. coli; all reagents, unless otherwise stated. Schricks Laboratories (Jona, Switzerland) (6S)-5-formyl-5,6,7,8-tetrahydrofolic acid, calcium salt. Pierce, Part of Thermo Fisher Scientific Inc. (Rockford, Illinois, USA) B-PER Bacterial Protein Extraction Reagent PerkinElmer Life And Analytical Sciences, Inc. (Waltham, Massachusetts, USA) 35  S-methionine (43.5 TBq/mmol).  Stratagene (La Jolla, California, USA) QuickChange® site-directed mutagenesis kit. TriLink Biotechnologies (San Diego, California, USA) Synthetic mRNA templates (Table 1).  17  Table 1. Template sequences. Template sequences used in this study and their corresponding full-length translation products are shown. Capital letters indicate DNA, small letters RNA, and underlined sequence corresponds to the first codon of the coding region. Each X indicates one Spacer 9 residue (triethylene glycol phosphate).  41P template and fusion product: 5'-ggaggacgaa aug AAA AAA AAA AAA AAA AAA AAA XXX ACC (puromycin)-3' Met-3’-(puromycin)-DNA/RNA-5' 1GCC template and fusion product: 5'-ggaggacgaa aug gcc AAA AAA AAA AAA AAA AAA AAA XXX ACC (puromycin)-3' Met-Ala-3’-(puromycin)-DNA/RNA-5' 5GCC template and fusion product: 5'-ggaggacgaa aug gcc gcc gcc gcc gcc AAA AAA AAA AAA AAA AAA AAA XXX ACC (puromycin)-3' Met-(Ala)5-3’-(puromycin)-DNA/RNA-5' CAC1 template and fusion product*: 5'-ggaggacgaa aug cac gcc gcc gcc gcc AAA AAA AAA AAA AAA AAA AAA XXX ACC (puromycin)-3' Met-Bio-(Ala)4-3’-(puromycin)-DNA/RNA-5' CAC2 template and fusion product*: 5'-ggaggacgaa aug gcc cac gcc gcc gcc AAA AAA AAA AAA AAA AAA AAA XXX ACC (puromycin)-3' Met-Ala-Bio-(Ala)3-3’-(puromycin)-DNA/RNA-5' CAC3 template and fusion product*: 5'-ggaggacgaa aug gcc gcc cac gcc gcc AAA AAA AAA AAA AAA AAA AAA XXX ACC (puromycin)-3' Met-(Ala)2-Bio-(Ala)2-3’-(puromycin)-DNA/RNA-5' CAC4 template and fusion product*: 5'-ggaggacgaa aug gcc gcc gcc cac gcc AAA AAA AAA AAA AAA AAA AAA XXX ACC (puromycin)-3' Met-(Ala)3-Bio-Ala-3’-(puromycin)-DNA/RNA-5' CAC5 template and fusion product*: 5'-ggaggacgaa aug gcc gcc gcc gcc cac AAA AAA AAA AAA AAA AAA AAA XXX ACC (puromycin)-3' Met-(Ala)4-Bio-3’-(puromycin)-DNA/RNA-5'  *In some experiments, sarcosine was used in place of alanine (with templates CAC1-5).  18  2.2  Methods  2.2.1 Ribosome purification Purified ribosomes were obtained from A19 (CGSC#5997) E. coli bacterial cells, as described [67]. Three-liter bacterial cultures were grown to mid-log phase in Lauria-Bertani (LB) broth and harvested by centrifugation in an Eppendorf 5810R centrifuge, rotor F34-6-38 (Eppendorf Canada, Mississauga, Ontario, Canada) for an hour at 12,000 rpm at 4˚C. The cell pellet was washed using a wash buffer (10 mM Tris-HCl pH 7.5, 10 mM Mg(OAc)2, 100 mM NH4Cl, 0.25 mM ethylene-diamine-tetraacetic acid (EDTA), and 7 mM β-mercaptoethanol (BME)) and resuspended in B-PER solution supplemented by 4 mM Mg(OAc)2, 100 mM NH4Cl, 0.25 mM EDTA, 200 µg/mL lysozyme, and 7 mM BME. Clear S30 fraction was obtained after three successive 30-minute spins at 15,600 rpm using a Beckman UltraCentrifuge (Beckman Coulter Canada, Inc., Mississauga, Ontario, Canada) with a SW41 Ti swinging bucket rotor at 4˚C. The obtained supernatant (30S fraction) was flash-frozen in liquid nitrogen and stored at minus 80˚C till further purification. Purified ribosomes were obtained from a 20-mL portion of the S30 fraction by three successive 18-hour salt washes at 28,500 rpm, in 30% (w/v) sucrose in buffer (10 mM TrisHCl pH 7.5, 10 mM Mg(OAc)2, 500 mM NH4Cl, and 7 mM BME) at 4˚C, using a Beckman UltraCentrifuge with a SW41Ti swinging bucket rotor as described [67]. Following the third salt wash, the cell pellet was re-suspended in storage buffer (10 mM Tris-HCl pH7.5, 10 mM Mg(OAc)2, 60 mM NH4Cl, 0.5 mM EDTA, 3 mM BME). The concentration of ribosomes in solution was determined as per Lambert-Beer’s law from the UV absorbance at 260 nm and standard OD260 values as described (1 OD260 = 23 pmoles of purified ribosomes) [93], with an  19  average stock concentration of approximately 5 µM. The solution was aliquoted, flash-frozen in liquid nitrogen and stored at minus 80˚C. 2.2.2 Translation factor purification Translation factors and other enzymes were expressed and purified as recombinant proteins bearing a six-histidine tag, as previously described [15]. The purified translation factors and enzymes included three initiation factors (IF), three elongation factors (EF,) methionyl-tRNA transformylase (MTF), T7 RNA polymerase, T4 RNA ligase and twenty aminoacyl tRNA synthetases (AARS), including MetRS and AlaRS, needed for initial experiments with natural amino acids. The clones of BL21(DE3) E. coli , kindly supplied by Dr. Jack Szostak (Harvard University) (Table 7, APPENDIX), containing plasmids coding for respective proteins coupled with six histidine residues, were grown in 1 liter LB broth each with the addition of the appropriate antibiotic [15]. The protein expression was induced by adding isopropyl-ß-D-thiogalactopyranoside (IPTG) to a final concentration of 1 mM to the culture in mid-log phase. The cells were harvested in a hanging basket rotor A4-81 spun in an Eppendorf 5810R centrifuge at 4˚C, 4000 rpm for 20 minutes and subsequently washed with the wash buffer (50 mM HEPES-KOH pH 7.6, 1 M NH4Cl2 and 10 mM MgCl2). A volume of 2-5 mL of lysis buffer (50 mM HEPES-KOH pH 7.6, 1 M NH4Cl2, 10 mM MgCl2, 0.3 mg/mL lysozyme, 0.1% Triton X-100, 0.2 mM phenylmethylsulphonyl fluoride (PMSF) and 7 mM BME) per gram of cell pellet was added to lyse the cells and incubated on ice for 30 minutes. The cells were additionally lysed by sonication three times in duration of one minute with a one minute of break in between the treatments, with Branson Sonifier 450 (Branson Ultrasonics Corp., Danbury, Connecticut, USA), using the following settings: timer: hold, duty cycle: 50%, and output control: 3.  20  The lysate was cleaned of cell debris by centrifugation using an Eppendorf 5810R centrifuge rotor F34-6-38 for 1 hour at 12,000 rpm at 4˚C. The proteins in the supernatant were purified on AKTA FPLC (GE Healthcare Bio-Sciences Inc., Baie d'Urfé, Québec, Canada), using freshly regenerated 2 x 1 mL HisTrap FF affinity columns (GE Healthcare), at linear gradient elution of 0% to 100% B in 20 minutes, typically at 1 mL/min flow rate. The binding buffer contained 50 mM HEPES-KOH pH 7.6, 1 M NH4Cl2, 10 mM MgCl2, 7 mM BME and 20 mM imidazole, and the elution buffer contained 50 mM HEPES-KOH pH 7.6, 1 M NH4Cl2, 10 mM MgCl2, 7 mM BME and 400 mM imidazole. The collected fractions were analyzed using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). SDS-PAGE gel was performed as per a modified LaemmLi method [94]. The discontinuous SDS-PAGE gel consisted of 5% acrylamide stacking gel and a 10% acrylamide separating gel. The 5% stacking gel contained 1.0 M Tris-HCl, pH 6.8, 0.1% (w/v) SDS, 16.3% (v/v) Protogel®, stabilized, ready-to-use 30% (w/v) acrylamide/methylene bisacrylamide solution (37.5:1 ratio), 0.1% (w/v) ammonium persulfate and 0.1% (v/v) TEMED. The 10% separating gel contained 1.5 M Tris-HCl, pH 8.8, 0.1% (w/v) SDS, 34% (v/v) Protogel®, 0.1% (w/v) ammonium persulfate and 0.04% (v/v) TEMED. The 0.75 mm thick gels were prepared and run using a BioRad Mini-PROTEAN 3 electrophoresis system. The samples were typically loaded in a 10 µL total volume, diluted two times by 2X SDS-PAGE sample loading buffer (0.0375 M Tris-HCl, pH 6.8, 10%(v/v) glycerol, 0.2% (w/v) SDS, 5% (v/v) BME, 0.02% (w/v) bromophenol blue (BPB)). Electrophoresis was performed at a constant voltage of 100 V per gel, through both the stacking gel and the separating gel, using ten times diluted 10X Tris-Glycine-SDS-PAGE  21  Buffer (0.25 M Tris base, 1.92 M glycine, and 1% (w/v) SDS). After the run, the gels were stained with Coomassie blue, dried onto a filter paper and examined. Upon determining that respective fractions were of appropriate purity and molecular weight (Figure 22, APPENDIX), the fraction contents were pooled and desalted using Amicon Ultra-15 centrifugal columns (Millipore, Billerica, Massachusetts, USA), by spinning at 4˚C and 4000 rpm three times in duration of half an hour. The solution volume was brought up to 10 mL with the storage buffer (50 mM HEPES-KOH pH 7.6, 100 mM KCl, 10 mM MgCl2, 30% (v/v) glycerol, and 7 mM BME) before each spin as a buffer exchange during protein concentration. After the third spin, the sample solutions were diluted to 10 mL and enzyme concentrations were determined based on UV absorbance at 280 nm and respective extinction coefficients [95] (as outlined in Table 7, APPENDIX). The protein solutions were aliquoted and stored at minus 80˚C. 2.2.3 Preparation of 10-formyl-tetrahydrofolate 10-Formyl-tetrahydrofolate (10-formyl-THF) was prepared as previously described [96]. Fifteen milligrams of starting material, 5-formyl-tetrahydrofolate (M.W. 511.5), was dissolved in 1 mL purified water while kept on ice. The pH was adjusted to 1.5 with 1M HCl and the solution volume made up to approximately 8 mL with undiluted BME, converting the starting material to 5,10-methylene-tetrahydrofolate. The pH was adjusted to 8.5 with 2 M KOH and the solution volume filled to 10 mL with BME. This last step converts 5,10formyl-THF to 10-formyl-THF. Two-milliliter aliquots were transferred into pre-chilled cryotubes and each aliquot overlaid with 300 µL mineral oil and stored at minus 80˚C. According to the method reference, estimated efficiency of this conversion is greater than 95%.  22  2.2.4 tRNA synthesis Full-length tRNA from the alaW gene in E. coli was amplified out of the respective clone plasmid prepared in-house by PCR with PWO SuperYield DNA polymerase, using the M13F primer and a custom reverse primer (5' - TGG TGG AGC TAA GCG G - 3'). The same tRNA minus 3’ -CA was made by amplifying the same cloned plasmid with M13F and a –CA custom reverse primer (5' - GTG GAG CTA AGC GGG - 3'). The PCR program used for both versions of tRNA was: 1) 95˚C, 1 minute 2) 95˚C, 0.5 minute 3) 50˚C, 1 minute 4) 72˚C, 1 minute, cycle 29 times to 2) 5) 72˚C, 10 minutes 6) 16˚C, indefinitely. The PCR product size was confirmed by analyzing on a 2% agarose gel made with 0.5X Tris/Borate/EDTA buffer (TBE) (10X TBE buffer is 0.089 M Tris base, 0.02 M EDTA and 0.089 M boric acid) and the addition of 0.03% (w/v) ethidium bromide for later visualizing under UV light. The gels were cast and run using a BioRad Mini-Sub Cell GT electrophoresis cell. The gels were typically programmed to operate at 100 V per gel and sample volume of 12 or 18 µL which contained an aliquot of PCR mixture brought up to volume with 6X sample loading buffer (10 mM Tris-HCl, pH 7.5, 0.03% (w/v) BPB, 0.03% (w/v) xylene cyanol, 60% (v/v) glycerol, 60 mM EDTA). Once it was confirmed that PCR products were of expected size (approximately 130 base pairs), the PCR reaction mixtures were extracted in an equal volume of phenol:CHCl3:isoamyl alcohol (25:24:1, pH 5.0) and ethanol precipitated by adding 0.1 volume 3 M sodium acetate, pH 5.2 and 2.5 23  volume ethanol. The solutions were placed at minus 80°C for 20 minutes (or minus 20°C overnight) to promote the precipitation of nucleic acid content, spun down at 14000 rpm in Eppendorf 5417R centrifuge at 4°C, rinsed with 70% (v/v) ethanol and dried under vacuum. The obtained DNA template was transcribed overnight using T7 RNA polymerase prepared in-house using 100-200 µg DNA template, 5X transcription buffer (400 mM HEPES-KOH, pH 7.5, 10 mM spermidine, 200 mM dithiothreitol (DTT) and 125 mM MgCl2), 5X NTP mix (20 mM each NTP) and 25X RNASecure®). Obtained nucleic acid material was cleaned from protein content by phenol/chloroform extraction and precipitated with ethanol as described, to be subsequently purified by analyzing on 10% urea-PAGE (prepared using National Diagnostics Sequagel® solutions and protocol) and eluted out of the excised gel bands in a buffer containing 0.5 M NH4OAc, 1 mM EDTA, pH 8.0 and 0.1% (w/v) SDS. The eluted tRNA was ethanol precipitated as described, dissolved in purified water and the concentration of tRNA in solution determined from the UV absorbance at 260 nm (ε260 nm ≈ 750 000 M-1*cm-1). The solution was stored at minus 20°C. For incorporation of unnatural amino acids, a library of 19 tRNA molecules based on the same alaW tRNA gene clone but lacking the two 3’ terminal bases, C75 and A76, was made using Stratagene QuikChange® site-directed mutagenesis kit and protocol and 19 pairs of custom-made primers (Table 8, APPENDIX). Following the site-directed mutagenesis procedure (to change the anticodon) and Dpn I treatment (to remove methylated and hemimethylated molecules of originally introduced plasmid template), obtained mutant pUC19 plasmids were used to transform XL-1 Blue competent cells made in-house using a standard protocol per Stratagene instructions (Stratagene, An Agilent Technologies Division, La Jolla, California, USA).  24  After sequences were confirmed, the plasmids were obtained from the respective clone cultures and amplified by PCR with PWO DNA polymerase, using M13F and a custom reverse primer (5' - GTG GAG CTA AGC GGG - 3') to obtain a -CA-3’ end. Amplified sequences were used for in vitro transcription by T7 RNA polymerase and transcripts extracted with phenol/chloroform, precipitated with ethanol, and further purified on 10% urea-PAGE gel, as described above. The bands of correct size were observed by UVshadowing, excised and eluted out of the gel and ethanol precipitated. The pellets were dissolved in purified water and the concentration of tRNA in solution determined from the UV absorbance at 260 nm (ε260 nm ≈ 750 000 M-1*cm-1). The solutions were stored at minus 20°C. For the purpose of this thesis research, in vitro made tRNAGGC full length was used enzymatically charged with alanine for alanine incorporation into fusions, and tRNAGTG and tRNAGGC lacking –CA-3’ were chemo-enzymatically charged and used for biocytin and sarcosine incorporation into fusions, respectively. 2.2.5 tRNA aminoacylation [35S]Met was enzymatically acylated to commercial initiator tRNA by MetRS, and incorporated in translation product in situ, during translation initiation. Full-length tRNAGGC as well as total E. coli tRNA for test translations were charged enzymatically with natural amino acids (tRNAGGC with alanine) as required, using a previously described protocol [97] with some modifications. To allow tRNA folding, solutions containing approximately 100 µg tRNA were heated in the presence of 15 mM MgCl2 at 90°C for 3 minutes and slowly cooled to ambient temperature in a beaker of water.  