Applications of controlled DNA assembly and disassembly in biosensing and responsive drug delivery

Deoxyribonucleic acid (DNA) provides many exciting functions outside of simply encoding genetic information. Herein, we report two techniques to control the assembly and disassembly of nucleic acids. Within these techniques, DNA is used as a programmable material for biosensing and drug delivery. In chapter 1, we report an effort to create a quantitative Polymerase Chain Reaction (qPCR)-compatible split aptamer assay. Split aptamers are single-stranded DNA molecules that reversibly assemble in the presence of a target molecule, which promotes reactivity by increasing the effective reactive group concentrations during aptamer assembly. Currently, our lab has pioneered the development of this technique, referred to as Split Aptamer Proximity Ligation (StAPL) technology, for small molecule detection. Previous techniques are limited due to the incompatibility of ligated backbones with qPCR amplification techniques, thus rendering this approach to quantification impractical. The goal of our research is to rectify this limitation and expand the scope of split aptamer chemistry using a triazole mimic of the phosphodiester backbone reported by Brown and coworkers, which would allow for real-time detection and quantification of small molecule targets. In chapter 2, we describe the application of nucleic acids to drive the formation of micelles for use as a programmable material for small molecule detection and responsive drug delivery. Micelles are spherical structures composed of amphiphilic monomers, which arrange their hydrophobic and hydrophilic regions to minimize contact between their hydrophobic sections with hydrophilic solvents. As result, micelles contain a hydrophobic pocket that can be used for hydrophobic drug delivery. However, this application is limited by the fact that guest molecules contained within the micelle are in equilibrium with the surroundings, thus allowing them to diffuse out, and micelles also dissociate at low monomer concentrations. We propose that DNA-erosslinking will increase micelle stability, allowing for guest molecule transport at lower concentrations of micelles without drug diffusion. DNA-crosslinking also allows for micelle disassembly control in response to environmental stimuli. Our current research efforts focus on generating and characterizing DNA-crosslinked micelles (DCM) by creating monomers comprised of DNA and hydrophobic polymer for biosensing and drug delivery.

APPLICATIONS OF CONTROLLED DNA ASSEMBLY AND DISASSEMBLY IN BIOSENSING AND RESPONSIVE DRUG DELIVERY by Annika Pecchia-Bekkum A Senior Honors Thesis Submitted to the Faculty of The University of Utah In Partial Fulfillment of the Requirements for the Honors Degree in Bachelor of Science In Chemistry Approved: ------- Dr. Cynjiiia Burrows ChairrDepartment of Chemistry Dr. Thomas Richmond Department Honors Advisor Dr. Sylvia D. Torti Dean, Honors College August 2013 ABSTRACT Deoxyribonucleic acid (DNA) provides many exciting functions outside of simply encoding genetic information. Herein, we report two techniques to control the assembly and disassembly of nucleic acids. Within these techniques, DNA is used as a program­mable material for biosensing and drug delivery. In chapter 1, we report an effort to create a quantitative Polymerase Chain Reac­tion (qPCR)-compatible split aptamer assay. Split aptamers are single-stranded DNA molecules that reversibly assemble in the presence of a target molecule, which promotes reactivity by increasing the effective reactive group concentrations during aptamer as­sembly. Currently, our lab has pioneered the development of this technique, referred to as Split Aptamer Proximity Ligation (StAPL) technology, for small molecule detection. Previous techniques are limited due to the incompatibility of ligated backbones with qPCR amplification techniques, thus rendering this approach to quantification impracti­cal. The goal of our research is to rectify this limitation and expand the scope of split ap­tamer chemistry using a triazole mimic of the phosphodiester backbone reported by Brown and coworkers, which would allow for real-time detection and quantification of small molecule targets. In chapter 2, we describe the application of nucleic acids to drive the formation of micelles for use as a programmable material for small molecule detection and responsive drug delivery. Micelles are spherical structures composed of amphiphilic monomers, which arrange their hydrophobic and hydrophilic