Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Ultraviolet absorbance and circular dichroism analysis of DNA oligomers containing adenine tracts Lim, Yee Chee 2007

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

Item Metadata

Download

Media
831-ubc_2007-0167a.pdf [ 8.57MB ]
Metadata
JSON: 831-1.0060162.json
JSON-LD: 831-1.0060162-ld.json
RDF/XML (Pretty): 831-1.0060162-rdf.xml
RDF/JSON: 831-1.0060162-rdf.json
Turtle: 831-1.0060162-turtle.txt
N-Triples: 831-1.0060162-rdf-ntriples.txt
Original Record: 831-1.0060162-source.json
Full Text
831-1.0060162-fulltext.txt
Citation
831-1.0060162.ris

Full Text

U L T R A V I O L E T A B S O R B A N C E A N D C I R C U L A R D I C H R O I S M A N A L Y S I S OF D N A O L I G O M E R S C O N T A I N I N G A D E N I N E T R A C T S by , Y E E C H E E L I M B . S c , The University of Saskatchewan, 2004 A THESIS S U B M I T T E D I N P A R T I A L F U L F I L M E N T OF T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F M A S T E R OF S C I E N C E in T H E F A C U L T Y OF G R A D U A T E S T U D I E S (Chemistry) T H E U N I V E R S I T Y OF B R I T I S H C O L U M B I A A p r i l 2007 © Yee Chee L i m , 2007 Abstract "A-tract" is defined as phased runs of at least four consecutive adenines, i.e. (dA) n-(dT) n , where n > 4. The B*-form of D N A characteristic of A-tracts is distinct from the canonical B - D N A , with high base propeller twist and a narrower minor groove. The B * -form of D N A was examined using U V absorption and circular dichroism (CD) spectroscopy in order to estimate the extent of B*-type conformation adopted by 12-mer D N A oligomers containing different A-tract lengths. The systematic variation aims to study how the propensity towards B * - D N A formation depends on different A-tract lengths and different base compositions flanking the A-tract. C D and U V melting experiments indicate that B*-form has distinctive spectral signatures. The structural formation o f B * - D N A increases with A-tract length, but can be affected by the location of the A-tract within the sequence as well as neighboring A A / T T , A T , and T A base pairs. The spectroscopic results generally correlate well with differential scanning calorimetry (DSC) data. The calorimetrically obtained results were compared with thermodynamic parameters predicted by the Santa Lucia nearest-neighbor (NN) model. Disagreements between experimental and predicted thermodynamic values exist particularly for mixed A T sequences and those with the same number of N N parameters. Such discrepancies may be caused by different stabilities resulting from various extent of B*-type formation within a given D N A sequence. Since N N estimates of the melting temperature do not adequately account for structural differences, the incorporation of additional structural information may have a pronounced impact on thermodynamic variables and w i l l help to improve the N N model considerably. Consequently, this allows for a more accurate prediction of the stability of short D N A sequences (< 25 base i i pairs), often used in molecular biology applications involving sequence dependent hybridization reactions. In light of the increasing interest in the development of locked nucleic acids ( L N A ) for probe and primer design and theurapeutic applications, the thermodynamics and spectroscopic studies on the structural effects of the incorporation of L N A nucleotides on the A-tract structure w i l l also be presented. 111 Table of Contents Abstract i i Table of Contents. • -iv List of Tables v i List of Figures v i i List of Abbreviations ix Acknowledgments x Chapter 1: Introduction 1.1 Canonical Conformations of D N A 1 1.1.1 B - D N A 4 1.1.2 A - D N A ; 4 1.1.3 Z - D N A 5 1.2 Non-canonical Conformation of D N A 6 1.2.1 B * - D N A ! 6 1.2.2 Relationship between B*-form and A-tracts 7 1.2.3 Origin and Stabilization of B*-form ,. 8 1.2.4 Biological Impact of A-tract D N A Oligomers 9 1.3 Spectroscopic Characterization of A-tract D N A Oligomers 10 1.3.1.1 U V - V i s i b l e Absorption Spectroscopy ....10 1.3.1.2 U V Absorbance Spectra of Nucleic Acids 14 1.3.1.3 Two-State Model of D N A Denaturation ....16 1.3.2 Circular Dichroism 17 1.3.2.1 Theoretical Background of Circular Dichroism 17 1.4 Importance and Applications of D N A Thermodynamics 19 1.5 Differential Scanning Calorimetry 21 1.6 Concept of the Unified Nearest-Neighbor Model 23 1.7 Objectives 24 1.8 Thesis Organization 26 Chapter 2: Materials and Methods 2.1 Introduction 27 2.2 D N A Oligomer Design ..27 2.3 Buffer and Sample Preparation 30 2.4 U V Absorbance Measurements 30 2.5 Circular Dichroism Measurements.... 31 Chapter 3: Results and Discussion 3.1 Thermal Difference Spectra by U V Measurements 32 3.2 Temperature-Dependent C D Spectra.. 37 iv 3.3 C D Spectral Characteristics of B*-form of A-tract Structure 39 3.4 C D Difference Spectra.. 41 3.4.1 Analysis of P A M Sequences .' 42 3.4.2 Analysis of P A E Sequences 45 3.4.3 Effects of Sequences Flanking the A-tract 48 3.5 Quantification of D N A Secondary Structure 50 3.6 Comparison of Spectroscopic and Thermodynamic Results 53 3.6.1 Correlation of U V Measurements with Thermodynamic Variables 53 3.6.2 Correlation of C D Measurements with Thermodynamic Variables. 55 3.7 Evaluation o f the Nearest-Neighbor Model for A-tract Oligomers. 59 Chapter 4: The Impact of L N A on A-tract Structure 4.1 Limitations of L N A on A-tract Structure 62 4.2 Nucleic A c i d Analogues 63 4.2.1 Peptide Nucleic Acids 63 4.2.2 PhosphOrothioate Oligonucleotides ....64 4.2.3 Locked Nucleic Acids 65 4.2.3.1 Chemical Properties of L N A 66 4.2.3.2 Applications of L N A ..67 4.3 Motivation 67 4.4 Effects of L N A Substitution on Thermodynamic Stability 68 4.5 U V Absorbance and C D Analysis of L N A : D N A Duplexes 70 4.6 Conclusion 72 Chapter 5: Conclusion 5.1 General Summary 73 5.2 B*-form of A-tract Oligomers 74 5.3 Factors Governing the Propensity for B*-form '. 75 5.3.1 Effect of A-tract Length 75 5.3.2 Positional Effects of A-tract 76 5.3.3 Sequence Context of Flanking Sequences 77 5.3.4 Nearest-Neighbor Base Pair Interactions 78 5.3.5 Conformational Effects 78 5.4 Impact of A-tract Structure on its Thermodynamic Properties 79 5.5 Future Directions 80 References 83 Appendix 94 v List of Tables Table 2.1: The list of A-tract D N A oligomers used in this study 29 Table 3.1: Summary of length and positional effects of A-tracts 36 based on the X^ax and peak intensities at of the U V absorbance difference spectra. Table 3.2: Summary of B*-form content in P A M and P A E sequences 51 Table 3.3: D S C results summarizing the thermodynamic variables 55 for D N A sequences containing A-tracts. Table 4.1: Thermodynamic parameters for two L N A : D N A duplexes 68 and its corresponding unmodified D N A duplexes. v i List of Figures Figure 1.1: The representative crystal structures of A - , B - , and Z - D N A 3 Figure 1.2: The C2'-endo and C3'-endo sugar conformations of D N A 4 Figure 1.3: Schematic energy level diagram for the U V absorbance process 11 Figure 1.4: Electronic molecular energy levels 13 Figure 1.5: A three dimensional diagram of U V absorbance as a function 14 of temperature and wavelength. Figure 3.1: (A) U V absorbance spectra of d ( A ) i 2 at 15°C and at 75°C 34 (B) T D S of d(A)i2 resulting from the subtraction of the low temperature U V spectrum from the high temperature U V spectrum. Figure 3.2: Comparison of the U V absorbance difference spectra of. 35 homopolymeric d(A)i2 and heteropolymeric d ( T A T T A T A A T A T A ) (NA01), each of which are assumed to adopt B*-form and B-form conformations, respectively. Figure 3.3: Correlation between the ~ 260 nm peak intensities obtained 35 from the U V difference spectra as a function of A-tract length. Figure 3.4: (A) Temperature-dependent C D spectra of d(A)i2 from 37 15°C to 75°C. (B) The change in ellipticity at 248 nm with respect to temperature. Figure 3.5: Comparison of C D spectral characteristics for B*-form 39 and B-form conformations. Figure 3.6: C D spectra o f P A M sequences with different .42 A-tract lengths. Figure 3.7: C D difference spectra of P A M sequences with different 42 A-tract lengths. Figure 3.8: C D spectra of P A E sequences with different A-tract lengths 45 Figure 3.9: C D difference spectra of P A M sequences, with different 45 A-tract lengths. v i i Figure 3.10: A plot of 248 nm peak intensities of the C D difference spectra 47 as a function of A-tract length for P A M and P A E series of sequences. Figure 3.11: Comparison of the C D difference spectra of 48 P A M 4, P A M 4b,and P A M 4c to evaluate the effect of different base pair steps flanking the A-tract site. Figure 3.12: A n illustration of the "flexrod" baseline used to quantify 52 the C D signal intensity at 260 nm. Figure 3.13: Quantification of B*-form content of P A M and P A E sequences 52 Figure 3.14: Correlation between T D S peak intensities at 260 nm 53-54 with (A) T m , (B) AH°, and (C) A S 0 . Figure 3.15: Correlation between C D difference spectra peak intensities. 56-57 at 260 nm.with (A) T m , (B) A H 0 , and (C) AS°. Figure 3.16: C D difference spectra of P A M 4, P A M 4b, and P A M 6 to 61 show that structural differences among these sequences contribute to the differences found between the predicted and experimental thermodynamic parameters. Figure 4.1: A diagram of the molecular structure of L N A and D N A . 65 nucleotides. Figure 4.2: U V absorbance difference spectra of A L N A 1 and A L N A 2 70 compared to their respective natural analogs, EE07 and EE08. Figure.4.3: C D spectra at 15°C of A L N A 1 and A L N A 2 compared ...71 to their respective natural analogs, EE07 and EE08. v i i i List of Abbreviations A Adenine bp Base pairs C Cytosine C D Circular Dichroism D N A Deoxyribonucleic A c i d D S C Differential Scanning Calorimetry E D T A Ethylenediaminetetraacetic A c i d G Guanine L N A Locked Nucleic A c i d N N Nearest-neighbor P C R Polymerase Chain Reaction P N A Peptide Nucleic A c i d R N A Ribonucleic A c i d SNP Single Polynucleotide Polymorphism T Thymine T m ' Melting temperature . T D S Thermal Difference Spectra U V Ultraviolet ix Acknowledgments Thanks are due to helpful guidance from my supervisors, Michael Blades and Robin Turner. I sincerely thank Georg Schulze, Curtis Hughesman, Fred Rossell for their extensive discussions and critical comments, and Marcia Y u for her involvement in the final stages of thesis. Special thanks to my brother, Yee Cheng L i m , who has always been a source of motivation. Finally, my heartfelt gratitude goes to Jeffrey Robert Johnston, for his feedback and continuous encouragement throughout the course of this study. x Chapter 1 Introduction Most of the work presented in this thesis is directed toward better understanding j the sequence-dependent determinants of secondary structure of D N A (and certain non-natural) oligonucleotides and concomitant effects on the stability of the duplex form. In this introductory chapter, the well-known right-handed anti-parallel double helical structure of B - D N A wi l l be addressed. Other D N A conformations such as A - D N A and Z - D N A w i l l also be discussed to exemplify other variant forms of D N A , which is largely attributed to the sequence-dependent nature of D N A structure. The unique structure of B * - D N A is of considerable interest because its relationship with the presence of adenine tracts (A-tracts) represents a good example of sequence specific conformational modulations. Structural studies of A-tracts form the basis for understanding D N A curvature and deformation, which has significant relevance in terms of sequence specific regulation of genetic issues at the molecular level. 1.1 Canonical Conformations of DNA The foundation of modern molecular biology starts with the discovery of the classic deoxyribonucleic acid ( D N A ) duplex structure, initially described in 1953 by 1 2 Watson and Crick ' . There are many excellent reviews, giving detailed descriptions of 3 10 the structure of D N A " . Therefore, only a brief summary of the D N A structure w i l l be provided here to facilitate discussion within the context of this thesis. 1 D N A consists of two anti-parallel polynucleotide strands, one in the 5' to 3' direction and the other in the 3' to 5' direction. The nucleotide building blocks consist of a sugar moiety (deoxyribose), a phosphate linker, and a nucleobase. Four different nucleotide bases occur in D N A : adenine (A), thymine (T), guanine (G), and cytosine (C). Each base within a rung of the D N A ladder is always paired with its complementary base, such that A always pairs with T through two hydrogen bonds, and G always pairs with cytosine (C) through three hydrogen bonds. According to Chargaff s rules, the complementary base pairing results in equal numbers of A and T, and equal numbers of G and C in a D N A molecule 4. The intertwined strands make two grooves of different widths, referred to as the major groove and the minor groove, which may facilitate binding with specific drugs. For example, the polyamide antibiotics netropsin 5 and distamycin 6 bind into the minor groove of D N A by recognizing AT- r i ch sequences. In the investigation of D N A secondary structures, early diffraction studies led to the identification of two distinct conformations of the D N A double helix, depending on factors such as base compositions and environmental conditions (i.e. relative humidity and salt concentration). The structural features of the three most common conformations of D N A , A - , B - , and Z - D N A , is shown in Figure 1.1. A t high humidity (92%) and low salt, the dominant structure is called B - D N A 7 " 9 and at low humidity (75%) and high salt, the favoured form is A - D N A 1 0 " n . In 1979, a left handed helix formed by a synthetic hexamer d ( C G C G C G ) was discovered and is now known as Z - D N A 1 3 . 2 Figure 1.1: The side projections of the crystal structures of A - , B - , and Z - D N A (shown from left to right). A - and B - D N A are right-handed helices, whereas Z - D N A is a left handed double helix with a backbone that follows a zig-zag pattern. Adapted from http://upload.wikimedia.Org/wikipedia/en/b/b9/A-B-Z-DNA_Side_View.png The morphological difference found in D N A originates from the variations in the preferred sugar conformations of the deoxyriboses and the orientation of the base relative to the sugar. The deoxyribose in D N A , where the stabilizing 2'-hydroxyl effects are absent, exists in fast equilibrium between C 2 ' - endo and C 3 ' - endo conformations, which are shown in Figure 1.2, with a relatively low energy barrier 1 4 (~2 kcal m o l - 1 ) . For deoxyribose, the C 2 ' - endo conformation is slightly lower in energy (0.5 - 1.0 kcal m o l - 1 ) 1 4 compared to the C 3 ' - endo and this explains why the C 2 ' - endo sugar conformation of D N A is preferred. In most double-helical nucleic acids, except Z - D N A , the rotation of a base around its glycosidic bond is in the anti conformation 4. 3 C3'-endo C2'.-endo Figure 1.2: The C2'-endo and C3'-endo sugar conformations of D N A . The preferred geometry of intrinsic sugar puckering and the decreased phosphate-phosphate repulsion in C2'-endo conformation results in it being a lower energy structure compared to C 3 ' -endo. 1.1.1 B - D N A Natural double-stranded D N A exists at physiological p H as a B - D N A (B-form) helix. The C 2 ' - endo sugar conformation gives rise to the B-form helix 4 . The conformation of B - D N A is defined by a helical repeat of 10.6 base pairs per t u rn 1 5 ' 1 6 , a narrow minor groove, and a widened major groove. In B - D N A , the major and minor grooves are of equal depth since the bases sit directly on the axis 1 7 . The B-form is favored by AT- r i ch D N A , indicative of the role of sequence in determining conformational stability. 