Fundamentals and applications of solid state and biological nanopores

Both biological and solid state nanopores have been used to study fundamentals of ion transport phenomena and potential applications in single molecule analysis. This dissertation presents investigations of the effect of the electric double layer in the activation energy of ion transport in conical shaped glass nanopores. Further, it presents the use of the a-hemolysin nanopore to detect the DNA cancer biomarker benzo[a]pyrene, and to detect structural differences between the A- and B- form duplexes. Chapter 1 overviews the solid state and biological nanopores, common transport phenomena observed in sold state nanopores, and detection of ssDNA, dsDNA, and biomarkers using biological nanopores. Chapter 2 describes the effect of the electric double layer on the activation energy of ion transport through conical shaped glass nanopores. The study shows that the activation energy values for transport within an electrically charged conical glass nanopore differ from the bulk values due to the voltage and temperature-dependent distribution of the ions within the double layer. Finite element simulations based on the Poisson-Nernst- Planck model semiquantitatively predict the measured temperature-dependent conductivity and dependence of activation energy (Ea) on applied voltage. The results highlight the relationships between the distribution of ions with the nanopore, ionic current, and E a, and their dependence on pore size, temperature, ion concentration, and applied voltage. Chapter 3 describes how the a-hemolysin (aHL) nanopore platform can be used to detect the benzo[a]pyrene diol epoxide (BPDE) adduct to guanine (G) in synthetic oligo deoxynucleotides. BPDE adducts are formed by exposure to the carcinogenic precursor benzo[a]pyrene (BP), a polycyclic aromatic hydrocarbon, and considered as a biomarker that can initiate cancers. Translocation of a 41-mer poly-2'-deoxycytidine strand with a centrally located BPDE adduct to G through aHL in 1 M KCl produces a unique multilevel current signature allowing the adduct to be detected from either the 5' or 3' directions. This result suggests that BPDE adducts and other large aromatic biomarkers can be detected with aHL, presenting opportunities for the monitoring, quantification, and sequencing of mutagenic compounds from cellular DNA samples. Chapter 4 describes the unzipping of double-stranded nucleic acids by an electric field applied across the membrane, providing structural information about different duplex forms. Comparative studies on A-form DNA-RNA duplexes and B-form DNA-DNA duplexes with a single-stranded tail identified significant differences in the blockage current and in the unzipping duration between the two helical forms. The effect of varying the length of the single-stranded overhang was investigated, and A-form DNA-PNA duplexes were studied to provide additional support for the proposed model. This result identifies key differences between A- and B-form duplex unzipping that will be important in the design of future probe-based methods for detecting DNA or RNA.

FUNDAMENTALS AND APPLICATIONS OF SOLID STATE AND BIOLOGICAL NANOPORES by Rukshan Tharanga Perera A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Chemistry The University of Utah May 2016 Copyright © Rukshan Tharanga Perera 2016 All Rights Reserved The Universi ty of Utah Graduate School STATEMENT OF DISSERTATION APPROVAL The dissertation of Rukshan Tharanga Perera has been approved by the following supervisory committee members: Henry S. White Chair 11/17/2015 Date Approved Cynthia Burrows Member 11/17/2015 Date Approved Shelley D. Minteer Member 11/17/2015 Date Approved Michael David Morse Member 11/17/2015 Date Approved Rajesh Menon Member 01/19/2016 Date Approved and by Cynthia Burrows Chair/Dean of the Department/College/School o f __________________ Chemistry and by David B. Kieda, Dean of The Graduate School. ABSTRACT Both biological and solid state nanopores have been used to study fundamentals of ion transport phenomena and potential applications in single molecule analysis. This dissertation presents investigations of the effect of the electric double layer in the activation energy of ion transport in conical shaped glass nanopores. Further, it presents the use of the a-hemolysin nanopore to detect the DNA cancer biomarker benzo[a]pyrene, and to detect structural differences between the A- and B- form duplexes. Chapter 1 overviews the solid state and biological nanopores, common transport phenomena observed in sold state nanopores, and detection of ssDNA, dsDNA, and biomarkers using biological nanopores. Chapter 2 describes the effect of the electric double layer on the activation energy of ion transport through conical shaped glass nanopores. The study shows that the activation energy values for transport within an electrically charged conical glass nanopore differ from the bulk values due to the voltage and temperature-dependent distribution of the ions within the double layer. Finite element simulations based on the Poisson-Nernst- Planck model semiquantitatively predict the measured temperature-dependent conductivity and dependence of activation energy (Ea) on applied voltage. The results highlight the relationships between the distribution of ions with the nanopore, ionic current, and E a, and their dependence on pore size, temperature, ion concentration, and applied voltage. Chapter 3 describes how the a-hemolysin (aHL) nanopore platform can be used to detect the benzo[a]pyrene diol epoxide (BPDE) adduct to guanine (G) in synthetic oligo deoxynucleotides. BPDE adducts are formed by exposure to the carcinogenic precursor benzo[a]pyrene (BP), a polycyclic aromatic hydrocarbon, and considered as a biomarker that can initiate cancers. Translocation of a 41-mer poly-2'-deoxycytidine strand with a centrally located BPDE adduct to G through aHL in 1 M KCl produces a unique multi­level current signature allowing the adduct to be detected from either the 5' or 3' directions. This result suggests that BPDE adducts and other large aromatic biomarkers can be detected with aHL, presenting opportunities for the monitoring, quantification, and sequencing of mutagenic compounds from cellular DNA samples. Chapter 4 describes the unzipping of double-stranded nucleic acids by an electric field applied across the membrane, providing structural information about different duplex forms. Comparative studies on A-form DNA-RNA duplexes and B-form DNA-DNA duplexes with a single-stranded tail identified significant differences in the blockage current and in the unzipping duration between the two helical forms. The effect of varying the length of the single-stranded overhang was investigated, and A-form DNA-PNA duplexes were studied to provide additional support for the proposed model. This result identifies key differences between A- and B-form duplex unzipping that will be important in the design of future probe-based methods for detecting DNA or RNA. iv TABLE OF CONTENTS ABSTRACT........................................................................................................................... iii LIST OF TABLES............................................................................................................... viii LIST OF FIGURES............................................................................................................... ix LIST OF ABBREVIATIONS............................................................................................ xvii ACKNOWLEDGEMENTS...................................................................................................xx Chapters 1. INTRODUCTION............................................................................................................. 1 1.1 Glass Nanopores to the Study Activation Energy of Ion Transport..................3 1.1.1 Electric Double Layer................................................................................4 1.1.2 Ion Current Rectification............................................................................8 1.2 a-Hemolysin Nanopore to Detect DNA Cancer Biomarkers and Structural Differences between A- and B-Form Duplexes................................................ 10 1.2.1 Nanopore Ion Channel Recordings and Single Strand Translocation ...12 1.2.2 Duplex Unzipping....................................................................................15 1.3 References............................................................................................................18 2. EFFECT OF THE ELECTRIC DOUBLE LAYER ON THE ACTIVATION ENERGY OF ION TRANSPORT IN CONICAL NANOPORES.............................. 22 2.1 Introduction.......................................................................................................... 22 2.2 Experimental Section.......................................................................................... 25 2.2.1 Chemicals and Materials..........................................................................25 2.2.2 Glass Nanopore Membrane (GNM) Fabrication....................................25 2.2.3 Cell Configuration and Data Acquisition............................................... 25 2.2.4 Computational Analysis and Simulations.............................................. 26 2.3 Results and Discussion....................................................................................... 26 2.3.1 Experimental Activation Energies...........................................................26 2.3.2 Finite Element Simulations..................................................................... 28 2.3.3 Dependence of Activation Energy as a Function of Applied Voltage ..35 2.4 Conclusions.......................................................................................................... 41 2.5 References............................................................................................................ 42 52.6 Supplemental Material...................................................................................... 46 52.1 Arrhenius Plots for a 35 nm Radius Pore at Different KCl Concentrations....................................................................................................46 52.2 Arrhenius Plot over an Extended Range of Temperature......................47 52.3 Assessment of the Uncertainty in Simulated Activation Energy due to Measurement Error in the Half-Cone Angle................................................... 48 52.4 Simulated Activation Energies as a Function of Voltage......................49 52.5 Temperature Dependence of Diffusion Coefficients............................. 51 52.6 Simulated Arrhenius Plots for 35 nm Pore at Different KCl Concentrations....................................................................................................52 52.7 Concentration Distribution of K+ in Vicinity of the Nanopore Orifice..53 52.8 Simulated Concentration Profiles of K+ and Cl" Along the z Axis of a 20 nm Nanopore......................................................................................................54 52.9 Electric Field Profiles for a 20 nm Pore at Different Temperatures.....55 52.10 Apparent Activation Energies Calculated at Different Applied Voltages.............................................................................................................. 56 3. DETECTION OF BENZO[A]PYRENE-GUANINE ADDUCTS IN SINGLE­STRANDED DNA USING THE a-HEMOLYSIN NANOPORE............................. 57 3.1 Introduction.......................................................................................................... 57 3.2 Experimental Section.......................................................................................... 60 3.2.1 Chemicals and Materials for Preparation of BPDE-DNA Adduct .....60 3.2.2 Preparation of BPDE-DNA Adduct ..................................................... 60 3.2.3 Glass Nanopore Membrane (GNM) and Bilayer Formation for Ion Channel Recording ............................................................................................ 62 3.2.4 Data Analysis ......................................................................................... 63 3.3 Results and Discussion....................................................................................... 63 3.3.1 Ion Channel Measurements..................................................................... 63 3.3.2 Translocation of 4-mer BPDE Adduct................................................... 64 3.3.3 Translocation of 41-mer BPDE Adduct................................................. 66 3.3.4 Deep Blockage Level Analysis...............................................................66 3.4 Conclusions.......................................................................................................... 71 3.5 References............................................................................................................ 72 53.6 Supplemental Material...................................................................................... 75 53.1 Sample i-t Trace for the Unmodified 41-mer in 1 M NaCl...................75 53.2 Sample i-t Trace for the Unmodified 41-mer in 3 M NaCl...................76 53.3 Sample i-t Trace for the Unmodified 41-BPDE Adduct in 3 M NaCl..77 53.4 Translocation Analysis of the 41-mer and 41-mer BPDE in 1 M KCl and 3 M NaCl....................................................................................................78 vi 4. SIZE-DEPENDENT UNZIPPING OF DUPLEXS OF A-FORM DNA-RNA, A-FORM DNA-PNA, AND B-FORM DNA-DNA IN THE ALPHA-HEMOLYSIN NANOPORE.....................................................................................................................79 4.1 Introduction.......................................................................................................... 79 4.2 Experimental Section.......................................................................................... 83 4.2.1 DNA and RNA Preparation..................................................................... 83 4.2.2 Chemicals and Materials..........................................................................83 4.2.3 Ion Channel Recordings...........................................................................84 4.2.4 Data Analysis........................................................................................... 85 4.3 Results and Discussion....................................................................................... 85 4.3.1 Unzipping of DNA-DNA versus DNA-RNA Duplexes.......................85 4.3.2 Unzipping of DNA-DNA versus DNA-RNA Duplexes with 10-nt Overhang.............................................................................................................92 4.3.3 Unzipping of DNA-DNA versus DNA-RNA Duplexes with no Overhang.............................................................................................................94 4.4 Conclusions.......................................................................................................... 99 4.5 References..........................................................................................................100 S4.6 Supplemental Material.................................................................................... 104 54.1 Sample i-t Trace of a Mixture Containing B-form Duplex.................. 104 54.2 Sample i-t Trace of a Mixture Containing A-form Duplex..................105 54.3 Sample i-t Trace of a Mixture Containing both A- and B-form......... 106 54.4 Voltage Dependence of Unzipping Times for DNA-RNA and DNA-DNA Duplexes............................................................................................. 107 54.5 Unzipping of DNA-RNA Duplex with 40-nt Overhang................. 