regions to minimize contact between their hydrophobic sections with hydrophilic solvents. As result, micelles contain a hydro-phobic pocket that can be used for hydrophobic drug delivery. However, this application is limited by the fact that guest molecules contained within the micelle are in equilibrium with the surroundings, thus allowing them to diffuse out, and micelles also dissociate at low monomer concentrations. We propose that DNA-erosslinking will increase micelle stability, allowing for guest molecule transport at lower concentrations of micelles with­out drug diffusion. DNA-crosslinking also allows for micelle disassembly control in re­sponse to environmental stimuli. Our current research efforts focus on generating and characterizing DNA-crosslinked micelles (DCM) by creating monomers comprised of DNA and hydrophobic polymer for biosensing and drug delivery. TABLE OF CONTENTS ABSTRACT ii CHAPTER 1 1 INTRODUCTION 1 METHODS 4 RESULTS AND DISCUSSION 8 REFERENCES 10 CHAPTER 2 12 INTRODUCTION 12 METHODS 14 RESULTS AND DISCUSSION 18 REFERENCES 22 CONCLUSION 23 1\/ CHAPTER 1: qPCR-Compatible Small-Molecule-Dependent Split Aptamer Assay INTRODUCTION Aptamers are nucleic acid sequences capable of selectively binding a small mole­cule and have been found useful in a variety of applications. Our laboratory has pio­neered one such application, referred to as Split Aptamer Proximity Ligation (StAPL) technology, in the detection of small molecules, combining it with colorimetric tech­niques and polyacrylamide electrophoresis.1,2 Split aptamers are aptamers that have been split into fragments that retain specificity for their target molecules.2 In this technique, the split aptamer strands have been functionalized with reactive groups that, in the pres­ence of the target molecule, are brought into close proximity due to split aptamer assem­bly (Figure 1). This proximity increases the reactive group's effective molarities, drasti- Figure 1: In the presence of the small molecule target, the split aptamer fragments will anneal together, bringing reactive groups X and Y into close proximity and promoting ligation. cally increasing the reaction rate over that of an untemplated reaction. The advantage of using split aptamers is that they are capable of "sandwiching" specific small molecules, unlike aptamers which require a target with larger surface area.2 However, current StAPL techniques do not offer real-time detection and quantification of small molecule concen­tration. The technique herein reported is an attempt to rectify this limitation by creating a split aptamer analogue to Proximity Ligation Assays (PLA) that is compatible with quan­titative Polymerase Chain Reaction (qPCR). PLA utilize primary antibodies specific for an antigen, which are then bound by •2 secondary antibodies (PLA probes). The PLA probes are conjugated with short oligonu­cleotide sequences that, upon coming into close proximity to one another, can be joined via enzymatic ligation, and amplified via rolling circle amplification. After amplification, labeled oligonucleotide probes are added to the solution, allowing for quantification through fluorescence intensity.3 However, antibodies are expensive, large in size, and dif­ficult to produce as they require animals when selecting for a target and thus have sto­chastic results.4 Furthermore, due to the use of animals to make antibodies, it is not pos­sible to generate antibodies for toxic molecules. Split aptamers have a greater specificity and affinity for a larger scope of possible products, and are cheaper and easier to pro­duce. 4 Therefore, our research focuses on utilizing the advantages of split aptamers to create an analogue to this technique that is additionally compatible with qPCR to obtain real-time quantitative data on small molecule concentrations in solution. qPCR uses polymerases to amplify the DNA sample, which is then heated to de­nature the DNA strands.5 The sample is then once again amplified, then heated, and the cycle repeated several times, resulting in exponential amplification.5 To quantify the re­action, SYBR Green, a fluorescent dye that intercalates double-stranded DNA, is added to the solution, thereby resulting in a fluorescent signal that increases with each PCR cy­cle. 5,6 The fluorescent signal is proportional to the sample concentration during the expo­nential stages, allowing for sample concentration determination after the fluorescence signal plateaus or through an external or internal standard.5 3 Our research utilizes the biocompatible triazole linkage