1.1.2 A - D N A The sugar pucker in A - D N A adopts the CV-endo conformation 4. In comparison to B - D N A , the helical structure of A - D N A is more uniform, characterized by a deeper 1 7 major groove, and a narrower minor groove . Besides a helical repeat of 11 base pairs per turn 3 ' 4 , the most distinguishing feature of A - D N A relates to the displacement of these base pairs at a tilt o f 20° with respect to the helix a x i s 3 ' 4 ' 1 1 . A - D N A is less flexible than B - D N A and the rigidity is proposed to have biological significance. For example, D N A 4 replication can occur at a higher degree of fidelity in the A-form due to the added stiffness5. Sequences containing G-tracts, which are defined by sequences with homo-G C in base composition and particularly rich in G G base steps (or C C base steps) such as poly(dG)-poly(dC), tend to adopt the A- fo rm 1 8 . A t low relative humidity, B - D N A undergoes a reversible conformational change to A - D N A . The B to A conformational change plays an important role in gene regulatory processes of D N A polymerase and HIV-1 reverse transcriptase, both of which are found in an A- l ike structure1 9. 1.1.3 Z-DNA Under high salt conditions, the alternating copolymer poly[d(C-G)] was observed by Wang et al.20 to adopt a conformation unexpectedly different from A - or B - D N A , called Z - D N A . Aside from its left-handed helical sense, Z - D N A has 12 base pairs per helical turn. The base pairs are displaced in the opposite direction from those in A - D N A , resulting in a deep minor groove and a flat major groove 4 ' 1 7 . The overall chain sense in Z - D N A is directed downward at the left of the minor groove and upward at the right, as opposed to the right-handed family of D N A duplexes. The sugar and glycosidic bond alternate in the following manner4: C 2 ' - endo in anti dC or dT and C 3 ' - endo in syn d G or dA. For Z - D N A , the requirement of purine/pyrimidine alternation is due to the 91 99 preference of the syn conformation ' . Some examples of sequences known to adopt the Z-form include poly (dG-dC)-poly(dG-dC) and poly(dG-dT)-poly(dA-dC), with exception of poly(dA-dT) - poly(dA-dT). The preference of alternating G and C over A and T does not yet have a clear explanation 1 7. 5 1.2 Non-canonical conformation of DNA 1.2.1 B*-DNA The formation of various structural forms of D N A , discussed up to this point, depend on factors such as salt, hydration, and base pair composition. A s more highly resolved methods for structural studies have become available, it has been found that some B - D N A sub-conformations also exist. Analysis of poly(dA)-poly(dT) helices have been observed to pack in the crystal lattice differently from ordinary B - D N A . This homopolymeric D N A duplex displays 10.1 ± 0.1 base pairs per turn 2 4 , setting it apart 25 from the common B - D N A structure with alternating or random distributions of bases . The minor groove is considerably narrower in poly(dA)-poly(dT) than in standard B-form D N A . Other unusual properties of poly(dA) -poly(dT) include resistance to transformation into other helical forms , unlike the standard B - D N A , under different conditions of relative humidity and salt concentration 2 4. If the salt concentration of the environment is raised (or the relative humidity lowered), the normal B —»• A transition does not occur. Instead, a disproportionation of the double helix into a triple helix and a single polynucleotide chain is observed, similar to the poly(A)-poly(U) system . Despite these differences, the structure of poly(dA)-poly(dT) differs only slightly from that of the canonical B - D N A , and thus, its conformation is referred to as B * - D N A . The unusual structure of poly(dA)-poly(dT) is stabilized by purine-purine base stacking interactions and by additional hydrogen bonds, both of which arise from the high propeller twist of the base pairs. The propeller twist feature refers to the rotation of the bases along their longitudinal axis 2 4 . The local, sequence-dependent distortions of structure are primarily associated with changes in the orientation of bases in order to 6 maximize base stacking interactions. A n energetically favorable hydration reaction in the • • 97 minor groove is thought to stabilize the B*-form of the polynucleotide ' . 1.2.2 Relationship between B*-form and A-tracts The helical irregularities found in the crystal structure of the Eco RI recognition site containing the dodecamer sequence, d ( C G C G A A T T C G C G ) , led to the realization that D N A secondary structure is indeed sequence-dependent 2 8 ' 2 9. The next important study of sequence effects came from the investigation of the intrinsically curved or bent kinetoplast D N A , in which regions of curvature were found to result from the presence of properly phased stretches of adenine, known as A- t rac t 3 0 ' 3 1 (dA n "dT n ) , within the helical repeat. Since then, additional experimental evidence has been accumulated to understand the behavior of A-tracts. The fundamental aspect of the A-tract requires that at least f o u r 3 2 - 3 5 adenine residues in tandem be present in a given oligonucleotide sequence, with only A A (equivalently TT) or A T base-steps, and no T A base-steps18. Molecular dynamics studies further identified the formation of sequence dependent hydration patterns in the minor groove, often called "spine of hydration", near the A-r ich 36 region . The spine of hydration is often suggested to play a critical role in the stabilization of the B*-form of D N A in so lu t ion 3 7 ' 3 8 . In addition, the high propeller twist of the A T base pairs found in the more rigid and less flexible B*-form structure differentiates it from the common B-form, thus maximizing purine-purine stacking interactions 3 2 ' 3 3 . Curved fragments of oligonucleotides containing A-tracts have been detected from its retarded movement during electrophoresis in polyacrylamide g e l s 3 9 - 4 1 because curved structures are hindered more than straight molecules when migrating through the 7 gel pores . The necessity for a continuous run of adenines in order to observe strong bending anomalies is supported by experimental evidence that the disruption of the run of A ' s within an A-tract by a G , C, or T base produces a near normal gel migration pattern of the respective substituted concatamers 4 2. The electrophoretic mobility of phased A -tracts is not drastically affected by the sequence within the G C containing regions connecting adjacent A-tracts 4 3 . Other properties of the A-tract include the progressive narrowing of the minor groove toward the 3' end 4 4 ~ 4 7 and the existence of bifurcated hydrogen bonds 2 4 , 3 7 . It should be noted that bifurcated hydrogen bonds are a consequence of the high propeller twisting in A-tracts, rather than being the driving force for the formation of A-tracts. Replacing A A A with the inosine (I) mutant A I A , which does not form bifurcated hydrogen bonds resulted in little change in anomalous gel . retardation , suggesting that conformational stability due to bifurcated hydrogen bonds may be slight, i f any 3 7 . 1.2.3 Origin and Stabilization of B*-form Several theoretical models have been developed to understand the sequence-dependent effects of A-tracts on conformational stability of D N A and to relate it to D N A bending. The junction model proposes that the A-tract region exists in a non B - D N A form with base pairs at a negative inclination relative to the overall helix a x i s 4 9 ' 5 0 . Parallel stacking of the highly inclined base pairs in an A tract causes the non-A-tract D N A helix axis to form an angle with the A-tract D N A helix axis. A deformation of the helix axis is then observed at the junctions between A-tracts and B - D N A 4 6 . 8 The wedge m o d e l 5 1 ' 5 2 attributes the non-parallel stacking of adjacent base pairs to deflections in every A A dinucleotide step 3 5. Each non-parallel A A step effectively forms a wedge that can curve the helix axis i f they occur within the D N A helical repeat. More recently, it has been proposed that A-tracts stabilize the B*-form through cations localized in the A-tract minor groove more dominantly than the A-tract major 21 groove . The localization of cations depends on electrostatic interactions with the functional groups of the D N A bases and backbone. Preferential localization of cations within the minor groove of A-tracts can cause asymmetric distribution of phosphate neutralization, which in turn can cause narrowing of the minor groove and bending of the helical axis. 1.2.4 Biological Impact of A-tract DNA Oligomers The biological relevance of A-tract structures is linked to its distinctive structural properties. A-tracts serve important roles in several biological processes, including D N A replication and packaging, chromosome segregation, and transcriptional regulation 2 4. For example, A-tracts have been found to act as upstream promoters that activate transcription in yeast 5 3. Another example involves that of histone H I which preferentially binds.and aggregates scaffold-associated regions via the numerous A-tracts present within these sequences in order to regulate transcriptional processes 5 4. In terms of biological recognition, the sequence-specific deformations caused by A-tracts have been found to be involved in activities of regulatory proteins. For example, the mechanism by which the Bacillus subtilis L rpC protein interacts with D N A depends on the curved characteristics of sequences containing phased A-tracts 5 5 . 9 1.3 Spectroscopic Characterization of A-tract DNA Oligomers 1.3.1.1 UV-Visible Absorption Spectroscopy U V - V i s i b l e absorption spectroscopy is based on the absorption of light due to the interaction of the oscillating electromagnetic field of the radiation with the electrons in the molecule. If the frequency of electromagnetic field corresponds to the energy difference between the ground and excited states, the electrons in the molecule are shifted to higher energy levels. This increase is equal to the energy E of the photon, which is related to the frequency u and the wavelength A, of the radiation by the following equation: E = ho = h c / X (1.1) where h is the Planck's constant and c is the velocity of light. The change in energy may be in the electronic, vibrational, or rotational energy of the molecule. During the U V light absorption process, the energy differences correspond to those of the electronic states of atoms and molecules and causes transitions of electrons from the ground state to an excited state5 6. The absorption process is extremely fast, occurring in about 10"1 5 seconds 5 7. A n illustration of the absorption process is shown schematically in Figure 1.3. 10 Electronic . . Excited State, Ei ! : 'A Electronic Ground State, E 0 : Figure 1.3: Schematic energy level diagram for the U V absorption process. Two electronic levels are shown, together with the corresponding vibrational sublevels which are represented by the horizontal lines. The arrows represent the transition of electrons from the ground state to the various vibrational levels of the first excited state upon absorption of U V photons. The amount of light absorbed by the sample is described by the Beer-Lambert law, which is the relationship between the light intensity entering the absorbing medium, I 0 , and the light intensity leaving the absorbing medium, I: log (I 0 /1) = A = eCl (1.2) where A is the absorbance, s is the molar absorption coefficient, C is the concentration of the sample, and / is the sample pathlength. s has the units L mol" 1 cm"1 and is characteristic of the molecule of interest. Apparent deviations from the Beer-Lambert law may arise from instrumental and sample artifacts such as light scattering, a non-monochromatic light source, inhomogeneous sample, non-linear photodetector, and absorption by the cuvette 5 6 ' 5 7 . Energy 11 Molecular properties of the sample, such as sample aggregation or complex formation as the concentration is changed, may also cause deviations from l inear i ty 5 6 ' 5 7 . The fundamental basis for the U V absorption bands of nucleic acids is correlated with the types of bonds they contain. Electrons forming single bonds are called a electrons 5 6. The characteristic functions and charge densities of a electrons are rotationally symmetrical with respect to the bond axis 5 6 . Electrons responsible for double bonds are called n electrons, the characteristic functions and charge densities of which have a nodal plane through the bond axis 5 6 . In unsaturated systems, n electrons predominantly determine the energy levels of the electrons, which are excited by the absorption of ultraviolet light, n electrons refer to the unshared electrons or non-bonded electrons in molecules containing atoms like nitrogen and oxygen. In bonding systems, a electrons are. more strongly bound than the n electrons while in the antibonding levels, the a* level has higher energy compared to the n* l eve l 5 6 . Possible electronic transitions involving a, n, and n electrons are shown in Figure 1.4. Based on the properties of the conjugated ring systems of nucleic acid bases, conformational changes in D N A can be followed by observing the shifts in their U V absorption spectra. The presence of chromophores in D N A molecules can be identified by examining the wavelength corresponding to the absorption maximum, A, m a x. A , m a x is affected by the type of chromophore present as a result of the difference in the electronegativities of the elements forming the multiple bonds 5 6 . U V absorption by a single C=C double bond exhibits, an absorption maximum at ~ 180 n m 5 6 . When the conjugation of double bonds increases, the energies of the molecular orbitals lie closer together and the U V peaks associated with the % —» %* transition are shifted towards 12 longer wavelengths. In general, the absorption bands of nucleic acid bases normally found in the near U V and visible regions arise from either n —> n* or n —* n* * 58 • ' transitions . The allowed n —> n* transitions at 260 nm are a lot more intense than the n —* 71* transitions which are symmetry forbidden and consequently of weak intensity. Energy fj* 71* n 71 J * 0 • t b * b t c J * t * t Antibonding Antibonding Non-bonding Bonding Bonding Figure 1.4: A n illustration of the energies for the various types of molecular orbitals. The arrows depict the four possible types of electronic transitions (o —*• a *, n —> a *, n —* 7c*, and 7t —* 7T*) among different molecular energy levels brought about by the absorption of radiation. (Reprinted with permission from Skoog, D. A., Holler, F. J., and Nieman, T. A.. Principles of Instrumental Analysis, 5th edition, p. 331. Copyright 1998 by Harcourt Brace & Company) 13 1.3.1.2 U V Absorbance Spectra of Nucleic Acids Temperaiira Figure 1.5: U V absorbance surface as a function of temperature and wavelength of the homopolymer d(A)i2 (EE07). See Table 2.1 on page 21 for a complete list o f other sequences used in this study. The ease of measurement, simplicity of the instrumentation, and requirement for low concentrations (~ 5 ug/ml) 5 9 of D N A oligonucleotide samples have made U V absorbance spectroscopy a common method for studying D N A transition thermodynamics. D N A duplex melting and formation can be easily monitored through changes in U V absorbance. The 3-dimensional melting surface o f d (A) i2 (EE07) shown in Figure 1.5 has a peak maximum positioned at 259 nm. Certain of the subunits of nucleic acids (purines) have an absorbance maximum slightly below 260 nm while others co (pyrimidines) have a maximum slightly above 260 nm . Therefore, the U V absorbance maxima of D N A polymers can vary somewhat depending on their sequence composition. 