In the present work a modified version of yellow fluorescent protein containing an unnatural structural homologue of the natural amino acid pyrrolysine with a norbornene moiety was produced by expression in Escherichia coli. The incorporation of the unnatural amino acid was achieved by amber stop codon suppression method. A bio-othogonal click reaction was performed, binding a synthetic fluorescent dye to the modified protein. All steps towards necessary for obtaining the genetically modified organism were performed and documented. The artificial amino acid, as well as the dye used in the click reaction were synthetically prepared. The success of the project was demonstrated by LC/MS studies of the products. Fluorescence spectroscopy of click reaction product and the protein was performed, but no conclusive proof of FRET effects could as yet be made. This point remains of interest for future studies.
Table of contents
Abstract
Glossary
1 Introduction: Oligonucleotide Synthesis
2 Objective
3 Procedures and Materials
3.1 Chemistry
3.1.1 Methods and Materials utilized
3.1.2 Syntheses
3.2 Biology
3.2.1 Methods and Materials utilized
3.2.2 Primer Synthesis
3.2.3 Plasmid Mini Prep
3.2.4 Agarose Gel Electrophoresis
3.2.5 Polymerase Chain Reaction
3.2.6 Ligation
3.2.7 Transformation
3.2.8 Colony PCR
3.2.9 Protein Expression
3.2.10 Protein Purification
3.2.11 SDS-PAGE
3.2.12 Protein Digest
3.2.13 Protein Click Reaction
4 Results and Discussion
4.1 Agarose Gel Electrophoresis
4.2 SDS PAGE
4.3 Fluorescence Spectroscopy
4.4 Mass Spectrometry
4.4.1 MALDI-TOF
4.4.2 LC / MS
5 References
Abstract
In the present work a modified version of yellow fluorescent protein containing an unnatural structural homologue of the natural amino acid pyrrolysine with a norbornene moiety was produced by expression in Escherichia coli. The incorporation of the unnatural amino acid was achieved by amber stop codon suppression method. A bio-othogonal click reaction was performed, binding a synthetic fluorescent dye to the modified protein. All steps towards necessary for obtaining the genetically modified organism were performed and documented. The artificial amino acid, as well as the dye used in the click reaction were synthetically prepared. The success of the project was demonstrated by LC/MS studies of the products. Fluorescence spectroscopy of click reaction product and the protein was performed, but no conclusive proof of FRET effects could as yet be made. This point remains of interest for future studies.
Glossary
illustration not visible in this excerpt
1 Introduction: Oligonucleotide Synthesis
As prescribed by course instructors, this introduction presents only one
of the theoretical aspects relevant to the course.
In this work custom-designed DNA primers for PCR were obtained by automated solid-state DNA synthesis using the phosphoramidite method. Allowing the artificial preparation of oligonucleotides of any desired sequence, and even the incorporation of unnatural synthetic nucleosides this technique has become a versatile tool for modern chemical biology and molecular medicine. Synthetic oligonucleotides are applied, among others as hybridization probes for detection of complementary DNA and RNA, tools for introduction of restriction sites, primers for PCR, siRNA and antisense oligonucleotides. Oligonucleotide synthesis requires a sequential coupling of each nucleoside unit to the growing polymer chain in the right order. It is essential that each step proceed rapidly, reliably and with extremely high yield, because the high number of nucleosides, up to 200 per molecule, would potentiate any chance of false coupling or side reaction, as only one missing nucleoside alters the entire oligonucleotide sequence. Early work in the 1950s[2],[3] developed oligonucleotide synthesis from the retrosynthetically logical synthetic equivalents of H-phosphonates and phosphate trieesters of nucleosides as naturally occurring 3’- and 5’- phosphates and phosphodiesters are not reactive enough for reliable coupling. Contemporary synthesis by the phosphoramidite method[4] makes use of an optimized set of protecting goups and reaction conditions, which are sufficiently reliable to make automation of the process possible.