reported by Brown and coworkers, which is an artificial mimic for native phosphodiester bonds found in DNA (Figure 2A).7 Brown reports that the triazole linkage is compatible with DNA polymeras- 2A Base # + N3 Base Biocompatible trizoie linkage Native D N A 2B H O NH -O-O H TB DM S C I, Pyridine, 0 °C, 48 h 53% NC Figure 2: A) Structural similarity of native DNA and triazole linkage. N2 and N3 are hydrogen bond accepters and thus substitute for phosphodiester oxygens in native DNA. B) Reaction scheme for preparation of 3'-propargyl thymidine-5'-0-(2-cyanoethyl-Ar,Ar-diisopropyl)phosphoramidite for DNA synthesis es and the native machinery of E. coli bacteria.7 The N2 and N3 atoms of the triazole linkage function as hydrogen bond accepters, allowing this linkage to substitute for the phosphodiester oxygens (Figure 2A).7 As such, using a triazole mimic promotes compat­ibility with qPCR amplification techniques, which can be further combined with the spec- ificity of split aptamers for their targets, allowing for specific and real-time determination of small molecule concentration at low cost. Therefore, our current research goals are to generate split aptamer strands with the 3'-alkyne necessary for triazole formation (Figure 2B). This will be accomplished by synthesizing a phosphoramidite containing the 3'- alkyne, which will be given to the HSC Core Facility at the University of Utah for re­verse DNA synthesis. The 5'-azide split aptamer fragment will be synthesized by the HSC Core Facility. For proof of concept, we propose using the cocaine split aptamer se­quence since it is a well-characterized split aptamer.8'9 The split aptamer fragments will anneal in the presence of cocaine, resulting in the formation of the triazole linkage, which can then be amplified by DNA polymerases. This reaction can then be quantified using SYBR Green, allowing for real-time detection and quantification of small molecule tar­gets. METHODS All reagents were purchased from Sigma Aldrich, Chem-Impex International, Fisher Scientific, Mallinckrodt Chemicals, Cambridge Isotope Laboratories, Inc. and Alfa Aesar. All chemicals were utilized without further purification, with the exception of tet-rahydrafuran (THF), which was distilled over calcium hydride and stored with grade 564 molecular sieves (3 A) (8-12 mesh). All water-sensitive reactions were conducted under nitrogen and the glassware was oven-dried overnight. Column chromatography was per­formed under nitrogen using EMD Millipore Silica Gel 60 with particle size 0.040 - 0.063 mm. Thin layer chromatography was carried out using EMD Silica Gel 60 F264 plates. 'H NMR was conducted in deuterated chloroform or methanol using a Unity-300 spectrometer. Synthesis of 5'-0-(tertbutyldimethyl)-2'-deoxy-thymidine (1):10,11 2'deoxythymidine (2.00 g, 8.26 mmol) and ter/-butyl dimethylsilychloride (1.36 g, 9.07 mmol, 1.1 eq.) were added to 50 mL round bottom flask and stirred at 0 °C in pyridine (10.0 mL) for 48 h. The precipitate was filtered, dissolved in DCM and washed with wa­ter (3 x 20 mL). The reaction mixture was then washed with saturated sodium bicar­bonate and dried using sodium sulfate. The solvent was removed in vacuo and the result­ing white precipitate was dissolved in hexanes (20 mL) and refluxed at 60 °C for 1 h. The white precipitate was then filtered and dried under vacuum. The title compound was iso­lated as a white powder (1.56 g, 4.38 mmol, 53% yield) of compound. !H NMR (300 MHz, CDCI3) 5 8.03 (s, IH), 7.38 (s, IH), 6.27 (m, 1 H), 4.36 (s, 1 H), 3.93 (m, IH), 3.75 (m, 2H), 2.25 (m, IH), 2.02 (m, IH), 1.96 (s, 3H), 0.82 (s, 9H), 0.01 (s, 3H) (Figure 3). S olvent: COC13 ««lu. daisy 1.000 Pulao *6.7 d «a r « «B Acq. elaut 4.000 a « Width *499.9 K* 32 ra p a t ltlo n s OBSERVE H I , 300.077 D*TA PMOCESetMQ FT o ila H5536 T o ta l t im 3 » l n . ir 1 / vrnmr ay a /data 1II-CDC13 L L OH O NH nA o ppm 6 .0 0 Figure 3: iII NMR of 5'-0-(tertbutyldimethyl)-2'-deoxy-thymidine (1) in CDC13. Synthesis of 5'-0-(tertbutyldimethyl)-3'-propargyl-thymidine (2):12 5'-0- (tertbutyldimethyl)-2'-deoxy-thymidine (0.792 g, 2.22 mmol) was coevaporated with dry THF (3x10 mL), and then dried under vacuum overnight. Dried THF (20 mL) and sodi-flagplq directory i SaQuofiCo: a2p tolvmat: CDC13 ftal**. dalay 1.000 aac Pulse *6.7 ‘Icjrcoa Ae<j. tine 4.090 sac Width «<»?.» B« 12 OBSXXVT Bl. S00.07713M M o NH N ^O S I. r r r J / / /. // . .......1 K ......._I. «. iJL 4. .X r 0 ppm «.ot Figure 4: 'H NMR of 5'-0-(tertbutyldimethyl)-3'-propargyl-thymidine (2) in CDC13. um hydride (60% in oil, 0.249 g, 6.23 