14 When a solution of duplex D N A is heated, the base pairs are separated as the hydrogen bonds between them are broken. This results in the formation of single-stranded D N A and this process is commonly referred to as " D N A denaturation" or "melting". The qualitative changes accompanying the denaturation process can be followed by monitoring the difference in U V absorbance between the two states at a single wavelength (usually 260 nm) as the temperature increases. D N A denaturation is a reversible process upon slow cooling conditions to below its T m . The stability of the double helix is not only determined by the hydrogen bonds between paired bases but also by the stacking of parallel planes of the bases 6 0. Base-stacking is due to the overlapping of the Tt-electron orbitals of the planar bases. Because of this 7t-electron interaction, double-stranded D N A exhibits a lower light absorbance at 260 nm than single-stranded D N A 6 1 . This hypochromicity effect, which occurs as a consequence of the disruption of the electronic interactions among neighboring bases, is often used to understand nearest-neighbor base pair interactions 6 2. For short D N A oligomers, the U V absorbance values at 260 nm monitored over a range of temperatures results in a sigmoidal-shaped "melting curve". The temperature at which 50% of the D N A is denatured is called the melting temperature, denoted as T m . The stability of the D N A duplex, and hence its T m , depends on several factors, including buffer composition (salt concentration and identities of the ions in solution), base composition (regions with alternating pyrimidine/purine steps and A T regions melt more readily than G G regions; T m increases with G C content), the length of the D N A sequence (shorter lengths of D N A wi l l have lower T m ) , and the secondary structure of the D N A . Due to the various factors that can affect T m , it is difficult to predict accurately the exact 15 melting temperature of a given sequence. The practical importance of having a good estimate for T m is exemplified in the determination, of appropriate P C R primer annealing temperatures. Temperatures too far beyond T m w i l l produce insufficient primer-template hybridization resulting in low P C R product yield. Temperatures significantly lower than T m may possibly lead to non-specific products caused by a high number of base pair mismatches. In addition, information about the T m can be used to determine the minimum length of an oligonucleotide probe needed to form a stable double helix with a target gene at a particular temperature . 1.3.1.3 Two-State Model of DNA Denaturation To probe the melting process of D N A oligonucleotides, an important assumption is that the melting process is a reversible transition between double-stranded and single-stranded D N A and is commonly referred to as the "two-state process" 6 4 ' 6 5 . In this model, each D N A molecule is considered to be either totally in the double helix form or totally dissociated. The melting temperature, T m , defined as the temperature at which half of the nucleic acids are in the single-strand form, can be extracted from the experimental data. The two-state model is usually regarded as valid for short (<12 bp) oligonucleotides 6 6. 16 1.3.2 Circular Dichroism 1.3.2.1 Theoretical Background of Circular Dichroism Circularly polarized light is produced from two linearly polarized and mutually perpendicular beams of equal maximum magnitude that are out of phase by JI/2. Circular dichroism (CD) is the difference in absorption for left and right circularly polarized light by the same transitions observed in normal absorption spectroscopy 6 7. Therefore, Beer's law wi l l be obeyed for either rotation of circularly polarized light, and the difference in absorption, A A , is given by: where / is the pathlength, C is the molar concentration, and the subscripts L and R denote the left and right rotation of the light, respectively. As is characteristic of the molecule being examined and depends on the wavelength of light. The C D bands may be either According to equation 1.3, C D can arise only in the spectral region where the sample absorbs light. There cannot be a measurable difference i f the absorbance is close to zero. A n optically active molecule wi l l absorb the left circularly polarized light differently from right circularly polarized light. Consequently, the transmitted light appears as elliptically polarized light. Thus, the ellipticity, 0, usually measured in units of millidegrees (mdeg), is defined by the ratio between the minor axis and the major axis of the elliptically polarized right 6 8: A A = A L - A R = ( e L - SR) IC = Ae/C (1.3) positive or negative, depending on which rotation of light is absorbed more strongly. tan.8 = (ER - EL) (ER + EL) (1.4) 17 where ER and EL are the magnitudes of the electric field vectors of the right-circularly and left-circularly polarized light, respectively. When ER equals EL (when the absorbance of right- and left-circular polarized light are the same), 0 is 0° and the light remains linearly polarized. When either ER or EL is equal to zero (when the circular polarized light in completely absorbed in only one direction), 0 is 45° and the light emerges as circularly polarized 6 8 . Since the intensity of light, I, is proportional to the square of the electric-field vector, the ellipticity becomes: 9(radians) {*R + £ L ) (1.5) The natural logarithm form of equation (1.2) is: 1 = he (1.6) B y substituting equation (1.5) into (1.6), ellipticity can be rewritten as: - A R [ . i n ~ A L i „ m - , A A M ° $(radians) = x_ 4 r — . 1P Equation (1.7) can be simplified using a first-order Taylor series approximation. Since A A « l , terms of A A can be discarded in comparison with unity. Converting from radians to degrees yields: . . A JnlO 18u\ Bidegrees) = AA(—)(—) 18 The linear dependence of solute concentration and pathlength is removed by defining molar ellipticity as: [9l~ CI (1.9) Then combining the last two expressions with Beer's Law, molar ellipticity becomes: • • . J n l O w 1 8 0 , [9] = lOOAef——)( ) = 3298Ae 4 / V 7T 7 (1.10) Therefore, the relationship between C D and normal absorption spectroscopy is embedded in equation (1.10). Since C D monitors transitions only at their wavelength of maximum absorption, C D shapes are simpler, and the spectra are easier to interpret. 1.4 Importance and Applications of DNA Thermodynamics The stability of duplex D N A strongly depends on temperature and solution conditions, particularly for short oligonucleotides. Therefore, thermodynamic profiles of D N A are of fundamental and practical importance in understanding and predicting the sequence-dependent secondary structures of nucleic acids. The application of thermodynamic parameters for nucleic acids are found in many molecular biology applications such as the design of hybridization probes'using natural or modified nucleic acid bases for D N A microarrays or P C R , which are used in a variety of genotyping and other genomic applications (e.g. gene expression profiling, hereditary disease diagnostics, etc). Successful probe and primer design for such applications relies heavily on the ability to predict the thermodynamic stability of the complexes formed by the oligonucleotide probes. The choice of optimal conditions for hybridization experiments, minimum length of a probe required for hybridization, and probe and primer 19 stability, are heavily dependent on accurate thermodynamic predic t ions 6 9 - 7 1 . For example, the melting temperature (T m ) of different probes in D N A microarray registers must be accurately estimated using an appropriate nearest-neighbor thermodynamic model that permits the iterative selection of a set of probes with a minimal T m 79 variability . In multiplex P C R , where all the amplifications must occur under the same conditions, inaccurate T m predictions lead to poor design of an optimal sequence and consequently, the amplification or detection of wrong sequences Modified, non-naturally occuring nucleic acids such as peptide nucleic acids ( P N A ) 7 7 ' 7 8 and locked nucleic acids ( L N A ) 7 9 > 8 0 are useful for. their enhanced hybridization affinities towards single-stranded D N A or R N A . For example, they have been used for their ability to distinguish between s s D N A and s sRNA in the development of non-labeled oligonucleotide probes and antisense technologies ' . Substitution of one or more D N A nucleotides with a locked nucleotide in oligonucleotide probes ( L N A ) has shown to significantly increase the thermal stability of the L N A : D N A duplex, and to improve the discrimination between perfectly matched and mismatched-target nucleic acids. These two features are exploited for enhancing hybridization efficiencies under any prescribed conditions. 20 1.5 Differential Scanning Calorimetry Calorimetric measurements allow for the evaluation of thermodynamic properties of a system, which is required to understand the relationship between energy changes involved in a chemical or physical-chemical process . Differential scanning calorimetry (DSC) is a type of calorimetric technique that measures the heat capacity, C p , of a system as it varies with temperature7 3. B y heating or cooling a sample and reference material (which consists of a sealed empty aluminium pan) under the condition that they are always maintained at the same temperature, it is possible to measure the changes in C p that accompany heat-induced structural transitions in the sample . D S C measurement is often used to determine the thermal stability and reversibility of D N A denaturation process. One advantage of D S C is that a direct measurement of thermodynamic parameters of nucleic acids can be obtained. However, compared to spectroscopic methods (typically U V measurements), a relatively larger amount of sample is required for D S C experiments 8 5. In a typical D S C experiment, the excess heat capacity curve of the D N A solution relative to the reference buffer, C p e x , is recorded as a function of temperature for the or thermally induced transition of an oligomeric D N A sample . Valuable thermodynamic information, such as the difference in heat capacity (AC P ) , enthalpy (AH°), and entropy (AS°), can be obtained from a single D S C curve. Integration of C p e x with respect to temperature, as shown in equation (1.11), yields AH°. A H ° = j C p e x d T (1.11) The difference between the initial and final baselines of the D S C profile gives a direct measure of the heat capacity change, A C P , accompanying the transition 7 3. B y 21 converting the experimental A C p e x versus T curve to a A C p e x / T versus T curve yields the value of AS" using equation (1.12): A S 0 = | ( C p e 7 T ) dT (1,12) . Based on equation (1.13), the corresponding value of free energy change, AG°, can also be determined at any temperature. AG° = AH°-TAS° (1.13) where T is in Kelv in , A H " is in cal mol" 1 , and AS° is in units of cal K" 1 mol" 1. Equation (1.13) assumes that A C P is zero, which means that AFT and AS° are assumed to be temperature independent. This approximation is commonly used for analyzing the O f hybridization thermodynamics of nucleic acids . The slope and intercept from a 1/Tm versus In C T plot has been shown to be equal to R(rc-l) A F T V . H . and [ A S ° V . H . - (« - l )R ln2 + R l n n ] A H ° V . H . , respectively 8 6. Here, n represents the molecularity of the complex (the number of strands that associate), R is the gas constant, A H ° v H. is the van't Hoff enthalpy, A S V . a is the van't Hoff entropy, and C T is the total strand concentration. Comparison between AH°v.H.(model-dependenf) and A H ° (model-independent) allows one to evaluate i f the two-state model holds for a given D N A melting transition, but often results in large discrepancies . If AH° cf. A H ° V . H . reveals that the two-state model does not hold, then one explanation would be that a non-two-state transition is involved. Another interpretation is related to the differences in hydration between the duplex-stranded groups and single-stranded groups giving rise to an increase in the heat capaci ty 8 9 ' 9 2 ' 9 4 . . The thermodynamic analysis of the oligonucleotides used in this study assumes a two-state approximation and a negligible change in heat capacity at constant pressure, A C P . A change in heat capacity is generally 22 regarded as a dominant factor accounting for a difference between thevan't Hoff enthalpy and the calorimetric enthalpy that can significantly affect the thermodynamic properties of duplex formation 9 5. 1.6 Concept of the Unified Nearest-Neighbor Model The first nearest-neighbor (NN) type model was initially used to predict the stability and predict the temperature-dependent behaviour of R N A 9 6 and was later extended to apply to D N A duplex stability 9 7. The basic concept of the N N model for nucleic acids describes how the thermodynamic stability of a given sequence is ' dependent on the identity and orientation of neighboring base pa i r s 9 7 ' 9 8 . The nearest-neighbor approximation assumes that the stability of a D N A duplex depends on the sum of neighboring base pair interactions that accounts for hydrogen bonds in a base pair and stacking interaction between nearest-neighbor bases. Thus, the prediction of 07 thermodynamic parameters is possible i f the primary sequence is known . The ten Watson-Crick D N A nearest-neighbor pair-wise combinat ions 9 7 ' 9 9 that are present in anti-parallel duplex D N A are: A A / T T ; A T / T A ; T A / A T ; C A / G T ; G T / C A ; C T / G A ; G A / C T ; C G / G C ; G C / C G ; G G / C C . Here, the nearest-neighbor sequences are represented with a slash, /, separating the strands in antiparallel orientation. For example, A T / T A means 5'-A T - 3 ' is paired with 3' -TA-5 The "unified" oligonucleotide N N model takes into account thermodynamic information from six research groups, with the aim to overcome the confusion arising for reasons such as differences in salt dependencies, data analysis and experimental approach 9 9. These parameters were derived from multiple linear regressions of 108 sequences and solving for 12 unknowns: the ten Watson-Crick D N A N N interaction 23 parameters, one initiation parameter, and one correction for terminal A T pairs ' . The reliability of the unified parameters for predicting a dataset of 264 sequences ranging in length from 4 to 16 bp has been tested 1 0 0, yielding an average deviation between experimental and predicted T m of 1.6°C. Although the unified N N model provides reasonably good thermodynamic predictions for most applications of nucleic acids 1 0 0 , there have been suggestions that the possible influence of non-adjacent base pairs should also be taken into consideration. However, no other model has yet been proposed and verified, hence, the unified N N model remains the best available and is widely used. Limitations of the unified N N model are reflected in its inadequacy to properly predict more complex sequence-dependent interactions involving D N A curvature 1 0 1 " 1 0 3 , fraying end effects 1 0 4, triplet repeats 1 0 5, mismatches 1 0 6, hairpin structures 1 0 7, chemically modified oligonucleotides 1 0 8, and D N A dumbbells 1 0 9 . In addition, a more refined predictive model for non-canonical B-form D N A , as well as other D N A deformations associated with A-tracts, w i l l be required to address the full dimensionality of D N A secondary structure and conformation. 1.7 Objectives The rapid growth in nucleic acids research is attributed to the wealth of sequence information generated by the human genome project and the recent advances in both theoretical and experimental methods used. The results from a variety of studies have indicated that a combination of structural and thermodynamic data w i l l be needed to fully understand the sequence-dependent structure of A-tract oligonucleotides for potential theurapeutic applications, given its role in several key biological processes. 