108 54.6 Thermal Melting Analysis of A-and B-form Duplexes.......................109 54.7 Continuous i-t Trace of DNA-RNA Duplex with 10-nt Overhang at 160 mV..............................................................................................................110 54.8 Voltage Dependent Trapping Time of the DNA-RNA Duplex with 10-nt Overhang.................................................................................................111 54.9 Sample i-t Trace of DNA-DNA Duplex with 10-nt Overhang at 160 mV 112 54.10 Sample i-t Trace of DNA-RNA Duplex with no Overhang...............113 54.11 Sample i-t Trace of DNA-DNA Duplex with no Overhang..............114 54.12 Sample i-t Trace of DNA-DNA Duplex with no Overhang at 200 mV.....................................................................................................................115 54.13 Sample i-t Trace of a Mixture Containing DNA-PNA Duplex......... 116 54.14 Comparison of Unzipping Times of DNA-RNA Duplexes with 3' and 5' Overhangs............................................................................................ 117 54.15 Sample i-t Trace of a Mixture Containing DNA-PNA Duplex......... 118 5. CONCLUSION.............................................................................................................. 119 5.1 References.......................................................................................................... 121 vii LIST OF TABLES Tables 2.1. The viscosity of water at different temperatures.........................................................50 3.1 Time constants measured for the 41 -mer standard and 41 -mer BPDE strand versus voltage..........................................................................................................................71 LIST OF FIGURES Figures 1.1. Schematic diagram of the conical shaped nanopore formed in a glass membrane. Values for r are typically between 10-1000 nm.................................................................... 4 1.2. A schematic diagram of the electric double layer at the charged surface/ electrolyte interface, as proposed by Gouy-Chapman model. The brown solid line shows exponential decay of potential across the solution.....................................................................................6 1.3. A schematic diagram of the overlapped electric double layer. The brown dashed lines show the electric potential due to each plane and the solid black line represents the overall potential between the two charged surfaces........................................................................... 7 + 1.4. A schematic diagram representing the ion current rectification in negatively charged glass nanopores. (A) Diode like i-V behavior observed in pores that show ICR. (B) K and Cl- moving across the pore at different applied voltage.................................................9 1.5. The structure of the heptameric aHL pore with the dimensions in the lumen. This figure is reprinted with the permission of reference 37. Gu, L.Q.; Shim, J. W. Analyst 2010, 135, 441 14............................................................................................................................. 11 1.6. A schematic diagram showing single strand DNA translocation through aHL nanopore under the applied voltage (right). A typical i-t trace observed for single strand DNA translocation (left).................................................................................................................. 13 1.7. Current defection observed during the DNA-crown ether adducts translocation. Sample i-t traces of 5' entry for (A) mono and (B) bis adducts (120 mV trans versus cis). reprinted with permission from the reference 51, An, N.; Fleming, A. M.; White, H. S.; Burrows, C. J. Proc. Natl. Acad. Sci. U.S.A. 2012, 109, 11504.......................................... 14 1.8. Duplex unzipping inside the aHL vestibule. (A) A duplex without an overhang entering the vestibule. The current time trace shows the strands does not unzip. and the molecule is removed by switching the potential. (B) A duplex with an overhang entering the vestibule and eventually unzipping under the applied electric field. The deep blockage current returns to the open channel currents suggests unzipping and translocation of the longer strand. Further, long strands of DNA were collected and identified via PCR on the trans side, proving that the longer strand actually translocates through the pore after unzipping. This study also reveals that duplexes that have a mismatch have different unzipping times, hence showing different unzipping kinetics ...........................................15 1.9. Monitoring the Uracil DNA glycosylase (UDG) enzyme activity for dsDNA using an aHL channel. (A) The structure of dsDNA with a 5'-polydT24 tail within a-HL. The red box indicates the location of the uracil (U) base or the abasic site (AP). (B) A Scheme of the UDG hydrolysis reaction. (C) The blockage current difference observed for uracil (U) and abasic site (AP)............................................................................................................... 17 2.1. (a) Experimental design for measuring the temperature- and electrolyte concentration-dependent i-E response for conical glass nanopores. KCl electrolyte is placed both inside the capillary and in the external reservoir, and a voltage applied across the two Ag/AgCl electrodes. Temperature is controlled by a Peltier heater/cooler, and measured via a K-type thermocouple. (b) Representative i-E curves recorded at a scan rate of 10 mV s-1 and temperatures between 10 and 35 °C for a 35 nm pore in 0.1 mM KCl electrolyte. E refers to the potential of the Ag/AgCl electrode in the internal solution relative to the external solution.................................................................................................................................. 24 2.2. Determining the activation energy of electrolyte transport through a conical glass nanopore. (a,b) Representative Arrhenius plots constructed from ionic currents measured at +0.35 V and -0.35 V for a 35 nm pore in 1 mM KCl electrolyte. (c,d) E a as a function of both KCl concentration and pore size at +0.35 V and -0.35 V, respectively. Error bars, representing the standard deviation of 3 repeated current recordings and the standard error arising from the least squares fit to ln (|i|) - T x data are smaller than the data points used to present E a values in this paper.............................................................................................. 29 2.3. The 2D axisymmetric finite element model used to simulate the current through a conical pore as a function of temperature, voltage, and ion concentration. The boundary conditions and mesh are shown in the figure. A potential is applied to IJ while AB is held at zero potential. The electrolyte concentration at AB and IJ is maintained at the bulk value. A surface charge of -2 mC/m2 is applied to EF and FG and a finer mesh size is used close to the pore orifice. The boundaries far from the pore orifice (GH, ED and CD) are set at zero charge, which does not appreciably affect the calculated currents. This allows for a coarser mesh in these regions, as no double-layer need be resolved. AJ is the symmetry axis. The total normal ion flux (mol/m2s) at the semi-infinite boundary AB was computed by an integration and converted to the ionic current (C/s) by multiplying by Faraday's constant (96500 C/mol).The inset shows the expanded area of the pore mouth................31 2.4. Simulated activation energies of electrolyte transport through a conical glass nanopore. (a,b) Representative Arrhenius plots constructed for currents measured at +0.35 V and - 0.35 V for a 35 nm pore in 1 mM KCl electrolyte, from which E a are calculated. (c,d) Simulated E a as a function of both KCl concentration and pore size at +0.35 V and -0.35 V, respectively. See Figure 2 for a direct comparison to the experimental Arrhenius plots and values of E a for the same conditions............................................................................. 34 x 2.5. Experimental (a) and simulated (b) activation energies for a 20 nm pore for different KCl concentration as a function of applied potential. The error bars are smaller than the data points..............................................................................................................................36 2.6. Concentration profiles of K+ and Cl" ions in the vicinity of a 20 nm nanopore orifice at (a) +0.35 V and (b) -0.35 V (relative to the internal electrode). The surface charge on the wall is -2 mC/m2 and the bulk concentration is 0.1 mM at 25 °C. Note, different color scales are used for the two species. Plots for different pore sizes and cross-sectional plots of the species concertation are shown in Figure S2.7.................................................................... 37 2.7. The total ionic concentration at the orifice of a 20 nm nanopore at 0.1 mM KCl, at 10 and 35 °C and for (a) -0.35 V and (b) +0.35 V applied potential. Contours have been added at 1.0 and 1.3 mM to aid interpretation.................................................................................39 2.8. The total ionic concentration at the orifice of a 20 nm nanopore at 0.1 mM KCl, at 10 °C and 35 °C. (a) Radial cross section plot taken inside the pore at z = -20 nm at -0.35 V and (b) +0.35 V. Individual concentration profiles along z axis are given in Figure S2.8.........................................................................................................................................42 52.1. A sample Arrhenius plot (ln \i| versus T l) for values of i measured at 0.35 V for a 65 nm pore at 100 mM KCl from 10 °C to 45 °C showing the nonlinearity of the curve over a wider temperature range. Non-linear behavior was observed for other pores and concentrations. The linear range 10 °C to 35 °C was chosen for all the experiments.....46 52.2. Arrhenius plots (ln \i\ versus T"1) for values of i measured at 0.35 V for a 35 nm pore. The above Arrhenius plots were used to extract the activation energy values presented in Figure 2.2 of the main text, at +0.35 V for a 35 nm pore.................................................... 47 52.3. (a) An optical microscopy image of the glass nanopore. The image is roughly 40 times enlarged of the actual size (b) radius versus half-cone angle pairs that corresponding to a nanopore with a 9.18 MQ resistance and (c) the variation in activation energy as a function of half-cone angle in 0.1 mM at 0.35 V.................................................................48 52.4. (a) The surface charge density as a function of temperature. (b) Simulated activation energies for a 20 nm pore for 0.1 mM KCl as a function of applied potential using temperature dependent surface charge (black line) and experimental data (red line). The Arrhenius plot were linear for the between 10 °C - 35 °C.................................................. 49 52.5. Diffusion coefficient as a function of temperature. The expression D = 4.03 x10-11 T + 1.02 x 10-9 m2/s (T in °C) for K+ was used for both K+ and Cl- because the diffusion coefficients vary less than 2 %. (b) Ionic mobility is derived from n =D/RT (T in K). The expression n = 1.5 x10 -14 T+ 3.94 x 10-13 s mol kg-1 (T in °C) describe the data and was used in the finite element simulations...................................................................................51 52.6. Arrhenius plots constructed from the simulated temperature-dependent currents xi obtained at +0.35 V for a 35 nm pore. These Arrhenius plots were used to extract the activation energy values shown in Figure 2.2 in main text.................................................... 52 52.7. Concentration distributions at 25 °C in the vicinity of the orifice of (a) 20 nm (b) 35 nm (c) 50 nm and (d) 65 nm nanopores. The surface charge on the nanopore wall is -2 mC/m2 and the bulk concentration is 0.1 mM. The applied voltage is +0.35 V............... 53 52.8. Concentration profiles of K+ and Cl- in the vicinity of 20 nm nanopore orifice at (a) +0.35 V and (b) -0.35 V. The surface charge on the wall is -2 mC/m2 and the bulk KCl concentration is 0.1 mM........................................................................................................ 54 52.9. The axial component of the electric field inside the pore at z = -20 nm for a 20 nm nanopore (0.1 mM KCl, 10 °C and 35 °C). At (a) -0.35 V and (b) +0.35 V......................55 52.10. Apparent activation energies calculated from finite element simulations at different applied voltages for a 20 nm pore.........................................................................................56 3.1. Benzo[a]pyrene metabolism leading to guanine adducts in DNA............................. 58 3.2. Ion-exchange HPLC traces for 4 mer-BPDE and 41-mer BPDE: The HPLC conditions utilized solvent A = 10% CH3CN, 90% ddH2O; B = 1 M NaCl in 10% CH3CN 90% ddH2O, 25 mM Tris pH 8; flow rate = 1 mL/min while monitoring the absorbance at 260 nm. The separation was initiated at 15% B followed by a linear increase to 100% B over 30 min.................................................................................................................................... 61 3.3. Proposed model for translocation of a 4-mer and 4-mer BPDE adduct through the aHL nanopore. (A) Representative i-t trace for the 4-mer (5'-CCGC-3') strand, (B) representative i-t trace for a 4-mer BPDE adducted oligomer. All data were recorded at 180 mV (trans versus cis) in 1 M KCl at 25.0 ± 0.5 oC with a 100 kHz low-pass filter and 500 kHz data acquisition rate................................................................................................ 65 3.4. Current versus time profile collected over 20 s for 41-mer BPDE (2 ^.M) in 1 M KCl. The data were recorded at 180 mV (trans versus cis) at 25.0 ± 0.5 oC. The red dotted lines indicate the places where long open channel blockages were removed............................... 67 3.5. Event types detected during translocation of the 41-mer BPDE sample. (A) Representative i-t traces for translocation of the 41-mer BPDE sample, (B) blowup of a 3'- entry event, and (C) blowup of a 5'-entry event. The data were recorded at 180 mV (trans versus cis) at 25.0 ± 0.5 oC. The data were refiltered to 50 kHz. Results from measurements are presented as percent ratio of the blockage current versus open channel current %(///0). The i-t traces for events >50 ^s were analyzed. Long open channel current segments (20 - 500 ms) were manually removed, as indicated on the i-t trace. A relatively low capture rate (~70 events/s) was observed due to the low concentration (2 ^M) of the 41-mer BPDE studied.(D) Proposed model for the translocation of a 41-mer BPDE adduct through aHL. (I) DNA enters from the cis side of the channel by threading either the 3' or 5' tail. (II) The xii BPDE adduct becomes caught at the 1.4 nm central constriction that gives rise to the deep blockage in the ion current recorded that marks the presence of the BPDE adduct. (III)The DNA translocates through the P-barrel.................................................................................68 3.6. Current histograms for the step-current levels monitored for the 41-mer BPDE events. (A) Plots of frequency distributions for the I2 and I2 ' current levels. (B) Plots of frequency distributions for the I3 current levels. The data were collected at 120, 160, and 180 mV (trans versus cis) in 1 M KCl at 25.0 ± 0.5 oC and plotted with a bin size of 0.5 pA. Population distributions represent 400-450 event................................................................70 53.1. Current versus time profile for the 41-mer standard (4 ^M) in 1 M KCl. The data were recorded at 180 mV (trans versus cis) at 25.0 ± 0.5 oC. Open channel baseline current intervals longer than 20 ms were removed from the following i-t traces and indicated by the red dashed lines........................................................................................................................75 53.2. Current versus time traces collected over 20 s for the 41-mer standard (4 ^.M) in 3 M NaCl. The data were recorded at 180 mV (trans versus cis) at 25.0 ± 0.5 o C........... 76 53.3. Current versus time profile for the 41-mer BPDE (2 ^.M) in 3 M NaCl. The data were recorded at 180 mV (trans versus cis) at 25.0 ± 0.5 o C ............................................... 