mmol, 2.8 eq.) were added and the reaction was stirred vigorously at 0 °C. Propargyl bromide (0.26 mL, 2.92 mmol, 1.3 eq.) was added to the solution after 1 h, and the reaction mixture was stirred for an additional 6 h. The reac­tion mixture was quenched with water, and product was extracted with EtOAc (3 x 20 mL), then washed with brine. The organic layer was isolated, dried with sodium sulfate, filtered, and concentrated in vacuo. The yellow oil was purified using silica gel column chromatography (5% MeOH:DCM with 0.5% pyridine) giving a light yellow oil (0.660 g, 1.67 mmol, 75% yield). *H NMR (300 MHz, CDCI3) 5 8.25 (s,lH), 7.43 (m, 1H), 6.31 (m, 1 H), 4.75 (t, 3 H), 4.39 (m, IH), 3.97 (m, IH), 3.79 (m, 2H), 2.25 (m, IH), 1.89 (s, 3H), 0.83 (s, 9H), 0.04 (s, 3H) (Figure 4). Synthesis of 3'-propargy 1-thymidine (3):13 S'-O-^ertbutyldimethyO-S'-propargyl-thymidine (0.661 g, 1.85 mmol) and ammonium fluoride (1.19 g, 32.1 mmol, 17 eq) were dissolved in methanol (15 mL) and refluxed un­der nitrogen overnight. The solvent was removed in vacuo and the resulting yellow oil was then re-dissolved in ethanol (100 mL). The solution was then washed twice with brine (100 mL), dried using magnesium sulfate, and then concentrated. The product was then purified through silica gel flash chromatography (5% MeOH:DCM with 0.5% pyri­dine). An off-white crystal was obtained (0.195 g, 0.696 mmol, 38% yield). *H NMR (300 MHz, CD3OD) 5 7.81 (s, IH), 6.35 (t, IH), 4.58 (m, 1 H), 4.23 (s, 2 H), 3.76 (t, 2H), 2.90 (s, IH), 2.41 (m, IH), 2.20 (m, IH), 1.87 (s, 3H) (Figure 5). UafOUh OjUtySOO E Rrebive d ire c to ry . f lU H /i 8 o |)U d ire c to ry ; n i a : PHOTOS Puls* Sequence: slpol Solvent! CDJOD Mblent c a o n t u n □ tcm -300 *unitr300r.«r* ■alax. dalay 1.890 sac Pulse 46.7 dagewfl Jtoj. tin 4.000 mc Width 4499.9 Hz 10 repetition* OBSERVE Bl, 3 0 0 .0 j» ja il M ur 1 / vnaray a / da ta rf / f jLjlU J NH N-^O HCL V > i Figure 5: NMR of 3'-propargyl-thymidine (3) in CD3OD. Synthesis of 3'-propargy 1 thymidine-5'-0-(2-cyanoethy1-TV, A-diisopropyl) phosphoramidite (4):7 3'-propargy 1-thymidine (0.0927 g, 0.332 mmol) was added to dried THF (10.0 mL) under nitrogen. DIPEA (0.15 mL, 0.86 mmol, 2.6 eq.) and 2-cyanoethyl-/V,AL diisopropylchlorophosphoramidite (0.10 mL, 4.48 mmol, 14 eq.) were added dropwise and the reaction was allowed to stir for 2 h at room temperature. The volume of the solu­tion was reduced to 3 mL by flushing with nitrogen, then dichloromethane (10 mL) was added and the mixture was washed with degassed aqueous potassium chloride solution (10 mL). The solution was concentrated in vacuo. Product purification was attempted us­ing silica gel column chromatographic (40% Hex:EtOAC), but product possibly decom­posed on column. RESULTS AND DISCUSSION Our current efforts have focused on the synthesis of the 3'-propargy 1 thymidine phosphoramidite for use in reverse phase DNA synthesis, which will be conducted by the HSC Core Research facility. 5'-0-(tertbutyldimethyl)-2'-deoxy-thymidine (1) was syn­thesized using reported procedures.10,11 This compound initially provided a purification problem due to unreacted 2'deoxy-thymidine remaining in solution, which reacted with propargy 1 bromide in the following step to form a mixture of possible products. This problem was solved by utilizing the procedure reported by Beigelman and coworkers and extending the reaction time to 48 h, giving the product in a 53% yield.11 The second synthesis step was conducting using a procedure reported by Beigel­man and coworkers, forming 5'-0-(tertbutyldimethyl)-3'-propargyl-thymidine (2) in a 75% yield. Initially, before the upstream purification problems of the first step were re­ solved, this contained a mixture of possible products, such as TBDMS protected and un­protected compounds, as well as propargylated and unpropargylated products possibly at the 3' and 5' ends as well. Once the purification problem in the first step was resolved, the pure propargylated compound was isolated as a yellow oil. Initial attempts to deprotect the thymidine group involved a TBAF reaction for 1 h.12 However, this reaction did not yield identifiable results by NMR, and the conditions used may have resulted in the decomposition of the thymidine reactant. However, the deprotected 3'-propargyl thymidine has been successfully synthesized using ammonium fluoride, obtaining the product in a 38% yield and confirmed by NMR.13 The final step in the synthesis of the phosphoramidite was attempted and yielded promising results by TLC. Using the procedure reported by Brown and