24 Thermodynamic variables obtained by D S C measurements have shown that discrepancies exist in thermodynamic characterizations of D N A containing A-tracts. Proper explanations are needed in order to advance the basic understanding of sequence-dependent structural phenomena and to assess the degree to which local D N A distortions may contribute to the forces that drive biologically significant events. Toward this end, the major goal of this thesis is to qualitatively determine the extent to which B*-form is influenced by the presence of A-tracts through temperature-dependent U V and C D methods. Since temperature wi l l influence the relative populations of B * - and B -forms 1 0 7 , the predisposition towards either conformation by different D N A oligonucleotides with different A-tract lengths w i l l be quantified. Next, the impact of other factors governing B*-form, such as position of A-tract, context of flanking sequences, and nearest-neighbor distribution w i l l also be evaluated. Correlation between the structure and thermodynamic studies can provide further insights into how B*-form may be stabilized by thermodynamic effects. It is hypothesized that the combination of both thermodynamic and structural studies can enhance D N A structure prediction of A-tract sequences. Further efforts are still required to assign universally appropriate parameter sets of the hybridization thermodynamics for A-tract oligonucleotides to improve the accuracy of secondary structure prediction. Several factors must be taken into consideration. For example, structural information connecting T m and B*-form may need to be included in N N models. In addition, accurate modeling of the effect of position of the "A-tract within a given sequence may be required, as well as the sequence context adjacent to the A-tract. Such refinement is necessary, not only to predict the threshold values for secondary structure stability, but also to elucidate 25 the mechanisms that control the formation of secondary structures associated with A -tracts. 1.8 Thesis Organization This chapter highlights the background information relating B*-form with A-tract sequences and covers the theoretical aspects of the spectroscopic methods used to probe factors modulating the non-canonical D N A conformation induced by A-tracts. Chapter 2 provides a description of the materials and methods used in this work. In Chapter 3, a detailed discussion, based on the experimental results obtained, w i l l focus on the.factors governing the propensity for B*-form structure and whether or not a correlation exists between the B*-form structure and its thermodynamic properties. Chapter 4 provides the some insights into the effect of conformation on the overall A-tract structure, induced by the incorporation of locked nucleic acid ( L N A ) bases. Finally, Chapter 5 summarizes the current results and provides some possible guidelines for further investigating the stability and dynamics of B*-form structure. 26 Chapter 2 Materials and Methods 2.1 Introduction The materials and methods used in the experiments performed in this work wi l l be presented in this chapter. The rationale for the design of A-tract D N A oligonucleotides used w i l l be discussed. In this study, the spectroscopic techniques, ultraviolet (UV) absorbance and circular dichroism (CD), used to monitor changes in absorbance and chirality in the U V region, w i l l be discussed. 2.2 DNA Oligomer Design The rationale for our oligomer design is based on assessing the structural influences of A-tracts on the extent of B * - D N A formation in a systematic fashion and to determine i f a correlation exists between thermodynamic stability and these structural findings. In Table 2.1, the five different sets of short D N A oligomers are summarized. These were designed to be 12 base pairs in length to ensure that duplex formation proceeds in a two-state manner. The A-tract site of each oligonucleotide shown in Table 2.1 is underlined for clarity. Set I oligomers, d ( T A T T A T A A T A T A ) and d(A) i2 , were used as control sequences to represent B-form and B*-form, respectively. Set II represents the P A M group of sequences, which contain a centrally located A-tract site of general sequence (dA) n , where n = 3, 4, 6, 8, and 10, flanked on either side by A T and T A base pairs. B y varying the length of the A-tract, this set of sequences 27 serve to investigate the effects of A-tract length on the extent of formation of B*-type helix. The minor groove at the 5'end of A-tracts is wide and it gradually narrows towards the 3'-end. The gel mobilities of sequences adjacent to the 3'-end, rather than the 5'-end, of an A-tract have been shown to be indistinguishable from one another 1 6 7. Therefore, set III oligomers (hereafter termed P A E sequences) were designed to position the A-tract site at the 3'-end of the sequence, with A T and T A base pairs flanking the 5'-end of the A-tracts. The A-tract lengths for the P A E sequences were designed to be the same as those of P A M sequences to enable direct comparisons between P A M and P A E sequences for the evaluation of positional effects of A-tract on the amount of B*-type helix formation. Set IV which consisted of P A M 4b and P A M 4c were designed to have the same number of consecutive adenine runs as P A M 4. With this design, any differences in the nearest neighbor contributions to D N A stability and/or subtle structural/conformational deviations from the standard B-form between duplexes of the same length should be attributable to difference in only flanking sequence contexts. It is well-established that sugar pucker of locked nucleic acids ( L N A ) is locked in a XZy-endo conformation and thus yields an A-form duplex. The locked ribose conformation enhances base-stacking and backbone re-organization, which significantly increases the thermal stability of oligonucleotides. Therefore, Set V was designed to examine the influence of the presence of the conformationally biased locked bases towards the overall A-tract structure through a direct comparison .between the L N A : D N A duplexes ( A L N A 1 and A L N A 2 ) and their respective natural analogs (EE07 and EE08). 28 The structures formed by D N A sequences containing A-tracts were characterized by U V absorbance and circular dichroism methods, which wi l l be addressed in the following sections. Table 2.1: D N A oligomers used in this study were designed to address the issues relating to factors that determine the propensity for B*-form conformation. A-tracts are underlined for clarity. These oligomers are grouped into four different sets: set I oligomers were used as control sequences for comparative purposes; set II oligomers contain centrally located A-tract sites; set III oligomers contain A-tract sites which are positioned at the 3'-end; set IV oligomers were used to determine the influence of neighboring base pairs which flank the A-tract; set V oligomers were evaluated for conformational effects on the A-tract structure. Set Name Sequence (12-mers) T m (°C) Set I N A 0 1 d ( T A T T A T A A T A T A ) 36.5 EE07 d ( A A A A A A A A A A A A ) 47.7 Set II P A M 3 d ( A T T A T A A A T A T T ) 37.8 P A M 4 d l T A T T A A A A T T A T ) . 42.6 P A M 6 d C T A T A A A A A A T A T ) 42.8 P A M 8 d ( T T A A A A A A A A T T ) 43.4 P A M 10 d(T A A A A A A A A A A T ) 45.5 Set III P A E 3 d ( T A T T A T A A T A A A ) 36.9 P A E 4 d f T A T T A T A T A A A A ) 35.5 P A E 6 d ( T A T T A T A A A A A A ) 39.5 P A E 8 d f T A T T A A A A A A A A ) - 40.2 P A E 10 d ( T T A A A A A A A A A A ) 41.2 Set IV P A M 4 d C T A T T A A A A T T A T ) 42.6 P A M 4b d f T A A T A A A A T A A T ) 40.0 P A M 4c d ( A T A T A A A A T A T A ) 38.9 S e t V EE07 d ( A A A A A A A A A A A A ) ." 47.7 A L N A 1 d ( A A A L A A A L A A A L A A A ) 52.3 EE08 d ( T T T T A T A A T A A A ) 36.2 A L N A 2 d ( T T T L T A T L A A T L A A A ) 43.1 29 2.3 Buffer and Sample Preparation The p H 7.0 buffer solution consisted of 1 M N a C l , 10 m M N a 2 H P 0 4 , and 1 m M Na2(EDTA). E D T A chelates divalent or trivalent cations that nucleases need to degrade D N A . E D T A also prevents the binding of these cations to D N A , which may adversely affect the melting process of the D N A during the experiments. Sodium chloride was used to adjust the ionic strength and total N a + concentration. The buffer was filtered through a 0.2 urn syringe filter to remove microbes and dust particles. D N A oligomers were generously contributed by C. A . Haynes, who purchased them from Integrated D N A Technologies, Inc. (Coralville, IA, U S A ) and were used in this work without further purification. The initial concentrations were estimated based on U V absorbance at 260 nm . Stock solutions of the double-stranded forms were prepared by mixing equimolar amounts of complementary strands. Samples of 5.0 u M D N A oligonucleotides were then prepared by diluting an aliquot of the stock solution with the buffer solution. Prior to use, each sample was annealed by heating to 80 °C, followed by cooling to room temperature at a rate of 0.5°C/min. 2.4 U V Absorbance Measurements For the purpose of obtaining thermal difference spectra (TDS), the protocols for obtaining U V absorbance measurements closely follows the method described by Mergny et al . Absorbance measurements were conducted using a Varian Cary 4000 U V - V i s double-beam spectrophotometer, equipped with a Peltier temperature control accessory. Using 1 cm pathlength cells, the U V absorbance measurements were acquired at the temperatures that correspond to the fractions of unfolded and folded D N A are < 1% and > 95%, respectively, as predetermined calorimetrically by members of the C. A . Haynes 30 laboratory. In other words, the T m o f the D N A must lie between these two temperature values. For all the samples, the choice of the upper and lower temperature was > 75°C and <15°C, respectively. U V absorbance spectra were collected at every 1 nm data interval with a scan speed of 600nm/min for the wavelength range of 240 - 340 nm. T D S were obtained by subtracting the low temperature spectrum from the high temperature spectrum. 2.5 Circular Dichroism Measurements C D spectra were acquired using a J A S C O 810 Spectropolarimeter equipped with Peltier devices. A l l measurements were obtained in a nitrogen environment to remove oxygen from the instrument because the high intensity xenon light converts oxygen to ozone which destroys the optics in the C D instrumentation. The D N A oligomers were prepared in buffer to a final total strand concentration of 25 p M . Each oligomer sample was placed in a 2.0 mm pathlength cuvette (Suprasil, Hellma, U K ) . The solution was heated to 75 °C for 10 minutes and slowly cooled to 15 °C. Samples were allowed to equilibrate at the particular temperature studied for 15 minutes prior to the measurement. The spectropolarimeter was scanned from 200 nm to 340 nm at a scan rate of 50 nm/min. Data points were acquired for every 1 nm with a response time of 4 seconds. A l l C D data were the average of five accumulations and were buffer-subtracted. To ensure little or no instrumental drift, baselines were recorded both before and after recording the sample spectrum and averaged. 31 Chapter 3 Results and Discussion 3.1 Thermal Difference Spectra by UV Measurements Figure 3.1 A shows the U V absorbance spectra of double-stranded and single-stranded forms of d(A)i2 at 15°C and 75°C, respectively. The change in U V light absorption during D N A denaturation is attributed to the change in e between the single-stranded and double-stranded forms. The major peak occurring at ~ 260 nm is ascribed to 7t —* 7t* transitions 5 9 ' 6 1 . The increased absorbance at high temperature indicates that the base-stacking interactions in the native structure are disrupted when the oligomer sample is heated. The subtraction of the low temperature spectrum from the high temperature spectrum resulted in the TDS displayed in Figure 3 . IB, characterized by a maximum peak . at 260 nm and a slight shoulder located at ~ 295 nm. While the difference absorbance band at 260 nm again reflects 7t —• 7r* transitions, the slight shoulder at -295 nm is indicative of the presence of n —* n* transitions 5 9 ' 6 1 . The shapes of all the resulting TDS are quantitatively similar to that of d(A)i2 (Figure 3. IB) since all of the oligonucleotide samples used in this study have 100% A T content. This finding is in agreement with a previous suggestion that TDS shapes depend on base composition and are useful in reflecting the subtleties of base-stacking interactions 1 1 0. For example, Figure 3.2 compares the TDS of the control sequences d(A)>2 (EE07) and d ( T A T T A T A A T A T A ) (NA01). Figure 3.2 shows how the two T D S curves differ, with d ( A ) i 2 having a substantially higher intensity than that observed for d ( T A T T A T A A T A T A ) . Since the n —* n* and n —> n* transition moments are perpendicular and parallel, respectively, to the helix axis, increased 32 stacking resulting from the formation of a folded structure of d(A)i2 may result in hyperchromism 5 9 , 1 1 0 . The differences between the T D S for all the samples in this study are mainly characterized by the exact location and intensities of the maximum peak, as summarized in Table 3.1. A s shown in Table 3.1, the wavelength maxima of the heteropolymeric d ( T A T T A T A A T A T A ) and the homopolymeric d ( A ) i 2 sequence are 259 nm and 263 nm, respectively. This result is in agreement with findings of Mergny et al. that the peak maxima positions of sequences containing alternating A T motifs are shifted towards shorter wavelengths compared to sequences with A n T n b locks 1 1 0 . This observation can be explained by the differences in base-stacking interactions that stabilizes the formation of different conformations, i.e. B*-form or B-form, as a result of different interactions between the A-tract region and the neighboring sequences. The plot shown in Figure 3.3 reveals that the difference in absorbance at 260nm increases as a function of A-tract length in the sequences studied. The nature of the observed trend for P A M sequences exhibits a significant increase in intensities going from n = 3 to n =6, where n is the number of consecutive adenines, but increases slowly as the A-tract length approaches that of d(A)i2 (Figure 3.3). Overall, P A M sequences displayed an additive behavior where the intensity of the T D S band at 260 nm increased in a near-linear fashion over the range of lengths examined. In the case of P A E sequences, the intensities increased going from n = 3 to n = 5, with n=6 having the greatest intensity. However, when n is increased to 8 and 10 consecutive adenines, the intensities decreased. This observation found for P A E sequences could result from base-stacking interactions due to narrowing of the groove within the A-tract being most efficient after approximately six consecutive adenines 1 1 1, but declining thereafter. 33 -0.1 • 240 —I— 260 —I— 280 300 320 340 Wavelength (nm) Wavelength (nm) Figure 3.1: Thermal denaturation of the control sequence d(A)i2 (EE07) monitored by U V absorbance measurements. (A) Comparison of the U V absorbance spectra collected at 15°C (blue line) and 75°C (red line). The major peak at 260 nm reflects the presence of 7i —> 7t* transitions as a result of base-stacking interactions. (B) Thermal difference spectrum of poly(dA)-poly(dT) (black line) results from the subtraction of the U V absorbance spectrum at 15°C from the spectrum at 75°C. 34 Figure 3.2: Comparison of the U V absorbance difference spectral signatures of the homopolymeric d(A) i2 (EE07) and heteropolymeric d ( T A T T A T A A T A T A ) (NA01) sequences. The former D N A sequence is representative of a B*-form conformation, whereas the latter is assumed to adopt B-form conformation. Number of consecutive (dA) residues Figure 3.3: Correlation between the ~260 nm peak intensities of the U V difference spectra as a function of A-tract length. The A absorbance values on the y-axis are obtained from the difference between the high (75 °C) and low (15°C) temperature U V absorbance spectra scanned at 260 nm. A-tract length is determined from the number of consecutive adenine residues within the A-tract site. 35 Table 3.1: The length and positional effects of A-tracts on the Xmax and peak intensities at ^max of the U V absorbance difference spectra. Name Sequence A , m a x b (nm) Differential Other (12-mer) absorbance intensity at '"-max peaks/features N A 0 1 a d ( T A T T A T A A T A T A ) 259 0.138 * E E 0 7 a d ( A A A A A A A A A A A A ) 263 0.190 ** P A M 3 d ( A T T A T A A A T A T T ) 263 0.146 * P A M 4 d ( T A T T A A A A T T A T ) 263 0.160 ** P A M 6 d ( T A T A A A A A A T A T ) 263 0.179 ** P A M 8 d ( T T A A A A A A A A T T ) 263 0.187 ** P A M 10 d ( T A A A A A A A A A A T ) 263 0.189 ** P A E 3 d ( T A T T A T A A T A A A ) 259 0.148 ** P A E 4 d ( T A T T A T A T A A A A ) 259 0.164 ** P A E 6 d ( T A T T A T A A A A A A ) 257 0.205 ** P A E 8 d ( T A T T A A A A A A A A ) 258 0.168, ** P A E 10 d ( T T A A A A A A A A A A ) 258 0.173 ** - indicates negative peak 283 nm * indicates shoulder located between 287 - 305 nm ** indicates shoulder located between 290- 300 nm a Control sequences b Standard deviations for Xmax at 260 nm is ± 2 nm c Standard deviation for differential absorbance intensities at 260 nm is 0.020 36 3.2 Temperature-Dependent C D Spectra E o 9. UJ 00 CM O Q. -10 <D T3 £ -12 -14H -16 •20 -22 Wavelength (nm) B EE07 10 20 —I— 30 40 50 60 70 80 Temperature (°C) Figure 3.4: Thermal denaturation of d(A)i2 (EE07) monitored by C D spectroscopy. (A) The C D spectra of d ( A ) n (EE07) over a temperature range of 15°C to 75°C. The decrease in ellipicity at elevated temperatures above the melting temperature of d(A)i2 (49.3°C) reflects a decrease in the level of secondary structure. (B) The change in ellipticity at 248 nm with temperature. The solid curve represents the best fit line of the data from 25°C to 75°C, resulting in a melting curve that illustrates the two-state melting behaviour by which the process of denaturation proceeds. The melting behavior of the oligomer at temperatures below 25°C is often referred as a pre-melting transition 1 1 3 . 