25  The reactions were initiated by adding approximately 1.2 µM of AARS in the presence of buffer that contains 30 mM HEPES, pH 7.5, 30 mM KCl, 15 mM MgCl2, 4 mM ATP and 5 mM DTT and in a total volume of 2 mL. The solutions were incubated at 37°C for 1 hour after which the reactions were quenched by placing on ice and subsequently extracted with an equal volume of phenol:CHCl3:isoamyl alcohol (25:24:1, pH 5.0), precipitated with ethanol as described in Section 2.2.4 and obtained pellets dissolved in purified water. The solutions were stored at minus 80°C. 2.2.6 Preparation of NVOC-amino acid-pdCpA conjugates In order to chemo-enzymatically charge unnatural amino acids onto designated tRNAs, NVOC-amino acid derivatives conjugated to a dinucleotide, pdCpA, were synthesized using a standard method [8, 71, 72] (Figure 2). Equimolar amounts of the amino acid (1) were protected on the N-terminus with NVOC-Cl (2) under mild basic conditions. The protected amino acid was activated on the C-terminus with chloroacetonitrile under dry conditions to form a reactive cyanomethyl ester (3). Obtained cyanomethyl ester was separated from unreacted starting material by silica gel chromatography in an optimized dichloromethane (DCM)/methanol mobile phase (typically 95:5 (v/v) = DCM:methanol). The purified product was confirmed by electrospray ionization mass spectrometry (ESI-MS) flow injection analysis (sample volume 2 µL), using Waters Acquity UPLC with Waters Micromass Quattro Premier XE (Waters Ltd., Mississauga, Ontario, Canada). Obtained compound (3) was reacted with the 5’ phosphorylated dinucleotide pdCpA (made in-house as described) [71] to form the chemically-acylated dinucleotide (4), which was purified on Waters HPLC system consisting of Waters 600 and Waters 2996 photodiode  26  array detector and a preparative C18 HPLC column SunFire C18 OBD (19 x 150 mm) (Waters Ltd., Mississauga, Ontario, Canada) by reverse phase with a mixture of ammonium acetate pH 4.5 and acetonitrile (ACN) as the mobile phase (Mobile Phase A: 10% ACN in 10 mM ammonium acetate, pH 4.5 and Mobile Phase B: 70% ACN in 10 mM ammonium acetate, pH 4.5), using a linear gradient from 0% to 100% B in 60 minutes and the flow rate of 10 mL per minute. The HPLC elution was monitored at 260 nm and 350 nm for the presence of dinucleotide (ε260 nm = 25000 M-1*cm-1) and NVOC (ε350 nm = 6336 M-1*cm-1), respectively. The main goal of HPLC purification of conjugates was to separate them from un-reacted pdCpA, so the dinucleotide ligates to the tRNA with the amino acid attached to it. Peaks at both wavelengths (260 nm and 350 nm) were collected and after evaporating the mobile phase out of collected fractions using lyophilizer Labconco FreeZone 4.5 (Labconco Corporation, Kansas City, Missouri, USA). Each lyophilized fraction was separately dissolved in 10 mM glacial acetic acid and lyophilized again to exchange the remaining salt from the mobile phase after which the respective analytes were identified by ESI-MS. The fraction containing the clean conjugate is dissolved in DMSO and the conjugate concentration adjusted to approximately 3 mM based on UV absorbance. All products were stored at minus 80° C. In the final step of chemo-enzymatic acylation, tRNA molecules that were lacking 3’ -CA ending were ligated to a molar excess of (4) with T4 RNA ligase, as per the procedure described in Section 2.2.7. Four unnatural amino acids were prepared for future use with reconstituted system as described here: sarcosine, N-methyl-L-glutamic acid, L-homoproline and t-Bu-L-glycine and  27  one natural amino acid, L-alanine. The NVOC-sarcosine-pdCpA conjugate was used in this thesis research for testing the system ability to incorporate an unnatural element into mRNA fusions (along with biocytin) in making of completely orthogonal products. The NVOCpdCpA conjugate of biocytin was synthesized by Michelle Zhou using the same methodology. The obtained ESI-MS results for synthesized amino acid conjugates were as follows: NVOCsarcosine-pdCpA, yellow powder, m/z calculated from C32H40N10O20P2 is 946, ESI-MS found (M+H)+ 947 and (M+Na)+ 969; NVOC-N-methyl-L-glutamate methyl ester-pdCpA, yellow solid, m/z calculated from C36H46N10O22P2 is 1032, ESI-MS found (M+H)+ 1033; NVOC-Lhomoproline-pdCpA, yellow liquid, m/z calculated from C35H44N10O22P2 is 986, ESI-MS found (M+H)+ 987 and (M+Na)+ 1009; NVOC-t-Bu-L-glycine-pdCpA, yellow liquid, m/z calculated from C35H46N10O20P2 is 988, ESI-MS found (M+H)+ 989; NVOC-L-alaninepdCpA, yellow powder; m/z calculated from C32H40N10O20P2 is 946, ESI-MS found (M+H)+ 947.  28  20 R1  1  OH  NH  H3C  +  H3C  O  H3C  O  O  O  3  O  O +  O  N  O +  N  O  R2  R1  1)10%Na 2CO3/THF 2)ClCH 2CN/DMF (dry)/TEA(24eq.)  N  O  N -  O  Cl -  O  2  R1= -H, -CH3  22 H3C  O  21  R2  pdCpA+TBA-OH/DMF (dry), TBA-OAc (catalytic amount)  O  O O  23 H3C H3C  O  O +  O  N  4  O  R2 O  N -  R1  pdCpA  O  O  Figure 2. Synthesis of the aminoacyl-pdCpA conjugate. The amino terminus of amino acid (1) is protected with NVOC group (2) and subsequently activated on its C terminus to form a cyano-methyl ester (3). Activated amino acid is conjugated to pdCpA to form an aminoacyl-dinucleotide (4). This conjugate can be ligated to tRNA by T4 RNA ligase.  29  Figure 3.  Preparative scale HPLC purification chromatogram of NVOCSarcosine-pdCpA. NVOC-Sarcosine cyanomethyl ester in large excess (210 µmol), previously purified using column chromatography, has been reacted with 7.2 µmol pdCpA in 80 µL DMF in the presence of catalytic amount of t-Bu-ammonium acetate, as described. After 6 hours, the reactions were quenched by addition of 1 mL 10% ACN in 10 mM ammonium acetate, pH 4.5, and purified using preparative scale HPLC. The peak at retention time 27.8 minutes was suspected to be the conjugate peak based on its approximate 4:1 ratio of absorbances at 260 and 350 nm, respectively (the figure shows the overlay of chromatograms at the two wavelengths of interest), and that was confirmed using ESI-MS analysis (Figure 4).  30  Figure 4. ESI-MS NVOC-Sarcosine-pdCpA analysis mass spectrum. The chromatogram confirms NVOC-Sarcosine-pdCpA analysis resulted in the calculated m/z as expected in positive ion mode: (M+H)+ 947 and (M+Na)+ 969.  31  2.2.7 Ligation of NVOC-amino acid-pdCpA conjugate to a respective tRNA(-CA) Chemo-enzymatically charged tRNAs were used to incorporate unnatural amino acids into the fusions, as well as incorporate biocytin (Bio), which allowed products of translation to be identified using a streptavidin gel-shift assay [87]. NVOC-amino acid-pdCpA conjugates were ligated onto respective tRNA lacking the 3’ C75 and A76 using previously described protocol [98] with some modifications. To allow tRNA folding, solutions containing approximately 40 µg -CA-3’ tRNA were heated in the presence of 2 volumes 10 mM HEPES, pH 7.5 at 90°C for 3 minutes and slowly cooled to ambient temperature in a beaker of water. The ligation reactions were initiated with the addition of 6.25 mM T4 RNA ligase (made in-house) to a solution of folded tRNA containing 50 mM Tris-HCl, pH 7.8, 10mM MgCl2, 1 mM ATP, 10 mM DTT and 0.3 mM NVOC-Bio-pdCpA, in a total volume of 160 µL at 25°C. The solutions were incubated at 37°C for 1 hour after which the reactions were made up to 250 µL with purified water and extracted with an equal volume of phenol:CHCl3:isoamyl alcohol (25:24:1, pH 5.0); nucleic acid content is precipitated with ethanol as described in Section 2.2.4 and obtained pellets dissolved in purified water, adjusted to approximately 40 or 80 µM (equivalent to 1 or 2 mg/mL, respectively) for each acylated tRNA (ε260 nm ≈ 769902 M-1*cm-1) and stored at minus 80°C. The solutions were photo-deprotected prior to use in translation (to remove the N-terminal NVOC group), using a Newport mercury-xenon lamp (Newport Corporation, Irvine, California, USA) with a 315-nm cut-off filter at 500 W for 7 min as previously described [8]. 2.2.8 Translation in vitro The translation reactions were performed as described [67] [99]. Polymix buffer pH 7.6 (5 mM Mg(OAc)2, 0.5 mM CaCl2, 8 mM putrescine, 1 mM spermidine, 5 mM K(PO4)3, 95 mM KCl, 5 mM NH4Cl, 1 mM dithiothreitol, 3 mM ATP, 2.5 mM GTP, 6 mM 32  phosphoenolpyruvate, and 0.01 mg/mL pyruvate kinase) was supplemented with 10 mM creatine phosphate and 30 µM 10-formyl-THF (3 mM stock solution of 10-formyl-THF was made in-house as described previously [96]). The reconstituted system (PURE) translation reactions contained a final concentration of approximately 0.5 µM of each MTF, Met-RS, IF1, IF2, IF3, EF-Ts, EF-G, and 6.6 µM EF-Tu, 1.2 µM purified ribosomes, 0.8 µM [35S]Met and 0.4 µM of fMet-tRNA. The reactions worked best with 0.8-1.0 µM template [100] in a volume of 50 µL and 1-3 µg of separately acylated tRNAs per template codon. Detailed PURE system standard conditions (composition) based on previous studies [67, 100] and the appropriate modifications based on experimental findings are shown in Table 9 in the APPENDIX. Translation reactions were assembled on ice by making a master mixture of all the common components, and adding each in 10% excess to account for loss in pipetting in that first step. This first step was regarded necessary not only to simplify and shorten the sample preparation but also to minimize the random error. Upon homogenizing the master mixture, the appropriate exact volumes of master mixture were aliquoted into individual tubes and the components that were varying for that experiment were added to the tubes. Finally, the mRNA template is added last and the samples were incubated at 37°C for 1 hour. Following the 1 hour incubation, KCl and Mg(OAc)2 concentrations were adjusted to 550 and 50 mM, respectively. A 10X stock of this solution (termed “stop salts”) containing 4.55 M KCl and 450 mM Mg(OAc)2, was added to each sample accordingly to facilitate mRNA display fusion formation. Upon adding the stop salts solution, the samples were incubated at room temperature for 1 hour and then at minus 20°C overnight [8, 67, 99].  33  Note: According to related literature [67], [35S]Met can be supplemented with 40 µM Met to maximize the yield, but the addition of non-radioactive methionine to the reaction also dilutes the translation product radioactive signal. For that reason, the experiments with nonradioactive methionine were not found useful in this research. 2.2.8.1  dT25-cellulose purification  Translated mRNA-puromycin-peptide fusions were purified on dT25-cellulose and quantified by measuring the amount of radioactive material using Beckman Coulter LS 6500 scintillation counter (Beckman Coulter Canada, Inc., Mississauga, Ontario, Canada). To perform dT25-cellulose purification, each 25- or 50-µL total volume reaction mixture was incubated for an hour at room temperature with 100 or 200 µL 25% dT25-cellulose slurry, respectively, in 600 µL incubation buffer (100 mM Tris-HCl, pH 8.0, 1 M NaCl, 0.2% (v/v) Triton X-100) to allow the binding between the poly(T) sequence on the dT25-cellulose beads and the poly(dA) tail present in the DNA linker sequence. Post incubation, the dT25-cellulose beads were washed at least five times to remove all unbound radioactivity and eluted several times with water. A small portion of total, mixed dT25-cellulose elution was aliquoted into vials with scintillation liquid and scintillation counts were measured. The remainder of the eluted product was precipitated with ethanol as described in Section 2.2.4 and obtained dried pellets dissolved in purified water. If storage was needed, the fusion solutions were stored at minus 80°C until further analysis. The dT25-cellulose purification not only allowed for removal of unused radioactivity and PURE system components after translation, it also provided an identification test of the product: if the product of translation indeed was made in an mRNA display format (i.e., contained peptide and nucleic acid), then the radioactivity would associate with material binding to dT25-cellulose as previously demonstrated. 34  2.2.8.2  Scintillation counts assay  Scintillation counting served to determine the reaction yield in counts per minute (CPM), which could be converted to pmol [35S]Met or pmol product. This assay is not specific solely to the full-length product, but rather it is reflective of all mRNA display fusions formed in the reaction. This includes a full-length product, as well as a range of incomplete translation fusions. Full-length fusions are translation products that have the whole length of template coding region translated to a polypeptide sequence, which then forms a fusion through the puromycin residue. Incomplete fusions are translation products that have only a part of the coding region translated into the peptide sequence before the puromycin residue prematurely enters the ribosome, closing the fusion and disabling further translation of the respective template. Determined CPM values were considered equal to the number of disintegrations per minute (DPM), as the counter efficiency was considered to be approximately 100% (instrument scintillation efficiency for [35S]Met is assumed above 0.95 CPM/DMP, as per calibration data for [14C]). CPM values were converted to pmoles of fusion product according to the specific activity of [35S]Met and the following formula: 1 mmol [35S]Met = 37 TBq = 37 *106 MBq (as per specific activity of [35S]Met) therefore: 10 pmol [35S]Met = 0.37 MBq 1 Bq = 1 s–1 = 60 DPM ≈ 60 CPM therefore: 10 pmol [35S]Met = 0.37 MBq = 22.2*106 CPM and 1 pmol [35S]Met = 22.2*105 CPM = 1 pmol fusion product The measured CPM values were corrected by the sample dilution factor and converted to pmol fusion product using the last proportion. The specific activity of [35S]Met was approximated to be equal to the one stated on the certificate of analysis, due to the high 35  turnover of [35S]Met in the laboratory as every solution would be used up within the period of a couple of weeks so the effect of decrease in [35S]Met specific activity due to its natural decomposition (t1/2 = 87.4 days) was negligible compared to that of the other, inherent sources of variability of the method. Note: If [35S]Met in the reactions is supplemented with “cold” Met, then the ratio of concentrations Met and [35S]Met is also treated as a dilution factor for the measured radioactivity of the fusion product. It is approximated that [35S]Met and Met have the same incorporation efficiency. However, the scintillation counts assay was only moderately useful as it could not differentiate between the two kinds of fusions, as the difference between full-length and incomplete fusions was solely in the length of the polypeptide portion of the molecule independent of [35S]Met content. The remainder of dT25-cellulose purified product was typically analyzed on a 12% urea-PAGE gel, which allowed estimating the percentage of full-length vs. incomplete fusions as well as establishing their radioactivity, and it is therefore used to assay the quality of the fusion product. 2.2.8.3  PAGE gel analysis  Fusion material was analyzed on urea-PAGE gels (Sequagel® reagents and protocol, 12% or more cross-linked) using a 1X TBE (Section 2.2.4) buffer and was visualized via GE Healthcare Bio-Sciences Inc. phosphor image screen via a Typhoon variable mode imager (GE Healthcare), typically after two-day exposure. In addition, 10% or more cross-linked Tris-Tricine-PAGE gel was performed as per modified method of Schagger and von Jagow [101] using Protogel® reagent and protocol and analyzed using 0.2 M Tris-HCl pH 8.9 as the anode buffer and 0.1 M Tris, 0.1 M Tricine and 0.1% (w/v) SDS as the cathode buffer.  36  Both urea- and tris-tricine-PAGE gels ware typically prepared as 0.75 mm thick gels and analyzed using BioRad Mini-PROTEAN 3 electrophoresis system at 100 V per gel. The samples were prepared for gel analysis with 2X RNA formamide sample buffer (95% (v/v) formamide, 0.025% (w/v) SDS, 0.025% (w/v) BPB, 0.5 mM EDTA) and loaded typically in 10-20 µL volume. The radioactivity allowed for phosphor imaging of the product while its migration on a gel, depending on the gel resolving ability, was a reflection of the product size; therefore, it was possible to estimate how much of the fusion product was prematurely linked to puromycin and how much of the product had a longer peptide part (larger in size). In order to ensure the sample aliquots carried enough radioactivity to produce an image, the amount of radioactivity loaded into each lane was at least 1,000 - 2,000 CPM to ensure the signal was sufficiently above the background of the phosphor image. In order to additionally confirm that the isolated product was in mRNA display format, the material obtained post dT25-cellulose purification was divided into two equal aliquots prior to analyzing on the gel. One of the two aliquots was treated with RNase A (4 µg RNase A, for 1 hour at 37°C), and the other aliquot kept untreated. Those two aliquots were typically analyzed side by side on the same gel and were expected to show a size difference (the treated sample migrates faster on the gel than the non-treated sample, due to the lack of RNA portion of the template). Streptavidin gel-shift assay. To determine if the biocytin residue was incorporated into translation products, 12% urea-PAGE gel analysis of the total, RNase treated product was performed as described but in the presence of streptavidin (protein, M.W. 52.8 kDa) in the sample loading buffer (2X sample loading buffer contained 0.2 mg/mL streptavidin in 37  37  mM piperazine pH 6.1, 37 mM EDTA and 6 M urea) [87]. This assay allowed an unambiguous determination of the fraction of the product into which biocytin got incorporated, where biocytin would serve as a molecular handle due to the high affinity of streptavidin for biocytin (KD = 1015 M-1). This interaction between biocytin and streptavidin would result in a slower migrating complex on the gel that is referred to as a gel-shift. This allowed for identification and separation of biocytin-containing product on a gel from the rest of the fusions generated in the same reaction. Finally, since biocytin was incorporated site specifically in certain positions of a peptide part of the fusion (according to its assigned codon position), the streptavidin gel-shift assay allowed for considerably more detailed analysis of elongated fusion products than the regular PAGE gel assay. Most gel images were evaluated for presence or absence of signal. Some of the images were evaluated using ImageQuant version 5.2 software (Molecular Dynamics, GE Healthcare BioSciences Inc., Baie d'Urfé, Québec, Canada) to determine the band volume. The band volume was used to calculate the total signal per lane as well as the percent of a certain band out of the total signal per lane. The top band percent out of the total signal per lane was calculated usually after streptavidin gel-shift assay to show how much of biocytin-containing product (the fraction of product that shifts with streptavidin, assay described in the next section) was generated relative to the total fusion product generated in a translation. That calculated percent was used against the total scintillation counts determined per sample to determine the pmoles of biocytin-containing product. In summary, the total signal of fusion product per sample (both, scintillation counts and/or calculated pmoles and arbitrary units of gel image analysis) as well as the percent of the top  38  band to the total signal per lane were used to qualitatively assess the fusion products. These two parameters were only used to compare and determine the trends between the series of samples made for a single experiment, while the absolute value of reaction would vary depending on the day of the experiment or the batch of starting materials used. Nonetheless, the trends and ratios are found repeatable in these assays and are used to make inferences about the reaction outcomes, as well as to determine what changes to the system composition were yielding the most fully elongated product (e.g., EF-Tu or tRNA titration experiments). In the case of magnesium titration experiments, band volume replicates were plotted using GraphPad Prism version 5.0a (GraphPad Software, Inc., La Jolla, California, USA) software to graphically represent the effect of Mg+2 ion increase on unnatural amino acid incorporation and to perform the same software (GraphPad Prism) one-way ANOVA analysis for repeated measurements to determine the statistical significance of the obtained results.  