coworkers, the reaction was conducted with oven-dried glassware, dried THF, and molecular sieves for 2 t n h under mtrogen. By TLC, a possible product spot was observed, but the crude material contained large amounts of unreacted material. However, attempts to purify the com­pound through silica gel column chromatography under nitrogen were unsuccessful as the possible product decomposed before analysis by NMR could be attempted. Future at­tempts with this synthesis will involve more washes to remove the N,N-diisopropyl chlo-rophosphoramidite, the use of a glove bag to provide a more carefully controlled inert atmosphere, and more rigorous attempts to dry the solvents for the reaction and subse­quent purification. In conclusion, steps 1-3 of the synthesis of the phosphoramidite have been ac­complished and confirmed by NMR. Once the phosphoramidite has been synthesized, it will be incorporated into the 3' end of the split aptamer sequence for cocaine by solid phase DNA synthesis. The HSC Core facility will also synthesize the 5'-azide strand nec­essary for further experiments. Upon synthesis of both cocaine aptamer strands, the speci­ficity and sensitivity of the assay will be tested and reaction conditions, such as split ap­tamer : catalyst ratios and reaction times, will be optimized. Various sample concentra­tions will be tested with the qPCR assay and used to determine the limit of detection and quantifiable limit. This assay functions as a proof of concept as conceivably almost any split aptamer can be functionalized with the necessary 5'-alkyne and 3'-azide necessary for compatibility with this assay, and thus could greatly expand the scope of detectable targets. REFERENCES 1. Sharma, A.; Heemstra, J. J. Am. Chem. Soc. 2011,133 (32), 12426-12429. 2. Sharma, A. K.; Kent, D. A.; Heemstra, J. M. Anal. Chem. 2012, 84, 6104-6109. 3. Gullberg, M. et al. PNAS, 2004, 22, 8420-8424. 4. Keefe, A.; Pai, S.; Ellington, A. Nat. Rev. Drug Discovery 2010, 9, 537-550. 5. Sigma Aldrich. (2008). qPCR Technical Guide. Retrieved from: <http -JIwww. sigmaaldrich. com/etc/medialib/docs/Sigma/ Generallnformation/qpcrt echnical_guide.Par.0001.File.tmp/qpcr_technical_guide.pdf> 6. Sigma Aldrich (n.d.). SYBR Green based qPCR. Retrieved from: <http://www.sigmaaldrich.com/life-science/molecular-biology/pcr/quantitative-pcr/ sybr-green-based-qpcr.html> 7. El-Sagheer et al. PNAS 2011,108 (28): 11338-11343. 8. Stojanovic, M. N.; de Prada, P.; Landrey, D. W. J. Am. Chem. Soc. 2000,122, 11547- 11548. 10 9. Zhang, J. et al; Small 2008, 4, 1196-1200. 10. Wuts, P. G. M.; Greene, T. W. (2006) Greene's Protective Groups in Organic Syn­thesis, fourth edition. New York City: John Wiley & Sons, Inc. 11. Serebryany, V.; Karpeisky, A.; Matulic-Adamic, J.; Beigelman,L. Synthesis 2002,12, 1652-1654. 12. Lolk, L et al. J. Med. Chem. 2008, 51, 4857-4967. 13. James, D. et al. Tet. Lett. 2010, 51, 1230-1232. CHAPTER 2 DNA-Crosslinked Micelles for Biosensing and Responsive Drug Delivery INTRODUCTION Micelles are three-dimensional spherical structures formed from the assembly of amphiphilic (containing both hydrophobic and hydrophilic regions) monomers.1'2 The driving force behind the formation of micelles is the minimization of contact between the hydrophobic region and hydrophilic solvent.1,2 In hydrophilic solvents, the hydrophobic region is contained within the micelle's interior, forming a pocket that could carry hydro-phobic guest molecules, and therefore micelles could be used in hydrophobic drug deliv-ery. 1,3,4 Micelles as hydrophobic drug delivery systems pose two problems. Below the critical micelle concentration (CMC), micelles break apart, dissociating into their mono­mers. 2,3,5 Additionally, micelles can house hydrophobic guest molecules, but this is lim­ited as a delivery device due to the dynamic interchange of guest molecules with the sur­roundings. 2,3,5 The guest molecules do not remain housed in the interior of the micelle, but diffuse into solution. However, using DNA base-pairing to form crosslinks can rectify these limitations by serving two important purposes in rendering micelle biosensing and drug delivery feasible. By combining the specificity and additional stability donated by DNA molecules in combination with hydrophobic polymers, we aspire to create mono­mers that are stable at lower concentrations and that prevent the escape of guest mole­cules. These DNA-crosslinked micelles (DCM) will be formed from monomers contain­ing complementary DNA strands, which anneal together to form the crosslinks, attached 12 13 to a conjugated