37 Figure 3.4A shows detailed structural information provided by the C D spectra of d(A)i2 obtained over a series of temperatures ranging from 15°C to 75°C. Changes in the shape and magnitude of the C D intensity extrema reflects progressive disruption of base-stacking and base-pairing interactions as the temperature increases from 15°C to 75°C. A t temperatures below the melting temperature of d (A)n (49.3°C), the C D spectra of the duplexes are similar, but not identical. A s the temperature increases beyond the melting temperature, more dramatic changes appear in terms of the magnitude and the shape of the peaks at 220 nm, 260 nm, and 280 nm, as well as the trough at 248 nm. These spectroscopic changes suggest that the duplexes initially are structurally or conformationally different at low temperatures and become structurally or conformationally similar at temperatures above the melting temperature 1 1 2. A t higher temperatures, the bases have a random orientation relative to one another, resulting in the decrease in ellipticity of the peaks and trough over the wavelength region studied. Also seen in Figure 3.4A is the presence of an isoelliptic point located at 264 nm, which indicates a two-state transition between the double-stranded and single-stranded states with no intermediate steps oyer the entire temperature range s tud ied 1 1 3 ' 1 1 4 . It is clear from the C D spectral data that the conformation adopted by the oligomer sample is dependent on temperature, as the relative peak magnitudes and shapes change with temperature. A t the appropriate wavelength, C D appears to represent the melting curve of the D N A oligomer. The largest temperature-dependent intensity change occurs at 248 nm (Figure 3.4A). Figure 3.4B illustrates the resulting plot of a C D melt performed at 248 nm reflects a sigmoidal curve which is qualitatively similar to a typical melting curve 38 obtained by U V measurements at 260 nm. This indicates that C D melting studies may also be used to analyze the thermal stability of D N A oligomers. 3.3. C D Spectral Characteristics of B*-form of A-tract Structure ' I ' 1 • 1 ' 1 ' 1 1 1 . r-200 220 240 260 280 300 320 Wavelength (nm) Figure 3.5: C D spectrum of d(A)i2 (EE07, red line) exhibiting B*-form spectral characteristics. The C D spectrum of B-form d ( T A T T A T A A T A T A ) ( N A 0 1 , blue line) is shown for comparison . The C D spectra of EE07 and N A 0 1 are measured at 15°C and 10°C, respectively. Structurally, the heteropolymeric d ( T A T T A T A A T A T A ) (NA01) , being in the classical B - D N A conformation, is significantly distinct from the homopolymeric d(A)n (EE07) , which adopts the B * conformation 1 1 5 ' 1 1 6 . Structural differences between the two oligomers are reflected, in particular, in the unusually high torsional rigidity of the homopolymeric d ( A ) n compared to the mixed d ( T A T T A T A A T A T A ) sequence 1 1 7" 1 1 9. Figure 3.5 compares the C D spectral features between d ( A ) ^ (EE07) and d ( T A T T A T A A T A T A ) (NA01), which are used as reference B*-form and B-form structures, respectively. The sequence d(A)i2 displays a C D spectrum with several 39 characteristic C D bands located at 209 nm, 220 nm, 248 nm, 261 nm, and 285 nm (red line). In contrast, the C D spectrum of d ( T A T T A T A A T A T A ) (blue line) indicates a conformation expected of a normal B-type helix , as evidenced by the absence of a ~ 260 nm band and its 'conservative' character composed of two bands at 248 nm and 280 nm of negative and positive ellipticity, and of nearly equal rotational strength (blue line). This is in good qualitative agreement with literature f i n d i n g s 1 2 1 - 1 2 3 : B-form of D N A have a positive ellipticity maximum at 260 nm, a weaker negative band at 250 nm, and a strong negative band at about 210 nm; B-form is defined by C D spectral characteristics of a null near 260 nm, with a broad positive peak in the region of 280 nm and a negative ellipticity maximum near 250 nm. The bands located near 258 nm and 272 nm have previously been assigned to adenine and thymine, respectively 1 2 4. The C D peak at ~ 260 nm for the d(A)i2 (EE07) spectrum is particularly distinct compared to the B-form C D spectrum of d ( T A T T A T A A T A T A ) (NA01). Thus, the 260 nm C D peak may reveal structural information specific to A-tracts, particularly the formation of B*-form structure, and can be attributed to adenine-adenine stacking. The absence of a 260 nm C D signal and a significant difference in the overall C D spectral features of d ( T A T T A T A A T A T A ) (NA01) mainly results from the absence of A-tracts. This comparative study of the two control sequences to distinguish between B*-form and B -form is required to evaluate whether A-tracts can produce a distinguishable fine structure in the C D of other sequences containing A-tracts (Set II, III, and IV, Table 2.1). 40 3.4 CD Difference Spectra In section 3.1, the analysis of the U V absorbance results for different sequences is mainly based on the T D S plots. In the same fashion, the C D difference spectra for all the oligonucleotides used in this study w i l l be generated. According to an earlier discussion by Greve et al., factors such as base-to-backbone and base-base interactions contribute to a smaller extent to the difference spectrum than to an original C D spectrum as they are largely subtracted 1 1 4. Difference C D spectra are better resolved than the original C D spectra and mainly reflect interstrand interactions since intrastrand interactions are largely subtracted. Thus, the advantage of using C D difference spectra lies in their simplicity and subsequent ease of interpretation 1 1 4. In sections that follow, it w i l l be shown that the C D difference spectra of set II, set III, and set IV sequences typically have a positive maximum at 248 nm and a global minimum at 260 nm. Having discussed the overall C D spectral features of conformations of B*-form, we w i l l now address the details of the conformational perturbations induced by the A-tract site as a function of the A-tract length and position within duplex D N A . The impact of the base pairs neighboring the A-tract site w i l l also be discussed. . 41 3.4.1 Analysis of P A M sequences Figure 3.6: C D spectra at 15°C of P A M sequences with different A n-tract lengths (where n = 3, 4, 6, 8, and 10). The A-tract is located within the central portion of the 12 bp long sequence. - 1 3 - 1 1 1 1 1 1 1 1 1 1 1 1 1 1 220 240 260 280 300 320 340 Wavelength (nm) Figure 3.7: C D difference spectra for P A M sequences with various A n-tract lengths, where n = 3, 4, 6, 8, and 10. These spectra are generated by subtracting the C D spectrum at 15°C from the C D spectrum at 75°C. The C D difference spectra of the control sequences d ( A ) i 2 (EE07) and d ( T A T T A T A A T A T A ) (NA01) are included for comparison. 42 The low temperature C D spectra for the P A M series with A n-tract where n = 3, 4, 6, 8, and 10 are shown in Figure 3.6. These spectra are characterized by peaks centered at 220 nm, 260 nm and between the region of 270 - 280 nm, as well as troughs located at 208 nm and 248 nm. The overall trends of the low temperature C D spectra suggest that the base-stacking interactions, as indicated by the increase in peak height and the increase in trough depth, becomes larger as the length'of the A-tract increases. The low temperature C D spectra of d ( T T A A A A A A A A T T ) ( P A M 8) and d ( T A A A A A A A A A A T ) ( P A M 10) resembles the reference B*-form C D spectrum of d(A)i2 (EE07), as indicated by the similarity in the shape of the ~ 260 nm band. Nevertheless, some differences occurring in the region of 270 - 280 nm between the C D spectra of EE07, P A M 8, and P A M 10, may be attributed to the influence of the thymine residues located at the 5'- and 3'- ends of the duplex. The shape of the low temperature C D spectra for d ( A T T A T A A A T A T T ) ( P A M 3), d ( T A T T A A A A T T A T ) ( P A M 4), and d ( T A T A A A A A A T A T ) ( P A M 6) approaches that of a B-form C D spectrum of d ( T A T T A T A A T A T A ) (NA01), as indicated by the formation of a broad C D band in the region between 270 - 280 nm. The intensity of the C D band located at 260 nm, which reflects the contribution from the adenine bases 1 2 2, increases as a function of A-tract length. It is interesting to note that this particular trend is also reflected by the band at 220 nm. The intensity and the shape of all the C D bands are distinct for all P A M oligomers, possibly due to the structural differences related to the relative amount of B * -form adopted depending on the sequence context of base pairs neighboring the A-tract site. 43 Figure 3.7 compares the C D difference spectra of the series of P A M sequences to further evaluate the extent of B*-conformation formed due to A-tract length effects. Compared to other P A M sequences, the C D difference spectrum of P A M 10 bears the closest resemblance to EE07 in terms of the shape and intensity of the bands at 248 nm and 260 nm. However, the difference between P A M 10 and EE07 occurs in the band in the region between 270 - 280 nm^ strongly suggestive of either a n —*• n* transition on adenine or n —»• n* transition on thymine 1 1 4 . In addition, the peak at ~ 270 nm is less pronounced in P A M 10 compared to EE07. This possibly indicates that the destabilizing feature of T A base step within the sequence 1 2 5 causes the adenine bases within the A-tract to be slightly less efficiently stacked in P A M 10 compared to EE07 1 2 6 ' 1 2 7. The generally observed increases in intensities for both the peaks and troughs in the C D difference spectra for the P A M series are consistent with increasing base-stacking interactions as the length of the A-tract increase. There is no clear correlation between C D difference band intensities at 260 nm and A-tract length. However, the C D difference band intensities at 248 nm increase in a non-linear fashion with the length of contiguous adenine residues within the A-tract, as seen in Figure 3.10. A s the length dependence approaches a maximum value, the ellipticity approaches that of EE07 slowly, in agreement with the finding by Gudibande et al126. In general, the difference in ellipticity at 248 nm derived from the C D difference spectra presumably reflect the effect of A-tract length on the overall conformation of a given oligomer sequence. 44 3.4.2 Analysis of P A E sequences Figure 3.8: C D spectra at 15°C of P A E sequences with different A n-tract lengths (where n = 3, 4, 6, 8, and 10). The A-tract is located at the 3'-end portion of the 12 bp long sequence. - i o i 1 1 1 1 1 1 1 1 1 1 1 ' 1 220 240 260 280 300 320 340 Wavelength (nm) Figure 3.9: C D difference spectra for P A E sequences with various A-tract lengths, n, where n = 3, 4, 6, 8, and 10. These spectra are generated by subtracting the C D spectrum at 15°C from the C D spectrum at 75°C. The difference spectra of the control sequences d(A),2 (EE07) and d ( T A T T A T A A T A T A ) (NA01) are included for comparison. 45 Figure 3.8 shows a comparison of the low temperature C D spectra of the P A E series of oligomers. The structures formed by the P A E series show a lower tendency towards exhibiting the B*-form spectral characteristics as the A-tract lengthens. For all the P A E sequences, the C D spectra remain relatively unchanged in terms of the magnitude of the ellipticity at 260 nm while differing substantially at the 220 nm peak and the 248 nm trough. This implies a lower tendency towards adopting B*-form structure for P A E sequences relative to the results seen for P A M sequences. Figure 3.9 shows the C D difference spectra for P A E sequences. Here, the C D difference band at 248 nm displayed a general increase in intensity with an increase in A -tract lengths of P A E sequences. In contrast to P A M sequences, the intensity of ~ 260 nm positive peak is somewhat less enhanced for P A E sequences. Thus, end effects did not significantly affect the spectral features of the oligonucleotides. Despite the existence of longer A-tracts which might be expected to be susceptible to local alteration of the overall conformation, the lack of spectral changes at 260 nm may suggest decreased propensity for the formation of B*-form duplex from the 3'-end junction of the A-tract . Based on the C D difference spectra in Figure 3:9, the peak intensities at 248 nm for P A E sequences are plotted as a function of A-tract length in Figure 3.10. It is interesting to note that this trend for P A E is similar to a corresponding trend observed in U V difference spectra (Figure 3.3), suggesting that the U V transition at ~ 260 nm can be detected by ellipticity at ~ 248 nm. For example, the data point for d ( T A T T A T A A A A A A ) ( P A E 6) displayed an unexpectedly higher intensity in the C D difference ellipticity at 248 nm compared to d(T A T T A A A A A A A A ) ( P A E 8) and d ( T T A A A A A A A A A A ) ( P A E 10). Like the P A M series, the P A E series of sequences 46 generally also show considerable differences between the 248 nm intensities of the C D difference spectra as a function of A-tract length. - • - P A M PAE 4 6 8 10 Number of consecutive (dA) residues 12 Figure 3.10: The plot of 248 nm peak intensities of the C D difference spectra as a function of A-tract length for P A M (squares) and P A E (diamonds) series. The A ellipticity values on the y-axis are obtained from the difference between the high (75 °C) and low (15°C) temperature C D spectra at 248 nm. The trend for both series of sequences show that the 248 nm intensities based on the C D difference spectra is useful in providing insight into effects of A-tract length. 