77 53.4. Translocation time analysis of the 41-mer and 41-mer BPDE in 1 M KCl. Only the events longer than 70 ^.s were used for translocation analysis of 41-BPDE. The data were recorded at 120,160, and 180 mV (trans versus cis) at 25.0 ± 0.5 oC. The time distribution for translocation of 41 -mer was fit with a Gaussian model. The modified 41 -mer BPDE showed longer translocation times (325-375 events were analyzed), and its duration histogram exhibits an exponential decay............................................................................................................... 78 4.1. Structures for aHL and the duplex nucleic acids studied. (A) The structure of wild-type aHL based on an x-ray crystal structure (pdb 7aHL) (reference 13). (B) The structure of a B-form DNA-DNA duplex (pdb 1BNA) (reference 28), (C) and the structure of an A-form DNA-RNA duplex (pdb 1RRR) (reference 43)...........................................................82 4.2. (A) A representative i-t trace showing uninterrupted data collected at 10 kHz at 120 mV. The mixture contained 8 ^.M of both A- and B-form duplexes in 1 M KCl, 10 mM PBS, pH 7.4 at 20 oC. (Bj The expanded window in Figure 4.3 A shows the deep-block current differences between A- and B-form duplexes. The red dashed line represents the blocking current of A- form duplex and the blue dashed line indicates the blocking current of B-form duplex during the unzipping process. The expanded trace in (B) is filtered to 1 kHz for presentation purposes.............................................................................................. 87 4.3. Current blockage, unzipping time duration, and i-t density plots for the duplex systems studied. (A) DNA-DNA duplex (B-form), (B) DNA-RNA duplex (A-form), and (C) A-and B-form duplexes analyzed as a 1:1 mixture. All experiments were performed at 120 mV (trans versus cis) in 1 M KCl (10 mM PBS, pH 7.4), at 20 °C in the presence of 8 ^.M xiii duplex 88 4.4. Proposed models for trapping and unzipping of DNA-RNA (A-form) and DNA-DNA (B-form) duplexes. The green color region show the highest voltage drop across the pore based on both experiments and molecular dynamics simulations (reference 54,58,59)....91 4.5. Studies for unzipping of A- and B-form duplexes with a shorter 10-nt tail. (A) Unzipping of DNA-RNA. (B) DNA-DNA duplexes. All experiments were performed at 120 mV in 1 M KCl (10 mM PBS, pH 7.4) at 20 °C. Event durations for Type 1 (C, left) and Type 2 (C, right) were recorded at voltages from 100-160 m V ............................... 93 4.6. Unzipping of A- and B-form duplexes without a single-stranded tail. (A) Unzipping of DNA-RNA blunt-end duplex, (B) DNA-DNA blunt-end duplex. All experiments were performed at 120 mV (trans versus cis) in 1 M KCl (10 mM PBS, pH 7.4) at 20 °C....95 4.7. A sample i-t trace showing uninterrupted data collected at 10 kHz at 120 mV. The mixture contained 8 |iM DNA-PNA duplexes in 1 M KCl, pH 7.4 at 20 oC....................97 4.8. Unzipping duration as a function of voltage for DNA-DNA (black), DNA-RNA (red) and DNA-PNA (blue). The data were recorded at 20 oC in 1 M KCl, 10 mM PBS, pH 7.4. The data were fit in to an exponential decay equation to obtain the unzipping time....... 98 54.1. A sample i-t trace showing uninterrupted data collected at 10 kHz for 20 s at 120 mV. The mixture contained 8 jjM B-form duplex in 1 M KCl, 10 mM PBS, pH 7.4 at 20 °C 104 54.2. A sample i-t trace showing uninterrupted data collected at 10 kHz for 20 s at 120 mV. The mixture contained 8 A-form duplex in 1 M KCl, 10 mM PBS, pH 7.4 at 20 oC 105 54.3. A sample i-t trace showing uninterrupted data collected at 10 kHz for 20 s at 120 mV. The mixture contained 8 A- and B-form duplexes in 1 M KCl, 10 mM PBS, pH 7.4 at 20 oC. The two expanded windows, A and B, show the deep block current differences between A- and B-form duplexes. The expanded area is filtered to 1 kHz for presentation purpose................................................................................................................................ 106 54.4. Unzipping duration histograms as a function of voltage for DNA-RNA (left) and DNA-DNA (right) duplexes. The data were recorded at 20 oC in 1 M KCl, 10 mM PBS, pH 7.4. An exponential decay was fit to the data to obtain the unzipping time.............. 107 54.5. (A) A sample i-t trace showing uninterrupted data collected at 10 kHz for 20 s at 120 mV. The mixture contained 10 DNA-RNA duplex with 40-nt overhang in 1 M KCl, 10 mM PBS, pH 7.4 at 20 °C. (B) Current blockage, unzipping time duration, and i-t density plots for DNA-RNA duplex with 40-nt overhang................................................ 108 xiv 54.6. Thermal melting analysis of the DNA-DNA and DNA-RNA duplexes. All measurements were performed in 10 mM PBS, pH 7.4. The absorbance at 260 nm, Abs260 nm, was monitored as the temperature was increased from 20 oC to 100 oC at a ramp rate of 1 oC/min. At each time interval, the temperature was equilibrated for 30 s prior to making each absorbance measurement. Each experiment was conducted in triplicate............. 109 54.7. A sample i-t trace showing uninterrupted data collection at 10 kHz for 20 s at 120 mV. The mixture contained 8 of DNA-RNA duplex with 10-nt overhang in 1 M KCl, 10 mM PBS, pH 7.4 at 20 oC. The two expanded windows (A and B) show the blockage due to occupation of the 10-nt overhang in the vestibule................................................... 110 54.8. Trapping time duration histograms as a function of voltage for DNA-RNA duplex with a 10-nt overhang. Only the events with %I/Io between 20 and 80 and t > 200 ^.s were analyzed as duplex unzipping events (single strand translocation is much faster). Data were recorded at 20 oC in 1 M KCl, 10 mM PBS, pH 7.4. An exponential decay was fit to the data to obtain the unzipping time........................................................................................ 111 54.9. A sample i-t trace showing uninterrupted data collection at 10 kHz for 20 s at 160 mV. The mixture contained 8 of the DNA-DNA duplex with 10-nt overhang in 1 M KCl, 10 mM PBS, pH 7.4 at 20 oC. The two expanded windows (A and B) show long-current blocks are due to unzipping of the duplex and the shorter blocks (less than 1 ms denoted by asterisks) are from translocation of the excess ssDNA.................................. 112 54.10. A continuous i-t trace showing uninterrupted data collection at 10 kHz for 20 s at 160 mV. The mixture contained 8 DNA-RNA blunt end duplex in 1 M KCl, 10 mM PBS, pH 7.4 at 20 oC. The expanded window A shows short translocation events (less than 500 |j,s) that are due to excess single strands present in the mixture.................................. 113 54.11. (A) A continuous i-t trace showing uninterrupted data collection at 10 kHz for 10 s at 120 mV. The cis side contained 8 DNA-DNA blunt end duplex in 1 M KCl, 10 mM PBS, pH 7.4 at 20 oC. Long-current blockages show the duplex occupying the vestibule and the short events (less than 1 ms) are due to translocation of excess single strands. Interruption of the current blockage was due to the polarity reversal of the channel to remove the duplex in the nanopore. (B) Residual current when a blunt end duplex is inside the vestibule as a function of voltage. (C) Frequency of the events between two current levels shown in Event Type 3b in the main text Figure 4.5B........................................... 114 54.12. A sample i-t trace showing uninterrupted data collection at 10 kHz for 20 s at 200 mV. The blunt end duplex unzips at 200 mV but not at 120 mV (see preceding section). The mixture contained 8 of the DNA-DNA duplex with 10-nt overhang in 1 M KCl, 10 mM PBS, pH 7.4 at 20 oC. The two expanded sections (A and B) shows long-current blocks are due to unzipping of the duple........................................................................... 115 54.13. A sample i-t trace showing uninterrupted data collected at 10 kHz for 20 s at 120 mV. The mixture contained 8 DNA-PNA duplex in 1 M KCl, 10 mM PBS, pH 7.4 at xv 20 oC 116 54.14. Unzipping duration histograms as a function of voltage for the DNA-PNA duplexes. Data were recorded at 20 oC in 1 M KCl, 10 mM PBS, pH 7.4. An exponential decay was fit to the data to obtain the unzipping time. The cis side of the protein channel contained 8 of DNA-PNA sample................................................................................................... 117 54.15. Comparison of the Unzipping duration histograms for the DNA-RNA duplexes. (A) For 5' poly C overhang. (B) For 3' poly C overhang Data were recorded at 20 oC in 1 M KCl, 10 mM PBS, pH 7.4. An exponential decay was fit to the data to obtain the unzipping time. The cis side of the protein channel contained 8 of DNA-RNA sample in each case 118 xvi LIST OF ABBREVIATIONS 2D : two-dimensional aHL: alpha hemolysin n : dynamic viscosity of the medium Ag/AgCl : silver/silver chloride BP : benzo[a]pyrene BPDE : benzo[a]pyrene diol epoxide °C : degree Celsius cm : centimeters DNA : deoxyribonucleic acid DPhPC: 1,2-diphytanoyl-s«-glycero-3 -phosphocholine E : electric potential EA : activation energy EDL: electric double layer ELISA : enzyme-linked immunosorbent assay FEM : finite-element method GO : gigaohm G-BPDE : guanine- benzo[a]pyrene diol epoxide adduct GNM : glass nanopore membrane HPLC: high performance liquid chromatography i : current ICR : ion current rectification I-E: current versus electric potential ilim :limiting current Io :open pore current i-t : current-time K+ : potassium ions KCl : potassium chloride kHz :kilohertz kJ : kilo Joul LC-MS : liquid chromatography coupled to mass spectrometry ln : natural logarithm M : moles per liter MQ : megaohm mC : millicoulomb mL : milliliter mmHg : millimeter mercury ms : millisecond mV : millivolt nA : nanoampere NDR : negative differential resistance nM : nanomolar xviii nt : nucleotide PAH : polycyclic aromatic hydrocarbons PNA : peptide nucleic acid Pt : platinum PA : picoampere pm : picometer RNA : ribonucleic acid THF : tetrahydrofuran Tm : melting temperature V : voltage xix ACKNOWLEDGEMENTS I would like to express my gratitude to my advisor Dr. Henry White for his guidance during my PhD study at University of Utah. Frequent advice based on his extensive experience in research has helped me in designing, implementing, and finishing a research project in an effective manner. I would also like to extend my gratitude towards my committee and especially Dr. Cynthia Burrows for helpful suggestions and guidance throughout my study. I would also thank Dr. Aaron Fleming, Dr. Robert Johnson and Dr. Martin Edwards for their useful contributions to my research work. I thank my past and present group members; Dr. John Watkins, Dr. Deric Holden, Dr. Qian Jin, Dr. Long Luo, Dr. Qianjin Chen, Dr. Kim McKelvey, Mr. Jewen Xiong, Ms. Cherry Tan, Mr. Sean German, Mr. Alan Zhang for being helpful and making a great working environment in the lab. I also must thank our collaborators, Dr. Na An, Dr. Yun Ding, Dr. Ania Wolna, Dr. Jan Riedl, and Mr. Lidong for useful discussions. Special thanks goes to Dr. Steven Feldberg for useful discussions during his visits to our lab. Finally yet importantly, I would also like to sincerely thank my father, mother, and my family members for their immense love, support, and encouragement. CHAPTER 1 INTRODUCTION Miniaturizing the dimensions of a sensor comparable to the size of individual molecules is a key feature of single-molecule sensing.1-3 This fundamental principle has led to the development of a new class of label-free sensors termed "nanopores". Nanopores have been investigated extensively for biosensing applications that play a significant role in medicine, analytical biochemistry, and biotechnology.4-6 Incorporation of fabrication and designing technologies from the semiconductor industry has helped to develop miniaturized sensing devices that allow the production of highly portable, low footprint, scalable devices.7 Nanopores may be confined to the nanoscale in depth, width, or both.8 Nanopores can be identified as either biological nanopores, extracted from bacteria, or solid-state, fabricated in thin membranes using modern nanotechnology.5,9 In each case, chambers on both sides of the pore are filled with an electrolyte and a voltage is applied across the nanopore to measure the conductance. When a charged analyte that is comparable to the size of the pore is introduced into one of the chambers, the electrophoretic force will drive it through the pore, giving rise to a brief change in conductance as the analyte passes through the pore. The change in conductance can provide useful information about the analyte, such as its size and charge. The origin of this idea dates back to the 1950s, when a similar method was used develop the Coulter counter technology that is today used to count red blood cells.10,11 The main advantage of this method is the label free detection of the analyte. Resistive pulse techniques have been used to detect virus particles in the 1970s, but any detection of single macromolecules such as DNA, RNA, or protein remained unexplored until the 1990s.12,13 Apart from sensing applications, nanopores have also received considerable attention due to the unique mass transport phenomena that arise because of their high surface to volume ratios and surface charge. Ion-current rectification,14-16 ion concentration polarization,17 and negative differential resistance 18 are some of the unique phenomena that can influence transport in nanopores and in nanofluidic structures. Understanding the unique properties at the nanoscale is important in developing nanopore sensors and micro-and nano-fluidic devices. Surprisingly, few studies have focused on the temperature dependence of the ion transport at confined geometries. We used the activation energy of ion transport (Ea) as a parameter to study the effect of temperature on ion transport in conical-shape glass nanopores. The Ea, measured as a function of pore size, electrolyte concentration, and applied voltage are presented in Chapter 2. Additionally, Chapter 3 and 4 present practical applications where a-hemolysin protein nanopores were used to detect a DNA cancer biomarker and the structural differences between different types of nucleic acid duplexes. 2 3 1.1 Glass Nanopores to the Study Activation Energy of Ion Transport The robustness and durability of solid-state pores compared to their biological counterparts, offer fine-tuning of the size and shape at subnanometer precision,19 permitting detailed studies of fundamental ion transport in well-defined nanopore geometries. Silicon nitride, silicon oxide, and metal oxides are the most widely used materials to fabricate nanopores using ion beam sculpting, atomic layer deposition, and e-beam drilling techniques.20-22 Our lab has developed a simple bench-top method to fabricate conical shape nanopores embedded in a thin glass membrane (Figure 1.1), referred to herein as glass nanopore membrane (GNM).23 The method does not require any sophisticated micro/nano fabrication techniques and also allows modification of the surface (via well-characterized silane chemistries) to introduce hydrophobicity or change the surface charge.24 Nanoscale conical-shaped glass nanopores prepared in our laboratory have been used to study ion transport phenomena and in particle translocation experiments. The fabrication of GNMs is reported elsewhere but briefly summarized here.23 As the first step, a 25 ^.m Pt wire is sharpened electrochemically and sealed into a glass capillary. Borosilicate or soda-lime glass is mainly used for this purpose. The sealed glass is then polished away until the tip of the Pt wire is exposed. In the last step, the sealed Pt is electrochemically etched out resulting in a conical-shaped nanopore embedded in the glass membrane. The radius of the pore mouth can be measured by the conductance across the nanopore when the electrolyte concentration, temperature, and the half cone angle of the nanopore are known. This simple but very effective method has been used to fabricate GNMs to investigate ion transport properties presented in Chapter 2. GNMs used as a platform to support lipid bilayers are presented in Chapter 3 and 4. 