polymer, which would function as the hydrophobic region of the micellar structure. Figure 6: Stimuli for micelle dissociation: A) aptamers, B) complementary nucleic acid strands, C) nu­cleases to cause micelle disassembly and release of hydrophobic guest molecule. DCM can be disrupted in a number of ways (Figure 6). Using an aptamer as one of the strands, the complementary DNA sequences will be disrupted by the addition of the target molecule (Figure 6A). Other methods include adding complementary DNA se­quences (Figure 6B) and nucleases. Upon disruption of the crosslinks, the CMC increas­es, leading to dissociation and release of guest molecules. Our current research goals involve the characterization and purification of mono­mers formed by reacting thiol-functionalized trebler DNA sequences with maleimide-terminated Poly(N-isopropylacrylamide) (PNIPAAm), a polymer (Figure 7). The trebler modifier enables synthesis of three joined DNA strands in parallel. Three strands are used 6A 6B 6C Figure 7: Maleimide-conjugated PNIPAAm:thiol DNA reaction to generate amphiphilic monomers to see if this promotes interactions between multiple monomers, as two DNA strands would likely form dimers. PNIPAAm is a temperature-responsive polymer that transi­tions from a hydrophilic state to a hydrophobic state about its lower critical solubility temperature (LCST) of 32 °C.6 Therefore, above temperatures of its LCST, the PNIPAAm should form the hydrophobic core of the micelles once conjugated to the tre-bler DNA. This should increase the stability of the micelles at concentrations, lowering the CMC below the non-crosslinked value. Once the methods for DCM generation are developed, different DNA strands can be used to generate bioresponsive assays. METHODS All reagents were purchased from Sigma Aldrich, Fisher Scientific, and the American Polymer Standards Corporation. The DNA sequences were synthesized by the HSC Cores Research facility at the University of Utah. All chemicals were utilized with­out further purification. High-Performance Liquid Chromatography (HPLC) was con­ducted using a 1260 Infinity HPLC system from Agilent Technologies and Matrix- Assisted Laser Desorption/Ionization (MALDI) was conducted using a Malvern Nano- Zetasizer Nano ZS. Dynamic Light Scattering (DLS) was conducted using a Waters Mi­cromass MALDI micro MX. 15 DNA Sequences: The following four sequences were used in the following experiments. DNA 4 is the complementary sequence for DNA 1, DNA 4 is the complementary strand to DNA 1, and DNA 5 is a sequence with mismatches (underlined). DNA Sequence FAm 5 7FAm/TTG TCT GCT TTT TTT TTT/S-S/3 ' 1 5'-/Cy3/CGT GCA GAT CGT CAT/trebler/sp9/S-S/-3' 4 5'-/Cy3/GCA CGT CTA GCA GTA/trebler/sp9/S-S/-3' 5 5'-/Cy3/GCT CGC CTT GCT GTA/trebler/sp9/S-S/-3' General Cleavage procedure: Disulfide protecting groups were cleaved by reaction with dithiothreitol (DTT). The DNA sequences were dissolved in H2O (1.00 mL), then a 500 jiL aliquot was incubated at room temperature with -10 mg of DTT for 2h. The resulting solution was purified with a Nap-5 column and lyophilized. General reaction conditions: Each cleaved DNA sequence (2-5 nmol) was mixed with maleimide-functionalized PNIPAAm in 1:10, 1:5, 1:2, 1:1, and 2:1 ratios of DNA:PNIPAAm in 0.1 M TEA at pH 7.0, 7.5, and 8.0. Each solution was then shaken at room temperature overnight, then ana­lyzed via HPLC with SEC and C l8 columns. The SEC columns used were SEC-3 col­umns purchased from Agilent Technologies with 3 pm particle sizes and 7.8 x 300 mm dimensions. A 100 A pore size guard column, 100 A pore size column, and 300 A pore size column was connected in series and heated to 25 °C. The method utilized for SEC columns was as follows: £ (L/mol*cm) 169800 147600 153500 135300 16 Time (min) Parameter 0 0.5 mL/min flow 0-120 66% methanol in H2O The C18 column used was a Zorbax Eclipse XDB-C18 analytical column purchased from Agilent Technologies with a 5 pm pore size and 4.6 x 200 mm dimensions. The op­timized method utilized for the C l8 column, which was also set to 25 °C, was as follows: Time (min) Parameter 0 2 mL/min flow 0 85% 0.1 TEAA in H2O, 15% Acetonitrile 2 80% 0.1 TEAA in H2O, 20% Acetonitrile 22 70% 0.1 TEAA in H2O, 30% Acetonitrile 30 0% 0.1 TEAA in H20 , 100% Acetonitrile 32 100% 0.1 TEAA in H2O, 0% Acetonitrile Reactivity of maleimide-functionalized PNIPAAm with sulfur groups using fluorescein (FAM) sequence: FAM DNA (5 nmol) was reacted with 34.2 mM maleimide-functionalized PNIPAAm (1.46 pL, 50 nmol, 5 eq.) and diluted to 10 pL total volume with methanol and allowed to sit at room temperature overnight. The resulting solution was then run on HPLC and the results compared to unreacted FAM DNA (5nmol) (10 pL total volume in MeOH). 