47 3.4.3 Effects of Sequences Flanking the A-Tract 10-, — i — 1 — i — 1 — i — i — i — — i — i — i — i -220 240 260 280 300 320 Wavelength (nm) Figure 3.11: Comparison of the C D difference spectra of d ( T A T T A A A A T T A T ) ( P A M 4), d ( T A A T A A A A T A A T ) ( P A M 4b), and d ( A T A T A A A A T A T A ) ( P A M 4c) to evaluate the effect of TT, A T and T A base pair steps flanking the A-tract site. These sequences contain the same A-tract length but have different sequence context in the base pairs neighboring the A-tract site. To properly investigate the influence of flanking sequences on A-tracts, three sequences with the same A-tract length are evaluated. The contrast between the T m values of sequences d ( T A T T A A A A T T A T ) ( P A M 4 ) and d ( T A A T A A A A T A A T ) (PAM4b) illustrates the presence of a base polarity effect, e.g. A T versus T A , and its influence on thermal stability. When T A base pairs are located adjacent to the A-tract site, towards both 5'- and 3'- directions, the T m decreased from 42.6°C ( P A M 4 ) to 40.0°C (PAM4b) . When two T A base pairs flanked both sides of the A-tract as seen in sequence d ( A T A T A A A A T A T A ) (PAM4c) , there is a further decrease in T m to 38.9°C. The decreasing trend in T m with the increasing number of flanking T A base pairs is also observed in the corresponding AS° and AFT values. These results suggest that a T A base 48 pair adjacent to the A-tract site has a destabilizing effect on the duplex, whereas greater stability is found with flanking T T (= A A ) and A T base pairs. These observations strongly agree with previous findings that emphasized the more efficiently stacked A T steps versus T A steps 1 2 9 ' 1 3 ° . Crystallography studies have shown that large conformational changes found at the T A and A T steps, especially at junctions of the A -tract and non-A-tract region, can lead to close cross-strand contacts between adenine bases across an A T step and may explain the different behavior of A T versus T A steps in 131 the context of A-tract induced curvature . The structural comparison based on C D difference spectra for P A M 4 , P A M 4 b , and P A M 4 c is illustrated in Figure 3.11. The band at 248 nm does not show significant differences in magnitude for all the sequences that are compared because all three of these sequences have the same A-tract length. This observation confirms that the intensity of the band at 248 nm is dependent on A-tract length. Knowing that these sequences contain different sequence context in the regions adjacent to the A-tract, the variability revealed in the C D difference band at ~ 260nm may reflect the detailed mutual orientations of these bases adjacent to the A-tract in the he l ix 6 7 . The band intensity obtained from the C D difference spectra at -260 nm decreased in the following order: P A M 4 c > P A M 4 b > P A M 4 , in agreement with the trends in thermodynamically observed results. A possible explanation for the observed v results is that the bases flanking the A-tract region may be a determining factor in the preferential stability of the overall conformation of the oligonucleotide. Since the 5' T A step has lower energy of stacking interaction ( A H = 6.0 kcal/mol) than A T ( A H = 8.6 kcal/mol) , less favorable stacking interaction exist for T A than for A T base steps, 49 causing the former to be more easily disrupted. This provides additional stabilization energy for the formation of B*-conformation induced by the adenine base-stacking within the A-tract in order to maximize n overlap 1 3 3 . Hence, a much more intense negative band at -260 nm resulted for P A M 4 c than P A M 4 and P A M 4 b . Overall, these results reveal a relationship between the extent of interaction with neighboring base pairs and the thermodynamic properties of the A-tract containing duplexes. The nature of flanking sequences could make the formation of B * - D N A more or less favorable. 3.5 Quantification of DNA Secondary Structure Although A-tract length information is conveniently accessed from the intensity of the bands at 248 nm from the C D difference spectra, its relationship to the amount of B*-form is complicated and can be significantly influenced by the effect of flanking sequences. In order to determine what truly determines B*-form content, it should be kept in mind that B*-form is a consequence of compact base-stacking arrangement2 4. Since we know that the sequences with A-tracts forming B*-form structures are characterized by a positive C D signal at ~ 260nm by referral to a reference spectrum, this peak is directly related to the type of secondary structure present and can be used to give an estimate for the relative population of B*-form in the oligomers studied. Analogous to the method used by Scarlett et al123 for determining the percentage A - or B - form duplex, d(A)i2 (EE07) is assumed to give the best estimate of 100% B * -form. N A 0 1 is not used to represent 0% B*-form since it has a slight signal at 260 nm, evident from its low temperature C D spectrum and C D difference spectrum. Rather, NA0.1 w i l l be used qualitatively as a reference B-forrri spectrum. A "flexrod" baseline, as illustrated in Figure 3.12, for CD260 of d(A)i2 (EE07) is used as a reference baseline for 50 calculating the relative peak heights of all the other sequences at -260 nm. The percentage B*-conformer is calculated by taking the ratio of the individual peak heights at -260 nm over the peak height of EE07. The percentage of B*-form content for both . P A M and P A E sequences, listed in Table 3.2, are plotted against A-tract length (expressed as the percentage consecutive (dA) residues), resulting in linear plots seen in Figure 3.13. B*-form content of a D N A duplex increases as a function of A-tract length. A good linear correlation is obtained for both P A M (R 2 = 0.912) and P A E (R 2 = 0.989) sequences. It should be noted, however, that this method of quantification is merely an estimation of B*-form content because these values are not corrected in terms of fraction D N A melted and the "flexrod" baseline is not the true baseline of the entire spectrum. Table 3.2: Summary of B*-form content in P A M and P A E sequences. The percentage B*-form for all the sequences are calculated based on the assumption that the control sequence d(A)i2 (EE07) best represents 100% B*-form. Name Sequence Peak C D 2 6 o % B*-form (12-mer) (As) N A 0 1 d ( T A T T A T A A T A T A ) 1.57 16.94 EE07 d ( A A A A A A A A A A A A ) ,9.27 100.00 P A M 3 d ( A T T A T A A A T A T T ) 2.38 25.67 P A M 4 d ( T A T T A A A A T T A T ) 1.60 17.26 P A M 6 d ( T A T A A A A A A T A T ) 3.00 32.36 P A M 8 d ( T T A A A A A A A A T T ) 4.80 51.78 P A M 10 d ( T A A A A A A A A A A T ) 8.76 94.50 P A E 3 d ( T A T T A T A A T A A A ) 1.95 21.04 P A E 4 d ( T A T T A T A T A A A A ) 2.88 31.07 P A E 6 d ( T A T T A T A A A A A A ) 4.15 44.77 P A E 8 d ( T A T T A A A A A A A A ) 5.61 60.52 P A E 10 d ( T T A A A A A A A A A A ) 6.93 74.76 51 Figure 3.12: A n illustration of the "flexrod" baseline (red line) used to obtain the magnitude of the C D signal (red arrow) at 260 nm. The 260 nm C D peak is attributed to adenine base stacking of the A-tract site. Since A-tracts are responsible for the formation of the overall B*-form structure, the magnitude of the 260 nm C D signal is used to estimate the amount of B*-form structure formed. • PAM • PAE Consecutive (dA) residues as % of total Figure 3.13: Quantification of B*-form content of P A M (squares) and P A E (diamonds) sequences with respect to A-tract length (expressed as consecutive (dA) residues as % of total). The correlation factor, R 2 , for P A M and P A E are 0.912 and 0.989, respectively. 52 3.6 C o m p a r i s o n o f S p e c t r o s c o p i c a n d T h e r m o d y n a m i c R e s u l t s 3.6.1 C o r r e l a t i o n o f U V M e a s u r e m e n t s w i t h T h e r m o d y n a m i c V a r i a b l e s 0.25 0.00 FT = 0 .866 35 37 39 41 43 45 47 T m (°C) B E c o CO CS CD o c CO J3 o CO < 0.25 0.20 0.15 0.10 0.05 0.00 F T = 0 .803 • FT = 0 .838 • PAM • PAE 58 60 62 64 66 68 70 72 74 76 78 AH 0 (kcal/mol) 53 2 0.05 0.00 FT = 0.820 160 170 180 190 200 A S 0 (cal/mol K) 210 220 Figure 3.14: Correlation between TDS peak intensities at 260 nm with (A) T m , (B) AFT, and (C) AS° for P A M (squares) and P A E (diamonds) sequences. The pink and blue lines represent the best linear fits for P A M and P A E data sets, respectively. The correlation factors for P A M and P A E , respectively, are: (A) R 2 ( P A M ) = 0.866 and R 2 (PAE) = 0.376, (B) R 2 ( P A M ) = 0.838 and R 2 (PAE) = 0.803, (C) R 2 ( P A M ) = 0.820 and R 2 (PAE) = 0.845. On the basis of T D S spectra obtained by U V measurements, the intensities of the maxima located at -260 nm reflects the increase in the absorbance of ultraviolet light that accompanies the unstacking of bases upon denaturation of duplex D N A . Correlation of differential absorbance values at 260 nm with T m (Figure 3.14A), AH° (Figure 3.14B), and AS° (Figure 3.14C) for P A M and P A E sequences generally show good linear correlations, as indicated by values of R > 0.8. A n exception is found for the relationship between T m o f d ( T A T T A T A A A A A A ) ( P A E 6) and its differential 260 nm absorbance value (Figure 3.14A), due to the possibility that there exists a critical A-tract length that most favors B*-l ike formation, particularly for the case where A-tract is positioned at the 3'end. 54 3.6.2 Correlation of C D Measurements with Thermodynamic Variables The thermodynamic parameters, T m , AH°, and A S 0 , for all the oligonucleotides used in this study are tabulated in Table 3.3. These results are obtained through differential scanning calorimetric experiments conducted by Curtis Hughesman of the C. A . Haynes lab. Thermodynamic analysis provides insight into the stability of J3*-form formation, thereby complementing the structural evidence of B*-form observed from U V and C D experiments. Table 3.3: Summary of the thermodynamic variables for D N A sequences containing A -tracts obtained by D S C measurements. Each of the 75 u M samples are suspended in a buffer solution containing 1- M N a C l , 10 m M N a 2 H P 0 4 , and 1 m M N a 2 ( E D T A ) (pH 7.0). (Reproduced with permission from Curtis Hughesman) Name Sequence T m a ( ° C ) AH° AS°, (12-mer) (kcal/mol) (cal/mol-K) N A 0 1 d ( T A T T A T A A T A T A ) 36.5 (35.9) b 58.4 (76.0) 167.1 (224.4) EE07 d ( A A A A A A A A A A A A ) 47.7 (46.3) 74.9 (82.3) 211.9 (236.1) P A M 3 d ( A T T A T A A A T A T T ) 37.8 (39.3) 63.5 (77.4) 182.6 (226.2) P A M 4 C d ( T A T T A A A A T T A T ) 42.6 (39.8) 72.1 (78.1) 206.6 (228.0) P A M 4 b c d ( T A A T A A A A T A A T ) 40.0 (39.8) 65.7(78.1) 188.2 (228.0) P A M 4 c d ( A T A T A A A A T A T A ) 38.9 (37.6) 58.5 (76.7) 165.8 (225.3) P A M 6 C d ( T A T A A A A A A T A T ) 42.8 (39.8) 72.4 (78.1) 207.6 (228.0). P A M 8 d ( T T A A A A A A A A T T ) 43.4 (44.2) 72.0 (80.9) 206.0 (233.4) P A M 10 d ( T A A A A A A A A A A T ) 45.5 (44.2) 76.5 (80.9) 218.5 (233.4) P A E 3 d ( T A T T A T A A T A A A ) 36.9 (38.1) 61.6 (77.4) 177.0 (227.1) P A E 4 d ( T A T T A T A T A A A A ) 35.5 (38.1) 60.5 (77.4) 174.5 (227.1) P A E 6 d ( T A T T A T A A A A A A ) 39.5 (40.4) 71.8(78.8) 208.1 (229.8) P A E 8 d ( T A T T A A A A A A A A ) 40.2 (42.5) 66.0 (80.2) 189.1 (232.5) P A E 10 d ( T T A A A A A A A A A A ) 41.2 (44.7) 64.5 (81.6) 183.4 (235.2) a Melting temperatures are calculated at the total oligomer strand concentration of 75 u M . Esimated errors in T m , A H , and AS are approximately 0.3 %, 15.1 %, and 16.5 % respectively. b The values in parentheses are predicted using the Santa Lucia nearest-neighbor model, whereas the values in non-parenthesis are experimentally determined values obtained from D S C plots. 0 D N A duplexes with identical nearest-neighbor base pairs. 55 A • PAM • PAE 58 6 0 6 2 64 6 6 6 8 7 0 7 2 7 4 7 6 7 8 AH° (kcal/mol) 5 6 c • PAM • PAE 160 170 180 190 200 210 220 AS 0 (cal/mol K) Figure 3.15: Correlation between C D difference spectra peak intensities at 248 nm with (A) T m (B) AFT and (C) AS° for P A M (squares) and P A E (diamonds) sequences. The pink and blue lines represent the best linear fits for P A M and P A E data sets, respectively. The correlation factors for P A M and P A E , respectively, are: (A) R 2 ( P A M ) = 0.821 and R 2 (PAE) = 0.675, (B) R 2 ( P A M ) = 0.652 and R 2 (PAE) = 0.811, (C) R 2 ( P A M ) = 0.612 and R 2 (PAE) = 0.789. The peak intensities at 248 nm obtained from the C D difference spectra of P A M and P A E series of oligomers are plotted against T m (Figure 3.15A), AFT (Figure 3.15B), and AS" (Figure 3.15C) in order to observe the correlation between structure and thermodynamic parameters. In each case, the increase in A-tract length induces thermal stabilization as seen with the increase in melting temperature and enthalpy values 1 3 4 . AFT increases more-or-less in parallel with the adenine content of the A-tract. This observation is reflective of increased base stacking and hydrogen bonding interactions 1 3 4 ' 1 5 . Overall, the larger entropy values for P A M sequences compared to P A E sequences suggest that the formation of B*-form is more enthalpically favored. The correlation of difference in ellipticity at 260 nm, (Ae)260, with thermodynamic variables lead to rather inconsistent results with small correlation coefficient values. This 57 situation illustrates the complexity behind contributions to thermodynamic stability and argues for more careful consideration of sequence effects. D N A sequence effects for the A-tract oligomers in this study can be divided into two groups: (i) sequence effects within the A-tract region, such as the interactions between all the neighboring base pairs, and (ii) sequence effects within the non-A-tract region, such as the differences in nucleotide composition of A-tract flanking sequences, and the influence of the base pairs at the 5'-and 3'-end. O f the 10 N N base pairs discussed in Chapter 1, A A (=TT) base steps takes on a particular significance because of its prominent role in the structure of A-tracts in solution and in A-tract-induced axis bending. Molecular dynamics results have shown that D N A bending and flexibility are highly correlated, i.e. steps that show the most intrinsic deformation from B-form D N A are also the most deformable 3 7. The trend for the flexibilities for base steps involving A and T is predicted to be in the following order, starting from the most flexible: T A > A T > A A 3 7 . The difference in the flexibilities is consistent with the results of gel electrophoretic mobility studies, where A4T4 showed a large gel migration anomaly indicative of D N A bending, whereas T4A4 shows normal B -D N A gel migration behavior 1 2 5 . In our work, d ( T A A A A A A A A A A T ) ( P A M 10) has the largest absolute (As)26o value, whereas d ( T T A A A A A A A A T T ) ( P A M 8) has the lowest absolute value of (Ae)26o, although they both contain identical number of N N base pairs, i.e. 9 A A / T T , 1 A T , and 1 T A . The reason for the two extreme values now depends on the evaluation of the sequence context at both the 5'- and 3'- end. Since P A M 10 has A T base steps and P A M 8 has T T base steps, it is obvious that A T base steps stabilizes B * -form structure of A-tract better than T T base steps. This results in a more intense negative C D band at 260 nm for P A M 10. Absolute values of (Ae)26o for P A M 58 sequences of n = 3, 4, and 6 range between that of P A M 8 and P A M 10 since these sequences have less than 9 A A / T T base pairs, and more than one each of A T and T A base pairs . P A E sequences show less difference in (Ae)26o values compared to P A M sequences, indicating that A-tract may have very little effect on the B*-form structure when it is located at the 3- end. Nevertheless, (Ae)26o values are useful for probing sequence effects. For example, P A M sequences of n = 3, 4, 6, and 8 have fewer A A / T T base steps compared to n = 10, and yet the difference between the largest and smallest (As)26o values for the P A E series is relatively less significant compared to the P A M series. Although (As)26o does not directly relate to thermodynamic values, it plays a role in representing the balance between the sequence context within the A-tract and non-A-tract region. It should be noted that the ability to understand and predict stability depends on knowing the sequence context and structural information that the molecule w i l l take based on the sequence. 