4 25-75 jam Figure 1.1. Schematic diagram of the conical shaped nanopore formed in a glass membrane. Values for r are typically between 10-1000 nm. Not drawn to the scale. 1.1.1 The Electric Double Layer Many of the intriguing mass transport properties of glass nanopores arise from their charged surfaces at near-neutral pH. Glass surfaces can acquire surface charge due to ionized surface groups or adsorbed ions when brought into contact with an aqueous phase. At neutral pH, glass acquires negative charge due to the SiO" groups. The charged surface influences the distribution of the ions and dipolar constituents in close proximity, which results in a net change in the positions of the ions to minimize their total free energy. As a result of the electrostatic interactions, ions of opposite charge to the surface charge (counter ions) will be attracted to the surface while ions of the same charge (co-ions) are repelled.25 The electrostatic interactions create a thin layer of counter ions at the charged surface. The charged layer that forms at the charged surface and electrolyte interface is termed the "electric double layer" (EDL).26,27 Helmholtz was the first to propose the idea of EDL in 1879. His model provided a good foundation to describe the ion distribution at the interface; however, it does not account for important factors including diffusion/mixing of ions in solution, adsorption, and the possibility of interaction between solvent dipole moments and the charged surface. Later, Gouy28 (1910) and Chapman29 (1913) developed a model to introduce the "diffuse layer" where counter ions in the electrolyte are not rigidly held at the interface but mobile due to the influence of diffusion and electrostatic forces. In this model, the ions are considered as point charges. The electric potential exponentially decreases away from the charged surfaces and obeys the Poison-Boltzmann distribution as shown in Figure 1.2.25,27 5 d y dx2 F z iCoe -zey ~w~ s s n (11) where Co i s the bulk concentration of ions, y is the electrostatic potential at point x, and z is the valency of the ions. F is the faraday constant and k is the Boltzmann constant. s and So represents the permittivity of the solvent and permittivity of vacuum, respectively. The composition of the counter ions at the charged surface is significantly different from the bulk solution and the net charge built up at the surface generates an electrical potential difference at any two points between the surface and bulk solution. The Debye length (r-1) is the parameter used to characterize the thickness of the EDL. The Debye length is a function of electrolyte concentration and the temperature, r = s„s RT 2 z2 F 2c (12) 6 Figure 1.2. A schematic diagram of the electric double layer at the charged surface/ electrolyte interface, as proposed by Gouy-Chapman model. The brown solid line shows exponential decay of potential across the solution. where £r is the relative permittivity, £o is the permittivity in a vacuum, R is the gas constant, T is the absolute temperature, z is the electrolyte valence, F is Faraday's constant, and c is the electrolyte concentration. For a 1:1 electrolyte at 25 °C, the above simplifies to: where c is in units of mol dm-3. For a 1:1 electrolyte at 25 °C, the Debye length in 0.01 mM solution is approximately 100 nm and at 100 mM it is approximately 1 nm. 7 The existence of the EDL structure alters the fluid flow at charged surfaces. When two or more surfaces are close to each other, that is, separated by a distance comparable to the Debye length, an overlapped EDL can be expected (Figure 1.3).30 The distance between the two charged surfaces facing each other and the Debye length can determine the degree of the overlap. The electro-neutrality of the system is also strongly effected due to counter ion enrichment between the surfaces and creates a nonuniform potential distribution. The difference in ion distribution at the pore mouth can result in different transport phenomena at the nanoscale compared to the bulk solution. CD c -t<-D» O CL o -t-» o Q LU --------------------------------------------------► Distance ------------- Total potential ------------ Potential due to each plane Figure 1.3. A schematic diagram of the overlapped electric double layer. The brown dashed lines show the electric potential due to each plane and the solid black line represents the overall potential between the two charged surfaces. 1.1.2 Ion Current Rectification The formation of overlapping double layers in nanopores, as a consequence of charged surfaces separated by distances comparable to the Debye length, gives rise to unique ion transport phenomenon not observed in macroscopic pores. One such unique transport phenomena shown by nanopores is ion current rectification (ICR).14-16 When the nanopore has an asymmetric charge distribution, the i-V curve shows a diode like behavior and does not obey Ohm's law (Figure 1.4). Wei et al. first reported ICR in 1997,15 demonstrating that ICR is strongly dependent on the pore size and the electrolyte concentration. Since then many reports on ICR have been published, demonstrating the uses of ICR for sensing applications.31-36 ICR can be rationalized using the ion accumulation and depletion model.16,37 The surface of the glass is negatively charged at neutral pH due to dissociation of protons from the silanol groups. The negatively charged walls will be screened by positive ions and form an EDL as described in the previous section. The length of the EDL is roughly 5k -1 where k -1 is the Debye length. At low electrolyte concentrations, the Debye length expands and the pore orifice is predominantly occupied by the counter ions (K+). When a positive voltage is applied to the electrode inside the nanopore (the outside electrode is grounded), K+ ions move outside and Cl- ions move inside the pore. However, due to accumulation of K+ ions at the pore orifice, Cl- ions get rejected moving in to the pore resulting in an ion depletion zone inside the pore. This ion depletion inside the pore gives rise to a lower current than expected from linear Ohmic behavior. On the other hand, when a negative voltage is applied to the electrode inside the nanopore, Cl- will be rejected moving outside, creating an ion accumulation zone inside the pore. Therefore, at negative applied potentials 8 9 Figure 1.4. A schematic diagram representing the ion current rectification in negatively charged glass nanopores. (A) Diode like i-V behavior observed in pores that show ICR. (B) K+ and Cl- moving across the pore at different applied voltage. the current observed is higher compared to the linear Ohmic behavior. In this thesis, Chapter 2 presents a study of ion transport at the nanoscale under ion current rectification conditions. Furthermore, our studies show the nanopore becomes less rectifying at higher temperatures, which intuitively agrees with the idea that at higher temperatures the thermal energy of ions can overcome the electrostatic barriers generated by the nanopore walls to a greater extent. 10 1.2 a - Hemolysin Nanopore to Detect DNA Cancer Biomarkers and Structural Differences between A- and B-form Duplexes Protein-based biological nanopores harness the reproducibility of biological systems to furnish well-defined channels, some of which have high-resolution crystal structures to provide a better understanding of their properties.9 Furthermore, biological nanopores can be modified to induce structural changes using site-directed mutagenesis. Examples of protein nanopores include alpha-hemolysin (aHL),9,38 MspA,39 aerolysin,39,40 and ClyA.41 The most studied protein nanopore is the aHL, a toxin secreted as a 33.2-kD water-soluble monomer by Staphylococcus aureus (Figure 1.5).9 This protein can self-assemble into a heptamer and spontaneously inserts into a lipid bilayer, producing a trans membrane channel. Due to the size and the comparatively higher stability of the aHL nanopore, it has been heavily explored as a biosensor platform for DNA, RNA, proteins, and small molecules.13,42-48 To study these molecules, a voltage is applied across the protein, which itself is supported in a lipid bilayer, to electrophoretically drive analytes into the nanopore. Interactions between the channel and molecule of interest yield characteristic deflections in the current and residence times. The size-limiting property of this channel is the central constriction (d = 1.4 nm), which allows single-stranded DNA (d = 1.0 nm) to pass through the P-barrel but not double-stranded DNA (d = 2.0 nm).49 This unique structure of aHL has been capitalized upon as a next-generation DNA sequencing platform.8 The idea of using aHL nanopore for DNA sequencing was first proposed by Deamer et al. with the ultimate aim of obtaining different conductance levels for different bases as each base moves through the tightest constriction of the nanopore. 50 11 Figure 1.5. The structure of the heptameric aHL pore with the dimensions in the lumen. This figure is reprinted with the permission of reference 37. Gu, L.Q.; Shim, J. W. Analyst 2010, 135, 441 The first demonstration of DNA translocation through aHL was published in 1996 by Kasianowicz and co-workers.13 The potential to sequence DNA with nanopores was well-recognized and many research groups are contributing to the development of methods towards the ultimate goal of sequencing DNA. Further, the aHL channel also provides an excellent system for monitoring reactions and conducting biophysical experiments to interrogate solutes in solution. Chapters 3 and 4 present studies that have been performed using aHL. Chapter 3 focuses on detecting a biologically relevant cancer biomarker using simple translocation experiments whereas Chapter 4 presents the use of unzipping experiments to distinguish between structural differences among A and B forms of duplexes. 1.2.1. Nanopore Ion Channel Recordings and Single Strand Translocation As a solid support to form the lipid bilayer, we used the GNM discussed in section 1.1. GNMs used in these experiments have significant advantages over SiN and Teflon solid supports that are commonly in use. The diameter of the GNMs we use here are at least one to two orders of magnitude smaller, and this results in more stable bilayers. The bilayers formed on GNMs are more resistive to environmental disturbances. Our studies have shown the bilayer can be stable for more than a week.51 Most importantly, the smaller surface area of the GNM can result in lower noise due to reduced charging effects on the bilayer and less capacitance across the solid support. The silanol groups on glass allow surface modification and introducing more hydrophobic functionality to the surface can enhance the bilayer stability.24,51,52 After the lipid bilayer is formed on the GNM, monomeric aHL is added to one side of the electrolyte reservoir. Monomeric aHL forms a stable heptametrical ion channel that immediately inserts in to the bilayer, forming a transmembrane ion channel. To study the desired analyte, for example, DNA, a voltage is applied across the protein to guide the analyte toward the nanopore by the electric field. The change in conductance is measured as a function of time as the DNA passes through the protein nanopore. This allows information to be obtained related to the structure of the DNA (Figure 1.6). The key information that can be extracted from a statistical analysis of a single strand DNA translocation experiment includes translocation time (t), current blockage (I), or residual current (I/Io, where Io is the open channel current), and event frequency f). Both translocation time and current blockage amplitude histograms exhibit Gaussian-like distributions; the peak value of the histograms are reported as the mean translocation time 12 13 Figure 1.6. A schematic diagram showing single strand DNA translocation through aHL nanopore under the applied voltage (right). A typical i-t trace observed for single strand DNA translocation (left) and the mean current blockage, respectively.13,46 After the first demonstration of DNA and RNA oligomer translocation was reported by Kasianowicz and coworkers in 1996, extensive studies have been performed on DNA and RNA translocation.53-55 In more recent work, the White and Burrows laboratories have demonstrated the use of translocation experiments to detect bulkier adducts.56,57 Interestingly, single-stranded DNA modified with one or two crown ether adducts showed distinct current blockage (Figure 1.7 (A) and (B)). These studies have concluded the nanopore can detect a single abasic site or two abasic sites per strand as the lesion-containing strand translocates through the pore. Inspired by the idea that bulky adducts can modulate the current, we have found that the introduction of the bulkier benzo[a]pyrene diol epoxide (BPDE) adduct can perturb translocation and give rise to unique current signals. Binding of BPDE to G residues in the TP53 gene can cause lung cancer and is therefore considered as a biomarker.49,58-60 14 (A) (B) Figure 1.7. Current defection observed during the DNA-crown ether adducts translocation. Sample i-t traces of 5' entry for (A) mono and (B) bis adducts (120 mV trans versus cis). reprinted with permission from reference 51, An, N.; Fleming, A. M.; White, H. S.; Burrows, C. J. Proc. Natl. Acad. Sci. U.S.A. 2012, 109, 11504. 15 The identification of adducts and their locations in the genome are highly important to locate hot spots for mutations. These results are presented in Chapter 3. 