39  3  RESULTS  3.1  Assembling the system and initiation study  As a first step, all protein factors of the reconstituted system were individually expressed in a hexahistidine-tagged format and purified on FPLC as described. Purified protein concentrations were determined spectrophotometrically and assumed to have a 100% activity. Ribosomes and other necessary components were also made as described in the methods section and synthetic DNA/RNA templates were purchased from a commercial source. The first translation reactions were performed using the 1GCC template. Immediately after translations, 2 µL aliquots of the reaction mixtures were analyzed on a 10% tricine-PAGE gel prior to any purification (Figure 5, Panel A), showing the template dependent radioactive product band creation, whereas translation without template showed no distinct bands. The product of the reaction was dT25-cellulose purified and ethanol precipitated as described in Section 2.2.8. An aliquot of each wash as well as elution were analyzed using scintillation counting to determine the presence of radioactivity in elution. The scintillation counts of the wash came down to the background after five washes and that number of washes was typically used in all further assays. The elution scintillation counts showed a radioactive signal pertaining to the fusion product (that stayed bound to the cellulose beads after repeated washes to be eventually eluted with water due to hydrogen bond breakage in a low ionic strength environment), which was confirmed further with ethanol precipitation of eluted product using PAGE assay of RNase treated and non-treated product (Figure 5, Panel B).  40  A RNase A AlaRS (µM) Ala (mM) Ala-tRNAGGC (µg/µL) Template (0.8 µM)  B -  -  -  +  0.5 40 0.0125 1GCC  Same as panel A  1  1  2  2  Figure 5. Translation of 1GCC template. Two 50-µL translation reactions were carried out, one in the absence of any template (Panel A, lane 1), one in the presence of 1GCC template (Panel A, lane 2). The Panel A gel shows analysis of each reaction mix without purification or additional treatments. The gel was dried and visualized using phosphor imaging. After dT25-cellulose purification of template reaction product, half of the product was treated with 4 µg of RNase for an hour, then analyzed on a 10% tricine-PAGE gel again, dried and visualized by phosphor imaging (the image shown in Panel B). After having RNA part of the molecule removed, the treated fusion migrated farther down the gel, while the radioactivity stayed preserved (in a non-treated fusion molecule, the peptide is bound to the template through the puromycin residue that continues into the DNA part of the template; DNA part of the template normally extends into the RNA part of the template, but being that that part was missing in the treated fusions, they were migrating notably lower on the gel).  41  The translation product elution was divided into two equivalent aliquots and one was treated with RNase A. Both aliquots were analyzed on the same tricine-PAGE gel. The migration position offset in size between the non-treated product (Figure 5, Panel B lane 1) and the treated product (Figure 5, Panel B lane 2) demonstrated that the isolated product was indeed in a peptide-nucleic acid fusion format (i.e., in an mRNA display format). After establishing that the system is able to make a product in mRNA fusion format, the next step was to ensure the maximal total fusion yield is achieved by adjusting the relative amounts of the ribosome and the template, as they are the main reaction components. By obtaining maximal total fusion yield of translation, we would ensure that the PURE system is adjusted to reach the maximum translation initiation under given conditions. For these experiments the 1GCC template was used again, as it is the simplest system. A series of translations was performed using varying amounts of ribosome (16, 32, 48 and 74 pmoles per series of reactions, respectively) and template (10, 50 and 100 pmoles per series of reactions, respectively), while the other reconstituted components were kept constant. The translation reactions were carried out on a scale of 50 µL total volume (as described in the methods section) by adding approximately 0.5 µM AlaRS, 0.4 mM alanine and 0.0125 µg/µL alanyl-tRNA to be acylated in situ (as opposed to performing the acylation separately and adding purified and pre-charged tRNA to the translation mix, which was done in later translation experiments). This way, it was assumed that elongation was not interfered by any limiting factors that might have been related to the elongation phase; as both alanyl-tRNA and AlaRS were sufficient and able to recycle, and translation reactions was assumed to be solely dependent on initiation efficiency, as determined by the relative amount and ratio of ribosome and template concentrations.  42  In addition, all materials used (e.g., proteins, tRNA, ribosomes) were out of the same stock solution for each of the respective materials. As the materials were all obtained using in vivo or in vitro biosynthetic processes, the differences in their stock solution composition and concentration between different batches were sometimes quite substantial. As their exact concentration is an important factor that influences the final result of the translation reaction, in order to make reasonably sound inferences out of this experiment it was important to keep using the same stock solutions of all the reaction components, while varying only ribosome and template concentrations accordingly. Upon translation and “stop salt” incubation, translation products were purified using dT25-cellulose. After five washes, the wash solution radioactivity was equal to background and several elutions were performed with a total volume of 800 µL water. An aliquot of each sample elution was used for scintillation counting and obtained CPM values were converted to pmoles of fusion product. The results were plotted using Microsoft Excel 2004 for Mac, version 11.5, to compare the change in the amount of the fusion product with the change in ribosome and template concentrations, as shown in the Figure 6. As a measure of total yield of translation, % translation efficiency was determined relative to the amount of template and concentration of ribosome used in translation. Percent translation efficiency (%TE) was calculated as a percent ratio of pmol product generated and pmol template used in the translation. Figure 6 shows on the y-axis %TE as a function of pmol ribosome (x-axis). The graph shows three curves, each corresponding to a different template concentration at different ribosome concentrations as indicated in the figure legend. Based on the obtained data, it was concluded that within the tested range of concentrations, the highest  43  relative yield as well as the respective signal intensity was obtained with approximately 50 pmol of each, ribosome and template per 50 µL reaction, (the graph line with a square label in the Figure 6) and those approximate concentrations were used in the remainder of the study experiments. The graph with the diamond label (obtained at 10 pmol template) as well as the graph with the triangular label (obtained at 100 pmol template), showed either a very poor radioactive signal (10 pmol template) or a too low of translation efficiency (100 pmol template), and those relative template concentrations were not used in further testing. In the next phase of the experiments, the intent was to study elongation, which is the next phase of translation.  44  Figure 6. Percent translation efficiency at various concentrations of ribosome. The translation reactions were carried out on a scale of 50 µL total volume, with varying amounts of ribosome and template while the other components of the system were kept constant. After dT25-cellulose purification, obtained elutions were evaluated for the presence of mRNA fusions by scintillation counting. The counts were corrected by the background radioactivity and converted to the pmoles of total product made based on the specific activity of radiolabel. The percent ratio of pmoles product to pmoles template used in the reaction (%TE) was plotted against the pmoles of ribosome used in respective reactions.  45  3.2  Elongation study  3.2.1 Translation of 5GCC template The elongation study was performed to estimate the relative amount of full-length and incomplete fusions created. After initial evaluation, the goal of the study was to vary the concentration of major factors that might effect elongation (e.g., tRNAGGC or EF-Tu) in order to determine their most favorable concentrations (contributing to the highest relative amount of full-length product generated), and to compare the efficiency of incorporation to codons relative to the AUG start codon. The first translation reactions performed in the elongation study were synthesis of Met(Ala)5-mRNA fusions using the 5GCC template. The translation reactions were performed as described, on a scale of 50 µL total volume, under standard conditions with alanyl-tRNA acylation in situ. As the concentration of tRNAGGC (alanyl-tRNA) was suspected to greatly affect the elongation, a series of translation reactions was performed with varying concentrations of tRNAGGC. Translation reactions were performed with approximately 0.5 µM AlaRS, 0.4 mM alanine and increasing concentrations of tRNAGGC to be acylated in situ (0.001, 0.01, 0.1 and 0.23 µg/µL). The result of this analysis (Figure 7) showed in the RNase non-treated product bands (lanes 4, 6, 8 and 10) that there is a change in the size and the amount of fusion product generated depending on the concentration of tRNAGGC added to the translation reaction. Lanes 1 and 2 show the size control, 41P fusion (coding for initiator methionine alone), untreated and after treatment with RNase A, respectively. Lanes 3, 5, 7 and 9 of the same figure represent RNase A digested product. The change in size of the product as well as the band intensity show that the most of the elongated product synthesized in the system occurs at 0.1 µg/µL tRNAGGC. Further increase of tRNA (0.23 µg/µL) caused a slight translation inhibition (the band of 46  weaker intensity) while the lower concentrations than 0.1 µg/µL caused a predominance of smaller, incomplete fusions (migrate lower on the gel). The same size difference in migration pattern was expected in the RNase treated bands, being that the only difference between the treated and non-treated products was just the mRNA portion of the fusion, while the DNA-puromycin-peptide part stayed intact after the treatment. However, this pattern was not observed in the treated bands, as all treated product migrated the same or very similar distance as the size control (RNase A treated 41P fusion, lane 2 in Figure 7). In conclusion, even though 5GCC template translations yielded some elongated fusion product that migrated slightly higher on tricine-PAGE than 41P (methionine-only size control), it remained unclear what was the exact elongation of the product (i.e., how many alanine residues got incorporated after methionine before the puromycin formed the fusion). Therefore, to further study elongation 5GCC experiments were replaced by CAC1-5 template translations.  47  RNase A (µg/µL) tRNACCG Template (1 µM)  -  + 0 41P  1  2  + 0.001  3  4  -  5  + 0.01  6  -  +  -  0.1 5GCC  7  8  9  + 0.23  10  Figure 7. Translation of 5GCC template. Five translation reactions were performed, a translation with template 41P, which codes for only initiator methionine fusion and serves as the size control, and four translation reactions with 5GCC template, each with an increasing amount of alanyl-tRNA. As indicated in the methods section, all the samples were obtained using the same, aliquoted reaction mixture apart from the addition of alanyl-tRNA and templates, which were added separately and last. Post-translationally, reactions were dT25-cellulose-purified and ethanol precipitated; after purification, each sample was split in half, and one half of each sample was treated with 4 µg RNase A for 1 hour; after RNase treatment, each sample half was analyzed on a 15% tricinePAGE gel, dried and visualized by phosphor imaging.  48  3.2.2 Translation of CAC1 and CAC5 templates Since we were not able to unambiguously evaluate template 5GCC fusion elongation using available assays, 5GCC was replaced by CAC1 and CAC5 in further testing. Their coding regions, aside from the initiating AUG and GCC codons, contained one additional CAC codon on the second and the sixth position of the coding sequence, respectively. The introduced CAC codon permitted an increase in the coded chemical space, allowing for an affinity handle (biocytin) site-specific incorporation for studying elongation. For CAC1 and CAC5 experiments, all aminoacyl-tRNAs were acylated separately as described in the methods section, preventing proofreading activity of AlaRS from hydrolyzing biocytin from tRNAGUG. All -CA tRNA molecules, including tRNAGUG, were derived from alanyl-tRNA by site directed mutagenesis, therefore they had the same sequence and structure as alanyl-tRNA except for the anticodon sequence. As a result, if AlaRS were present in the same reaction, it would hydrolyze any non-alanine residues from the mutated alanyl-tRNAs, including tRNAGUG. 3.2.2.1  Specificity of streptavidin gel-shift assay  Four translation reactions were performed under standard conditions. Two were performed using the CAC1 template, both under standard conditions with the exception of tRNA contents – one CAC1 translation reaction had biocytin tRNA added but didn’t have any alanyl-tRNA while the other CAC1 translation reaction had alanyl-tRNA but didn’t have any biocytin tRNA (Figure 8, lanes 1-4). Both aminoacyl-tRNAs were separately acylated as described (biocytin tRNA using the method of chemoenzymatic acylation and alanyl-tRNA using enzymatic acylation), prior to addition to translation reactions. The other two translation reactions were performed using CAC5 and following the same pattern of tRNA contents per reaction (Figure 8, lanes 5-8). 49  The gel-shift occurs due to the specific binding between the biocytin residue and streptavidin present in the sample buffer. This specific binding causes that biocytin-containing fusion products migrate slower on a gel than non-biocytin fusions. Translation reactions were performed on a scale of 50 µL total volume as described, purified on dT25-cellulose, analyzed on a 12% urea-PAGE gel with and without streptavidin and observed by phosphor imaging. The result shows no gel-shift in the absence of biocytin regardless of the template used (Figure 8, lanes 4 and 8), as well as the absence of a gel-shift if no elongation took place after initiator methionine (due to alanine absence), even though biocytin was available for incorporation (Figure 8, lane 6). The assay confirmed specificity of the streptavidin gel-shift assay and the next step was to repeat the translations, but with both aminoacyl-tRNAs present in the same reaction.  50  RNase A Streptavidin Bio-tRNAGTG (µg/µL) Ala-tRNAGGC (µg/µL) Template (0.8 µM)  + + + + + + 0.06 0 0 0.15 CAC1  + + + + + + 0.06 0 0 0.15 CAC5  1  5  Elongated product containing biocytin (streptavidin bound) →  Elongated product without biocytin → Initiator methionine only product →  Figure 8.  