3.7 Evaluation of the Nearest-Neighbor Model for A-Tract Oligomers In Table 3.3, the comparison between experimental and predicted values shows that the N N model generally gives results consistent with the expected trend with respect to secondary structural determining sequences, i.e. there is a general relationship between length of A-tract and stability of an oligomer. For instance, the T m increases from 39.3°C to 44.2°C for P A M sequences, and 3 8 . T C to 44.7°C for P A E sequences, going from the . shortest to longest A-tract. However, there are very large differences between the thermodynamic parameters for identical nearest-neighbor interactions. Consider the 59 sequences d ( T A T T A A A A T T A T ) ( P A M 4). d(T A A T A A A A T A A T ) ( P A M 4b), and d ( T A T A A A A A A T A T ) ( P A M 6), all of which contain the same nearest-neighbor distributions, i.e. five A A / T T , three A T , and three T A base pair interactions. Experimentally determined values of T m , A H " , and AS° for these three oligomers do not match their N N model predicted values. The average percentage error between the measured values and predicted values in T m , A H , and AS are estimated to be approximately 5.0 %, 10.3 %, and 11.9 %, respectively. These differences reflect the complexity underlying D N A secondary structure, which can contribute to the stability of each of these sequences. The differences in structures of P A M 4, P A M 4b, and P A M 6, give rise to the differences in shapes of the C D difference spectra seen in Figure 3.16, and are likely the reason why all three of these sequences differ in terms of enthalpy and entropy. This evidence shows that the current N N model cannot estimate the thermodynamic values of A-tract oligonucleotides with the required degree of accuracy. These data are highly suggestive that improvement made to the current prediction model of A-tract oligonucleotides must not only consider the sequence context, but must also account for structural differences that a D N A molecule can adopt. 60 10-1 -1 ' 1 ' 1 1 1 ' 1 1 1— 220 240 260 280 300 320 Wavelength (nm) Figure 3.16: C D difference spectra of d ( T A T T A A A A T T A T ) ( P A M 4), d ( T A A T A A A A T A A T ) ( P A M 4b), and d ( T A T A A A A A A T A T ) ( P A M 6), all o f which contain the same nearest-neighbor distributions, i.e. five A A / T T , three A T , and three T A base pair interactions, and are thus predicted to have the same thermodynamic parameters. Structural differences in each o f these sequences give rise to different enthalpy and entropy values based on experimental D S C results. 61 Chapter 4 The Impact of L N A on A-Tract Structure The practical advantages of locked nucleic acids ( L N A ) have received considerable attention due to their therapeutic and diagnostic properties. In this chapter, a preliminary insight into the structural influence of L N A on the formation of B*rform characteristic of A-tracts w i l l be examined. On the basis of our analysis, it is observed that the B*-form characteristics of poly(dA)-poly(dT) are attenuated in the presence of L N A . 4.1 Limitations of DNA Probes and Primers The Human Genome Project and the need to elucidate the many molecular pathways that underlie all aspects of human health have motivated the development of many new oligonucleotide-based technologies for genome analyses, diagnostics, or therapeutics. For example, antisense nucleic acids have been used for silencing the expression of specific genes ' and D N A microarrays provide a way to measure gene expression 1 3 8 . Ideal oligonucleotide probes and primers w i l l be able to discriminate between its intended target and all other targets under a given hybridization condition with minimal variation . This would involve the ability to identify an optimal set of conditions for hybridization and to accurately predict the interaction with their respective targets of probes and primers of any length, any sequence, and any chemical composition. However, in practice, DNA-based technologies have not reached their full potential due to the challenges faced in the process of optimal probe and primer design. For example, during performance testing of probes and primers, it is difficult to measure the 62 accumulation of the oligonucleotide in the target site and uncertainties in delivery may lead to toxic effects. Low-probe hybridization efficiency due to specificity and biostability problems 1 4 0 caused by changes in secondary structure of the target sequence can also limit the potential of D N A probes and primers. These issues, as well as the increasing demand for potential antisense agents and diagnostic probes which have high . selectivity and high-affinity recognition of complementary nucleic acid sequences has motivated an intensive search for nucleic acid analogues that have better properties than • natural DNA-based oligonucleotides. 4.2 Nucleic Acid Analogues Some D N A and R N A analogs exhibit improved stability and specificity compared to natural D N A and R N A , and they are becoming more commonly used for probes and primers and for antisense therapeutics 1 4 1 ' 1 4 2 . Several such chemically modified oligonucleotides have been developed and some are commercially available. It is beyond the scope of this chapter to present a thorough review, but the main types w i l l be briefly described in order to provide a context for the locked nucleic acids ( L N A ) , which is the type of primary interest in this work. 4.2.1 Peptide Nucleic Acids Peptide nucleic acids (PNA) have the same base structure as D N A or R N A , but the sugar phosphate backbone consists instead of repeating N-(2-aminoethyl)-glycine units and the nucleobases are attached with methylenecarbonyl l inkers 1 4 3 . The neutral character of the backbone of P N A eliminates the Coulombic repulsion that occurs in natural nucleic acid hybridization. Therefore, P N A binds with higher affinity to D N A and 63 readily forms P N A / D N A hybrid duplexes . P N A has also been found to be stable against nuclease, protease, and peptidase activity, indicating that it is more robust in cells than D N A , R N A , and proteins. Together, these key features are the reasons why P N A is sometimes used in biosensors and as molecular recognition probes used in hybridization experiments 1 4 5. However, the potential applications of P N A are limited due to its poor water solubility, synthesis complexity and expense, and tendency to form tertiary and quaternary structures in solution. 4.2.2 Phosphorothioate Oligonucleotides Phosphorothioate oligonucleotides are chemically modified D N A or R N A oligonucleotides where a phosphate-oxygen bond is replaced by a phosphate-sulfur bond in the nucleic acid backbone 1 4 6 . The advantage of phosphorothioate oligonucleotides is that, like P N A oligonucleotides, they are more resistant to nuclease degradation 1 4 6 than D N A or R N A . Therefore, phosphorotioate oligonucleotides are extremely useful as antisense molecules inhibiting gene expression. A n example of the first phosphorothioate oligonucleotide drug, Vitravene, has been used for the treatment of cytomegalovirus ( C M V ) retinitis in A I D S patients 1 4 7. However, they bind poorly to target cells, exhibit lower hybridization affinity than P N A , and appear to show sequence non-specific toxicity 146 in some systems . 64 4.2.3 Locked Nucleic Acids Locked nucleic acids ( L N A ) is a novel class of conformationally restricted R N A nucleotide analogue in which the ribose ring is modified by an additional methylene linkage between the 2'-oxygen and the 4 ' - ca rbon 1 4 8 ' 1 4 9 (Figure 4.1). The methylene linkage ' locks' the sugar moiety into a C3'-endo conformation, resulting in the high affinity hybridization of L N A nucleotides to complementary D N A and R N A strands. Generally, synthetic L N A nucleotides can be inserted into any position within a D N A or R N A oligonucleotide by standard phosphoramidite chemistry using commercial D N A synthesizers 1 4 8 ' 1 4 9 . Since the synthesis and physical properties, such as solubil i ty 1 5 0 , o f L N A closely resemble those for D N A , existing protocols for creating arrays need only minimal modifications 1 5 1 . J W A ( A A A 0 = P - 0 0 = P - 0 I I ( A A A ( A A A LNA DNA Figure 4.1: The molecular structures of a locked nucleic acid (LNA).nucleotide and a D N A nucleotide. The comparison between the two molecular structures shows that the ribose ring in L N A contains a 2 ' -0 , 4 ' -C methylene linkage. 65 4.2.3.1 Chemical Properties of L N A Besides its resistance to nuclease degradation , L N A contributes to the highest affinity ever obtained by Watson-Crick hydrogen bonding due to exceptionally high sequence specificity for the fully matched nucleic acid target 1 5 3 ' 1 5 4 . Oligonucleotides containing L N A exhibit enhanced thermal stabilities compared to unmodified D N A / D N A and D N A / R N A duplexes. The increased thermal stability is proposed to be the result of the bridging methylene group of L N A that confers conformational rigidity and the local organization of the phosphate backbone, leading to improved base stacking interactions 1 5 5. LNA-incorporated duplexes have shown melting temperature increases relative to n o n - L N A duplexes of between 4°C to 10°C per L N A m o n o m e r 7 9 ' 1 4 8 ' 1 5 5 , depending on the oligomers' length, sequence, and number of L N A bases in the oligomers. 4.2.3.2 Applications of L N A The high affinity of hybridization demonstrated by pure L N A oligomers may allow substantial improvement of D N A microarray technology. In addition, L N A has been widely used in probe molecules in hybridization-based diagnostics such as SNP genotyping 1 5 6 . L N A is also recommended for use in any hybridization assay that requires high specificity and/or reproducibility e.g. dual labeled probes, in situ hybridization probes, molecular beacons and P C R primers 1 5 7 . Since L N A offers the possibility to adjust T m values or primers and probes in multiplex assays, it provides a new means of primer and probe design that can overcome many of the design limitations associated with D N A oligonucleotides. Particular benefits of incorporating L N A into probes and primers include shorter lengths, increased selectivity, tight clamping of the 3'-end 66 irrespective of base composition, and the ability to hybridize at a specified temperature regardless of sequence 1 5 8. 4.3 Mot iva t ion Detailed N M R studies have shown that fully modified L N A : D N A and L N A : R N A hybrids adopt a canonical A-type duplex morphology 1 5 9 . L N A modifications have been found to impact the overall conformation by changing the distribution of functional groups in the minor groove and the overall helical geometry relative to unmodified D N A 1 6 0 , and by altering the counterion uptake and hydration pattern in the modified duplexes 1 4 0 . However, the structure and thermodynamics regarding the influence of L N A modification, particularly in A-tract containing oligonucleotides, is not very well known. Nevertheless, by keeping in mind that A-tracts play a role in replication and in transcriptional regulatory regions, the ability to control the properties of A-tracts through the use of L N A modifications may add to the potential of L N A as theurapeutic agents in antisense and antigene strategies. Using U V absorbance and C D methods, the structural perturbation of B*-form associated with the incorporation of LNA-modi f ied duplex w i l l be examined. It is also of interest to characterize and understand the thermodynamics of L N A modification in A-tract containing oligonucleotides. 67 4.4 Effects of L N A Substitution on Thermodynamic Stability Table 4.1: Thermodynamic parameters of two L N A : D N A duplexes and their corresponding unmodified D N A duplexes. These results are obtained by differential scanning calorimetry experiments for 75 u M samples, each of which is suspended in a buffer containing 1 M N a C l , 10 m M N a 2 H P 0 4 , and 1 m M N a 2 ( E D T A ) (pH 7.0). (Reproduced with permission from Curtis Hughesman) Name Sequence3 T m CC) AH° A S 0 (kcal/mol) (cal/mol-K) EE07 d ( A A A A A A A A A A A A ) 47.7 74.9 211.9 A L N A 1 d ( A A A L A A A L A A A L A A A ) 52.3 71.4 197.7 EE08 d ( T T T T A T A A T A A A ) 36.2 65.0 188.5 A L N A 2 d ( T T T L T A T L A A T L A A A ) 43.1 67.5 191.8 Superscript L refers to L N A modifications. The T m prediction tool for D N A - L N A duplexes exist 1 5 9 , but since it has been reported to have a relatively large standard deviation of 1.6°C and 5.0°C for D N A and D N A - L N A mixmer oligonucleotides , respectively. Hence, the predicted values are not listed in the way that has been done for sequences listed in Table 3.3. The higher prediction error for L N A oligonucleotides may be attributed to their more complex chemical properties. , Based on the thermodynamic results in Table 4.1, the comparison between T m o f A L N A 1 and A L N A 2 with their respective natural analogs, EE07 and EE08, reveals that the incorporation of three locked bases in A L N A 1 and A L N A 2 leads to a considerable increase in T m , indicating that the L N A modified duplexes are much more stable than their corresponding unmodified duplexes. The increased thermal stability resulting from the incorporation of L N A monomers can be analyzed through the values for the change in AH° (AAH°) and change in AS° (AAS°), respectively, based on equations (4.1) and 68 A A H ° - A H ° L N A + D N A : D N A - A H ° D N A : D N A ( 4 - 1 ) and A A S ° = A S ° L N A + D N A : D N A " A S ° D N A : D N A ( 4 . 2 ) where the subscript L N A + D N A : D N A refers to the duplex incorporated with L N A monomers and D N A : D N A refers to the control D N A duplex. A negative A A H " indicates a favorable enthalpy change, whereas.a negative A A S ° represents an unfavorable entropy loss 1 4 0 . Therefore, the enhanced thermal stability in A L N A 1 can be attributed to its more enthalpically favorable configuration, as implicated by the decrease in the enthalpic contribution ( A A H ° = - 3 . 5 kcal/mol). This decrease is accompanied by a smaller entropy value relative to the unmodified D N A duplex ( A A S ° = - 1 4 . 2 cal /mofK). The opposite is true for the entropically favored configuration of A L N A 2 , with A A H ° = 2 .5 kcal/mol and A A S " = 3.3 cal /moLK. The enthalpically driven stability for A L N A 1 and entropically driven stability for A L N A 2 is in agreement with the net result observed for other L N A : D N A duplexes 1 4 0 . The thermodynamic behavior exhibited by the two L N A : D N A duplexes relative to their corresponding D N A : D N A duplexes can be explained by the general phenomenon of enthalpy versus entropy compensation. The enthalpic term is mainly associated with hydrogen bonding energies and van der Waals interactions, whereas rearrangements of the molecules, solvent water, and counterions are manifested in the entropic term 1 4 0 . The formation of hydrogen bonds and base stacking interactions during the annealing process of D N A duplexes are enthalpically favored, but the loss of degree of freedom is entropically unfavorable and is manifested, in a negative entropy value. The compensating nature observed in terms of enthalpy and entropy ensures that the changes 69 in overall binding free energy, AG°, are small, permitting readily reversible associations in solution 1 6 1 . The thermodynamic results for A L N A 1 and A L N A 2 show the impact of L N A substitution to either enthalpy or entropy relative to the unmodified duplexes. For A L N A 1 and A L N A 2 , the locked C3'-endo conformation of the locked adenine bases locally organizes the phosphate backbone in order to increase the stacking efficiency of the nucleobases which is enthalpically favored 55 4.5 U V Absorbance and C D Analysis of L N A / D N A duplex 0.20 340 Wavelength (nm) Figure 4.2: U V absorbance difference spectra of d ( A A A L A A A L A A A L A A A ) ( A L N A 1 ) and d ( T T T L T A T L A A T L A A A ) ( A L N A 2 ) obtained by subtracting the U V absorbance spectrum at 15°C from the spectrum at 75 °C. Comparison is made with the U V absorbance difference spectra of the control sequences d ( A ) i 2 (EE07) and d ( T T T T A T A A T A A A ) (EE08) to show the effect of the incorporation o f L N A nucleotides. Although A L N A 1 and A L N A 2 shows enhanced thermal stabilities, the U V absorbance difference spectra in Figure 4.2 displays lower peak maxima intensities for A L N A 1 and A L N A 2 compared to EE07 and EE08, respectively, indicating increased 70 hypochromicity due to the greater effect of base-stacking in A L N A 1 and A L N A 2 . The general shapes of the difference spectra of A L N A 1 and A L N A 2 are very similar to that of their respective natural analogs, EE07 and EE08, indicating similar conformational characteristics. 