Phosphoramidites are derivatives of nucleosides that possess a 3’- O -(N,N-diisopropyl phosphoramidite) as the activated ester equivalent. The other oxygen of the phosphoramidite group is bound to a 2-cyanoethyl residue serving as a base-labile protective group (cleavage as acrylonitrile) [4]. In order to secure the synthesis against any side reactions, all other reactive functional groups need to be protected. The DNA bases require different treatment: while thymine is naturally unreactive, the exocyclic amines need protection. This is effected by using the benzoyl amide of adenine and the acetyl or benzoyl amide of cytosine and the isobutyryl amide of guanine, all of which are labile under basic conditions. The 5’ hydroxyl group of desoxyribose is protected by an acid labile 4,4’-dimethoxytrityl group (DMT).[7] If RNA is to be synthesized, the 2’ hydroxyl group is protected with TBDMS, which can be removed by fluoride ions. The solid support upon which the oligonucleotide chain remains bound during the entire synthesis may be either controlled-pore glass or macroporous polystyrene carrying aminopropyl or aminomethyl residues, respectively. In order to permit binding of the first nucleoside, linker molecules like e.g. succinyl amide are attached to these, whose carboxyl group in turn forms an ester with the 3’ hydroxyl group of the first nucleoside.
The synthetic cycle, an overview of which is given in figure 1, proceeds through four steps. The first step (15 seconds) is deprotection of the 5’ hydroxyl group of the terminal nucleoside. The DMT group is removed by treatment with a solution of 3 % dichloroacetic acid in methylene chloride or toluene, in the form of the trityl cation, whose amount can easily be measured as the conductivity of the solution. These “trityl values” indicate the amount of molecules taking part in the reaction and should be constant over all synthetic steps. The next step is the coupling step. For this a solution of the appropriate nucleoside phosphoramidite in acetonitrile is mixed with a solution of tetrazole directly before being brought into contact with the solid phase. The resulting tetrazole phosphoramidate is extremely reactive and reacts immediately with the free hydroxyl group of the oligonucleotide on the solid support. Step 2 takes about 30 seconds. The subsequent step is capping of the (small percentage) unreacted free 5’ hydroxyl groups. This is indispensible because synthesis would otherwise be continued on these strands during the next cycle, resulting in chains with a missing base and hence false sequence by frame-shift. It also serves to remove phosphoramidites erroneously bound to O[6] in guanidine.[8] This third step takes about 10 s and is effected by treatment with a solution of acetic anhydride and 1-methylimidazole. Step four is the oxidation of the trivalent phosphite triester linkage created when the chain was elongated. Oxidation is achieved by adding iodine and water in acetonitrile in the presence of a weak base like lutidine. The phosphite first attacks iodine with its lone pair, forming a tetra-coordinated phosphor cation with one iodine atom bound. Iodine is then substituted by the nucleophile water, which is deprotonated, resulting in pentavalent phosphate.[1]
The cycle may in theory be repeated infinitely, but practical experience is that the yield of each step even in automated synthesis is close to, but not exactly quantitative, limiting the procedure to a maximum of 200mer oligonucleotides.
Abbildung in dieser Leseprobe nicht enthalten
Figure 1: phosphoramidite synthesis cycle. [1]
When the last nucleoside has been added to the chain the complete polymer needs to be cleaved from the solid support and the remaining protective groups, i.e. the various acyl residues on the bases and the 2-cyanoethyl residue on the phosphates need to be removed. As all these bonds are base-labil this can be achieved by treatment with an aqueous solution of methylamine and/or ammonia. For many applications the product is then purified via reverse-phase HPLC.[1]
2 Objective
The goal of this study is to achieve bacterial expression of a modified version of yellow fluorescent protein (YFP), a well-known derivative of the green fluorescent protein (GFP) from the jellyfish, Aequorea victoria capable of binding in a bio-orthogonal manner to a fluorescent molecule. In order to introduce bio-orthogonal reactivity into YFP a synthetic structural homologue of the natural amino acid, pyrrolysine was included at a defined position in the amino acid sequence of the protein by means of amber stop codon suppression technique. The pyrrolysine derivative used for this purpose possesses a norbornene moiety, which readily undergoes cycloaddition reactions. Under bio-orthogonal reaction conditions, the norbornene reacts with tetrazines in an inverse electron demand Diels Alder cycloaddition and subsequent elimination of nitrogen. The velocity and reliability of this reaction under such mild conditions as are suitable for proteins have inspired the term “click chemistry” for this and similar reactions. Thus any compound with a tetrazine residue may be covalently bound to the modified protein. Here a fluorescent rhodamine dye was chosen as a binding partner, whose absorption maximum is located at a longer wavelength than that of YFP, a circumstance that potentially allows fluorescence resonance energy transfer (FRET) from the fluorophore within the YFP protein onto the rhodamine. Hence excitement of YFP should result in emission from the rhodamine dye, which would be an elegant demonstration of the project’s success.