2  3  4  6  7  8  Translation of CAC1 and CAC5 templates with a single aminoacyltRNA. Translation reactions were carried out under standard conditions with either CAC1 or CAC5 templates. Varied assay components are indicated. Post-translationally, reactions were dT25-cellulose-purified, ethanol precipitated and each treated with 4 µg RNase A for 1 hour; after RNase treatment each sample was split in half and each half run on a 12% urea-PAGE gel, in the presence of streptavidin (lanes 1, 3, 5, and 7) and without streptavidin (lanes 2, 4, 6, and 8); gel was dried and all samples were visualized by phosphor imaging.  51  3.2.2.2  Biocytin incorporation comparison between templates CAC1 and 5  Ten translation reactions were performed under standard conditions, five with CAC1 and five with CAC5. The main goal was to compare biocytin incorporation with the two templates, as well as to estimate the relative extent of elongation. The concentration of Bio-tRNAGTG was gradually increased in each consecutive sample, to visually estimate if this change caused a trend in the amount of full-length product generated with both templates. The reactions were performed on a scale of 50 µL total volume as described, purified on dT25-cellulose, analyzed on a 12% urea-PAGE gel with and without streptavidin and observed by phosphor imaging. The result showed that a fraction of fully elongated product was synthesized with both templates, but considerably more so with CAC1 (Figure 9, lanes 1-10) than with CAC5 (Figure 9, lanes 11-20) (as evidenced by comparing the top bands of the non-streptavidin gel samples). The reason for this weaker CAC5 signal was unclear, but one of the potential explanations was that when an amino acid is getting incorporated in the finishing steps of translation (closer to puromycin), the natural amino acid (in CAC1) will compete better against puromycin (incorporate with more success) than an unnatural amino acid such as biocytin (in CAC5). The second conclusion was that the increase in Bio-tRNAGTG was sufficient to show a slight increase in full-length product formed for both template reactions, indicating that further testing was needed to determine the most favorable concentration of aminoacyl-tRNAs. Also, it was notable that CAC1 yielded an overall stronger signal on the phosphor image than CAC5, and that trend was reoccurring in further tests when the two were assayed under the same conditions. The next step was to determine how a change in EF-Tu concentration affects the elongation of both templates.  52  RNase Streptavidin Bio-tRNAGTG (µg/µL) Ala-tRNAGGC (µg/µL) Template (0.8 µM)  + + - + 0.04 0  1  + + - + 0.06 0.15  2 3  4  + + + + + + 0.08 0.10 0.15 0.15 CAC1  5  6  7  8  + + - + 0.12 0.15  + + + 0.04 0  + + - + 0.06 0.15  + + - + 0.08 0.15 CAC5  + + - + 0.10 0.15  + + - + 0.12 0.15  9 10 11 12 13 14 15 16 17 18 19 20  Figure 9. Translation of CAC1 and CAC5 templates. Translation reactions were carried out under standard conditions with CAC1 and CAC5 templates. Varied assay components are indicated. Post-translationally, reactions were dT25cellulose-purified, ethanol precipitated and each treated with 4 µg RNase A for 1 hour; after RNase treatment, each sample was split in half and each half run on a 12% urea-PAGE gel, in the presence of and without streptavidin; gel was dried and all samples were visualized by phosphor imaging.  53  3.2.2.3  EF-Tu concentration change effect on CAC1 and 5 elongation  Ten translation reactions were performed under standard conditions, five using the CAC1 template, four with increasing concentrations of EF-Tu and standard concentrations of both elongation tRNAs acylated previously, and one in the absence of alanyl-tRNA and standard concentration of EF-Tu (Figure 10, lanes 1-10). The other five translation reactions were performed using CAC5 and followed the same pattern of tRNA and EF-Tu concentrations per sample reaction, respectively (Figure 10, lanes 11-20). The image indicated that among CAC1 samples, gel-shift band in lane 4 of Figure 10 (corresponding to the sample obtained using 6.6 µM EF-Tu) was the most intense one. This suggested that the most suitable EF-Tu concentration for our system is slightly more than 10 times in excess to other translation factors in the mixture: while the concentrations of other factors were at 0.5 µM, 6.6 µM EF-Tu seemed to give the strongest gel-shift signal with CAC1. Concurrently, the signal of CAC5 biocytin-containing product of the figure was too weak to make any conclusions regarding EF-Tu concentration based on CAC5 translations. In order to determine whether the factor that contributes to this effect is solely the different length of the template 3’untranslated region (3’UTR) that positions puromycin closer to the coding region at the site of unnatural amino acid incorporation, the next experiment was designed to compare CAC1 and 1CAC translation for biocytin incorporation under the same conditions. If in fact these compared translation reactions yielded similar signal profiles, then a more probable cause for the difference seen between CAC1 and CAC5 biocytin incorporations was the requirement for alanine incorporation to reach biocytin during elongation while competing against puromycin peptidyl transfer.  54  RNase Streptavidin Bio-tRNAGTG (µg/µL) Ala-tRNAGGC (µg/µL) EF-Tu (µM) Template (0.8 µM)  + + - + 0.06 0.15 5  1  2  + + - + 0.06 0.15 6.6  3  4  + + + 0.06 0.15 10 CAC1  5 6  + + - + 0.06 0.15 15  7  8  + + - + 0.06 0 5  + + - + 0.06 0.15 5  + + - + 0.06 0.15 6.6  + + - + 0.06 0.15 10 CAC5  + + - + 0.06 0.15 15  + + - + 0.06 0 5  9 10 11 12 13 14 15 16 17 18 19 20  Figure 10. Titration of EF-Tu in CAC1 and CAC5 template translation. Translation reactions were carried out under standard conditions: five translation reactions with template CAC1 and five with CAC5. Varied assay components are indicated. Posttranslationally, reactions were dT25-cellulose-purified, ethanol precipitated and each treated with 4 µg RNase A for 1 hour; after RNase treatment each sample was split in half and each half run on a 12% urea-PAGE gel, in the presence of streptavidin (lanes numbered with odd numbers) and without streptavidin (lanes numbered with even numbers); gel was dried and all samples were visualized by phosphor imaging.  55  3.2.3 Translation of CAC1 and 1CAC templates In order to check the effect of puromycin residue proximity to the coding region (in this case for both templates the only coding sequence was a CAC codon, situated immediately after the start codon), the following experiment was performed: two translation reactions were performed under standard conditions, one with template CAC1, the other with 1CAC, and the only tRNAGTG made available in both translation reactions was biocytinyl-tRNAs, acylated previously. This way, the part of the coding region of CAC1 that would normally be coded for by Ala-tRNAGGC was artificially turned into the part of 3’UTR. In this one and in all future experiments, the standard condition for EF-Tu concentration of was set at 6.6 µM, based on the preceding experimental findings (Section 3.2.2.3). However, even though it was expected to note a trend in the result, no difference was observed in the incorporation efficiency of biocytin into two mRNA fusions, despite the difference in length of 12 nucleotide bases in 3’UTR of two templates (Figure 11). After analyzing the gel image visually, the estimated total signal in lanes 1 and 2, as well as the percent of the top band to the total signal per lane was very similar. After evaluating the band signal volumes more quantitatively using ImageQuant software, the percent top band of the total signal per sample lane for CAC1 and 1CAC was 28 and 25%, respectively, meaning that close to 30% of the total fusion made contained biocytin, while the non-shifted bands, corresponding to methionine-only fusions, appeared to have the same intensity too. This also supported the fact that not only were the product profiles the same for the two templates of different 3’UTR, but also the total fusion signal was close to be the same as well. This result showed that there was no difference between biocytin incorporation in two templates that have 3’UTR lengths different by four codons (12 bases). It also brought back into consideration the hypothesis that the reason for different biocytin incorporation between 56  CAC1 and CAC5 is the difference in the ability of the amino acid to compete against puromycin once the puromycin residue is relatively close to the respective codon in the final stages of elongation. Whether more fusion product created with CAC1 than with CAC5 under same conditions was solely a result of this or some other factor, remains to be seen in the future research. Finally, like in the next experiment, which involved comparison of CAC1 through 5 fusion products, the 30% approximate percent value was only representative of the batch of ribosome and tRNA molecules used for the experiment. Using different stock solutions of these materials can affect the absolute value of signal intensities or the percent of the top band to the total signal per lane, but the relative intensities of fusion signal obtained with different templates under same conditions are maintained the same (data not shown).  57  RNase A + + Bio-tRNAGTG (µg/µL) 0.06 Ala-tRNAGGC (µg/µL) 0 Template (0.8 µM) CAC1 1CAC  A Percent top band to → total signal per lane__  28  25  1  2  B  Figure 11. 3’UTR-independent biocytin incorporation. The templates CAC1 and 1CAC were translated under the same conditions; the concentration of tRNA coding for biocytin was kept constant, while tRNA coding for GCC was not made available in any of the reactions, artificially making a part of CAC1 coding region into a part of the 3’UTR. 50-µL reactions were post-translationally dT25-cellulose-purified and treated with 4 µg RNase A for 1 hour, then split in half and each half run on a 12% urea-PAGE gel, with streptavidin (Panel A) and without streptavidin (Panel B); gels were dried and all samples were visualized by phosphor imaging.  58  3.2.4 Translation of templates CAC1 through 5 In order to further explore the change in biocytin incorporation dependence on the codon position in a hexapeptide template, translations of templates CAC1 through 5 were performed. Five translation experiments were performed under standard conditions, each with one of the CAC1 through 5 templates and standard concentrations of both tRNAs (alanyl and biocytinyl) acylated prior to the reaction. The observed gel image signal intensities once again confirmed that when the CAC codon was the first after AUG, the highest relative amount of biocytin-containing product was being generated (Figure 12). However, after evaluating the band signal volumes using ImageQuant software, the percent of the top band to the total signal per lane for all five templates was found again gravitating toward 30%, (including CAC4, whose suspected poor quality caused out-of-pattern band profile). Therefore, more biocytin-containing product was not a result of uneven translation efficiency of CAC codon differently placed in the coding region, it was more likely a result of overall more fusion product created with CAC1 as opposed to all other template reactions. In order to make more definite inferences about this clearly observed, repeated trend, more experiments with different amino acids and/or tRNA molecules are required, but at this stage of system testing it was concluded that the system is able to generate a full-length hexapeptide containing a single unnatural amino acid, biocytin. In the next phase of testing it was attempted to utilize the system in generating a completely orthogonal fusion product using an N-methyl amino acid, sarcosine along with biocytin.  59  RNase A + + + + + Bio-tRNAGTG (µg/µL) 0.06 Ala-tRNAGGC (µg/µL) 0.15 Template (0.8 µM) CAC1 CAC2 CAC3 CAC4 CAC5  A Percent top band to → total signal per lane__  30  22  25  30  28  1  2  3  4  5  B  Figure 12. Codon position-dependent biocytin incorporation. All templates are translated using reconstituted system with aminoacyl-tRNAs (acylated prior to translation). Like in the previous one, varied assay components for this experiment were just the mRNA templates. Post-translationally, 50-µL reactions were dT25-cellulose-purified and treated with 4 µg RNase A for 1 hour, then split in half and each half run on a 12% ureaPAGE gel, with streptavidin (Panel A) and without streptavidin (Panel B); gels were dried and all samples were visualized by phosphor imaging. Note: The intensity of all CAC4 product bands is the low due to the poor quality of the template, as per analysis by TriLink (template manufacturer).  60  3.3  Sarcosine incorporation study  3.3.1 Sarcosine incorporation under standard conditions The next and final phase of the project was to test the ability of the system to incorporate Nmethyl amino acids, since they will be the first building blocks to be used in future library syntheses. As the simplest representative of N-methyl amino acids, sarcosine was selected and its CAC templates translation result was compared to the corresponding translation result of alanine. Similarly to previous CAC1 through 5 template reactions with alanine, a series of five reactions was performed using Sar-tRNAGGC instead of Ala-tRNAGGC. As shown in Figure 13, not only did the result show that the gel-shift with CAC2 through5 was not detectable in these reactions, but also initiator methionine-only band intensity was decreased as well. This was also reflected in very low scintillation counts of these reactions samples. In conclusion, in this and other initial tests (data not shown) with sarcosine using the reconstituted system under standard conditions resulted in virtually no sarcosine incorporation and there was a weaker fusion signal overall when compared to alanine CAC1 through 5 translation result. In the following experiments, attempts were made to improve sarcosine incorporation by varying standard conditions.  61  RNase A + + + + + Bio-tRNAGTG (µg/µL) 0.06 Sar-tRNAGGC (µg/µL) 0.15 Template (0.8 µM) CAC1 CAC2 CAC3 CAC4 CAC5  A  B 1  2  3  4  5  Figure 13. Undetectable incorporation of sarcosine under standard conditions. All templates are translated using reconstituted system with annotated tRNAs acylated previously. The samples were all obtained using the same, aliquoted reaction mixture, apart from the addition of templates. Post-translationally, 50-µL reactions were dT25-cellulose purified, split in half and each half run on a 12% urea-PAGE gel, with streptavidin (Panel A) and without streptavidin (Panel B); all samples were visualized by phosphor imaging.  62  3.3.2 Effect of magnesium ion concentration increase on alanine and sarcosine incorporation After the unsuccessful attempts to incorporate sarcosine into CAC1-5, we proceeded to check the effect of magnesium ion concentration on sarcosine incorporation, as the Mg+2 ion is known to play a major role in achieving stability of the ternary ribosome/aminoacyl tRNA/EF-Tu complex [102]. The test titration of magnesium was first performed with alanine and the CAC5 template to test the effect of magnesium concentration on the overall functioning of the system and formation of the full-length product, which we have already established was consistently being made using the biocytin incorporation assay. Seven CAC5 translation reactions were performed under standard conditions with increasing concentrations of Mg+2 ion for consecutive reactions (0, 1, 5, 10, 15, 20 and 50 mM, respectively). As shown in Figure 14, the increase of overall signal intensity in the resulting gel image indicated that there was a correlation between the increase of Mg+2 ion and the amount of generated biocytin-containing product, as well as the total fusion product made per lane with alanine.  63  RNase A Mg(OAc)2 (mM) Bio-tRNAGTG (µg/µL) Ala-tRNAGGC (µg/µL) Template (0.8 µM)  + 0  + 1  + 5  1  2  3  + + 10 15 0.06 0.15 CAC5  + 20  + 50  6  7  A  B  4  5  Figure 14. Titration of Mg(OAc)2 in CAC5 template translation. CAC5 template was translated using reconstituted system with added tRNAs acylated prior to translation and an increasing amount of Mg(OAc)2. The samples were obtained using the same, aliquoted reaction mixture, apart from the addition of template and Mg(OAc)2. Posttranslationally, 50-µL reactions were dT25-cellulose-purified and treated with 4 µg RNase A for 1 hour, then split in half and each half run on a 12% urea-PAGE gel, with (Panel A) and without streptavidin (Panel B); gels were dried and all samples were visualized by phosphor imaging.  