5-(iodoacetamido)-fluorescein reaction to analyze trebler DNA reactivity: DNA 5 (3.18 nmol) was reacted with 0.13 pM 5-(iodoacetamido)-fluorescein (2.03 pL and 4.07 pL, 15.9 mmol and 31.8 mmol, 5 eq. and 10 eq. respectively) in 0.2 M phos­phate buffer at pH 7 and 0.1 M TEA at pH 8.0 and 8.4. The solution was brought to 10 pL and allowed to shake overnight. The samples were then analyzed via the C l8 HPLC procedure. To confirm a reaction occurred, DNA 5 and 5-(iodoacetamido)-fluorescein were mixed and immediately injected on the HPLC column for analysis. Purification of 5:1 maleimide-functionalized PNIPAAm:DNA 5: Using a solution of DNA 5 prepared in a 1:5 ratio with maleimide-functionalized PNIPAAm as reported above, the reaction solution was then analyzed via HPLC using a C l8 column and possible product-containing fractions were collected. A peak was ob­served at 23.5 min, as it absorbed at 260 nm (A.max of DNA), 310 nm (Xmax of PNIPAAm), and 550 nm (Xmax of Cy3), possibly indicating product formation. However, due to ioniza­tion difficulties, it was not possible to confirm a reaction occurred through MALDI. DLS analysis of possible PNIPAAm-trebler DNA monomers: Using the procedure as described above, 1:5 DNA: PNIPAAm reactions were conducted with DNA 1 (1 (iL, 3.42 nmol), 4 (1 }xL, 2.74 nmol) and 5 (1 |iL, 3.18 nmol). These three samples were analyzed via dynamic light scattering at 25 °C, 32 °C, and 37 °C. 18 RESULTS AND DISCUSSION Reactivity of maleimide-functionalized PNIPAAm with sulfur groups using fluorescein (FAM) sequence: 8 A MWD1 A, Stg*494,S0 (ASH-32-8 0> mAU 3 0 - 25 20* 1 5 - 1 0 - 5 - L 0 - T----------- ----■---1------- ----■---■---1---1--- ------- ----r 5 10 IS 20 8 B MWDl A. &e*4W,S0 Rot-o!f (ASH-32-8 O) Figure 8: HPLC traces for A) Unreacted FAM DNA. B) 5:1 PNIPAAm: FAM DNA reaction monitor­ing at 494 nm for fluorescein. To confirm the reactivity of maleimide-functionalized PNIPAAm with thiol groups of DNA, the reaction was carried out using fluorescein labeled DNA as this can be easily detected by analyzing the HPLC eluant 494 nm. As demonstrated above (Figure 8 A), the unreacted FAM DNA contains a single, distinct peak at 7.1 min. Upon addition of PNIPAAm, the 494 signal contains additional peaks, suggestive of product formation (Figure 8B). 19 5-(iodoacetamido)-fluorescein reaction to analyze trebler DNA reactivity: 9 A «wp< e. H i h j u iw u iD w u m a .»_________ M________ «________ a S 10 1S » Figure 9: HPLC traces for A) DNA 5:5-(iodoacetamido)fluorescein reaction immediately upon mixing, B) after overnight reaction, monitoring at 494 nm for fluorescein and at 550 nm for Cy3 DNA modification. Due to difficulties verifying reaction occurring between the trebler DNA and the malei-mide- functionalized PNIPAAm, we used 5-(iodoacetamido)fluorescein, which absorbs at 494 nm and can be detected in HPLC traces, to ensure that the trebler DNA is reactive with maleimide-functionalized groups. Contrasting the results of the overnight reaction with that of those immediately upon mixing of 5-(iodoacetamido)fluorescein with DNA 5 does not clearly indicate reaction as no significant differences exist between the HPLC traces and are thus the results are inconclusive (Figure 9A and 9B). Purification of 5:1 maleimide-functionalized PNIPAAm:trebler DNA 5 Reaction: The HPLC procedure was modified to maximize the differences between the peaks. The peaks visible between 3 and 12 minutes in the 550 nm channel (the wavelength for Cy3) do not have a significant 310 nm signal (Figure 10). A slight signal is present at 310 nm, which was collected through HPLC purification. Attempts to analyze the possible product peak were unsuccessful through MALDI as it was not possible to get the sample to ion­ize. Further analysis by dynamic light scattering (DLS) was used to examine the size of particles present. MWD1 B. Sig=310,5D Ref=off (ASH-4B-1D) Figure 10: HPLC purification of PNIPAAm: trebler DNA reaction monitoring at 310 nm for PNIPAAm and 550 nm for Cy3 Modification 21 DLS analysis of possible PNIPAAm-trebler DNA monomers: The PNIPAAm- trebler 0 Figure 11: Progression of micelle formation upon increasing tem- Sltion temperature (32 C) perature. At 25 °C, monomers present (red and green). At 32 °C, structures in the size range of micelles present (blue). At 37°C, free of PNIPAAm, the particles PNIPAAm present forming larger structures