The first step towards implementation of this plan was the chemical synthesis of the norbornene amino acid needed for protein expression. This was achieved by converting l-Lysine in three synthetic steps. A carboxytetramethylrhodamine “TAMRA” fluorophore with a tetrazine residue intended for reaction with the norbornene amino acid was also prepared, as well as a fluorescein dye with a norbornene residue intended as a model of norbornene-modified YFP allowing comparison of click reaction with the tetrazine fluorophore. Solid phase phosphoramidite DNA synthesis was employed for the preparation of primers designed to allow amplification of the YFP coding sequence by PCR, incorporating suitable restriction sites.
The second part of the course was biological work, beginning with the PCR of the YFP gene and its integration in an expression vector capable of being expressed in competent Escherichia coli cells. Upon genetic modification these bacteria were cultivated in a medium containing the norbornene amino acid. Expression of YFP was induced and the protein afterwards extracted and purified by affinity chromatography. All of these operations necessitated analytical control by gel electrophoresis and mass spectrometry. Finally the click reaction between YFP and TAMRA-tetrazine was performed and the product examined by fluorescence spectroscopy and mass spectrometry.
3 Procedures and Materials
3.1 Chemistry
3.1.1 Methods and Materials utilized
The following reactions were performed with commercial grade reagants and dry solvents without further purification all of which were purchased from Acros Organics or Sigma Aldrich. Technical grade solvents were distilled prior to use as eluents in column chromatography. Solvents were evaporated under reduced pressure using Heidolph Laborota 4001 rotary evaporators and Vacuubrand PC3001 Vario vacuum pumps. Reactions were stirred magnetically and monitored by crude NMR or thin layer chromatography using Merck TLC Silica gel 60 F254 aluminium backed plates. TLC plates were viewed under ultraviolet illlumination (λ = 254 nm and 366 nm) or stained with either cerium molybdate stain (prepared by adding to 235 mL of distilled water 12 g of ammonium molybdate, 0.5 g of ceric ammonium molybdate, and 15 mL of concentrated sulfuric acid) or ninhydrin stain (1.5 g ninhydrin was dissolved in 100 mL of n-butanol and 3.0 mL acetic acid added). Flash chomatography was carried out on Merck Geduran Si 60 silica gel (particle size 40-63 μm) using a forced flow at 1.3-1.5 bar pressure. Yields refer to spectroscopically pure products. NMR spectra were recorded on Varian XL 400 (400 MHz) or Bruker AMX 600 (600 MHz) spectrometers. Chemical shifts (δ) are reported in parts per million (ppm) relative to tetramethylsilane, using solvent as the internal standard Signal multiplicities are depicted as s (singlet), d (doublet), t (triplet), q (quartet), m (multiplet), br (broad) and their combinations. High resolution mass spectra were recorded on Thermo Finnigan LTQ FT (ESI) and Finnigan MAT 95 (EI) devices.