64  The next step was to test whether this positive effect on fusion generation and elongation would be observed with an unnatural amino acid sarcosine as well. In addition, to ensure that the aminoacyl-tRNAGGC and aminoacyl-tRNAGTG concentrations were appropriate for the next experiments, another ten CAC5 translations were performed, five with alanine (Figure 15, lanes 1-5) and five with sarcosine (Figure 15, lanes 6-10), while titrating in aminoacyltRNA and increasing the magnesium concentration up to 20 mM, (four times higher than standard conditions 5 mM) in order to try to amplify the sarcosine signal. The titrations were performed under standard conditions except for the magnesium and aminoacyl-tRNA concentrations. The result showed for the first time fully elongated, completely orthogonal fusion product with sarcosine, clearly visible on a PAGE gel as a gel-shifting band (Figure 15, lanes 6-10). Using software ImageQuant for the band volume evaluation, the percent of the top (gelshifted) band to the total signal per lane was estimated and scintillation counts of the total fusion product per sample were determined. As the percent of the top band to the total signal per lane was increased with an increase in acylated tRNA but total fusion created was decreasing at the same time, in order to estimate the most suitable concentration of tRNA the two parameters were used to calculate the number of full length fusion molecules created (Table 2).  65  RNase A Mg(OAc)2 (mM) Bio-tRNAGTG (µg/µL) Ala-tRNAGGC (µg/µL) Sar-tRNAGGC (µg/µL) Template (0.8 µM)  +  +  +  +  +  + 20 0.03 0.04 0.05 0.06 0.07 0.03 0.08 0.1 0.13 0.15 0.18 0 0.12 CAC5  +  +  +  +  0.04  0.06  0.07  0.15  0.05 0 0.19  0.23  0.27  4  6  8  A Percent top band to → total signal per lane__  7  10  14  17  16  5  5  2  3  4  5  6  7  B  1  8  9  10  Titration of Ala-tRNAGGC and Sar-tRNAGGC in CAC5 translation, at a high Mg(OAc)2 concentration. The samples were obtained using the same, aliquoted reaction mixture apart from the addition of template and previously acylated tRNAs that were added in annotated amounts. Post-translationally, 50-µL reactions were dT25-cellulose-purified and treated with 4 µg RNase A for 1 hour, then split in half and each half run on a 12% urea-PAGE gel, with (Panel A) and without streptavidin (Panel B); gels were dried and all samples were visualized by phosphor imaging. Scintillation counts and fusions yield calculations shown in Table 2. Figure 15.  66  Yield estimate calculations of Ala-tRNAGGC and Sar-tRNAGGC titration experiment with CAC5 template at a high Mg(OAc)2 concentration. Upon dT25-cellulose purification the sample aliquots were used to determine scintillation counts (CPM) of sample solutions and used to calculate total yield of fusions per sample. Also, the band evaluation results were used to calculate the percent of full-length fusions out of the total band signal per lane and subsequently to calculate the number of full-length product molecules in the total yield based on scintillation counts. The estimated optimal concentration of aminoacyl-tRNAs has been determined as the one that yielded the highest number of full-length fusions relative to the amount of total fusions (i.e., sample 4 for alanyland sample 10 for sarcosyl-tRNA). As stated in Section 2.2.8.2, the specific activity of [35S]Met declared by the manufacturer was assumed unchanged at the time of experiment. Table 2.  Sample number  Aminoacyl-tRNA concentrations (µg/µL)  Total scintillation counts (CPM)  Total Percent fullNumber of Full-length Number of fusion length from the full-length product total product product total product product (pmol) molecules (pmol) (%) molecules  Ala tRNA/Bio tRNA 1  0.08/0.03  175316  0.0790  7.08  0.0056  7.897E+09  5.588E+08  2  0.10/0.04  130708  0.0589  9.64  0.0057  5.888E+09  5.674E+08  3  0.13/0.05  95448  0.0430  13.77  0.0059  4.299E+09  5.921E+08  4  0.15/0.06  89052  0.0401  17.17  0.0069  4.011E+09  6.888E+08  5  0.18/0.07  74661  0.0336  16.38  0.0055  3.363E+09  5.508E+08  Sar tRNA/Bio tRNA 6  0.12/0.03  211929  0.0955  5.52  0.0053  9.546E+09  5.269E+08  7  0.15/0.04  207214  0.0933  4.49  0.0042  9.334E+09  4.188E+08  8  0.19/0.05  201966  0.0910  4.20  0.0038  9.098E+09  3.819E+08  9  0.23/0.06  198645  0.0895  6.11  0.0055  8.948E+09  5.466E+08  10  0.27/0.07  170683  0.0769  7.64  0.0059  7.688E+09  5.872E+08  67  Based on the data shown in Table 2, the samples that gave the highest number of fully elongated fusion molecules with CAC5 were samples in lane 4 (for Ala- and Bio-tRNA, 0.15 and 0.06 µg/µL, respectively) and lane 10 (for Sar- and Bio-tRNA, 0.27 and 0.07 µg/µL, respectively). The tRNA concentrations used for those two samples were chosen as estimated optimal tRNA concentrations under standard conditions. The concentrations provided for much more adaptor (tRNA) molecules of unnatural aminoacyl-tRNAs than necessary to cover the sequence space of the standard conditions template concentration (1 µg of aminoacylated tRNA molecules approximates to 40 pmoles, and 40 pmoles of tRNA adaptors is required for a single codon incorporation when 40 pmoles of template is used per reaction), but this excess of tRNA accounts for imperfections in the in vitro reconstituted system (e.g., not all tRNA molecules were aminoacylated successfully, etc.) and for lower affinity of unnatural amino acids for incorporation than of natural amino acids. It also accounted for lower estimated relative efficiency of chemo-enzymatic acylation of sarcosyltRNA (~50%) when compared to enzymatic acylation of alanyl-tRNA (~70%). Introducing purification or standardization of acylated tRNAs was not considered at the time due to their fragile, hydrolysable nature. Finally, the percentages of full-length product created for alanine changed upon this increase of magnesium from the percentages achieved previously under standard conditions (in previous experiments, under standard conditions the percent gel-shift was found to be around 30% and in Figure 15, at 20 mM magnesium it decreased to only 17%). This inconsistency was most likely due to the fact that both purified ribosomes and all aminoacyl-tRNA molecules are made anew for this experiment. As already mentioned, the exact concentration and composition of starting material stock solutions largely affects translation and both  68  ribosomes and aminoacyl-tRNAs are products of series of variable, biological in vitro and in vivo processes. Therefore, the absolute values of the total signal per gel lane or the percent top band to total signal per lane or the total fusion scintillation counts per sample are only evaluated relative to other values obtained in the same experiment. Since in this experiment we were able to observe for the first time a clear gel-shift band of fully elongated product with sarcosine and biocytin using CAC5, subsequent experiments were performed in an attempt to quantify the effect of magnesium concentration on incorporation of either amino acids alanine or sarcosine with biocytin. The next experiment consisted of a series of seven point titrations of Mg+2 ion with each amino acid (alanine or sarcosine) to establish for each amino acid whether the Mg+2 ion increased the percent of the top band to the total signal per lane, or also whether the apparent increase of the top band was really a result of the increase in the overall signal that results in a concomitant increase in the top band. The final four titration experiments, one testing alanine incorporation, the other testing sarcosine, were performed over varying Mg+2 ion concentrations at aminoacyl-tRNA concentrations closest to the ones that previously gave the most product at 20 mM magnesium, as annotated in the legend of Figure 16. The translation reactions were performed on four different days, on a scale of 25 µL total volume. The order of alanine and sarcosine translation reactions was alternated on each test day (e.g., on the first day – alanine translations were performed first, the sarcosine translations second). To keep the random error down to a minimum (occurring due to pipetting or uneven deprotection of NVOC conjugates of aminoacyl tRNA), all common reaction ingredients for all titrations performed on the same day were aliquoted into the same  69  reaction master mix and further aliquoted into sample tubes from there. Bio-tRNAGTG was added to the master mix after the whole amount (needed for both series of translations) was deprotected at once. After aliquoting the common reagent master mixture into tubes, the varying ingredients were added to each sample solution (e.g., alaninyl- and sarcosyl-tRNA), respectively. Once the master mixture was aliquoted and tRNA solutions added, Mg(OAc)2 was titrated into each reaction mix, and in the end 20 pmol of CAC5 template was added. The reactions were incubated for an hour and the same volume of “stop salt” solution, as outlined in the methods section, was added to each to facilitate the fusion formation. Each reaction fusion product was purified on dT25-cellulose, analyzed on a 12% urea-PAGE gel in the presence of streptavidin and observed by phosphor imaging (Figure 16 shows two of eight gel images obtained). Using software ImageQuant for the band volume evaluation, all band volumes were evaluated and the percents of the top (gel-shift) band to the total signals per lane were determined. The obtained data were quantified and compared to determine the average intensities and standard deviations for both parameters over all four replicates for each different magnesium condition and for each amino acid. The data values were plotted using GraphPad Prism to produce column graphs and observe the patterns of each parameter change. Based on the all four graphs (Figures 17-20), it seemed that there was an increase in total product created with alanine and sarcosine. However, the percent top band of the total signal per sample lane are different for each amino acid series: for alanine, it did not make a consistent trend of increase in gel-shifted signal across the concentrations of magnesium that  70  were tested, while for sarcosine the increase in magnesium ion concentration appeared to correlate with a substantial increase in the percent top band of the total signal per lane. In order to determine the statistical significance of these observations, the data were evaluated using GraphPad Prism one-way ANOVA for repeated measurements. 3.3.2.1  One-way ANOVA analysis of magnesium titrations  The data obtained by the band volume analysis was processed using GraphPad Prism oneway ANOVA analysis for repeated measurements. To better evaluate the effect of magnesium ion concentration on incorporation of each amino acid, two described parameters were processed by ANOVA: total signal (band volume) per sample lane (all fusions) (Tables 3 and 5) and the percent of the top band to the total signal per lane (only the fully elongated fusions, as the template used was CAC5) (Tables 4 and 6). The goals of each ANOVA analysis (for each parameter and amino acid) were: (1) to determine if the differences between the group of result series means are greater than one would expect to see by chance assuming Gaussian distribution (comparing variability between the groups of results obtained at the same magnesium concentrations), and (2) to determine the confidence interval within which the matching (pairing) of means was effective (comparing variability between the groups of results obtained in repeated experiments). The first data point, acquired at 0 mM Mg(OAc)2 was not taken in consideration for ANOVA analysis, as the absence of any magnesium ion in the translation reaction will likely be causing more than one factor change; magnesium ion is not only critical for normal structure and function of translation reaction components, but their interactions as well, potentially causing more than one factor change under extreme conditions (such as extremely low concentration). Since one of the premises of ANOVA is that the investigated change in data 71  values is caused by a single factor change, the ANOVA results obtained by excluding the first data point bear more merit in evaluating this experiment than if ANOVA was performed across the whole tested range of magnesium concentrations. The result of one-way ANOVA analysis of the total signal per lane (Table 3 for alanine and Table 4 for sarcosine) showed that the increase in total signal per lane (sample) for both amino acids is significant under compared test conditions (treatments). However, the increase in percent top to total band volume/signal for alanine was not found significant (Table 5), due to unsuccessful pairing of the means in repeated experiments. Finally, the increase in percent top to total band volume/signal for sarcosine (Table 6) was found significant under the given test conditions.  72  RNase A Mg(OAc)2 (mM) Bio-tRNAGTG (µg/µL) Ala-tRNAGGC (µg/µL) Template (0.8 µM)  + 0  + 1  + 5  2  3  + 1  + 5  2  3  + + 10 20 0.06 0.15 CAC5  + 30  + 50  5  6  7  + + 10 20 0.06 0.27 CAC5  + 30  + 50  6  7  A  1 RNase A Mg(OAc)2 (mM) Bio-tRNAGTG (µg/µL) Sar-tRNAGGC (µg/µL) Template (0.8 µM)  + 0  4  B  1  4  5  Figure 16. Mg(OAc)2 effects on CAC5 template translation. Testing the effects of Mg(OAc)2 increase on alanine (Panel A) and sarcosine (Panel B) incorporation into respective CAC5 fusions, samples were obtained using the same, “master” mixture for both amino acids, apart from the addition of increasing volumes of Mg(OAc)2 stock solution, acylated alanyl-tRNAs and template. Post-translationally, 25 µL reactions were dT25-cellulose-purified and treated with 4 µg RNase A for 1 hour, then analyzed on a 12% urea-PAGE gel, in the presence of streptavidin; gels were dried and all samples were visualized by phosphor imaging and the band volume for each evaluated using ImageQuant (one of four experiments gel image shown, an outline made on each image to show that the biocytin-containing product was obtained under standard magnesium ion concentration with alanine while at the same conditions no biocytin-containing product was detectable with sarcosine). 73  A  B  Figure 17.  Magnesium effect on total fusion material production based on scintillation counts. The average amount of pmoles total fusion and respective standard deviations calculated based on scintillation counts of dT25-cellulose elution for four alanine CAC5 translations (Panel A) and for four sarcosine CAC5 translations (Panel B). These graphs show that increasing magnesium concentration results in more total peptide-nucleic acid material being made, for both amino acid sample series (alanine and sarcosine). The shown error bars are calculated standard deviations.  74  Figure 18. Magnesium effect on total band volume signal using alanyl-tRNA. The four series of translation reactions were performed under standard conditions with the exception of Mg(OAc)2 concentration, which was ranged starting from 0 mM, over 1, 5, 10, 20 and 30 mM to 50 mM, as described under Figure 16. Using software ImageQuant for the band volume evaluation, the total signal per lane was determined. The column diagram of average values and standard deviations for the total signal is shown here and respective oneway ANOVA analysis in Table 3. The shown error bars are calculated standard deviations.  75  Table 3.  One-way ANOVA analysis of total band volume per lane data for alanine translations (0 mM Mg(OAc)2 data point omitted).  Repeated Measures ANOVA for Alanine, Total Band Volume, P value 0.0001 P value summary Very highly significant Are means signif. different? (P < 0.05) Yes Number of groups 6 F 10.92 R squared 0.7845 Was The Pairing Significantly Effective? R squared 0.3281 F 11.33 P value 0.0004 P value summary Very highly significant Is there significant matching? (P < 0.05) Yes ANOVA Table Treatment (between columns) Individual (between rows) Residual (random) Total  SS 9.116E+13 5.675E+13 2.504E+13 1.73E+14  76  df 5 3 15 23  MS 1.823E+13 1.892E+13 1.669E+12  Figure 19. Magnesium effect on percent top band to total signal using alanyl-tRNA. The four series of translation reactions were performed under standard conditions with the exception of Mg(OAc)2 concentration, which was ranged starting from 0 mM, over 1, 5, 10, 20 and 30 mM to 50 mM, as described under Figure 15. Using software ImageQuant for the band volume evaluation, the total signal per lane was determined. The column diagram of average values and standard deviations for the total signal is shown here and respective oneway ANOVA analysis in Table 4. The shown error bars are calculated standard deviations.  77  Table 4.  One-way ANOVA analysis of percent top band to total signal per lane data for alanine translations (0 mM Mg(OAc)2 data point omitted).  