1.2.2. Duplex Unzipping The size-limiting properties of aHL at the central constriction (d = 1.4 nm) only permit single-stranded DNA (d = 1.0 nm) to pass through the P-barrel but not double­stranded DNA. While a duplex cannot translocate through aHL, it can unzip within the vestibule into two strands at moderate applied voltages (Figure 1.8 (A)). Both experimental and molecular dynamics simulation studies have suggested that the majority of the voltage drop occurs at the P-barrel.47,61 Figure 1.8. Duplex unzipping inside the aHL vestibule. (A) A duplex without an overhang entering the vestibule. The double strand does not unzip, and the molecule is removed by switching the potential. (B) A duplex with an overhang entering the vestibule and eventually unzipping under the applied electric field. The deep blockage current returns to the open channel current suggesting unzipping and translocation of the longer strand. Further, long strands of DNA were collected and identified via PCR on the trans side, proving that the longer strand actually translocates through the pore after unzipping. This study also reveals that duplexes that have a mismatch have different unzipping times, hence showing different unzipping kinetics. Therefore, an overhang attached to one of the strands can occupy the total cross section of the aHL, and the electric field inside the P- barrel can pull the strands that can result in unzipping of the two strands. Typically, a homo-polymeric overhang greater than 20 nucleotides is required to induce unzipping of the double-stranded DNA.62 Unzipping eventually permits the longer strand to pass through the P-barrel, giving rise to a current blockage. The unzipping time is at least an order of magnitude higher compared to simple translocation due to the longer residence time inside the vestibule. Branton and coworkers demonstrated the first example of duplex unzipping.63 In their study, they captured a duplex with an overhang inside aHL and subjected it to unzipping by applying an electric field. Statistical analysis of the unzipping times shows first order kinetics, which suggests the unzipping is an energy dependent process. Other duplex unzipping studies have been performed to understand unzipping kinetics,64 probe base pairing energy,65 and to detect microRNAs66 and oxidative damage.67 Recent unzipping studies performed in the White and Burrows laboratories revealed a new sensing zone of the aHL.68-71 A missing base or a lesion placed in the duplex that sits close to the sensing zone of the nanopore can produce different unzipping currents and durations (Figure 1.9). The discovery has led to the detection of damage in DNA and base-flipping kinetics using duplex unzipping inside aHL. Studies were extended to detect damage in RNA by utilizing the latch-sensing zone, and the unzipping behavior of DNA-RNA duplexes were carefully studied. In this dissertation, Chapter 4 presents unzipping studies of DNA-DNA, DNA-RNA, and DNA-PNA to demonstrate the use of aHL to distinguish between different forms of duplexes (A-and B-forms). In these studies, we built a physical 16 17 (A) (B) Blockage Current (pA) Figure 1.9. Monitoring the uracil DNA glycosylase (UDG) enzyme activity for dsDNA using an aHL channel. (A) The structure of dsDNA with a 5'-polydT24 tail within a-HL. The red box indicates the location of the uracil (U) base or the abasic site (AP). (B) A Scheme of the UDG hydrolysis reaction. (C) The blockage current difference observed for uracil (U) and abasic site (AP). model for the interaction of the DNA duplexes with the nanopore, which were tested by changing the length of the ssDNA overhang. This work provides useful insight on the interaction between dsDNA/DNA-RNA/DNA-PNA molecules and the aHL nanopore and is important in designing probes. 1.3. References (1) Wanunu, M.; Dadosh, T.; Ray, V.; Jin, J.; McReynolds, L.; Drndic, M. Nat. Nanotechnol. 2010, 5, 807. (2) Miles, B. N.; Ivanov, A. P.; Wilson, K. A.; Dogan, F.; Japrung, D.; Edel, J. B. Chem. Soc. Rev. 2013, 42, 15. (3) Venkatesan, B. M.; Bashir, R. Nat. Nano. 2011, 6, 615. (4) Vercoutere, W.; Winters-Hilt, S.; Olsen, H.; Deamer, D.; Haussler, D.; Akeson, M. Nat. Biotechnol. 2001, 19, 248. (5) Li, J.; Stein, D.; McMullan, C.; Branton, D.; Aziz, M. J.; Golovchenko, J. A. Nature 2001, 412, 166. (6) Han, J.; Craighead, H. G. Science 2000, 288, 1026. (7) Firnkes, M.; Pedone, D.; Knezevic, J.; Doblinger, M.; Rant, U. Nano Lett. 2010, 10, 2162. (8) Wanunu, M. Phys. Life. Rev. 2012, 9, 125. (9) Song, L.; Hobaugh, M. R.; Shustak, C.; Cheley, S.; Bayley, H.; Gouaux, J. E. Science 1996, 274, 1859. (10) Coulter, W. H. Means for Counting Particles Suspended in a Fluid. U.S. Patent 2,656,508, 1953. (11) Cornell, B. A.; Braach-Maksvytis, V. L. B.; King, L. G.; Osman, P. D. J.; Raguse, B.; Wieczorek, L.; Pace, R. J. Nature 1997, 387, 580. (12) Poutrel, B.; Lerondelle, C. J. Dairy Sci. 1983, 66, 2575. (13) Kasianowicz, J. J.; Brandin, E.; Branton, D.; Deamer, D. W. Proc. Natl. Acad. Sci. U. S. A. 1996, 93, 13770. (14) Siwy, Z. S. Adv. Funct. Mater. 2006, 16, 735. 18 (15) Wei, C.; Bard, A. J.; Feldberg, S. W. Anal. Chem. 1997, 69, 4627. (16) White, H. S.; Bund, A. Langmuir 2008, 24, 2212. (17) Gong, M. M.; Nosrati, R.; San Gabriel, M. C.; Zini, A.; Sinton, D. J. Am. Chem. Soc. 2015, 137, 13913. (18) Luo, L.; Holden, D. A.; White, H. S. ACS Nano 2014, 8, 3023. (19) Dekker, C. Nat. Nanotechnol. 2007, 2, 209. (20) Siwy, Z.; Fulinski, A. Phys. Rev. Lett. 2002, 89, 198103. (21) Storm, A. J.; Chen, J. H.; Ling, X. S.; Zandbergen, H. W.; Dekker, C. Nat. Mater. 2003, 2, 537. (22) Chen, P.; Mitsui, T.; Farmer, D. B.; Golovchenko, J.; Gordon, R. G.; Branton, D. Nano Lett. 2004, 4, 1333. (23) Zhang, B.; Galusha, J.; Shiozawa, P. G.; Wang, G.; Bergren, A. J.; Jones, R. M.; White, R. J.; Ervin, E. N.; Cauley, C. C.; White, H. S. Anal. Chem. 2007, 79, 4778. (24) White, R. J.; Ervin, E. N.; Yang, T.; Chen, X.; Daniel, S.; Cremer, P. S.; White, H. S. J. Am. Chem. Soc. 2007, 729, 11766. (25) Israelachvili, J. I, Intermolecular and Surface Forces, 2nd ed.; Academic Press: London, 2003. (26) Helmholtz, H. Ann. Phys. 1879, 243, 337. (27) Bard, A. J.; Faulkner, L. R. Electrochemical Methods; John Wiley and Sons: New York, 1980. (28) Gouy, M. J. Phys. Theor. Appl. 1910, 9, 457. (29) Chapman, D. L. Philos. Mag. 1913, 25, 475. (30) Baldessari, F.; Santiago, J. G. J. Colloid Interface Sci. 2008, 325, 526. (31) Cheng, L. J.; Guo, L. J. Chem. Soc. Rev. 2010, 39, 923. (32) Vilozny, B.; Wollenberg, A. L.; Actis, P.; Hwang, D.; Singaram, B.; Pourmand, N. Nanoscale 2013, 5, 9214. (33) He, H.; Xu, X.; Jin, Y. Anal. Chem. 2014, 86, 4815. (34) Deng, X. L.; Takami, T.; Son, J. W.; Kang, E. J.; Kawai, T.; Park, B. H. Sci. Rep. 2014, 4, 4005. 19 20 (35) Actis, P.; Vilozny, B.; Seger, R. A.; Li, X.; Jejelowo, O.; Rinaudo, M.; Pourmand, N. Langmuir 2011, 27, 6528. (36) Umehara, S.; Pourmand, N.; Webb, C. D.; Davis, R. W.; Yasuda, K.; Karhanek, M. Nano Lett. 2006, 6, 2486. (37) Kubeil, C.; Bund, A. J. Phys. Chem. C 2011, 115, 7866. (38) Gu, L.-Q.; Shim, J. W. Analyst 2010, 135, 441. (39) Manrao, E. A.; Derrington, I. M.; Laszlo, A. H.; Langford, K. W.; Hopper, M. K.; Gillgren, N.; Pavlenok, M.; Niederweis, M.; Gundlach, J. H. Nat. Biotechnol. 2012, 30, 349. (40) Fennouri, A.; Przybylski, C.; Pastoriza-Gallego, M.; Bacri, L.; Auvray, L.; Daniel, R. ACS Nano 2012, 6, 9672. (41) Biesemans, A.; Soskine, M.; Maglia, G. Nano Lett. 2015, 15, 6076. (42) Choi, L. S.; Mach, T.; Bayley, H. Biophys. J. 2013, 105, 356. (43) Cracknell, J. A.; Japrung, D.; Bayley, H. Nano Lett. 2013, 13, 2500. (44) Rodriguez-Larrea, D.; Bayley, H. Nat. Nanotechnol. 2013, 8, 288. (45) Boersma, A. J.; Bayley, H. Angew. Chem. Int. Ed. Engl. 2012, 51, 9606. (46) Meller, A.; Nivon, L.; Branton, D. Phys. Rev. Lett. 2001, 86, 3435. (47) Howorka, S.; Bayley, H. Biophys. J. 2002, 83, 3202. (48) Shim, J. W.; Tan, Q.; Gu, L. Q. Nucleic Acids Res. 2009, 37, 972. (49) Denissenko, M. F.; Pao, A.; Tang, M.; Pfeifer, G. P. Science 1996, 274, 430. (50) Church, G.; Deamer, D. W.; Branton, D.; Baldarelli, R.; Kasianowicz, J.; Characterization of individual polymer molecules based on monomer-interface interactions. U.S. Patent 5,795,782. Mar 17, 1995 (51) White, R. J.; Zhang, B.; Daniel, S.; Tang, J. M.; Ervin, E. N.; Cremer, P. S.; White, H. S. Langmuir 2006, 22, 10777. (52) Seed, B. Silanizing Glassware; John Wiley & Sons, Inc., 2001. (53) de Zoysa, R. S.; Krishantha, D. M.; Zhao, Q.; Gupta, J.; Guan, X. Electrophoresis 2011, 32, 3034. (54) Guy, A. T.; Piggot, T. J.; Khalid, S. Biophys. J. 2012, 103, 1028. 21 (55) Mathe, J.; Aksimentiev, A.; Nelson, D. R.; Schulten, K.; Meller, A. Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 12377. (56) An, N.; Fleming, A. M.; White, H. S.; Burrows, C. J. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 11504. (57) Schibel, A. E.; An, N.; Jin, Q.; Fleming, A. M.; Burrows, C. J.; White, H. S. J. Am. Chem. Soc. 2010, 132, 17992. (58) Pfeifer, G. P.; Besaratinia, A. Hum. Genet. 2009, 125, 493. (59) Godschalk, R. W.; Van Schooten, F. J.; Bartsch, H. J. Biochem. Mol. Biol. 2003, 36, 1. (60) Chuang, C. Y.; Tung, J. N.; Su, M. C.; Wu, B. C.; Hsin, C. H.; Chen, Y. J.; Yeh, K. T.; Lee, H.; Cheng, Y. W. Arch. Oral. Biol. 2013, 58, 102. (61) Aksimentiev, A.; Schulten, K. Biophys. J. 2005, 88, 3745. (62) Henrickson, S. E.; DiMarzio, E. A.; Wang, Q.; Stanford, V. M.; Kasianowicz, J. J. J. Chem. Phys. 2010, 132, 135101. (63) Sauer-Budge, A. F.; Nyamwanda, J. A.; Lubensky, D. K.; Branton, D. Phys. Rev. Lett. 2003, 90, 238101. (64) Jin, Q.; Fleming, A. M.; Burrows, C. J.; White, H. S. J. Am. Chem. Soc. 2012, 134, 11006. (65) Viasnoff, V.; Chiaruttini, N.; Bockelmann, U. Eur. Biophys. J. 2009, 38, 263. (66) Zhang, X.; Wang, Y.; Fricke, B. L.; Gu, L. Q. ACS Nano 2014, 8, 3444. (67) Schibel, A. E. P.; Fleming, A. M.; Jin, Q.; An, N.; Liu, J.; Blakemore, C. P.; White, H. S.; Burrows, C. J. J. Am. Chem. Soc. 2011, 133, 14778. (68) Jin, Q.; Fleming, A. M.; Johnson, R. P.; Ding, Y.; Burrows, C. J.; White, H. S. J. Am. Chem. Soc. 2013, 135, 19347. (69) Jin, Q.; Fleming, A. M.; Ding, Y.; Burrows, C. J.; White, H. S. Biochemistry 2013, 52, 7870. (70) Johnson, R. P.; Fleming, A. M.; Burrows, C. J.; White, H. S. J. Phys. Chem. Lett. 2014, 5, 3781. (71) Johnson, R. P.; Fleming, A. M.; Jin, Q.; Burrows, C. J.; White, H. S. Biophys. J. 2014, 107, 924. CHAPTER 2 EFFECT OF THE ELECTRIC DOUBLE LAYER ON THE ACTIVATION ENERGY OF ION TRANSPORT IN CONICAL NANOPORES 2.1 Introduction This chapter presents a fundamental study of ion transport in conical-shaped nanopores using the activation energy (E a) of ion transport as a parameter. The results highlight the relationships between the distribution of ions with the nanopore, ionic current, and Ea, and their dependence on pore size, temperature, ion concentration, and applied voltage. Ion transport phenomena in nanopores have attracted significant attention over the past 15 years in the context of the development of nanoscale resistive pulse sensors1-8 and in fundamental studies of transport in confined spaces.9-15 The push towards high performance energy storage devices also requires a fundamental understanding of ion transport phenomena near charged surfaces on the nanometer length scale.16-23 Typically, nanopores are constructed within polymers, glass, or silica, and have internal and external surfaces that are electrically charged. In an electrolyte, these charged surfaces attract counter ions from solution, resulting in the creation of an electric double layer (EDL) that screens the surface charge over a distance that is dependent primarily on the electrolyte concentration, but also the temperature. At the nanoscale domain, where the length of the EDL is comparable to that of the pore diameter, the consequences of the EDL on the ion transport, and subsequently the current-voltage (i-E) characteristics, are significant, with non-linear i-E behavior typically observed.24-34 However, the effect of temperature on ion transport in nanoscale domains has not been investigated in detail with the exception of the recent work published by Taghipoor and coworkers.35 These researchers examined the temperature dependence of ion transport through uniform (35 nm height) nanochannels. In this chapter, we examine how the presence of EDLs inside a conical glass nanopore affect the temperature dependence of ion transport, and by extension, the energy required for electrolyte transport through the nanopore. In any electrolyte solution, the movement of an ion by diffusion and migration along a potential gradient is associated with an activation energy, EA. We experimentally measured EA for electrolyte transport through conical glass nanopores of several pore sizes and varying electrolyte concentrations, as schematically shown in Figure 2.1. In conjunction with our experiments, we have performed a systematic series of finite element simulations to compute EA in conical nanopores based on the Poisson- Nernst-Planck model for ion transport. We show that at low electrolyte concentrations and in small pores, the confined geometry alters EA for electrolyte transport relative to that in bulk solution. In agreement with expectations based on the Gouy-Chapman theory,36 our simulations show that the electric potential profile perpendicular to the pore walls is strongly dependent on the temperature, resulting in significant changes in the ion concentration profile as the temperature is altered. This temperature dependence results in unusual potential-dependent values of EA for ion transport through a nanopore. 23 24 (a) (+) Ag/AgCI Internal solution On Thermocouple -Xr Ag/AgCI External solution © (b) 2 0 < S-2 -4 -6 \ Increasing T 10 - 35 °C - 0.4 - 0.2 0.0 0.2 0.4 E vs. Ag/AgCI (V) Figure 2.1. (a) Experimental design for measuring the temperature- and electrolyte concentration-dependent i-E response for conical glass nanopores. KCl electrolyte is placed both inside the capillary and in the external reservoir, and a voltage applied across the two Ag/AgCl electrodes. Temperature is controlled by a Peltier heater/cooler, and measured via a K-type thermocouple. (b) Representative i-E curves recorded at a scan rate of 10 mV s-1 and temperatures between 10 and 35 °C for a 35 nm pore in 0.1 mM KCl electrolyte. E refers to the potential of the Ag/AgCl electrode in the internal solution relative to the external solution. 2.2 Experimental Section 2.2.1 Chemicals and Materials All solutions were prepared using water (18.1 MQ cm) obtained from a Banstead E-pure system. KCl (99.8%, Fisher Scientific) was used without further purification. All electrolyte solutions were filtered using a 0.22 |im filter (Millipore, Inc.) prior to use. 2.2.2 Glass Nanopore Membrane (GNM) Fabrication Glass nanopore membranes were prepared from soda-lime glass capillaries (Dagan) as previously reported by our lab.37 The half-cone angle of the glass nanopore was estimated by optical microscopy and found to be 10.0 ± 1.5°. The radius of the orifice was determined from the ionic resistance (R) of the pore in 1 M KCl solution at a temperature of 25 °C using the expression r = 18.5/R.26 Four nanopores with radii of 20 nm, 35 nm, 56 nm, and 2000 nm were used. The uncertainty in the radii is approximately ~10 %. 2.2.3 Cell Configuration and Data Acquisition Current-voltage (i-E) curves were recorded using a CH Instruments 1030A potentiostat (Austin, TX). One Ag/AgCl Electrode was placed inside the capillary and the other (taken to be ground in these experiments) was placed in the external electrolyte reservoir. The voltage was scanned between 0.5 V and -0.5 V, starting from 0.5 V, at a scan rate of 10 mV s-1. Temperature control was achieved by a Peltier heater/cooler (custom PID control of a small-scale thermoelectric cooler (CUI Inc., CP20151)) situated directly below the cell, while the temperature was measured by a K-type thermocouple immersed in the solution of the external reservoir. i-E curves were recorded at 25 temperatures between 10 and 45 °C (results for temperatures between 10 and 35 °C are reported in the main text, the full range is considered in the Figure S2.1), and in electrolyte concentrations of 0.1 to 50 mM (with the 1 M solution used only for measuring nanopore sizes). The temperature is measured by a thermocouple (0.1 °C precision) residing in the same solution as the pore (volume 350 |iL) which has a lid on top. The distance between the pore orifice and the thermocouple is <2 mm, which will lead to minimal temperature differences between them. These components lie within a much larger polycarbonate block (~1 inch cube). The relatively large thermal mass of the block results in stable temperatures of the solution. A thermal equilibrium is reached in ~2 min and each measurement reported herein was taken after standing a minimum of 5 min to equilibrate, after which the solution (internal and external) and thermocouple will be at the same temperature. 2.2.4 Computational Analysis and Simulations Finite element simulations were performed with COMSOL Multiphysics 4.3 (COMSOL Inc.) on a desktop computer (Intel Core i7 CPU with 8 GB RAM). Details of the boundary conditions and meshing are presented later. 