3.1.2 Syntheses
3.1.2.1 Preparation of Norbornene Amino Acid
Bicyclo[2.2.1]hept-5-en-2-ylmethyl carbonochloridate (1)
illustration not visible in this excerpt
Under nitrogen 5-norbornene-2-methanol (621 mg, 0.605 mL, 5.00 mmol, 1.0 eq.) was added dropwise at 0 °C to a solution of bis(trichloromethyl)carbonate “triphosgene” (742 mg, 2.50 mmol, 0.5 eq.) and activated charcoal (10.0 mg) in dry tetrahydrofuran (10 mL) over a period of 30 min. The reaction mixture was stirred overnight while temperature increased from 0 °C to 20 °C.
The solution was filtered through celite, washed with diethyl ether. All volatile components were evaporated from the filtrate under reduced pressure and the liquid residue dried for 1 h in high vacuum affording 1 as colourless oil, which was used in the subsequent reaction without further purification.
N [6] -((bicyclo[2.2.1]hept-5-en-2-ylmethoxy)carbonyl)- N [2] -(tert -butoxycarbonyl)- L -lysine (2)
illustration not visible in this excerpt
1 was taken up in dry tetrahydrofuran (1.5 mL) and slowly added to a solution of Boc- L -Lysine (1.23 g, 5.0 mmol, 1.0 eq.) in 1.0 M NaOH/THF 2:1 v / v (15 mL) at 0 °C. After addition, the reaction mixture was allowed to warm to 20 °C and stirred overnight. Ethyl acetate (40 mL) was added and the aqueous layer was acidified to pH below 4 by addition of conc. hydrochloric acid. After separation the aqueous layer was extracted with ethyl acetate (3 x 40 mL). Combined organic layers were washed with brine and dried over Na2SO4 before evaporating the solvent under reduced pressure. The crude product thus obtained was subjected to flash chromatography (DCM/MeOH/AcOH 96:3:1 v/v/v, co-evaporated with toluene) to yield a yellow solid.
R f (DCM/MeOH/AcOH 90:8:2 v/v/v) = 0.56
[1] H NMR (MeOH-d4): d = 0.51-0.57 (m, 1H), 1.11-1.19 (m, 1H), 1.20-1.61 (m, 10H), 1.65-1.92 (m, 6H), 2.70 (s, 1H), 2.78-2.85 (m, 1H), 2.86 (s, 1H), 3.10-3.23 (m, 3H), 3.58-3.69 (m, 1H), 3.80-3.99 (m, 1H), 4.07-4.19 (m, 1H), 4.24-4.34 (m, 1H), 5.23-5.30 (m, 1H), 5.46-5.52 (m, 1H), 5.93-5.94 (dd, 1 H [3] J(H,H) = 5.4 Hz, 2.6 Hz), 6.06-6.11 (m, 1H), 6.14 (dd, 1H, [3] J(H,H) = 4.9 Hz, 2.8 Hz).
[13] C NMR (MeOH-d4): d = 21.6, 28.2, 28.6, 28.8, 29.8, 37.8, 38.1, 40.0, 41.5, 42.1, 43.4, 43.7, 44.6, 53.6, 54.3, 69.0, 69.6, 132.3, 136.6, 137.6, 138.1, 159.1, 173.2.
N [6] -((bicyclo[2.2.1]hept-5-en-2-ylmethoxy)carbonyl)- L -lysine (3)
illustration not visible in this excerpt
Boc-deprotection of compound 2 was effected by dissolving 2 (1.0 eq.) in 60% formic acid in chloroform 6:4 v/v (20 mL), and stirring the mixture for 24 h at 20 °C. Dimethylformamide (10 mL) was added and all volatile components were removed under reduced pressure. The residue was taken up in 50 mM hydrochloric acid and evaporated. Treatment with dimethylformamide and hydrochloric acid in this manner was repeated four times in order to remove residual formic acid and to arrive at the hydrochloride salt of 3. The product thus obtained after high vacuum drying was a yellow, highly viscous oil. Yield over all three steps: 1.12 g (3.7 mmol, 76%).