Repeated Measures ANOVA for Alanine, % Top Band to Total Signal P value 0.0438 P value summary Significant Are means signif. different? (P < 0.05) Yes Number of groups 6 F 3.027 R squared 0.5022 Was The Pairing Significantly Effective? R squared 0.2219 F 2.864 P value 0.0717 P value summary Not significant Is there significant matching? (P < 0.05) No ANOVA Table Treatment (between columns) Individual (between rows) Residual (random) Total  SS 10.06 5.713 9.972 25.75  78  df 5 3 15 23  MS 2.012 1.904 0.6648  Figure 20. Magnesium effect on total band volume signal using sarcosyl-tRNA. The four series of translation reactions were performed under standard conditions with the exception of Mg(OAc)2 concentration, which was ranged starting from 0 mM, over 1, 5, 10, 20 and 30 mM to 50 mM, as described under Figure 15. Using software ImageQuant for the band volume evaluation, the total signal per lane was determined. The column diagram of average values and standard deviations for the total signal is shown here and respective oneway ANOVA analysis in Table 5. The shown error bars are calculated standard deviations.  79  Table 5.  One-way ANOVA analysis of percent total band volume per lane data for sarcosine translations (0 mM Mg(OAc)2 data point omitted).  Repeated Measures ANOVA for Sarcosine, Total Band Volume, P value 0.0125 P value summary Significant Are means signif. different? (P < 0.05) Yes Number of groups 6 F 4.311 R squared 0.5897 Was The Pairing Significantly Effective? R squared 0.2246 F 3.529 P value 0.0409 P value summary Significant Is there significant matching? (P < 0.05) Yes ANOVA Table Treatment (between columns) Individual (between rows) Residual (random) Total  SS 4.343E+13 2.134E+13 3.023E+13 9.5E+13  80  df 5 3 15 23  MS 8.687E+12 7.112E+12 2.015E+12  Figure 21.  Magnesium effect on percent top band to total signal using sarcosyltRNA. The four series of translation reactions were performed under standard conditions with the exception of Mg(OAc)2 concentration, which was ranged starting from 0 mM, over 1, 5, 10, 20 and 30 mM to 50 mM, as described under Figure 15. Using software ImageQuant for the band volume evaluation, the total signal per lane was determined. The column diagram of average values and standard deviations for the total signal is shown here and respective oneway ANOVA analysis in Table 6. The shown error bars are calculated standard deviations.  81  Table 6.  One-way ANOVA analysis of percent top band to total signal per lane data for sarcosine translations (0 mM Mg(OAc)2 data point omitted).  Repeated Measures ANOVA for Sarcosine, % Top Band to Total Signal P value 0.0001 P value summary Very highly significant Are means signif. different? (P < 0.05) Yes Number of groups 6 F 11.39 R squared 0.7916 Was The Pairing Significantly Effective? R squared 0.1349 F 3.742 P value 0.0345 P value summary Significant Is there significant matching? (P < 0.05) Yes ANOVA Table Treatment (between columns) Individual (between rows) Residual (random) Total  SS 733.7 144.6 193.2 1071  82  df 5 3 15 23  MS 146.7 48.2 12.88  4  DISSCUSIONS AND CONCLUSION  4.1  Initiation study  After making the system components and establishing that the system was able to make a product in the mRNA fusion format, the first step in attempt to optimize the system performance was to ensure the maximal total yield of translation for the respective batch of reconstituted system components was achieved by adjusting the relative amounts of ribosome and template. By obtaining maximal total yield of translation under given conditions, we would approximate that the PURE system was suited for maximum translation initiation, even though each different batch of reconstituted system components would result in different absolute values of fusion signal obtained. However, the absolute value of the fusion signal would show the same trend of increase and decrease once the system was subjected to the same change in reaction conditions or composition (e.g., using more template would cause stronger overall fusion signal but lower translation efficiency while adding more than a certain concentration of tRNA would cause a decrease of the elongated fusion signal, etc.), therefore this initial testing provided with a good estimate of conditions favoring successful initiation. For these experiments 1GCC template was used as the simplest template system. The translation protein factors were kept at levels found in the literature [103] as they are capable of recycling in the system, while both ribosome and template are both required one per each formed fusion molecule, so their ratios were varied and total yield of translation observed. The experiments were performed using a single batch of ribosomes and the result (Figure 6) showed that the most suitable concentration of template under given reaction conditions was around 50 pmol, where it was possible to reach the peak in translation efficiency by  83  maximally utilizing present ribosomes (also approximately 50 pmol). Additionally, the results of experiments indicated that an increase in template concentration caused an increase in radioactivity of the dT25-cellulose purified fraction (an apparent increase of pmoles of product), but at the same time a decrease of %TE. The results obtained confirmed what was already indicated in relevant literature [67] that, when a PURE system is utilized to synthesize mRNA display products, ribosomes present in the system are not recycled and each ribosome is responsible for generating a single fusion molecule. Therefore, template concentrations less than 50 pmol may not utilize all available ribosomes, but may yield higher %TE, as the probability of all molecules of template getting paired successfully with the most viable ribosome concentration increases. Accordingly, higher template concentrations cause a decrease in %TE, as well as a potential for increase of initiator methionine-only product, since it maximizes the ribosome template pairing, while the other necessary factors, especially the ones responsible for elongation, may become a limiting factor. Additionally, experiments with other, more concentrated batches of purified ribosomes showed that preparations of higher ribosome concentration yield more translation product when more template is used (data not shown) [67]. The main conclusion after this part of the study was that the concentration of template introduced to the reaction should approximately follow the concentration of ribosome added to the reaction to result in the highest concentration of total fusions formed relative to the concentration of used ribosome and template, assuming that other limiting factors are present in sufficient excess (e.g., [35S]fMet-tRNAi). In order to maximize the yield however, using more concentrated purified ribosome stock solutions goes in favor of obtaining a higher  84  yield, as the chance that template molecules will encounter better, more viable ribosomes in the translation reaction mix is higher. 4.2  Elongation study  4.2.1 Translation of template 5GCC The result of 5GCC translation (Figure 7) suggested that tRNA concentration is one of the factors that influence the course and outcome of elongation of mRNA fusions. However, even though 5GCC template translations yielded some elongated fusion product that migrated higher on tricine-PAGE than 41P (methionine-only size control), it remained unclear what was the exact elongation of the product (i.e., how many alanine residues got incorporated after methionine before the puromycin formed the fusion), due to assay limitations. Therefore to further study elongation, 5GCC experiments were replaced by CAC1-5 template translations. 4.2.2 Translation of templates CAC1 and CAC5 CAC1 and CAC5 templates were first tested out of CAC series of templates. CAC1 and CAC5 templates were designed to verify the specificity of the new assay, streptavidin gelshift assay (Figure 8), then to check the efficiency of incorporation of an unnatural amino acid (biocytin) depending on its respective codon place in the coding sequence (Figure 9), and to be used for further system evaluation (e.g., to check the effect of ET-Tu concentration on elongation, Figure 10). These two templates also served for comparison of incorporation efficiency of biocytin residue in all templates of CAC series with alanine (Figure 11) and with sarcosine (Figures 13), respectively. Aside from initial AUG and four GCC codons, CAC1 and CAC5 contained a single CAC codon, which was used to site specifically and in situ incorporate an affinity handle biocytin into the fusion product. This was made possible by using biocytin chemo-enzymatically 85  charged on tRNAGUG, and it allowed for unambiguous evaluation of fusions by use of a streptavidin gel-shift assay [87] (Figure 8). The switch to using separately acylated tRNAs for translation reactions was an important step in preparing the system for future unnatural amino acid incorporations, since using separately acylated tRNA with unnatural amino acids by a chemo-enzymatic approach was the intended way to charge the tRNA molecules in the reconstituted system. In addition, AARSs reportedly have limited affinity for unnatural, especially N-methyl amino acids [89] and cannot be used to charge these orthogonal elements onto tRNAs. Initial testing of the system with natural amino acids and AARS (alanine and AlaRS) was intended to serve as a system suitability test in later experiments (for positive controls) and it had its purpose in evaluating the initiation phase of translation. The typical profile of CAC1 and CAC5 products on a 12% urea gel appeared as two bands: a bottom band with the higher signal intensity that corresponded to initiator methionine-only mRNA fusion, and a top band, which exhibited a gel-shift if the gel was run in the presence of streptavidin (Figure 9). The top band corresponded to completely or partially elongated product that contained a biocytin residue. Another conclusion of this experiment was that the closer the CAC codon was to the start codon, the template resulted in more material gel-shifted as compared to when the CAC codon was in the ultimate coding position (at the very end of the coding region). This suggested that elongation was largely affected by the position of the unnatural amino acid in the coding sequence but it was unclear whether the factor was the position or the kind of amino acid in the coding sequence or solely the proximity of the puromycin residue to the coding region in the moment of respective amino acid incorporation. Finally, there is a  86  possibility that tRNAGGC adaptor is stalling on the ribosome under given experimental conditions more than tRNAGUG causing the difference in CAC1-5 fusion profiles, and that possibility certainly deserves further elaborate investigation. As shown in Figures 8 and 9, not only was the CAC1 total band signal intensity and the amount of elongated product (gel-shift band) consistently much stronger than that observed for CAC5 product, but the scintillation counts of total fusions obtained with CAC1 were consistently higher than the ones obtained with CAC5 too. Since the main difference between CAC1 and CAC5 is the position of CAC codon in the coding sequence, and biocytin incorporation happens much more readily when the CAC codon is positioned closer to initiator methionine (i.e., template 5’ end) and farther away from the puromycin residue (positioned at the template 3’ end), it led to the next consideration: the proximity of the puromycin residue to the coding sequence itself may be the factor that affects translation of mRNA fusions, against the aminoacyl-tRNA for the A site within the ribosome. Therefore a comparison translation was performed with templates CAC1 and 1CAC, for which we hypothesized that solely by changing puromycin proximity to the CAC codon we can affect/enhance the yield of biocytin incorporation into the fusion product. 4.2.3 Translation of templates CAC1 and 1CAC In order to evaluate the impact on elongation that puromycin proximity to the ribosome plays in the moment of biocytin incorporation, a comparison translation was performed with templates CAC1 and 1CAC (Figure 11). In both templates, biocytin was incorporated as the first residue after methionine, no other aminoacyl-tRNAs were made available except BiotRNA, and the effect of different puromycin distance from the ribosome was observed. It was expected that the shorter distance between the biocytin codon and puromycin in the 1CAC template would make the puromycin residue closer to the ribosome peptidyl transfer site and 87  therefore more available to terminate translation prematurely than when that region is longer, as it is in CAC1. However, the expected effect of a shorter distance between the biocytin codon and puromycin was not confirmed experimentally. In addition, the percent of the top band to the total signal was found to be similar (around 30%) for both templates. The conclusion was that the length of 12 bases in 3’UTR for CAC1 doesn’t make a notable effect on the efficiency of translation, and that there is at least that much flexibility lengthwise in designing future library templates. Also, within one batch of ribosomes and under standard conditions, the relative percent of the top band (fully elongated product) to the total signal per sample between different templates (CAC1 through 5 and 1CAC) remained the same. It is the total band volume per lane that was found different for each of the templates (higher for CAC1 and 1CAC, and lower for all other tested CAC templates). 4.2.4 Translation of templates CAC1 through 5 In the next phase of testing we analyzed translation products of CAC2 though 4 side by side with CAC1 and CAC5. Based on what has been observed with CAC1 and CAC5, we anticipated a decrease in biocytin-containing fusion signal with templates CAC2-4, but it was difficult to predict whether the relative full-length product amount will drop precipitously or gradually with the increase in the number of necessary alanine incorporations prior to biocytin incorporation. The result in Figure 12 showed that the amount of gel-shift product precipitously drops between CAC1 and CAC2, and remains similarly low from CAC2 through CAC5 when compared to CAC1. However, as mentioned in the previous section, the band volume analysis showed that the percent of biocytin-containing fusions generated is fairly similar between templates, again 88  approaching 30%, even though the relative signal intensity of the top band is much lower for CAC2-5 than it is for CAC1. Therefore, what was actually lower for CAC2-5 was the total fusion signal per lane, also reflected in lower CPM of total eluted product, not the percent of the gel-shifting signal. As mentioned before, this obtained percent of biocytin-containing fusions absolute value was much higher than the one obtained during some later experiments (the titration of magnesium experiment involving CAC5 and alanine, Figure 19 with approximately 3%, or the titration of acylated tRNA experiment, Figure 15 with 17% of biocytin-containing fusions) and was a reflection of the current batch of ribosome and acylated tRNA quality. Ultimately, the absolute value of the yield represents a factor only as far as knowing how many times does translation reaction need to be scaled up with the respective batch of system components in the execution of the template library synthesis, as that is the intended application of the reconstituted system. Finally, it was noted that the amount of partially elongated product appeared to be higher with CAC4 and CAC5 than with other templates. This was reflected not only in a weaker gel-shift band but also the appearance of incomplete fusion products with CAC4 and CAC5 that do not contain biocytin, migrating slightly higher than initiator methionine-only product but lower than the top band in both presence and absence of streptavidin. CAC1 could theoretically be also producing incomplete fusion products to run between the fully elongated and initiator methionine-only product in the absence of streptavidin, but when incomplete fusions get created with CAC1, they will all move with streptavidin as they will all contain biocytin. It was the non-streptavidin gel (Figure 12, Panel B), which also showed higher intensity of the top band that demonstrated more elongated product of any length can be formed with CAC1 than with any other CAC template. Whether more fusions were made  89  with bulky biocytin coded for early in the coding sequence is a result of its unique chemical nature or other factors remains to be further investigated. 