suggestive of agglom­eration (black). present are monomers and free in solution. At 32 °C, there are structures present at the size range indicative of mi­celle formation.1 At 37 °C, a temperature higher than that of the PNIPAAm's critical transition temperature at which PNIPAAm becomes hydrophobic, even larger species are observed, indicative of the formation an agglomeration.6 While these results do suggest the formation of micelles from the amphiphilic monomers, this data is not conclusive. Further analysis is required through the use of a PNIPAAm control at the same condi­tions. The HPLC analysis of the reaction of the FAM DNA sequence with the PNIPAAm polymer and the trebler DNA sequences with fluorescein suggest the reactivi­ty of each of these species. However, it was not possible to visualize any PNIPAAm-trebler DNA conjugates through these methods. DLS was used to analyze the size distri­bution of the maleimide-functionalized PNIPAAm-trebler DNA reaction mixtures. The data at 32 °C is suggestive of micelle formation, while, upon increase in temperature to 37°C, larger agglomerates are formed. The current challenge posed by this technique is the difficulty in purifying and characterizing the reaction mixtures, which resulted in peak overlap in SEC column and C l8 data. Ongoing efforts focus on verifying the for­mation and practicality of the PNIPAAm-DNA reactions researching other means of pu­rification, such as thiol-modified beads to remove excess PNIPAAm. Once the micelle structure has been synthesized and characterized, future attempts will focus on exploring DCM guest binding and dissociation, as well as developing toxin-specific aptamers. By utilizing aptamers, the micelles can respond to their environments and release anti-toxins in the event of toxin exposure. Thereby, DNA-crosslinked mi­celles can not only increase the stability of these structures beyond that of the non­crosslinked CMC, but can also be utilized to selectively respond to specific stimuli in their environments. REFERENCES 1. Jeong, J. H.; Park, T. G. Bioconjugate Chem. 2001,12, 917. 2. Hait, S. K.; Moulik, S. P. J. Surfactants Deterg., 2001, 4, 303-304. 3. Bronich, T. K.; Keifer, P. A.; Shlyakhtenko, L.S.; Kabanov, A. V. J. Am. Chem. 2005,127, 8236-8237. 4. Allen, R ; Bandyopadhyay, S.; Klein, M. Langmuir 2000,16, 10547-10552. (CMC) 5. Yuting, L. et al. Macromolecules 2006, 39, 2726-2728. 6. Panambur, G.; Koltover, I.; Batchellar, S. (2013). Designing temperature and pH sen­sitive NIP AM based polymers. Retrieved from: <http://www.sigmaaldrich.com/materials-science/polymer-science/nipam-polymers. html> 22 CONCLUSION In conclusion, we present two projects that aim to control the assembly and disas­sembly of DNA for use as a small molecule assay and bioresponsive drug delivery sys­tem. In chapter 1, we describe an effort to combine Split Aptamer Proximity Ligation (StAPL) technology with quantitative Polymerase Chain Reaction (qPCR). Ongoing ef­forts using a synthetic phosphodiester backbone mimic will enable the combination of StAPL technology with qPCR for real-time detection and quantification of small mole­cule targets at low cost. Currently, the synthesis has progressed to the formation of the 3'- propargyl thymidine species, but efforts are underway to synthesize the final thymidine phosphoramidite for use in solid phase DNA synthesis of the cocaine split aptamer. In chapter 2, we report efforts to create DNA-erosslinked micelles (DCM) for use as a bioresponsive hydrophobic drug delivery system. DNA-crosslinking should stabilize the mieellar structure, as well as prevent hydrophobic guest molecules from diffusing out from the interior of the micelle. By lowering the critical micelle concentration (CMC) below that of the non-crosslinked species, the DCM will be stable at lower concentra­tions. Additionally, the DNA-crosslinking allows for the control of micelle disassociation through a variety of methods, including functionalization with split aptamers, addition of a complementary oligonucleotide sequence, or the use of nucleases, thereby promoting micelle decomposition into monomers. Ongoing efforts aim to synthesize, characterize, and verify the practicality of using the trebler DNA and maleimide-functionalized PNIPAAm polymer in the formation of ampiphilic monomers and mieellar structure. Once ampiphilic monomers and micelles have been generated, work will continue to im- prove the responsivity of this system to specific stimuli in the environment, thereby ena­bling its use as a biosensor and responsive drug delivery platform.