R f (DCM/MeOH/AcOH 90:8:2 v/v/v) = 0.08
[1] H NMR (D2O): d = 0.45 (dd, 1H, [3] J(H,H) = 11.5, 4.2 Hz), 1.11-1.15 (m, 1H), 1.18-1.24 (m, 1H), 1.26-1.40 (m, 4H), 1.52-1.81 (m, 4H), 2.24-2.33 (m, 1H), 2.64 (s, 1H), 2.74-2.80 (m, 2H), 2.92 (q, 2H, [3] J(H,H) = 5.2 Hz), 3.40-3.58 (m, 3H), 3.62-3.69 (m, 1H), 3.79-3.86 (m, 1H), 3.95-4.02 (m, 1H), 5.86 (dd, 1H, [3] J(H,H) = 5.6, 2.8 Hz), 6.01-6.05 (m, 1H), 6.10 (dd, 1H, [3] J(H,H) = 5.6, 2.9 Hz).
[13] C NMR (D2O): d = 172.96, 158.91, 158.87, 137.92, 137.36, 136.42, 132.13, 69.36, 68.82, 54.32, 53.40, 48.80, 44.37, 43.47, 43.22, 41.92, 41.30, 39.81, 37.89, 37.54, 29.57, 28.62, 28.42, 28.00, 21.44.
HR MS (ESI) m/z calculated for C15H25N2O4 [M+H]+: 297.18088, measured: 297.18052, m/z calculated for C15H24N2O4Na [M+Na]+: 319.16338, measured: 319.16278, m/z calculated for C15H26N2O4Cl [M+HCl+H]+: 333.15811, measured: 333.15765.
3.1.2.2 Preparation of fluorescein with norbornene residue (for comparison of click reaction)
N -(3',6'-dihydroxy-3-oxo-3 H -spiro[isobenzofuran-1,9'-xanthen]-6-yl)bicyclo[2.2.1]hept-5-ene-2-carboxamide (4)
illustration not visible in this excerpt
In a 1.5 mL plastic reaction vessel dry N,N-diisopropylethylamine “Hünig base” (8.0 mL, 46 mmol, 2.0 eq.) and benzotriazol-1-yl-oxytripyrrolidinophosphonium hexafluorophosphate “PyBop” (15 mg, 29 mmol, 1.4 eq.) were added to a shaking solution of norbornene carboxylic acid (3 mg, 22 mmol, 1.0 eq.) in dimethylformamide (300 mL). After shaking the mixture for 24 h at 35 °C, aminofluorescein (5 mg, 15 mmol, 0.7 eq.) was added and stirring continued for 5 h at 35 °C. The solution was concentrated in high vacuum and the product was subjected to flash column chromatography (1 x 15 cm, DCM:MeOH v/v 10:1 200 mL, 5:1 250 mL, 1:1 200 mL) yielding 5 mg (10.7 mmol, 92 %)
R f (DCM/MeOH v/v 5:1) = 0.85
HR MS (ESI) m/z calculated for C28H22NO6 [M+H]+: 468.14471, measured: 468.14398.
3.1.2.3 Preparation of Tetrazine TAMRA
Tetrazine derivative of 5(6)-carboxytetramethylrhodamin (8)
illustration not visible in this excerpt
In a 1.5 mL plastic reaction vessel “Hünig base” (7.9 mL, 46.5 mmol, 4.0 eq.) and benzotriazol-1-yl-oxytripyrrolidinophosphonium hexafluorophosphate “PyBop” (7.2 mg, 13.9 mmol, 1.2 eq.) were added to a solution of 5(6)-carboxytetramethyl rhodamine (5 mg, 11.6 mmol, 1.0 eq.) in dry dimethyl formamide (300 mL). After 1 h of stirring at 25 °C Tetrazine amine TFA salt 7 was added (4.8 mg, 12.8 mmol, 1.1 eq.) and stirring was continued for 4 h. Crude product was purified by flash column chromatography (1 x 15 cm, DCM:MeOH 10:1 à 5:1 v/v)
DC ( DCM/MeOH 4:1 v/v) Rf = 0.12.