4.3  Sarcosine incorporation study  To evaluate the incorporation of unnatural amino acids (other than biocytin), we used an unnatural amino acid, sarcosine, and tried to incorporate it in CAC1-5 templates and compare the relative translation product signal and the full-length product ratio with the same template fusions of alanine, under the same conditions. It seemed that under the standard conditions sarcosine could not produce any product except for a single biocytin-containing product with CAC1 (Figure 13). Not only was the amount of sarcosine-containing product with the other templates (CAC2-5) so low that it was below the detection limit of the method, but the initiator methionine-only product was lower than in the reaction mixture with alanine under the same conditions. This was not only obvious by the lower intensity of the respective gel signal, it was also reflected in low scintillation count of CAC fusions with sarcosine. The following step in adjusting the system for unnatural amino acid incorporation was to investigate the effect of different Mg+2 concentrations on translation product signal. High concentrations of Mg+2 (over 5 mM, which is Mg+2 ion concentration in standard conditions as well as physiological Mg+2 ion concentration for bacterial cells), are known to decrease the fidelity of enzymatic binding of tRNA to ribosome by increasing the stability of ternary, tRNA/Ef-Tu/ribosome complex [102]. Even though initial mRNA display translation experiments were performed at 5 mM Mg(OAc)2 [67], other researchers in the field have been reporting concentrations higher than 5 mM for unnatural amino acid incorporation in peptides and mRNA display: Suga’s group reported using 20 mM [68], while Merryman and Green, Szostak and Forster reported using 10 mM Mg(OAc)2 in their in vitro translation  90  reactions [66, 88, 103]. Finally, Ueda’s group, the one that originated the PURE method used 9 mM Mg(OAc)2 for their protein synthesis [15]. Therefore, we titrated Mg+2 to our system to find the concentration of Mg+2 that would work best for our in vitro translation. The first test translation performed with CAC5 and alanine showed a substantial increase of translation product signal correlating with the increase of Mg+2 from 5 to 50 mM (Figure 14), which was also reflected in an increase of fusion scintillation counts. As expected, the sarcosine translation using 20 mM Mg+2 shown in Figure 15 yielded clearly detectable full-length product with CAC5, meaning that sarcosine not only got incorporated, but it elongated over four consecutive positions of the peptide part of the fusion possibly due to a four-fold increase of magnesium ion. This way, CAC5 sarcosine incorporation became comparable with alanine incorporation under the new magnesium concentration conditions. To determine scientific importance of this finding, both CAC5 alanine and sarcosine translation reactions under increasing magnesium ion concentrations conditions were repeated in four replicates, each on the different day and in alternating order, as described in Results section. The overall conclusion after assessing gels (Figure 15) and quantifying pmol total fusion (Figure 16) was that there is an increase of total radioactive fusions made with an increase of Mg+2 for both amino acids. This conclusion was confirmed by one-way ANOVA analysis for repeated measures for both amino acids after excluding the data acquired at 0 mM Mg(OAc)2, as explained in Results section. As per performed ANOVA analysis, the increase of total fusion signal for alanine reactions was highly significant (p<0.0001). Also, the pairing of the means for respective parameter of repeated experiments was found significant (p value of the same order of magnitude).  91  Similarly to this finding, the increase in the percent of the top band to the total signal per lane for alanine with the increase of the magnesium ion concentration was significant (p=0.0438), but the pairing between the means was also not effective (the difference in variance between repeated experiments was too large), therefore invalidating the analysis of variance for this parameter for alanine. The same result was found after several different attempts to transform the data for ANOVA, when y (% top band to total signal) was plotted as a log(y) or ln(y) or a reciprocal value of y. Finally, the increase of total fusion signal for sarcosine reactions with the increase of magnesium ion concentration was shown to be highly significant (p=0.0047). The increase in the percent of the top band to the total signal per lane for sarcosine with the increase of the magnesium ion concentration was shown to be very highly significant (p<0.0001), and the pairing between the means for that parameter was found also effective (p=0.0345) across the analyzed range of magnesium concentrations. 4.4  Conclusion  The magnesium titration experiment result showed that the total fusion product created with CAC5 template, which requires four consecutive incorporations of elongating amino acid, is increasing with an increase in magnesium ion concentration for alanine, and the same goes for sarcosine. However, the percent of fully elongated fusions out of total fusions is increasing with an increase of magnesium concentration only in sarcosine translations, while with alanine that trend was not observed. This demonstrates that the reconstituted system we made in-house works better for incorporation of this unnatural amino acid under higher magnesium concentration and the increased concentration of magnesium at 20 mM can be used as a starting point in testing  92  other unnatural amino acids incorporation as well. However, the incorporation of the natural amino acid alanine could be boosted too, but solely due to the positive effect of magnesium on overall fusion formation. Also, the total fusion product of any of the tested amino acids was highly dependent on the quality of starting materials (such as ribosomes and tRNA) and respective amino acid codon position in the coding sequence. This thesis research has shown that it is possible to generate a hexapeptide fusion that will contain all unnatural amino acids in its sequence (apart from initiator methionine), using an in-house prepared reconstituted system for in vitro translation. Unnatural hexapeptide and hexapeptide fusion synthesis has been reported by two other research groups who used different adaptor tRNA systems that reportedly introduced a considerable bias into the reconstituted system, affecting its ability to evenly incorporate different unnatural elements into peptides/fusions [67, 68]. Our system however, up to this phase of testing, did not show any notable differences in its ability to incorporate different unnatural elements into hexapeptide fusions under the same conditions (unpublished data), setting a good start for future synthesis of good quality mRNA display libraries of peptidomimetics. In addition, four successive incorporations of the same unnatural amino acid into an mRNA display fusion attained in this thesis research have not been reported previously. Also, most components of the in vitro translation system were made from basic starting materials, using custom working procedures, thus forming the basis for future relatively inexpensive regeneration of system components. Since we have demonstrated the ability of the reconstituted system to incorporate multiple residues of unnatural amino acid sarcosine into an mRNA display molecule, the system is  93  now ready to be taken further into testing, with other unnatural amino acids and the library of templates. 4.5  Future experiments  The first step in forthcoming system evaluations is to check the incorporation of other unnatural amino acids that will be used in library design, again using CAC5 template and biocytin as an affinity handle. This step has already been done by other laboratory members. They have performed a series of CAC5 template translations with the following N-methyl and cyclic amino acids charged onto tRNAGGC: N-Me-L-Alanine, N-Me-L-Valine, LAzetidine-2-Carboxylic Acid, N-Me-L-Glutamate, N-Me-L-Phenylalanine, L-Homoproline, N-Me-L-Ornithine, L-Abrine, t-Bu-L-Glycine, N-Me-L-Aminohexanoic Acid, LOctahydroindole-2-Carboxylic Acid, L-Biocytin, and L-Proline. The percent top band to the total signal at 20 mM magnesium ion concentration obtained with all of the tested unnatural amino acids was very similar, around or slightly higher than 20%, which is consistent with the percent top band obtained with alanine under the same conditions (when using the same batch of ribosomes). This demonstrates that even though labor intensive, modification of the method to chemo-enzymatically charge tRNAs does not introduce any notable bias in charging adaptors with unnatural elements to be incorporated into an mRNA display molecule of the methods tested so far. This is the most prominent feature of the selected methods for our reconstituted system assembly and it may offer a more unbiased system for library synthesis than other competing methods. In the final phase of testing with single templates, these fully orthogonal, N-methyl-amino acid-containing products may be tested for proteolytic resistance to gain a desirable attribute of bearing druglike properties.  94  The next step will be designing the template libraries based on CAC1-5 sequences, with the addition of minimally 4 codon sequences downstream from the library coding region, to serve as a constant part of each library template for a primer that will be used post selection to amplify selected sequences. Once the templates are obtained, a test translation with one amino acid chemo-enzymatically charged on the pool of tRNAs will be needed to show the ability of the system to generate elongated product with the tested template library. This will test whether the templates are in good working condition and if they are, the template library can be taken to the next phase. The next phase of system testing will include translation of template libraries with a pool of unnatural (N-methylated) amino acids charged each on a designated tRNA of the assigned codon sequence for that amino acid. In this phase, the exact yield under constant conditions will need to be determined in order to estimate as close as possible the amount of sequence space that the libraries will be able to cover using the reconstituted system. This test will also validate the de novo genetic code, important for future decoding of fusion sequences collected after selection. Most importantly, once libraries are synthesized they will be tested for their resistance to proteolysis using various proteases. As a positive control on proteolytic digestion, a set of natural fusion libraries will be made using the same template libraries and reconstituted system with natural amino acids and AARSs, and will be treated with proteases in parallel with the unnatural fusion libraries. It is highly anticipated that the libraries made out of N-methyl amino acids will show substantially more resistance to proteolysis than their natural counterparts [8, 10]. Upon confirming that elongation of the fusion of appropriate size takes place with a respective template and that the libraries are relatively resistant to proteases, the system will  95  be ready to be used in new drug discovery. The libraries synthesized using an optimized system may be used to perform selections against immobilized protein target and choose the best binding fusions for the protein, and will serve as a proof-of-concept for this new drug discovery tool.  96  BIBLIOGRAPHY 1.  Hruby, H.J., Gehrig de Chavez, C., Kavarana, M., Peptide and Protein Hormones, Peptide Neurotransmitters, and Therapeutic Agents, in Burger's Medicinal Chemistry and Drug Discovery. 2003, John Wiley & Sons, Inc.  2.  Smith, G.P. and V.A. Petrenko, Phage Display. Chem Rev, 1997. 97(2): p. 391-410.  3.  Hanes, J. and A. Pluckthun, In vitro selection and evolution of functional proteins by using ribosome display. Proc Natl Acad Sci U S A, 1997. 94(10): p. 4937-42.  4.  Roberts, R.W. and J.W. Szostak, RNA-peptide fusions for the in vitro selection of peptides and proteins. Proc Natl Acad Sci U S A, 1997. 94(23): p. 12297-302.  5.  Lam, K.S., et al., The "One-Bead-One-Compound" Combinatorial Library Method. Chem Rev, 1997. 97(2): p. 411-448.  6.  Brenner, S. and R.A. Lerner, Encoded combinatorial chemistry. Proc Natl Acad Sci U S A, 1992. 89(12): p. 5381-3.  7.  Nazif, T. and M. Bogyo, Global analysis of proteasomal substrate specificity using positional-scanning libraries of covalent inhibitors. Proc Natl Acad Sci U S A, 2001. 98(6): p. 2967-72.  8.  Frankel, A., et al., Encodamers: unnatural peptide oligomers encoded in RNA. Chem Biol, 2003. 10(11): p. 1043-50.  9.  Frankel, A., et al., Unnatural RNA display libraries. Curr Opin Struct Biol, 2003. 13(4): p. 506-12.  10.  Finking, R. and M.A. Marahiel, Biosynthesis of nonribosomal peptides1. Annu Rev Microbiol, 2004. 58: p. 453-88.  11.  Uy, R. and F. Wold, Posttranslational covalent modification of proteins. Science, 1977. 198(4320): p. 890-6.  12.  Gevers, W., et al., The activation of amino acids for biosynthesis of gramicidin S. Proc Natl Acad Sci U S A, 1968. 60(1): p. 269-76.  13.  Baltz, R.H., Molecular engineering approaches to peptide, polyketide and other antibiotics. Nat Biotechnol, 2006. 24(12): p. 1533-40.  14.  Menzella, H.G. and C.D. Reeves, Combinatorial biosynthesis for drug development. Curr Opin Microbiol, 2007. 10(3): p. 238-45.  15.  Shimizu, Y., et al., Cell-free translation reconstituted with purified components. Nat Biotechnol, 2001. 19(8): p. 751-5. 97  16.  Wilson, D.N. and K.H. Nierhaus, The weird and wonderful world of bacterial ribosome regulation. Crit Rev Biochem Mol Biol, 2007. 42(3): p. 187-219.  17.  Munro, J.B., et al., Identification of two distinct hybrid state intermediates on the ribosome. Mol Cell, 2007. 25(4): p. 505-17.  18.  Ramakrishnan, V., Ribosome structure and the mechanism of translation. Cell, 2002. 108(4): p. 557-72.  19.  Zucker, F.H. and J.W. Hershey, Binding of Escherichia coli protein synthesis initiation factor IF1 to 30S ribosomal subunits measured by fluorescence polarization. Biochemistry, 1986. 25(12): p. 3682-90.  20.  Hartz, D., et al., Domains of initiator tRNA and initiation codon crucial for initiator tRNA selection by Escherichia coli IF3. Genes Dev, 1990. 4(10): p. 1790-800.  21.  Karimi, R., et al., Novel roles for classical factors at the interface between translation termination and initiation. Mol Cell, 1999. 3(5): p. 601-9.  22.  Rodnina, M.V. and W. Wintermeyer, Ribosome fidelity: tRNA discrimination, proofreading and induced fit. Trends Biochem Sci, 2001. 26(2): p. 124-30.  23.  Rodnina, M.V., et al., Structural dynamics of ribosomal RNA during decoding on the ribosome. Biochimie, 2002. 84(8): p. 745-54.  24.  Wintermeyer, W., et al., Mechanisms of elongation on the ribosome: dynamics of a macromolecular machine. Biochem Soc Trans, 2004. 32(Pt 5): p. 733-7.  25.  Kisselev, L.L. and R.H. Buckingham, Translational termination comes of age. Trends Biochem Sci, 2000. 25(11): p. 561-6.  26.  Zavialov, A.V., et al., A posttermination ribosomal complex is the guanine nucleotide exchange factor for peptide release factor RF3. Cell, 2001. 107(1): p. 115-24.  27.  Kaji, A., et al., The fourth step of protein synthesis: disassembly of the posttermination complex is catalyzed by elongation factor G and ribosome recycling factor, a near-perfect mimic of tRNA. Cold Spring Harb Symp Quant Biol, 2001. 66: p. 515-29.  28.  Selmer, M., et al., Crystal structure of Thermotoga maritima ribosome recycling factor: a tRNA mimic. Science, 1999. 286(5448): p. 2349-52.  29.  Joseph, S. and H.F. Noller, EF-G-catalyzed translocation of anticodon stem-loop analogs of transfer RNA in the ribosome. Embo J, 1998. 17(12): p. 3478-83.  30.  Crick, F.H., On protein synthesis. Symp Soc Exp Biol, 1958. 12: p. 138-63.  31.  Crick, F., Central dogma of molecular biology. Nature, 1970. 227(5258): p. 561-3. 98  32.  Söll, D. and T.L.R. Bhandary, tRNA: Structure, Biosynthesis, and Function. 1995, Washington, DC: American Society for Microbiology Press.  33.  Yarus, M., The accuracy of translation. Prog Nucleic Acid Res Mol Biol, 1979. 23: p. 195-225.  34.  Kurland, C.G., Translational accuracy and the fitness of bacteria. Annu Rev Genet, 1992. 26: p. 29-50.  35.  Jacob, W.F., et al., A single base change in the Shine-Dalgarno region of 16S rRNA of Escherichia coli affects translation of many proteins. Proc Natl Acad Sci U S A, 1987. 84(14): p. 4757-61.  36.  Steitz, J.A. and K. Jakes, How ribosomes select initiator regions in mRNA: base pair formation between the 3' terminus of 16S rRNA and the mRNA during initiation of protein synthesis in Escherichia coli. Proc Natl Acad Sci U S A, 1975. 72(12): p. 4734-8.  37.  Olins, P.O. and S.H. Rangwala, A novel sequence element derived from bacteriophage T7 mRNA acts as an enhancer of translation of the lacZ gene in Escherichia coli. J Biol Chem, 1989. 264(29): p. 16973-6.  38.  Zhang, J. and M.P. Deutscher, A uridine-rich sequence required for translation of prokaryotic mRNA. Proc Natl Acad Sci U S A, 1992. 