2.3 Results and Discussion 2.3.1 Experimental Activation Energies Using the experimental configuration shown in Figure 2.1(a), the current-voltage (i-E) response of conical glass nanopores of 20, 35, 56, and 2000 nm radius were recorded as a function of both temperature (10 - 35 °C) and KCl concentration (0.1 - 50 mM). The observed non-ohmic i-E response (Figure 2.1b), or ion-current rectification (ICR), is more 26 prominent for lower electrolyte concentrations and smaller pores. ICR occurs in asymmetric charged nanopores and is due to the generation of an EDL at the interface between the nanopore walls and the electrolyte, and can be explained by an ion accumulation-depletion model.38-40 Due to the conical shape of the nanopore, when a negative potential is applied to the interior of the nanopore relative to the exterior, positively charged K+ ions move from the bulk solution towards the interior of the pore while negatively charged Cl- ions move in the opposite direction. However, since the pore wall is negatively charged, K+ accumulate at the pore orifice while the transit of Cl- ions is partially blocked, resulting in an accumulation of both ions and a conductance higher than the bulk value. Conversely, when a positive voltage is applied to the interior of the pore, Cl- ions will be rejected by the pore and ions will be depleted at the pore orifice giving rise to a lower conductivity. Accumulation of ions may still occur when a pore is geometrically symmetric; however, the i-E response also is symmetric.41-44 From the individual i-E traces, we extracted values of the current between -0.5 and 0.5 V as a function of both KCl concentration and temperature. At negative voltages, a slight hysteresis of ~2% current (at -0.35 V) is observable between the forward and backward scan and thus the average of the current on the forward and backward scans was used. The current is proportional to the flux of the ions through the pore (diffusion and migration), and there is an energy penalty associated with moving ions through the nanopore. This activation energy (EA) is related to the measured current at a given voltage by the Arrhenius equation: 27 i = A exp E- RT (2.1) where R is the gas constant, T is the temperature and A is the pre-exponential factor. The resulting Arrhenius plots are linear (Figure 2.2a and b and S2.2) over the range 10 - 35 °C. However, at temperatures > 35 °C, slight nonlinear behavior becomes apparent (see Figure S2.1); only data between 10 - 35 °C were used in the analysis presented here. Values of Ea measured as a function of electrolyte concentration and pore size are shown in Figures 2.2c and 2.2d. For the 2000 nm pore, Ea does not show any significant change with the electrolyte concentration above 1 mM KCl and is equal to 13.4 ± 0.3 kJ mol-1, comparable to Ea (13.5 ± 0.2 kJ mol-1) based on temperature-dependent conductance values reported for bulk solution.45 As the pore diameter is reduced, the charge on the nanopore surface begins to influence ion transport, resulting in Ea values different from that observed in bulk solution. The effect is greatest when small nanopores are employed in conjunction with low electrolyte concentrations. At positive potentials, as shown in Figure 2.2c, Ea increases above the value observed in bulk solution, whereas at negative potentials, as shown in Figure 2.2d, a decrease in Ea is observed. This dependence of Ea on the voltage is discussed below. A similar dependence on Ea on the electrolyte concentration and at different voltages for a 20 nm radius nanopore is presented in the S2.10. 2.3.2 Finite Element Simulations Finite element simulations were used to understand how the nanopore geometry influences Ea for ion transport relative to that observed in bulk solution, and to explain the dependence of Ea on the applied potential, nanopore radius, and KCl concentration. The 28 29 (a) (b) -22.2 -22.4 < -22.6 -22.8- c -23.0 -23.2 -23.4 -23.6- +0.35 V -22.2 35 nm pore £5 -22.4 1 mM KCl ^ -22.6 -22.8 £ -23.0- -0.35 V -23.2 35 nm pore -23.4 1 mM KCl -23.6 ---- '----- 1----- '----- 1----- '----- 1----- '-- 0.0033 0.0034 0.0035 -ii (C) , p 20.0 OE 18.0 -> 16.0 , < LU 14.0 12.0 1 /7 (K ) +0.35 V ' 20 nm 35 nm 56 nm 2000 nm 10 20 30 40 [K C l] (mM) 50 O E LLl 0.0033 0.0034 0.0035 1 /7 (K '1) (d) 13.0 12.0 11.0 10.0 9.0 8.0 ^ * -0.35 V i f ■ 20 nm • 35 nm i A 56 nm i ▼ 2000 nm 10 20 30 40 [K C l] (mM) 50 Figure 2.2. Determining the activation energy of electrolyte transport through a conical glass nanopore. (a,b) Representative Arrhenius plots constructed from ionic currents measured at +0.35 V and -0.35 V for a 35 nm pore in 1 mM KCl electrolyte. (c,d) Ea as a function of both KCl concentration and pore size at +0.35 V and -0.35 V, respectively. Error bars, representing the standard deviation of 3 repeated current recordings and the standard error arising from the least squares fit to ln (|i|) - T l data are smaller than the data points used to present experimental Ea values in this paper. ion fluxes from each simulation were integrated over the boundaries of the simulation domain to compute the ionic current at different temperatures. An Arrhenius analysis of the dependence of the simulated current on temperature was used to compute values of Ea for comparison to experimental data. The 2D axisymmetric finite element model used to simulate the ionic current in conical nanopores is shown in Figure 2.3. A fine mesh is necessary to resolve the EDL potential and ion distribution on the charged walls. As such, the mesh size was fixed at 0.35 nm on the charged walls near the pore orifice (boundaries EF and FG in Figure 2.3), which is ~1/4 of the reciprocal Debye length (~1.3 nm) for the conditions giving the smallest double layer (50 mM, 10 °C). All other meshing parameters were left at the default values defined in the software for a ‘normal' mesh. We confirmed that our mesh was sufficiently fine by observing that a mesh consisting of elements of half the size gave no change in the solution. The half-cone angle of the glass nanopore was estimated by optical microscopy and found to be 10.0 ± 1.5°. Since the cone angle can vary from pore-to-pore by 15%, we also conducted simulations over a range of cone-angles between 8-12° while keeping the electrolyte concentration and voltage constant (see Figure S2.3). Calculated values of Ea were found to vary by less than 2% over this range of angles. The surface charge of the pore wall was kept constant at -2 mC/m2 for all simulations, which is reasonable for the range of electrolyte concentrations used (0.02 mM - 50 mM) at pH 7 0 . 39,40,45-47 Changes in surface charge density as a function of electrolyte concentration were not included as they have been shown to be negligible tor the range of experimental conditions used (i.e., <0.1 M, 0 - 35 °C); however, some dependence of the 30 31 Applied potential=Vap Symmetry axis C - Cbulk Figure 2.3. The 2D axisymmetric finite element model used to simulate the current through a conical pore as a function of temperature, voltage, and ion concentration. The boundary conditions and mesh are shown in the figure. A potential is applied to IJ while AB is held at zero potential. The electrolyte concentration at AB and IJ is maintained at the bulk value. A surface charge of -2 mC/m2 is applied to EF and FG and a finer mesh size is used close to the pore orifice. The boundaries far from the pore orifice (GH, ED and CD) are set at zero charge, which does not appreciably affect the calculated currents. This allows for a coarser mesh in these regions, as no double-layer need be resolved. AJ is the symmetry axis. The total normal ion flux (mol/m2s) at the semi-infinite boundary AB was computed by an integration and converted to the ionic current (C/s) by multiplying by Faraday's constant (96500 C/mol). The inset shows the expanded area of the pore mouth. surface charge on electrolyte concentration would be expected at higher concentrations and temperatures.35 A recent study by Taghipoor et al. has reported the dependence of surface charge of silicon dioxide with temperature (~16 mC/m2 in 0.1m M KCl at 25 °C).35 An equation for the dependence of the surface charge on temperature was extracted from values reported in this work to study the effect of temperature on Ea. However, Ea values calculated by this method are not in reasonable agreement with the experimental values (see Figure S2.4). These findings suggests that the glass nanopores used in these experiments do not bear a surface charge as high as -16 mC/m2 (at 25 °C) and/or the dependence of the surface charge on temperature is not as significant as reported for silicon dioxide by Taghipoor et al. All simulations are based on solving the coupled Poisson and Nernst-Planck equations, which describe the local ion distributions and the ion fluxes, respectively, as a function of the local electrical potential, O. The ionic flux, J, of species i (K+ or Cl-) is given by: J t = - D tV c i - z i jujF c j V O (2.2) where Dj, Cj, and Zj are the diffusivity, concentration, and charge of ionic species, respectively, and jUj represents the ionic mobility. T, F, and R are the temperature, Faraday's constant, and the gas constant, respectively. The relationship between the local ion distributions and potential is given by: 32 v 2 o = - F y ztct £ j (2.3) where s = 78 is the dielectric constant of the medium. The use of the Nernst-Planck (NP) equation to describe ion fluxes is reasonable when the applied pressure is zero and the electro-osmotic flow is negligible, conditions that match those used in this experiment.48-51 Solving the coupled Poisson and Nernst-Planck equations as a function of temperature requires that Di for K+ and Cl- are known as a function of temperature. We used the Stokes-Einstein relationship to estimate the ion diffusion coefficients: D = k J Di ~ 7 ---- (2 4) 33 where kB is the Boltzmann constant, n(T) represents the dynamic viscosity of the medium and ri is the radii of the ions. Since r/(T) is temperature dependent, literature values were used to construct an expression that accurately describes the dependence of r/(T) over the 10 - 35 °C range studied53. Details of this calculation are presented in the S2.5. Representative simulated Arrhenius plots of ln|i| vs T x are shown in Figure 2.4a and b. Additional Arrhenius plots at different concentrations are given in Figure 2.6. For the experimental Arrhenius plots (Figures 2.2a and b), we observe a linear trend within this temperature range. Linear fits to these plots were used to calculate Ea for electrolyte transport. This process was repeated for a range of electrolyte concentrations, pore sizes and potentials, chosen to match the experimental conditions for direct comparison. Calculated values of Ea are displayed in Figure 2.4c and d. The trends for simulated activation energies (Figure 2.4c and d) are in qualitative agreement with the experimental values (Figure 2.2c and d); activation energies increased 34 Figure 2.4. Simulated activation energies of electrolyte transport through a conical glass nanopore. (a,b) Representative Arrhenius plots constructed for currents measured at +0.35 V and -0.35 V for a 35 nm pore in 1 mM KCl electrolyte, from which Ea are calculated. (c,d) Simulated Ea as a function of both KCl concentration and pore size at +0.35 V and - 0.35 V, respectively. See Figure 2.2 for a direct comparison to the experimental Arrhenius plots and values of Ea for the same conditions. at positive external potentials and decreased at negative potentials. The slight difference in the values of experimental and simulated values of Ea at higher concentrations (13.4 ± 0.1 kJ mol-1 and 14.7± 0.1 kJ mol-1) arises because the diffusion coefficient depends on concentration as well as temperature.52 The simulations were performed using the temperature-dependent but concentration-independent diffusion coefficient (limit at infinite dilution, see Figure S2.5). The limiting value of Ea at high concentration derived from simulations is precisely as expected when compared to the experimental values at infinite dilution, upon which the model is based, thus confirming correct implementation of the model. 2.3.3 Dependence of Activation Energy as a Function of Applied Voltage Both experimental and simulated values of Ea as a function of voltage for a 20 nm pore are given in Figure 2.5. In both the experiments and the simulations, Ea decreases relative to bulk values at negative potentials and increases relative to bulk values at positive potentials. The resistance on the nanopore is determined by the solution at of the pore orifice and it is the potential- and temperature-dependent concentration distribution within this region that determines the dependence of Ea on the applied voltage. For low concentrations of KCl and small pores, where the deviation of activation energy from its value in bulk solution is largest, the concentrations of K+ and Cl- in the vicinity of the nanopore orifice are strongly dependent on the polarity of the applied potential. This is shown in Figure 2.6 through color distribution plots of the concentrations of K+ and Cl- for a 20 nm radius pore with 0.1 mM KCl at +0.35 V and -0.35 V. At both positive and negative applied voltages, 35 36 (a) 22.0 T- 20.0 1 o 18.0 E 16.0 " 5 14.0 - - lu 12.0 10.0 8.0 17.0 T-1 16.0 o E 15.0 £ 14.0 . < LU 13.0 12.0 ' --■- ■ -- •- . A 0.1 mM f 1 mM / 50 mM A . A 4 * A A Experiment -0.4 -0.2 0.0 0.2 0.4 E (V) (b) -«-0.1 mM 1 mM a 50 mM * S? ■y Simulated -0.4 -0.2 0.0 0.2 E (V) 0.4 Figure 2.5. Experimental (a) and simulated (b) activation energies for a 20 nm pore for different KCl concentration as a function of applied potential. The error bars are smaller than the data points. 37 Figure 2.6. Concentration profiles of K+ and Cl- ions in the vicinity of a 20 nm nanopore orifice at (a) +0.35 V and (b) -0.35 V (relative to the internal electrode). The surface charge on the wall is -2 mC/m2 and the bulk concentration is 0.1 mM at 25 °C. Note, different color scales are used for the two species. Plots for different pore sizes and cross-sectional plots of the species concertation are shown in Figure S2.7. the K+ concentration close to the pore opening is significantly higher than its value in bulk solution (0.1 mM) due to accumulation of the cation at the negatively charged pore wall. A slight inward shift and spreading of the zone of high K+ concentration at the pore mouth is observed when changing from positive to negative applied potentials. Conversely, the concentration of Cl- ions is lowest on the pore walls; this is due to electrostatic repulsion from the negative surface charge. At positive applied potentials the concentration of Cl- within the pore is less than that in bulk solution (~half its value in bulk solution for the conditions shown), whereas at the negative applied potentials an accumulation of Cl- in the center of the pore is observed (~3.5 enhancement for these conditions) as they are simultaneously repelled by the charged walls and the applied potential. However, it is important to note that these Cl- concentrations are smaller than those of the K+ ion. The total concentration of K+ and Cl- at the vicinity of the pore orifice is also shown in Figure 2.6. From this plot we can observe that the total ion concentration is greater for negative applied potentials compared to positive potentials for all but a small region from z = 0 to -15 nm. The comparatively higher concentration of ions when a negative potential is applied results in to the higher current magnitudes at negative potentials, for example, ICR, which can also be observed experimentally in the i-E curves of Figure 2.1b. Ea values are calculated directly from the current measured at different temperatures. Figure 2.7 shows plots of the simulated total ionic concentration at +0.35V and - 0.35V in the vicinity of a 20 nm nanopore orifice at two different temperatures. At -0.35 V, the ionic concentration in the vicinity of the pore orifice decreases as the temperature is increased from 10 °C to 35 °C. This can be seen in the concentration plot (Figure 2.7a), 38 39 (a) -0.35 V (b) +0.35 V 20 nm Figure 2.7. The total ionic concentration at the orifice of a 20 nm nanopore at 0.1 mM KCl, at 10 and 35 °C and for (a) -0.35 V and (b) +0.35 V applied potential. Contours have been added at 1.0 and 1.3 mM to aid interpretation. where contours have been added to aid in this interpretation. The opposite situation is observed at positive potentials, where an increase in ion concentration at the pore orifice is observed with increasing temperature. The activation