[1] H-NMR (DMSO- d6): δ (ppm) = 9.61 (t, J = 5.75 Hz, 0.5 H), 9.44 (t, J = 5.75 Hz, 0.4 H), 9.20 (t, J = 4.75 Hz, 2 H), 8.62–8.51 (m, 7 H), 8.33 (dd, 3J(H) = 8.00 Hz, 4J(H) = 1.40 Hz, 0.7 H), 8.28–8.24 (m, 0.7 H), 8.12 (t, J = 8.17 Hz, 0.7 H), 7.86–7.81 (m, 1.7 H), 7.74–7.69 (m, 2.2 H), 7.64–7.55 (m, 3 H), 7.39–7.16 (m, 4.2 H), 6.58–6.45 (m, 9 H), 5.82 (s, 0.8 H), 4.71 (d, J = 5.83 Hz, 1.1 H), 4.58 (d, J = 5.83 Hz, 1.1 H), 3.51 (s, 2.2 H), 3.35 (s, 34 H), 2.95 (d, J = 2.2 Hz 18 H)
(Preparation of compound 5 carried out by “orange” group!)
3.2 Biology
3.2.1 Methods and Materials utilized
Unless otherwise indicated below small volumes of liquids were measured using Eppendorf micropipettes and reactions were performed in Eppendorf polypropylene reaction vessels. PCR reactions were performed using Thermo Scientific Arktik Thermal Cycler. Freeze-drying of polar solvents was carried out on Christ alpha 2-4 LD plus lyophilizers. Purified water ddH2O (impedance value 18.2 MΩ) was obtained using a Milli-Q-Plus unit by Millipore. All Enzymes used were commercial preparations by New England Biolabs Inc. All solutions and buffers were water-based unless otherwise noted.
3.2.2 Primer Synthesis
Primer sequences for PCR amplification of the YFP gene were designed to accommodate the restriction sites of the plasmid pET28a(+) in order to allow transfer of the YFP coding sequence into the plasmid and to have similar annealing temperatures in order to facilitate PCR:
Forward primer: 5‘-TATCATCCATGGTGAGCAAGGGC-3‘
GC content = 52.0 %, annealing temperature = 56.11 °C, MW = 7731.1 g/mol
Reverse primer: 5‘-GGTGTGTCCGAGGATAGATCTCTTGTACAGCTCGTC-3‘
GC content = 55.4 %, annealing temperature = 55.32 °C, MW = 11098.3 g/mol
Synthesis was performed using the automatic DNA/RNA synthesizer ABI394 (Applied Biosynthesis) with nucleoside phosphoramidites as starting material. The details of the DNA synthesis cycle are explained in the introduction to this text (cf. chapter 1). Upon completion of primer synthesis the acetonitrile solvent was removed in high vacuum. Each step of DNA synthesis can easily be monitored by measurement of conductivity since the trityl protective group is released in cationic form, increasing conductivity proportional to reaction efficiency.
illustration not visible in this excerpt
Table 1: Trityl values.
The product was at this stage still bound to the solid phase support, which was placed in a 1.5 mL Eppendorf vessel and concentrated aqueous solutions of ammonia (400 mL) and methylamine (400 mL) were added, the mixture was heated to 65 °C for 10 min. The vessel was then cooled on ice and centrifuged at 16217 g for 5 min. The supernatant was filtered through a syringe filter and all solvents and liquid reagents were removed by SpeedVac (1.5 h) treatment and subsequent lyophilisation overnight. In order to precipitate DNA, ddH2O (100 mL), 3M aqueous solution of sodium acetate (60 mL) and ethanol (640 mL) were added and the mixture was cooled to –80 °C for 1 h before centrifugation for 1 h at –4 °C (16217 g) and removal of the supernatant. The precipitate was resuspended in ddH2O (100 mL) and dried by SpeedVac (1.5 h) treatment. The final product was suspended in ddH2O (25 mL) and concentrations were measured using Nanodrop photospectrometre with 1.5 mL for each measurement (results shown in table 2).
[...]