89(7): p. 2605-9.  39.  Zhelyabovskaya, O.B., et al., Artificial genetic selection for an efficient translation initiation site for expression of human RACK1 gene in Escherichia coli. Nucleic Acids Res, 2004. 32(5): p. e52.  40.  Sprengart, M.L., et al., The downstream box: an efficient and independent translation initiation signal in Escherichia coli. Embo J, 1996. 15(3): p. 665-74.  41.  O'Connor, M. and A.E. Dahlberg, Enhancement of translation by the epsilon element is independent of the sequence of the 460 region of 16S rRNA. Nucleic Acids Res, 2001. 29(7): p. 1420-5.  42.  Nelson, D.L. and M.M. Cox, Lehninger Principles of Biochemistry. 2005, New York: W.H. Freeman and Company.  43.  Sprengel, R., et al., Translationally coupled initiation of protein synthesis in Bacillus subtilis. Nucleic Acids Res, 1985. 13(3): p. 893-909.  44.  Karatan, E., et al., Molecular Display Technologies, in Encyclopedia of Molecular Cell Biology and Molecular Medicine. 2006, John Wiley & Sons, Inc.: Weinheim.  45.  Smith, G.P., Filamentous fusion phage: novel expression vectors that display cloned antigens on the virion surface. Science, 1985. 228(4705): p. 1315-7.  99  46.  Little, M., et al., Bacterial surface presentation of proteins and peptides: an alternative to phage technology? Trends Biotechnol, 1993. 11(1): p. 3-5.  47.  Georgiou, G., et al., Display of heterologous proteins on the surface of microorganisms: from the screening of combinatorial libraries to live recombinant vaccines. Nat Biotechnol, 1997. 15(1): p. 29-34.  48.  Francisco, J.A., et al., Production and fluorescence-activated cell sorting of Escherichia coli expressing a functional antibody fragment on the external surface. Proc Natl Acad Sci U S A, 1993. 90(22): p. 10444-8.  49.  Boder, E.T. and K.D. Wittrup, Yeast surface display for screening combinatorial polypeptide libraries. Nat Biotechnol, 1997. 15(6): p. 553-7.  50.  Cho, B.K., et al., A yeast surface display system for the discovery of ligands that trigger cell activation. J Immunol Methods, 1998. 220(1-2): p. 179-88.  51.  Mattheakis, L.C., et al., An in vitro polysome display system for identifying ligands from very large peptide libraries. Proc Natl Acad Sci U S A, 1994. 91(19): p. 9022-6.  52.  Nemoto, N., et al., In vitro virus: bonding of mRNA bearing puromycin at the 3'terminal end to the C-terminal end of its encoded protein on the ribosome in vitro. FEBS Lett, 1997. 414(2): p. 405-8.  53.  Stiege, W. and V.A. Erdmann, The potentials of the in vitro protein biosynthesis system. J Biotechnol, 1995. 41(2-3): p. 81-90.  54.  Jermutus, L., et al., Recent advances in producing and selecting functional proteins by using cell-free translation. Curr Opin Biotechnol, 1998. 9(5): p. 534-48.  55.  Kitaoka, Y., et al., Cooperativity of stabilized mRNA and enhanced translation activity in the cell-free system. J Biotechnol, 1996. 48(1-2): p. 1-8.  56.  Matveev, S.V., et al., Effect of the ATP level on the overall protein biosynthesis rate in a wheat germ cell-free system. Biochim Biophys Acta, 1996. 1293(2): p. 207-12.  57.  Kung, H.F., et al., DNA-directed in vitro synthesis of beta-galactosidase. Studies with purified factors. J Biol Chem, 1977. 252(19): p. 6889-94.  58.  Levin, A.M. and G.A. Weiss, Optimizing the affinity and specificity of proteins with molecular display. Mol Biosyst, 2006. 2(1): p. 49-57.  59.  Takahashi, T.T., et al., mRNA display: ligand discovery, interaction analysis and beyond. Trends Biochem Sci, 2003. 28(3): p. 159-65.  60.  Frankel, A., Personal communication, V. Stojanovic, Editor. 2006.  100  61.  Hanes, J., et al., Ribosome display efficiently selects and evolves high-affinity antibodies in vitro from immune libraries. Proc Natl Acad Sci U S A, 1998. 95(24): p. 14130-5.  62.  Lin-Goerke, J.L., et al., PCR-based random mutagenesis using manganese and reduced dNTP concentration. Biotechniques, 1997. 23(3): p. 409-12.  63.  Zaccolo, M., et al., An approach to random mutagenesis of DNA using mixtures of triphosphate derivatives of nucleoside analogues. J Mol Biol, 1996. 255(4): p. 589603.  64.  Stemmer, W.P., Rapid evolution of a protein in vitro by DNA shuffling. Nature, 1994. 370(6488): p. 389-91.  65.  Forster, A.C., et al., Programming peptidomimetic syntheses by translating genetic codes designed de novo. Proc Natl Acad Sci U S A, 2003. 100(11): p. 6353-7.  66.  Merryman, C. and R. Green, Transformation of aminoacyl tRNAs for the in vitro selection of "drug-like" molecules. Chem Biol, 2004. 11(4): p. 575-82.  67.  Josephson, K., et al., Ribosomal synthesis of unnatural peptides. J Am Chem Soc, 2005. 127(33): p. 11727-35.  68.  Kawakami, T., et al., Messenger RNA-programmed incorporation of multiple Nmethyl-amino acids into linear and cyclic peptides. Chem Biol, 2008. 15(1): p. 32-42.  69.  Li, S. and R.W. Roberts, A novel strategy for in vitro selection of peptide-drug conjugates. Chem Biol, 2003. 10(3): p. 233-9.  70.  Millward, S.W., et al., A general route for post-translational cyclization of mRNA display libraries. J Am Chem Soc, 2005. 127(41): p. 14142-3.  71.  Noren, C.J., et al., A general method for site-specific incorporation of unnatural amino acids into proteins. Science, 1989. 244(4901): p. 182-8.  72.  Roesser, J.R., et al., Ribosome-catalyzed formation of an abnormal peptide analogue. Biochemistry, 1986. 25(21): p. 6361-5.  73.  Saks, M.E., et al., An engineered Tetrahymena tRNAGln for in vivo incorporation of unnatural amino acids into proteins by nonsense suppression. J Biol Chem, 1996. 271(38): p. 23169-75.  74.  Wang, L., et al., Expanding the genetic code of Escherichia coli. Science, 2001. 292(5516): p. 498-500.  75.  Wu, N., et al., A genetically encoded photocaged amino acid. J Am Chem Soc, 2004. 126(44): p. 14306-7.  101  76.  England, P.M., Unnatural amino acid mutagenesis: a precise tool for probing protein structure and function. Biochemistry, 2004. 43(37): p. 11623-9.  77.  Bain, J.D., et al., Ribosome-mediated incorporation of a non-standard amino acid into a peptide through expansion of the genetic code. Nature, 1992. 356(6369): p. 537-9.  78.  Hohsaka, T., et al., Five-base codons for incorporation of nonnatural amino acids into proteins. Nucleic Acids Res, 2001. 29(17): p. 3646-51.  79.  Magliery, T.J., et al., Expanding the genetic code: selection of efficient suppressors of four-base codons and identification of "shifty" four-base codons with a library approach in Escherichia coli. J Mol Biol, 2001. 307(3): p. 755-69.  80.  Hohsaka, T., et al., Incorporation of nonnatural amino acids into proteins by using various four-base codons in an Escherichia coli in vitro translation system. Biochemistry, 2001. 40(37): p. 11060-4.  81.  Anderson, J.C., et al., An expanded genetic code with a functional quadruplet codon. Proc Natl Acad Sci U S A, 2004. 101(20): p. 7566-71.  82.  Cornish, V.W., et al., Site-specific incorporation of biophysical probes into proteins. Proc Natl Acad Sci U S A, 1994. 91(8): p. 2910-4.  83.  Forster, A.C., et al., A simplified reconstitution of mRNA-directed peptide synthesis: activity of the epsilon enhancer and an unnatural amino acid. Anal Biochem, 2001. 297(1): p. 60-70.  84.  Hartman, M.C., et al., Enzymatic aminoacylation of tRNA with unnatural amino acids. Proc Natl Acad Sci U S A, 2006. 103(12): p. 4356-61.  85.  Wang, L. and P.G. Schultz, Expanding the genetic code. Chem Commun (Camb), 2002(1): p. 1-11.  86.  Fahnestock, S. and A. Rich, Ribosome-catalyzed polyester formation. Science, 1971. 173(994): p. 340-3.  87.  Murakami, H., et al., A highly flexible tRNA acylation method for non-natural polypeptide synthesis. Nat Methods, 2006. 3(5): p. 357-9.  88.  Subtelny, A.O., et al., Ribosomal synthesis of N-methyl peptides. J Am Chem Soc, 2008. 130(19): p. 6131-6.  89.  Hartman, M.C., et al., An expanded set of amino Acid analogs for the ribosomal translation of unnatural peptides. PLoS ONE, 2007. 2(10): p. e972.  90.  Tan, Z., et al., Amino acid backbone specificity of the Escherichia coli translation machinery. J Am Chem Soc, 2004. 126(40): p. 12752-3. 102  91.  Hecht, S.M., et al., "Chemical aminoacylation" of tRNA's. J Biol Chem, 1978. 253(13): p. 4517-20.  92.  Robertson, S.A., et al., The use of 5'-phospho-2 deoxyribocytidylylriboadenosine as a facile route to chemical aminoacylation of tRNA. Nucleic Acids Res, 1989. 17(23): p. 9649-60.  93.  Spedding, G., Isolation and analysis of ribosomes from prokaryotes, eukaryotes, and organelles. Ribosomes and protein synthesis: a practical approach. 1990, New York: Oxford University Press. 1-27.  94.  LaemmLi, U.K., Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 1970. 227(5259): p. 680-5.  95.  Gasteiger, E., et al., ExPASy: The proteomics server for in-depth protein knowledge and analysis. Nucleic Acids Res, 2003. 31(13): p. 3784-8.  96.  Stover, P. and V. Schirch, Synthesis of (6S)-5-formyltetrahydropteroylpolyglutamates and interconversion to other reduced pteroylpolyglutamate derivatives. Anal Biochem, 1992. 202(1): p. 82-8.  97.  Pleiss, J.A. and O.C. Uhlenbeck, Identification of thermodynamically relevant interactions between EF-Tu and backbone elements of tRNA. J Mol Biol, 2001. 308(5): p. 895-905.  98.  Li, S., et al., In vitro selection of mRNA display libraries containing an unnatural amino acid. J Am Chem Soc, 2002. 124(34): p. 9972-3.  99.  Liu, R., et al., Optimized synthesis of RNA-protein fusions for in vitro protein selection. Methods Enzymol, 2000. 318: p. 268-93.  100.  Forster, A.C., et al., Pure translation display. Anal Biochem, 2004. 333(2): p. 35864.  101.  Schagger, H. and G. von Jagow, Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa. Anal Biochem, 1987. 166(2): p. 368-79.  102.  Thompson, R.C., et al., Effect of Mg2+ concentration, polyamines, streptomycin, and mutations in ribosomal proteins on the accuracy of the two-step selection of aminoacyl-tRNAs in protein biosynthesis. J Biol Chem, 1981. 256(13): p. 6676-81.  103.  Tan, Z., et al., De novo genetic codes and pure translation display. Methods, 2005. 36(3): p. 279-90.  103  APPENDIX Table 7. The list of required purified components for the translation system. The list contains all protein components of the PURE system made in the course of this thesis research. T7 RNA polymerase clone donated by Dr. Richard Fahlman (University of Alberta); all other protein and tRNA clones kindly supplied by Dr. Jack Szostak (Harvard University).  Protein name  E.coli clone (resist.)  MW (kDa)  ε (M-1∗ cm-1)*  T7 RNA Polymerase  #193 (AmpR)  98.9  141 010  Methionyl-tRNA Transformylase  pET24a-MTF (KanR)  34.2  44 585  Initiation Factor 1  pET28a-IF1(KanR)  8.2  2 980  Initiation Factor 2  pET28a-IF2 (KanR)  97.3  27 515  Initiation Factor 3  pET28a-IF3 (KanR)  20.6  4 470  Elongation Factor Tu  pET24a-EFTu (KanR)  43.3  20 525  Elongation Factor Ts  pET24a-EFTs (KanR)  30.4  4 595  Elongation Factor G  pET24a-EFG (KanR)  77.6  61 435  Alanyl-tRNA Synthetase  pET28a-AlaRS (KanR)  96.0  71 195  Arginyl-tRNA Synthetase  pET16b-ArgRS (AmpR)  64.7  62 020  Aspartyl-tRNA Synthetase  pQE30-AsnRS (AmpR)  65.9  43 110  Asparaginyl-tRNA Synthetase  pET24a-AspRS (KanR)  52.6  62 590  Cysteinyl-tRNA Synthetase  pET21a-CysRS (AmpR)  52.2  62 590  Glutaminyl-tRNA Synthetase  pET21a-GlnRS (AmpR)  65.6  82 195  Glutamyl-tRNA Synthetase  pET24a-GluRS (KanR)  63.5  Glycyl-tRNA Synthetase  pET21a-GlyRS (AmpR)  34.8(α) 76.8(β)  Histidinyl-tRNA Synthetase  pET21a-HisRS (AmpR)  47.0  77 405 60 070(α) 57 995(β) 55 600  Isoleucyl-tRNA Synthetase  pET21a-IleRS (AmpR)  104.3  176 545  Leucyl-tRNA Synthetase  pET21a-LeuRS (AmpR)  97.2  169 765  Lysyl-tRNA Synthetase  pET3a-LysRS (AmpR)  57.8  33 350  Methionyl-tRNA Synthetase  pET24a-MetRS (KanR)  76.6  Phenylalanyl-tRNA Synthetase  pET28a-PheRS (KanR)  36.8(α) 87.4(β)  Prolyl-tRNA Synthetase  pET21a-ProRS (AmpR)  63.6  96 760 21 430(α) 65 610(β) 54 320  Seryl-tRNA Synthetase  pET24a-SerRS (KanR)  48.4  34 630  Threonyl-tRNA Synthetase  pET28a-ThrRS (KanR)  74.0  97 010  Triptophanyl-tRNA Synthetase  pET21a-TrpRS (AmpR)  37.4  32 110  Tyrosyl-tRNA Synthetase  pET20-TyrRS (AmpR)  45.9  31 400  Valyl-tRNA Synthetase  pET21a-ValRS (AmpR)  108.2  171 240  104  Figure 22.  A selection of SDS-PAGE gel analysis of FPLC purified components of reconstituted system. The proteins were expressed and purified using FPLC as described. 10% SDS-PAGE gel analysis of FPLC fractions confirmed the purity and the correct size of the purified proteins. The appropriate fractions were pooled as outlined, the proteins were desalted and stored at minus 80° C.  105  Table 8. Set of DNA primers used for E. coli tRNA library synthesis. The nineteen sets of reverse and forward DNA primers were designed based on the alaW gene sequence and instructions provided in the Stratagene QuickChange® protocol for sitedirected mutagenesis and used as descibed in the methods section in making the library of E. coli tRNA molecules lacking -CA on their 3’ end.  Primer name AAC2F AAC2R AAG2F AAG2R ACG2F ACG2R AUC2F AUC2R CAC2F CAC2R CAG2F CAG2R CCC2F CCC2R CGG2F CGG2R CUC2F CUC2R GAC2F GAC2R GAG2F GAG2R GGG2F GGG2R GUG2F GUG2R UAC2F UAC2R UAG2F UAG2R UCG2F UCG2R UGC2F UGC2R UGG2F UGG2R UUC2F UUC2R  Primer 5’ to 3’ sequence GCT GGG AGA GCG CTT GCA TGT TAT GCA AGA GGT CAG CGG CCG CTG ACC TCT TGC ATA ACA TGC AAG CGC TCT CCC AGC GCT GGG AGA GCG CTT GCA TCT TAT GCA AGA GGT CAG CGG CCG CTG ACC TCT TGC ATA AGA TGC AAG CGC TCT CCC AGC GCT GGG AGA GCG CTT GCA TCG TAT GCA AGA GGT CAG CGG CCG CTG ACC TCT TGC ATA CGA TGC AAG CGC TCT CCC AGC GCT GGG AGA GCG CTT GCA TGA TAT GCA AGA GGT CAG CGG CCG CTG ACC TCT TGC ATA TCA TGC AAG CGC TCT CCC AGC GCT GGG AGA GCG CTT GCA TGT GAT GCA AGA GGT CAG CGG CCG CTG ACC TCT TGC ATC ACA TGC AAG CGC TCT CCC AGC GCT GGG AGA GCG CTT GCA TCT GAT GCA AGA GGT CAG CGG CCG CTG ACC TCT TGC ATC AGA TGC AAG CGC TCT CCC AGC GCT GGG AGA GCG CTT GCA TGG GAT GCA AGA GGT CAG CGG CCG CTG ACC TCT TGC ATC CCA TGC AAG CGC TCT CCC AGC GCT GGG AGA GCG CTT GCA TCC GAT GCA AGA GGT CAG CGG CCG CTG ACC TCT TGC ATC GGA TGC AAG CGC TCT CCC AGC GCT GGG AGA GCG CTT GCA TGA GAT GCA AGA GGT CAG CGG CCG CTG ACC TCT TGC ATC TCA TGC AAG CGC TCT CCC AGC GCT GGG AGA GCG CTT GCA TGT CAT GCA AGA GGT CAG CGG CCG CTG ACC TCT TGC ATG ACA TGC AAG CGC TCT CCC AGC GCT GGG AGA GCG CTT GCA TCT CAT GCA AGA GGT CAG CGG CCG CTG ACC TCT TGC ATG AGA TGC AAG CGC TCT CCC AGC GCT GGG AGA GCG CTT GCA TCC CAT GCA AGA GGT CAG CGG CCG CTG ACC TCT TGC ATG GGA TGC AAG CGC TCT CCC AGC GCT GGG AGA GCG CTT GCA TCA CAT GCA AGA GGT CAG CGG CCG CTG ACC TCT TGC ATG TGA TGC AAG CGC TCT CCC AGC GCT GGG AGA GCG CTT GCA TGT AAT GCA AGA GGT CAG CGG CCG CTG ACC TCT TGC ATT ACA TGC AAG CGC TCT CCC AGC GCT GGG AGA GCG CTT GCA TCT AAT GCA AGA GGT CAG CGG CCG CTG ACC TCT TGC ATT AGA TGC AAG CGC TCT CCC AGC GCT GGG AGA GCG CTT GCA TCG AAT GCA AGA GGT CAG CGG CCG CTG ACC TCT TGC ATT CGA TGC AAG CGC TCT CCC AGC GCT GGG AGA GCG CTT GCA TGC AAT GCA AGA GGT CAG CGG CCG CTG ACC TCT TGC ATT GCA TGC AAG CGC TCT CCC AGC GCT GGG AGA GCG CTT GCA TCC AAT GCA AGA GGT CAG CGG CCG CTG ACC TCT TGC ATT GGA TGC AAG CGC TCT CCC AGC GCT GGG AGA GCG CTT GCA TGA AAT GCA AGA GGT CAG CGG CCG CTG ACC TCT TGC ATT TCA TGC AAG CGC TCT CCC AGC  106  Table 9. Standard conditions (composition) of PURE translation system. An outline of protein and other components concentrations used in translation reactions when standard conditions annotated (NA = not applicable).  Protein or other  Current stock conc. (µM)  Working conc., approx.  10xTransl.buff.  NA  NA  10xCaCl2  5 000  500 µM  10xMg(OAc)2  50 000  5 mM*  IF1  45.00  0.5 µM  IF2  24.82  0.5 µM  IF3  115.53  0.5 µM  EF-Tu  82.19  6.6 µM  EF-Ts  53.92  0.5 µM  EF-G  38.25  0.5 µM  MTF  45.06  0.5 µM  10-formyl-THF  3 000  30 µM  MetRS  23.33  0.5 µM  Other RS, if used  NA  0.5 µM  fMet-tRNA  20.86  0.4 µM (or 0.02 µg/µL)  1 000  80 µM  1 000 - 40 000  10 - 400 µM  Cold Met (added when needed to maximize the fusion yield) Natural amino acids (if tRNA acylated in situ) Natural aminoacyl-tRNA (acylated in situ) Alanyl-tRNA (acylated separately from translation) Biocytinyl-tRNA (acylated separately from translation) Sarcosyl-tRNA (acylated separately from translation) Ribosomes  various  0.8 µM (or 0.02 µg/µL) per template codon 1.6 µM (or 0.04 µg/µL) per template codon 2.4 µM (or 0.06 µg/µL) per template codon 2.8 - 3.2 µM (or 0.07 - 0.08 µg/µL) per template codon 1 µM or more  [35S]Met  10  0.8 µM  Template  200  Purified water  NA  40 or more 40 or more 40 or more 40 or more  0.8 µM (40 pmol per 50-µL reaction) NA  *Current modification to standard conditions is in Mg(OAc)2 working concentration for unnatural amino acid incorporation: 20 mM instead of 5 mM.  107  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0066852/manifest

Comment

Related Items