energy for transport through the pore describes the rate of change of current with temperature. The current, in turn, is affected by two factors, (i) the ion mobility, and (ii) the ion concentration. The increase in the mobility of the ions with increasing temperature is the main driving factor in the current increasing when the temperature is increased from 10 °C to 35 °C, and is the sole factor in bulk solution, where ion concentrations remain constant. If the current scaled solely by the increase in mobility we would expect the calculated Ea would be precisely the value for bulk solution (14.7 kJ mol-1). Instead the change in nanopore current as a function of temperature is effectively modulated by the concentration of the ions at the vicinity of the nanopore orifice (where most of the resistance drop occurs). The lower concentration of ions with increasing temperature observed at negative voltages leads to a lower conductivity, a lower current, and hence, a value of Ea = 12.7 kJ mol-1 that is below that for bulk solution. Implicit in this reasoning is that the electric field is largely unchanged with changing temperature, as is shown in the Figure S2.9 At an applied potential of +0.35 V the ionic concentration within the nanopore increases as the temperature is increased, as seen in Figure 2.7b. The spreading of the region of higher concentration in the center of the pore can be visualized most easily by the 1.0 mM contour in the color plot (Figure 2.7b). In contrast to the situation with a negative applied potential, the increased concentration within the pore leads to an increase in current as the temperature is increased over and above that predicted by purely 40 considering the increase in mobility with temperature and explains the Ea=15.7 kJ mol-1 being greater than the value of Ea in bulk solution. Line graphs of total ion concentration at different temperatures inside a 20 nm nanopore is shown in Figure 2.8. For larger pore geometries and/or higher electrolyte concentrations the Debye length represents a smaller proportion of the pore radius. The deviation of ion concentration from the bulk is confined closer (relatively) to the pore walls and as it is this deviation that is the source of the behaviors discussed above, activation energies in large pores and at high electrolyte concentrations remain similar to those in the bulk. 2.4 Conclusions The apparent activation energies of ion transport through conical-shaped glass nanopores of varying size and over a range of electrolyte concentrations have been measured experimentally. At low electrolyte concentrations (< 50 mM) and small pore radii (< ~100 nm), Ea deviates from the value observed in bulk, increasing at positive applied potentials and decreasing at negative potentials. Finite element simulations support our interpretation that the change in the Ea can be explained by changes in the ion concentration profile within the pore as a function of temperature and voltage. At negative potentials, the ion concentration inside the pore decreases as a function of temperature, moderating the mobility-driven increase in conductivity with temperature, and hence decreasing the activation energy relative to bulk solution. At positive potentials, the rate of change in conductivity with temperature is enhanced by an increase in ion concentration within the pore as a function of temperature. These two dependencies suggest that the 41 42 (a) -0.35 V (b) +0.35 V ^ ■[ 2 - ~ z ^xis Pore wall -0.8 ► " 0.8- Pore wall ^ 0 5 10 15 20 r (nm) 0 5 10 15 20 r (nm) Figure 2.8. The total ionic concentration at the orifice of a 20 nm nanopore at 0.1 mM KCl, at 10 °C and 35 °C. (a) Radial cross section plot taken inside the pore at z = -20 nm at - 0.35 V and (b) +0.35 V. Individual concentration profiles along z axis are given in Figure agreement with the expectation that the effect of the electrostatic fields in determining ion distributions is reduced at higher temperatures. These effects may be applied to conical nanopores if their surface charge and geometry are fixed in the temperature range studied. (1) Lan, W. J.; Holden, D. A.; Liu, J.; White, H. S. J. Phys. Chem. C 2011, 115, 18445. (2) DeBlois, R. W.; Wesley, R. K. J. Virol. 1977, 23, 227. (3) Sexton, L. T.; Mukaibo, H.; Katira, P.; Hess, H.; Sherrill, S. A.; Horne, L. P.; Martin, C. R. J. Am. Chem. Soc. 2010, 132, 6755. (4) Jin, P.; Mukaibo, H.; Horne, L. P.; Bishop, G. W.; Martin, C. R. J. Am. Chem. Soc. 2010, 132, 2118. (5) Han, J.-H.; Kim, K. B.; Kim, H. C.; Chung, T. D. Angew. Chemie. Int. Ed. 2009, 48, 3830. (6) Umehara, S.; Pourmand, N.; Webb, C. D.; Davis, R. W.; Yasuda, K.; Karhanek, M. Nano Lett. 2006, 6, 2486. S2.8 2.5 References (7) Siwy, Z.; Trofin, L.; Kohli, P.; Baker, L. A.; Trautmann, C.; Martin, C. R. J. Am. Chem. Soc. 2005, 127, 5000. (8) Wanunu, M.; Morrison, W.; Rabin, Y.; Grosberg, A. Y.; Meller, A. Nat. Nanotecnol. 2010, 5, 160. (9) Pu, Q.; Yun, J.; Temkin, H.; Liu, S. Nano Lett. 2004, 4, 1099. (10) Van der Heyden, F. H.; Stein, D.; Dekker, C. Phys. Rev. Lett. 2005, 95, 116104. (11) Plecis, A.; Schoch, R. B.; Renaud, P. Nano Lett. 2005, 5, 1147. (12) Pennathur, S.; Santiago, J. G. Anal. Chem. 2005, 77, 6772. (13) Pennathur, S.; Santiago, J. G. Anal. Chem. 2005, 77, 6782. (14) Baldessari, F.; Santiago, J. G. J. Nanobiotechnol. 2006, 4, 12. (15) Hibara, A.; Saito, T.; Kim, H.-B.; Tokeshi, M.; Ooi, T.; Nakao, M.; Kitamori, T. Anal. Chem. 2002, 74, 6170. (16) Balke, N.; Kalnaus, S.; Dudney, N. J.; Daniel, C.; Jesse, S.; Kalinin, S. V. Nano Lett. 2012, 12, 3399. (17) Okubo, M.; Tanaka, Y.; Zhou, H.; Kudo, T.; Honma, I. J. Phys. Chem. B 2009, 113, 2840. (18) Rolison, D. R.; Long, J. W.; Lytle, J. C.; Fischer, A. E.; Rhodes, C. P.; McEvoy, T. M.; Bourg, M. E.; Lubers, A. M. Chem. Soc. Rev. 2009, 38, 226. (19) Long, J. W.; Dunn, B.; Rolison, D. R.; White, H. S. Chem. Rev. 2004, 104, 4463. (20) Gowda, S. R.; Reddy, A. L. M.; Shaijumon, M. M.; Zhan, X.; Ci, L.; Ajayan, P. M. Nano Lett. 2011, 11, 101. (21) Xiong, J.; Chen, Q.; Edwards, M. A.; White, H. S. ACS Nano 2015, 9, 8520 (22) Taghipoor, M.; Bertsch, A.; Renaud, P. Phys. Chem. Chem. Phys 2015, 17, 4160. (23) James, T.; Kalinin, Y. V.; Chan, C. C.; Randhawa, J. S.; Gaevski, M.; Gracias, D. H. Nano Lett. 2012, 12, 3437. (24) Wei, C.; Bard, A. J.; Feldberg, S. W. Anal. Chem. 1997, 69, 4627. (25) Siwy, Z. S. Adv. Funct. Mater. 2006, 16, 735. (26) White, R. J.; Zhang, B.; Daniel, S.; Tang, J. M.; Ervin, E. N.; Cremer, P. S.; White, H. S. Langmuir 2006, 22, 10777. 43 44 (27) Sa, N.; Baker, L. A. J. Am. Chem. Soc. 2011, 133, 10398. (28) Xia, F.; Guo, W.; Mao, Y.; Hou, X.; Xue, J.; Xia, H.; Wang, L.; Song, Y.; Ji, H.; Ouyang, Q.; Wang, Y.; Jiang, L. J. Am. Chem. Soc. 2008, 130, 8345. (29) Guerrette, J. P.; Zhang, B. J. Am. Chem. Soc. 2010, 132, 17088. (30) Vlassiouk, I.; Kozel, T. R.; Siwy, Z. S. J. Am. Chem. Soc. 2009, 131, 8211. (31) Yameen, B.; Ali, M.; Neumann, R.; Ensinger, W.; Knoll, W.; Azzaroni, O. J. Am. Chem. Soc. 2009, 131, 2070. (32) Cheng, L.-J.; Guo, L. J. Chem. Soc. Rev. 2010, 39, 923. (33) Kosinska, I. D. J. Chem. Phys. 2006, 124, 244707. (34) Siwy, Z.; Heins, E.; Harrell, C. C.; Kohli, P.; Martin, C. R. J. Am. Chem. Soc. 2004, 126, 10850. (35) Taghipoor, M.; Bertsch, A.; Renaud, P. ACS Nano 2015, 9, 4563. (36) Bard, A. J.; Faulkner, L. R. Electrochemical Methods, Fundamentals & Applications, 2nd ed.; John Wiley & Sons: New Jersey, 2001. (37) Zhang, B.; Galusha, J.; Shiozawa, P. G.; Wang, G.; Bergren, A. J.; Jones, R. M.; White, R. J.; Ervin, E. N.; Cauley, C. C.; White, H. S. Anal. Chem. 2007, 79, 4778. (38) White, H. S.; Bund, A. Langmuir 2008, 24, 2212. (39) Kubeil, C.; Bund, A. J. Phys. Chem. C 2011, 115, 7866. (40) Lan, W.-J.; Kubeil, C.; Xiong, J.; Bund, A.; White, H. S. J. Phys. Chem. C 2014, 118, 2726. (41) Hou, X.; Liu, Y.; Dong, H.; Yang, F.; Li, L.; Jiang, L. Adv. Mater. 2010, 22, 2440. (42) Cheng, L.-J.; Guo, L. J. Nano Lett. 2007, 7, 3165. (43) He, Y.; Gillespie, D.; Boda, D.; Vlassiouk, I.; Eisenberg, R. S.; Siwy, Z. S. J. Am. Chem. Soc. 2009, 131, 5194. (44) Karnik, R.; Duan, C.; Castelino, K.; Daiguji, H.; Majumdar, A. Nano Lett. 2007, 7, 547. (45) Wu, Y. C. K., W. F. Feng, D. Holland, L. A.; Juhasz, E.; Arvay, E. Tomek, A J. Res. Natl. Inst. Stand. Technol. 1994, 99, 241. (46) Behrens, S. H.; Grier, D. G. J. Chem. Phys. 2001, 115, 6716. (47) Sa, N.; Lan, W.-J.; Shi, W.; Baker, L. A. ACS Nano 2013, 7, 11272. (48) Ai, Y.; Zhang, M.; Joo, S. W.; Cheney, M. A.; Qian, S. J. Phys. Chem. C 2010, 114, 3883. (49) Daiguji, H.; Yang, P.; Majumdar, A. Nano Lett. 2003, 4, 137. (50) Vlassiouk, I.; Siwy, Z. S. Nano Lett. 2007, 7, 552. (51) Kubeil, C.; Bund, A. J. Phys. Chem. C 2011, 115, 7866. (52) Handbook of Chemistry and Physics; 95th ed.; CRC Press: Florida, 2014. (53) Bakajin, O.; Noy, A.; Fornasiero, F.; Park, H. G.; Holt, J.; Kim, S. Membranes with functionalized carbon nanotube pores for selective transport. US8940173. Dec 10, 2009. (54) Robinson, R. A.; Stokes, R. H. Electrolyte Solutions; 2nd ed.; Dover Publications: London, 2002. (55) Swindells, J. F.; Coe, J. R.; Godfrey, T. B. J. Appl. Phys. 1944, 15, 625. 45 46 S2.6 Supplemental Material S2.6.1 Arrhenius Plot over an Extended Range of Temperature Figure S2.1. A sample Arrhenius plot (ln |i| vs T"1) for values of i measured at 0.35 V for a 65 nm pore at 100 mM KCl from 10 °C to 45 °C showing the nonlinearity of the curve over a wider temperature range. Nonlinear behavior was observed for other pores and concentrations. The linear range 10 °C to 35 °C was chosen for all the experiments. 47 S2.6.2 Arrhenius Plots for a 35 nm Radius Pore at Different KCl Concentrations < c -23.8 -24.0 -24.2 -24.4 -24.6 -23.4 -23.6 -23.8 -24.0 -22.4 22.6 22. 8 ­- 23.0 - 22.0 - 22.2 -22.4 - 22.6 V 0.1 mM KCl _ -23.4 0.2 mM KCl < -23.6 - -23.8 N . -24.0 0.0033 0.0034 0.0035 1/r (K1) 0.0033 0.0034 0.0035 1/7 (K‘1) -23.2 3 -23.4 = -23.6 c -23.8 0.0033 0.0034 0.0035 0.0033 0.0034 0.0035 1/7 (K1) 1/7 (K1) V 5.0 mM KCl N. 10 mM KCl -21.4 < N . "T -21.6 N. C -21.8 ■ 0.0033 0.0034 0.0035 i/r (K '1) 0.0033 0.0034 0.0035 4 /t -20.2-j < -20.4 C -20.6 -20.8- ■ 0.0033 0.0034 0.0035 1/r (K1) 0.0033 0.0034 0.0035 1/7 (K'1) Figure S2.2. Arrhenius plots (ln |i| vs T 1) for values of i measured at 0.35 V for a 35 nm pore. The above Arrhenius plots were used to extract the activation energy values presented in Figure 2.2 of the main text, at +0.35 V for a 35 nm pore. 48 The half-cone angles of the pores were determined by optical microscopy and were estimated to be 10 ±1.5°. The radii of the pores were then determined from i-E measurements as described in our previous work, assuming a half-cone angle of 10°.26 Uncertainty in the half-cone angle gives rise to uncertainty in the pore radius. Figure S2.3b plots pairs of radius and half-cone angle that all give a resistance of 9.18 MQ (9.18 MQ corresponds to the measured resistance of a 20 nm pore of a cone angle 10° in 0.1 mM KCl at 25 °C). In this example, a half-cone angle of 10° gives a radius of 20 nm. Figure S2.3c shows the simulated activation energies for several radius/half-cone angle pairs. (only the half-cone angle is labelled). The ±1.5° uncertainty in the half-cone angle gives an uncertainty of 0.3 kJ/mol the simulated activation energies, indicating this is not a significant source of uncertainty. S2.6.3 Assessment of the uncertainty in simulated activation energy due to measurement error in the half-cone angle Figure S2.3. (a) An optical microscopy image of the glass nanopore. The image is roughly 40 times enlarged of the actual size (b) radius vs half-cone angle pairs that corresponding to a nanopore with a 9.18 MQ resistance and (c) the variation in activation energy as a function of half-cone angle in 0.1 mM at 0.35 V. 49 S2.6.4 Simulated Activation Energies as a Function of Voltage An equation (-0.0002044 x (7[°C]) -0.010469) for the dependence of the surface charge on temperature was extracted from values reported by Taghipoor et al.J5 to study the effect of temperature on surface charge. Activation energies calculated as a function of voltage assuming surface charge using this relation are shown in Figure S2.4 (b) as a black line. When compared with the red line that represents experimental data for the same conditions (20 nm pore, 0.1 mM KCl), we see the curve is offset considerably (~7 kJ mol" 1, at 0 V). A difference in the magnitude of the change in Ea with E is observed as well, although the trend of increasing Ea values with increasing potential is observed. Figure S2.4. (a) The surface charge density as a function of temperature. (b) Simulated activation energies for a 20 nm pore for 0.1 mM KCl as a function of applied potential using temperature dependent surface charge (black line) and experimental data (red line). The Arrhenius plot were linear for the between 10 °C - 35 °C. S2.6.5 Temperature Dependence of Diffusion Coefficients Values of the ion diffusion coefficients (Di) as a function of temperature were estimated using the Stokes-Einstein relationship (2.4) Dl =----k-BJ------ (2.4) 6 nr/ (T)rt where kB is the Boltzmann constant, T is absolute temperature, y(T) represents the dynamic viscosity of the medium, and r is the radius of the ion. Values of rK+ = 1.25 x 10-10 m and ra- = 1.21 x 10-10 m were used for the radii of K+ and Cl-, respectively.53 Because the dynamic viscosity changes with the temperature, literature values listed (given in Table S2.1) for the viscosity of water were used.54,55 The dependence of ^(T) on ion concentration is negligibly small, varying by only 0.8 % between pure water and 500 mM KCl.54,55 50 Table S2.1. The viscosity of water at different temperatures. T / K q (T / mPa s 283 1.3077 288 1.1527 293 1.0227 298 0.9177 303 0.8377 308 0.7827 51 In the Comsol finite element simulations, the Nernst-Plank equation (Equation 2 in the main text) is written in terms of the ionic mobilities. As the ionic radii of the two species are very similar we choose to assume that they53 both have the same diffusion coefficients and mobilities (those calculated using the ionic radius for K+).Values of h as a function of T computed from the corresponding values of Di ( h = Di/RT) are shown in Figure S2.5. NB: The unit, s mol kg-1 (used in Comsol simulations) can be converted to the conventional mobility (units m2 V-2s-1) by multiplying by Faraday's constant. (a) V) 2.40 2.20 2.00 E 1.80 b 1-60 - 1.40 Q 1.20 (b) ■ K+ T^">1.00- . K+ • cr iCb> • cr O 0.80- E« £ 0.60- s ir oT" ' ^0 .4 0 - 10 15 20 25 30 35 10 15 20 25 30 35 T (°C) T (°C) Figure S2.5. (a) Diffusion coefficient as a function of temperature. The expression D = 4.03 x10-11 T + 1.02 x 10-9 m2/s (T in °C) for K+ was used for both K+ and Cl- because the diffusion coefficients vary less than 2 %.(b) Ionic mobility is derived from h =D/RT (T in K). The expression h = 15 x10 -14 T+ 3.94 x 10-13 s mol kg-1 (T in °C) describe the data and was used in the finite element simulations. ((V) I/I) U| ((V) I/I) U| 52 S2.6.6 Simulated Arrhenius plots for 35 nm Pore at Different KCl Concentrations -23.4- -21.0 . 0.02 mM KCl 1.0 mM KCl 10 mM KCl < -23.6 < -212 = -23.8 S -21.4 C _c -21.6 ■ -24.0- 0.0033 0.0034 0.0035 0.0033 0.0034 0.0035 0.0033 0.0034 0.0035 1/r (K*1) 1/r (K'1) 1/r (K ') 1/r (K1) 1/7-(K-1) Figure S2.6. Arrhenius plots constructed from the simulated temperature-dependent currents obtained at +0.35 V for a 35 nm pore. These Arrhenius plots were used to extract the activation energy values shown in Figure 2.2 in main text. 53 S2.6.7 Concentration Distribution of K+ in Vicinity of the Nanopore Orifice Figure S2.7. Concentration distributions at 25 °C in the vicinity of the orifice of (a) 20 nm (b) 35 nm (c) 50 nm and (d) 65 nm nanopores. The surface charge on the nanopore wall is -2 mC/m2 and the bulk concentration is 0.1 mM. The applied voltage is +0.35 V. 54 S2.6.8 Simulated Concentration Profiles of K+ and Cl" Along the z Axis of a 20 nm Nanopore E r m * (a) +0.35 V 1.0­0.8­0.6­0.4­0.2 0.0^ +0.35 V » - 10 °c 0.14- +0.35 V -10°C (1 - 1 5 c 0.12- J ---- 15 °C j \ ---- 20 ° C S 0.10-- - ' ---- 20 °C I \ ---- 25 °C E, 0.08- ---- 25 °C | \ ---- 30 °C ---- 30 °c ^ I - 35 °C C 0.06- Si 0.04- ---- 35 0.02- 0.00- --------------- -----------------1---------------------------------1-------------------------------- 400 200 0 -200 z (nm) (b) -0.35 V -400 2000 0 -2000 z (nm) -4000 0.5- .____ . 0 4 S 0.3- o 0.2- 0 1- -0.35 V - i o °c A - 15 °c 1 -------- 20 °C % - 2 5 °c --------30 °C ^ - 35 °C 400 200 0 -200 z (nm) -400 2000 0 -2000 z (nm) -4000 Figure S2.8. Concentration profiles of K+ and Cl" in the vicinity of 20 nm nanopore orifice at (a) +0.35 V and (b) -0.35 V. The surface charge on the wall is -2 mC/m2 and the bulk KCl concentration is 0.1 mM. 55 S2.6.9 Electric Field Profiles for a 20 nm Pore at Different Temperatures Figure S2.9. The axial component of the electric field inside the pore at z = -20 nm for a 20 nm nanopore (0.1 mM KCl, 10 °C and 35 °C). At (a) -0.35 V and (b) +0.35 V 56 S2.6.10 Apparent Activation Energies Calculated at Different Applied Voltages Figure S2.10. Apparent activation energies calculated from finite element simulations at different applied voltages for a 20 nm pore. CHAPTER 3 DETECTION OF BENZO[A]PYRENE-GUANINE ADDUCTS IN SINGLE­STRANDED DNA USING THE a-HEMOLYSIN NANOPORE 3.1 Introduction In this chapter, we report that the a-hemolysin (aHL) nanopore platform can be used to detect a BPDE adduct to guanine (G) in synthetic oligodeoxynucleotides. Polycyclic aromatic hydrocarbons (PAHs) emitted into the environment by the incomplete combustion of coal, crude oil, and gasoline were reported to have carcinogenic properties in humans as early as 1876.1,2 In 1930, the PAH benzo[a]pyrene (BP) was identified as the carcinogen in these substances. Workers in tar distilleries, aluminum production, fossil fuel processing, and road paving are exposed to high levels of BP, as are smokers and consumers of grilled meats.3-6 Exposure to BP has been shown to increase susceptibility to lung and colon cancers.7 Cellular studies have demonstrated that one of the principal pathways through which BP is removed from the body is via cytochrome p450s (CYP450), yielding the final product benzo[a]pyrene diol epoxide (BPDE, Figure 3.1). BPDE exists in four isomeric forms with the (+)-a«ft'-7a,8P-dihydroxy-9a,10a-epoxy-7,8,9,10-tetrahydro-benzo[ a]pyrene (BPDE) isomer being the predominant one observed from enzymatic studies.8,9 BPDE is electrophilic and susceptible to nucleophilic attack from DNA, where 58 O N NH 'N^ N ^ " "N H R HO, HO'' G"BPDE Adduct Figure 3.1. Benzo[a]pyrene metabolism leading to guanine adducts in DNA. the base guanine (G) is a chief site for adduction of BPDE, yielding a stable adduct at the N2 position (G-BPDE, Figure 3.1). Because the epoxide ring-opening can occur by either Sn1 or Sn2 mechanisms, two diastereomers of G-BPDE are formed (Figure 3.1), leading to different structural perturbations of the DNA double helix. Moreover, mutations at specific G residues in the TP53 gene are responsible for the mutagenic properties of BPDE that lead to lung cancer.10-12 Therefore, identification of these adducts and their locations in the genome are critical to addressing an individual's susceptibility to cancers caused by BPDE. Several methods have been developed for quantification of BPDE adducts in genomes.13-15 The most commonly used methods include [32P]-postlabeling16 enzyme-linked immunosorbent assay (ELISA),16 liquid chromatography coupled to mass spectrometry (LC-MS),17 capillary electrophoresis MS, or HPLC coupled with a fluorescence detector.18 Analysis of BPDE adducts in DNA by these methods requires exhaustive nuclease digestion of the DNA sample to the nucleoside monomers. There are two major drawbacks with this step: (1) digestion of these adducts in DNA to the nucleoside monomers is often incomplete because the lesion is not a good substrate for any nucleases, and (2) digestion of the DNA causes all sequence information to be lost.19 Methods for quantification of G-BPDE adducts by LC-MS have identified this lesion to exist at a concentration of <10 adducts/108 nucleotides in the human genome,19,20 and a method that can directly analyze these adducts in the genome would be advantageous for quantification of BPDE adducts. In addition, a single-molecule method would have the added advantage of addressing the question of the distribution of adducts and identifying any hotspots for adduct formation. A powerful strategy for analyzing DNA is achieved by electrophoretically driving single-stranded DNA through the a-hemolysin (aHL) nanopore.21,22 Studies with this nanopore have demonstrated the potential for sequencing the four DNA bases,23 epigenetic markers,24 damage to DNA resulting from oxidation25-27 or deamination,28,29 photochemical damage,27 and base release that yields abasic sites.29,30 Herein, we demonstrate that the nanopore ion channel method can be applied to the direct detection of a G-BPDE adduct. In these studies, short (4-mer) and long (41-mer) synthetic DNA oligomers with a centrally located BPDE adduct were electrophoretically driven through the aHL nanopore while monitoring the current fluctuations and event times. These studies demonstrate that the BPDE adduct is capable of passing through the pore while producing a current blockage signature characteristic of the biomarker. These observations represent the initial step toward applying the nanopore method for the detection and quantification, and ultimately for reading the sequence, in which BPDE adducts reside in the genome. 59 3.2. Experimental Section Caution: All PAHs are potentially carcinogenic and should be handled in accordance with NIH Guidelines for the Use of Chemical Carcinogens. 3.2.1 Chemicals and Materials for Preparation of BPDE-DNA Adduct All DNA strands were synthesized from commercially available phosphoramidites by the DNA/peptide core facility at the University of Utah. (±)-Benzo[a]pyrene-7a,8P-dihydrodiol- 9a,10a-epoxide was purchased from MRIGlobal and used as received. All other chemicals were used without further purification. 3.2.2 Preparation of BPDE-DNA Adduct DNA samples were purified by ion-exchange HPLC prior to their use via the following method: solvent A = 10% CH3CN, 90% ddH2O; B = 1 M NaCl in 10% CH3CN 90% ddH2O, 25 mM Tris pH 8; flow rate = 1 mL/min while monitoring the absorbance at 260 nm. The method was initiated at 15% B followed by a linear increase to 100% B over 30 min. Synthesis of the BPDE adducted DNA strands were carried out according to a literature protocol.31 Briefly, the BPDE stock solution was made by dissolving BPDE in 19:1 THF and 1.5% aqueous triethylamine. Reactions were performed in 100-^L aliquots in Eppendorf tubes containing 2 mM DNA and 1 mM BPDE stock solution in 25 mM Tris, 1.5% aqueous triethylamine, 200 mM NaCl all at pH 9.2. The reaction was carried out overnight in the dark at 37 oC. The reaction mixture was neutralized by adding 3 mL of 20 mM PBS buffer (pH 7.5) before purification. Products were purified by ion-exchange HPLC running the same solvent system as reported above and shown in Figure 3.2. 60 61 Figure 3.2. Ion-exchange HPLC traces. (A) For 4 mer-BPDE. (B) 41-mer BPDE. Green color traces represent the reaction mixture, whereas red traces represent the standard DNA mixture. The HPLC conditions utilized solvent A = 10% CH3CN, 90% ddH2O; B = 1 M NaCl in 10% CH3CN 90% ddH2O, 25 mM Tris pH 8; flow rate = 1 mL/min while monitoring the absorbance at 260 nm. The separation was initiated at 15% B followed by a linear increase to 100% B over 30 min. All isomeric products were collected together and analyzed by ESI-MS to give the following result: 41-mer BPDE calcd mass = 12136.9, expt mass = 12139.2. Reaction yields were ~5%. 3.2.3. Glass Nanopore Membrane (GNM) and Bilayer Formation for Ion Channel Recording The method for fabrication of a conical-shaped nanopore in a thin glass capillary membrane has been previously reported.32 The nanopores used for these studies had an orifice with a 300 to 600 nm radius. Silanization of the glass surface was achieved with 2% (v:v) 3-cyanopropyldimethylchlorosilane in CH3CN for 6 h at room temperature to introduce a hydrophobic surface to which the lipid bilayer could form. Two Ag/AgCl electrodes were placed in solution on the inside (trans) and outside (cis) of the capillary. The electrolyte solution was comprised of either 1 M KCl or 3 M NaCl in 10 mM PBS pH 7.4. Current-time (i-t) recordings were performed using a custom built high-impedance, and low-noise system (Electronic BioSciences Inc., San Diego, CA). The lipid bilayer was formed with 1,2-diphytanoyl-sn-glycero-3-phosphochline across the silanized GNM; bilayer formation was indicated by a resistance increase from ~10 MQ to ~100 GQ. A gas-tight syringe was used to apply a pressure of 20-40 mmHg to the inside of the GNM capillary that facilitated protein insertion into the lipid bilayer.33 Wild type aHL was reconstituted from the monomer peptide added to the cis side of the GNM (0.2 ^L of a 1 mg/mL solution). Formation of a properly functioning nanopore was determined by an Io at 120 mV of 122 pA or 244 pA at 25 °C in 1 M KCl or 3 M NaCl, respectively. Ion channel measurements were performed at 120, 140, 160, and 180 mV (trans versus cis), while 62 recording the data with a 100 kHz low-pass filter and at a 500 kHz data acquisition rate. All experiments were performed at 25.0 ± 0.5 °C. 3.2.4 Data Analysis Events were extracted using QUB 2.0.0.29 and data were analyzed using OriginPro 9.1 and software donated by Electronic BioSciences Inc. (San Diego, CA). The i-t traces presented were refiltered to 50 kHz for presentation purposes unless stated otherwise. 3.3 Results and Discussion 3.3.1. Ion Channel Measurements Two DNA oligomers were chosen for study, a 4-mer and a 41-mer, with the sequences 5'-CCGC-3' and 5'-C20-G-C20-3', respectively. These oligomers were allowed to react with (±)-benzo[a]pyrene-7a,8P-dihydrodiol-9a,10a-epoxide following a literature protocol to yield an adduct at G.31 The presence of a single G ensured that only one adduct was formed per strand; hereafter, the adducted oligomers are referred to as 4-mer BPDE and 41-mer BPDE. A single aHL ion channel was inserted into a lipid bilayer spanning a glass nanopore membrane. 34 The DNA analyte was added to the cis side of the channel in a buffered (25 mM PBS, pH 7.4) 1 M KCl or 3 M NaCl solution. A voltage was applied to electrophoretically drive the DNA from the cis to trans side of aHL, while monitoring the ion current as a function of time. 63 3.3.2. Translocation of 4-mer BPDE Adduct Previous studies conducted in our laboratories have monitored translocation of DNA strands modified by a broad range of molecular adducts through the nanopore.29,30 In these studies, some adducts were found to be too large to translocate through the central constriction of aHL (1.4 nm in diameter).29,35 Due to the size and hydrophobic nature of the BPDE adduct, a short 4-mer BPDE strand was chosen for our initial experiments to determine if the adduct was too large to pass through the narrow constriction zone. The short modified strand was advantageous because of its ease of synthesis and characterization (Figure 3.2A), and this simplified study provided the basis for understanding how the BPDE adduct interacts with the aHL channel. A comparison of ion current versus time (i-t) recordings for the unmodified 4-mer and the 4-mer BPDE DNA oligomers recorded at 180 mV (trans versus. cis) in 1 M KCl solution is shown in Figure 3.3. Based on previous studies, the anticipated residence time for the C-rich 4-mer strand in the aHL nanopore is predicted to be ~4-8 ^s, and events of >50% blockage to the current in this time range were measured.36 Translocation of the 4- mer BPDE oligomer resulted in longer (>10 ^.s) events and exhibited unique current patterns that were not observed for the unmodified 4-mer strand (Figure 3.3B). All events initiated with a decrease in the open channel current (Figure 3.3B, Io) to a mid-level current blockage (Ia) that was centered at ~10% Io and lasted 10-200 ^.s. Next, the events progressed to a noisy deep-level current blockage (Ib) that was centered at ~75 % Io and lasted from 10 to 100 ^.s (Figure 3.3B). All events returned to Io (Figure 3.3B) without the appearance of another midlevel current suggesting the oligomer moved through the P-barrel and exited the trans side of the pore. 37 64 65 Figure 3.3. Proposed model for translocation of a 4-mer and 4-mer BPDE adduct through the aHL nanopore. (A) Representative i-t trace for the 4-mer (5'-CCGC-3') strand, (B) representative i-t trace for a 4-mer BPDE adducted oligomer. All data were recorded at 180 mV (trans versus cis) in 1 M KCl at 25.0 ± 0.5 oC with a 100 kHz low-pass filter and 500 kHz data acquisition rate. Based on the i-t traces above, we propose the following model to describe how the 4-mer BPDE strand translocates through the aHL pore. The initial midlevel current, Ia, is established when the 4-mer BPDE strand enters the vestibule of aHL from the cis side (Figure 3.3B, lA).Next, the BPDE strand enters the P-barrel and the current drops to a deep blockage level (Figure 3.3B, Ib), during which the 4-mer BPDE strand is driven through the narrow P-barrel (d ~1.4 nm) to the trans side of aHL. The sharp return to Io, and not back to the mid-level current Ia indicates that the 4-mer BPDE exited the trans side of the channel and does not return through the cis opening, because exit from the cis opening would give a second mid-level current (Ia). This promising result suggests that BPDE adducts can be detected in longer oligomer model systems that would occupy the full length of the channel while the adduct interacts with the protein walls. 3.3.3. Translocation of 41-mer BPDE Adduct In the second study, a 41-mer poly-2'-deoxycytidine (poly-C) strand with a centrally located G was synthesized and allowed to react with BPDE to give an adduct yield of ~5%. After HPLC purification, reinjection of the adduct sample established that it contained 30% unreacted DNA. Therefore, when analyzing the 41-mer translocation data, events shorter than 50 ^.s were attributed to unreacted starting material (tmax = 44 ^s at 120 mV) and discarded. Analysis of the 41-mer strand with a BPDE adduct (2 ^M) was conducted in buffered solutions (25 mM PBS, pH 7.4) containing 1 M KCl and 3 M NaCl. Typical i-t translocation events that are characteristic of the unique pattern observed for the DNA-BPDE adduct in 1 M KCl electrolyte are shown in Figure 3.4, and data collected in 3 M NaCl are presented in Figure S3.3. The time traces for unmodified 41-mer strands are presented in Figure S3.1 and S3.2 3.3.4. Deep Blockage Level Analysis Translocation events for the 41-mer BPDE are initiated when the open channel current is reduced to a shallow shoulder level (Ii) that has a %(Ii/Io) = 55 ± 5% (Figure 3.5). The entry of either the 3' or 5' end of the strand into the vestibule leads to varying lengths of the tail in the vestibule resulting in a shallow and broad distribution in current levels. Currents of this magnitude were previously described to result from partial entry of the DNA strand into the channel,38 similar to our results. Next, the event transitions to a deep level blockage current with values of I2 measured as 18 ± 2% and 21 ± 2% (Figure 3.5B and C). Based on previous studies of poly-C translocation, these two current distributions represent the entry directionally (5' or 3') into the P-barrel.37,39 Current histograms for the I2 currents can be deconvoluted into two pseudo-Gaussian distributions 66 67 Figure 3.4. Current versus. time profile collected over 20 s for 41-mer BPDE (2 ^M) in 1 M KCl. The data were recorded at 180 mV (trans versus. cis) at 25.0 ± 0.5 oC. The red dotted lines indicate the places where long open channel blockages were removed. 68 Figure 3.5. Event types detected during translocation of the 41-mer BPDE sample. (A) Representative i-t traces for tr