2 0 - , 220 240 260 280 300 320 340 Wavelength (nm) Figure 4.3: C D spectra at 15°C of d ( A A A L A A A L A A A L A A A ) ( A L N A 1 ) and d ( T T T L T A T L A A T L A A A ) ( A L N A 2 ) in comparison to their respective natural analogs, d(A)i2 (EE07) and d ( T T T T A T A A T A A A ) (EE08). The features of the peaks located between 260 - 300 nm for A L N A 1 and A L N A 2 are very similar to that of EE07 and EE08, respectively, but with relatively higher intensities in this region. Figure 4.3 shows the C D spectrum of A L N A 1 and A L N A 2 compared to EE07 and EE08, respectively. The similarity in spectral shapes indicates that the modified duplexes share certain common characteristics with their corresponding unmodified duplexes in terms of base stacking geometries. However, in both cases, the spectra of the L N A modified duplexes exhibit a higher intensity for the bands located in the region of 260 - 300 nm. The ratio between the intensity of the 260 nm C D bands of A L N A 1 and EE07 reveals an enhancement by a factor of 3.3, implying that the presence of the locked bases in A L N A 1 and A L N A 2 increases the conformational stability within the molecule. 71 Given that a duplex conforms to an A-form helix at the site of the locked base 1 4 0 , it is expected that the structures of A L N A 1 and A L N A 2 would exhibit spectral characteristics of A-form. This is in contrast to our findings since the C D spectra of A L N A 1 seem to accentuate the conformation adopted by their respective unmodified sequences. The interpretation of this result is most likely that the C D shoulder at 260 nm indicates the presence of A-tracts rather than helix conformation. Also , the bands of enhanced intensities in those sequences containing L N A nucleotides indicates the fact that the sugar-phosphate rearranges in the vicinity of L N A nucleotides so as to further enhance base-stacking interactions. 4.6 Confirmation of the Presence of A-tracts The incorporation of L N A monomers shows a marked increase in T m , compared with the corresponding unmodified D N A duplexes, in accordance with the ability of L N A s to form the extremely stable Watson-Crick base-pairs 1 5 3. The thermal stability for A L N A 1 is due to an enthalpically favored conformational change in the presence of locked nucleotide bases, whereas the thermal stability for A L N A 2 is entropically driven. Based on the analysis of our results, the B*-form characteristics of A-tracts are enhanced in the presence of L N A , although this observation requires further verification. Nevertheless, the presence of A-tracts is marked by the appearance of the shoulder at 260 nm in the C D spectra. 72 Chapter 5 Conclusion 5.1 General Summary The B*-form structure induced by the presence of A-tracts can be characterized spectroscopically using U V absorbance and C D techniques. Although the analysis of U V difference spectra provides a rapid and simple method for qualitative structural information of D N A oligonucleotides, relying on a single optical technique may lead to incomplete characterizations and incorrect conc lus ions 1 6 2 ' 1 6 3 . Thus, C D measurements are carried out in addition to U V absorbance measurements to further evaluate the influence of A-tracts on the formation of B * - conformation, which may or may not be revealed using U V spectroscopy alone. For secondary structure determination, C D spectroscopy is particularly useful in detecting small changes in mutual orientation of neighboring bases in duplex D N A . Our U V absorbance and C D analysis suggest that the inherent tendency of forming B*-form by A-tract D N A oligomers can be affected, to different extent, by differences in A-tract length, the relative position of the A-tract position, presence of flanking sequences, and nearest-neighbor base pair interactions. The general objective of this structural study is to understand how each of these factors determine the extent of B*-form structure and how its formation is correlated with its thermodynamic properties. 73 5.2 B * - f o r m S t r u c t u r e o f A - t r a c t O l i g o m e r s In summary, the B*-form structure of A-tracts can be described as being rigid and less flexible than the standard B-form D N A . Its distinctive characteristics are slight base-pair inclination, high propeller twist, bifurcated hydrogen bonding, and a minor groove narrowing in the 5' to 3' direction. The most striking feature for the presence of B*-form is indicated by a 260 nm positive peak of the low temperature C D spectrum. Generally, the C D of A-tract containing oligonucleotides decreases with increasing temperature, suggesting a loss of secondary structure. Temperature variations perturb the A-tract D N A duplex structure, resulting in pronounced changes in their C D and U V absorbance spectra. One of the unique properties of A-tracts duplexes regards its spine of hydration in its minor groove, which has been proposed to provide an energetically favorable contribution that stabilizes the B*-form of the duplex structure2 7. Thus, the loss of secondary structure observed from our results may indicate a disruption of the spine of hydration, with a concomitant conformational alteration of the base pairs. The optical changes observed presumably arise from the latter event. The disruption of the spine of hydration consequently widens the minor groove geometry , based on the study of the nature by which daunomycin binds to the minor groove of poly(dA)-poly(dT). The binding interaction proceeds when daunomycin perturbs the structure of the polynucleotide in a manner similar to temperature. A substantiated explanation for secondary structure loss is based on the junction model 4 7 , in which deflection of the helix axis is localized at junctions between B*-form structure of A-tracts and B-form of non-A-tracts. A t low temperature, a structural bend forms at the interphase between the A-tract and random sequence section because the 74 secondary structure of the A-tract is atypical. A t high temperature, the secondary structure of an A-tract segment becomes similar to that of random sequence D N A , the junction disappears and the helix straightens4 6. 5.3 F a c t o r s G o v e r n i n g the P r o p e n s i t y f o r B * - f o r m 5.3.1 E f f e c t o f A - t r a c t L e n g t h The extent of B * - D N A conformation can be variable depending on the length of A-tracts. The change in U V absorbance at 260 nm is indicative of base-stacking interactions, resulting in increased magnitude in the 260 nm peak intensities of T D S when A-tract length increases. The C D spectra also show an increase in both peak height and the increase in trough depth, also indicating base-stacking interactions become more pronounced as the length of the A-tract increases. Dependence of C D spectra on A-tract length manifested in the 248 nm peak intensity of the C D difference spectra. In our experiments, the enhanced 248 nm peak intensity is a result of the increase in the B*-form character in oligomers with longer A-tracts. Base-stacking interactions stabilizing the B*-form is embedded in the shape and intensity of a peak at 260 nm in the original C D spectra, i.e. oligomers with longer A-tracts (n > 6) adopt geometries closer to B*-form as a consequence of increased base-stacking interactions. Structural characteristics of A -tracts are conferred by the stacking of adenine bases so as to maximize 7t overlap, high propeller twist, and localization of cations within the minor groove 2 4 . Another particularly important aspect of B*-form may be the existence of an intrahelical cross-strand bifurcated hydrogen bond 1 6 4 . Favorable n overlap of adenine-adenine stacking affects the A-tract region, causing high propeller twist along the longitudinal axis, which 7 5 in turn is responsible for the formation of additional non-Watson-Crick hydrogen bonds across the major g r o o v e 1 3 3 ' 1 6 4 ' 1 6 5 and for the stabilization of the spine of hydration 1 6 6 . The bifurcated hydrogen bonds are formed when the N 6 atom of adenine comes into close proximity with the 0 4 atom of thymine 1 6 4 . The existence of bifurcated hydrogen bonds implies that at least three adenines in tandem are needed to stabilize the bifurcated hydrogen bond interaction 1 6 4, but interestingly, at least four consecutive adenines are required to show the anomalously slow gel electrophoretic mobility of A-tracts. There is no clear evidence for the extent of each of these factors that comes into effect in an A -tract group of sequences, but it is believed that together, they contribute to the stabilization of B*- fo rm 1 6 6 . In general, there is a reasonably good quantitative agreement between trends observed for T D S peak intensities at 260 nm (Figure 3.3) and results from C D difference peak intensities at 248 nm (Figure 3.10). 5.3.2 Positional Effects of A-tract The relative positioning of A-tract has been observed to affect the amount of B * -form structure that is formed. When A-tract is adjacent to the 3'-end of non-A-tract regions ( P A E series), the extent of changes in the C D spectra is minimal. The C D spectral feature of the P A E series of sequences appears to resemble that of a reference B -form D N A structure, even in sequences with longer A-tracts. For the P A E series, the overall effect on the formation of B*-form is minimal or suppressed and the non-A-tract region becomes dominant in determining the overall D N A conformation. When A-tract is located between two non-A-tract regions ( P A M series), there exists a more pronounced propensity for B*-form with increasing A-tract length. 76 The positional effects of A-tracts based on the comparison between P A E and P A M sequences validates previous hypothesis that B*-form builds up progressively from the 5'-end 1 2 8 , mainly because A-tracts displays a unique and nonuniform structure 1 6 7. 5.3.3 Sequence Context of Flanking Sequences Sequences containing the same central A-tract flanked symmetrically on both sides by A T and T A base pairs in different order, d ( T A T T A A A A T T A T ) (PAM4) , d ( T A A T A A A A T A A T ) (PAM4b) , and d ( A T A T A A A A T A T A ) (PAM4c) , display differences in the C D difference spectra, particularly in the region of 260 - 280 nm. This implies the differential effect of flanking sequences on the overall formation of B*-form. A unique feature associated with A-tract D N A is that a 5 ' -AT-3 ' step does not disrupt the narrowing of the minor groove width of an A-tract and other properties associated with 97 the A-tract . Since the 5'-end of an A-tract is wider than the 3'-end, the width of the minor groove at the 5'- A T step may allow for a more optimally stacked interaction within the A-tract region, as evidenced by a more enhanced negative band at 260 nm in the C D difference spectrum of d ( A T A T A A A A T A T A ) ( P A M 4c). In contrast, a 5 ' -TA-3' step is disruptive to an A-tract and thus, C D conformational changes approaches that of a canonical B-form D N A structure. Qualitative comparison based on our results have shown that flanking sequences can modulate the extent of formation of B*-form. The results presented here suggest that the propensity for B*-form formation is increased in the following flanking sequence context: T A < T T < A T . Such influence is possibly due to the interactions between A -tracts and flanking (non-A-tract) regions because A-tracts are not polymorphic in nature 77 and their effect extends into the flanking sequences neighboring it on its 3'-side . Our results have shown that the effect of the flanking sequence is structural in nature and can be explained by the nonuniformity of the A-tract structure. 5.3.4 N e a r e s t - N e i g h b o r B a s e P a i r I n te rac t ions D S C studies by Curtis Hughesman of the Haynes laboratory have demonstrated that the thermodynamic parameters of A-tract oligonucleotides, especially those with identical nearest-neighbor pairs, are not well predicted by the most current nearest-neighbor model. This is clear evidence for the complexity of D N A structure, which may be overlooked by simple nearest-neighbor analysis. The comparison of C D difference spectra for sequences which have the same nearest-neighbor interactions have shown that stability prediction must be based on the type of structure w i l l take, besides simply knowing the sequence of oligonucleotides containing A-tracts. Specifically, a new and improved model must consider the properties of A-tract oligonucleotides, which have been the underlying factor for the differences in thermal stability. 5.3.5 C o n f o r m a t i o n a l E f fec t s In addition to the increasing the thermal stability, our structural data shows that the presence of L N A provides the possibility of controlling the effects of stacking in the B*-form structure of A-tracts. This implies that the constrained conformation induced by the additional methylene linkage in L N A further contributes to the rigidity of the A-tract structure, and helps to strengthen the stacking interactions of the adenine residues within the A-tract. Hence, the enthalpically-driven stabilization due to L N A compensates for 78 the loss in entropy due to decreased degrees of freedom in the well-stacked structure. It is, however, difficult to draw any final conclusions on this factor involving L N A because L N A is known to stabilize the duplex in an A-form conformation. The next obvious step is to determine the spectral characteristics that distinguishes B*-form from A-form and to carry more experiments comparing both L N A hybrid duplexes and their corresponding unmodified D N A duplexes. It is important to note that our results-to-date relating to the conformational effects on A-tract structure using L N A bases confirms that the characteristic shoulder at 260 nm of the C D spectrum indicates the presence of A-tract. 5 . 4 Impact of A-tract Structure on its Thermodynamic Properties Although the available quantitative data on B*-form is limited, we have shown that the unique structure attributed to the presence of A-tracts is generally responsible for the increased thermal and thermodynamic stabilization. The stability of A-tract containing D N A oligomers has been ascribed to the extent of B*-form structure, which has been explained in terms of A-tract length, position, flanking sequences, number of nearest-neighbor interactions, and the presence of locked nucleotide bases. This stresses the significance of the overall B*-form structure. The B*-form structure is stabilized by the compact base stacking interactions of several adenine bases in longer A-tracts. Efficient base stacking relates to better overlap of the Ti-electrons. In the thermodynamic sense, higher B*-form content results in increased thermal stabilization, which is governed by enthalpy-entropy effects, because more energy is required to destroy the highly efficient base-stacking arrangement in A-tracts. 79 5.5 Future Directions Our current results have opened new directions for an outlook on both fundamental and practical concerns. Undoubtedly, the current database of A-tract sequences must first be expanded and tested experimentally in order to further evaluate and improve the reliability of spectroscopic methods for secondary structure determination. A t this stage, attention is focused on AT- r i ch sequences containing A -tracts. This list o f sequences w i l l be expanded to accommodate other nearest-neighbor effects, involving for instance C G and G C base steps, to observe further influences o f sequence context on B*-form conformation. Accumulation of these systematic studies would provide a more solid framework for rapid predictions for most, i f not all , mixed sequences based on comparative sequence analysis. It has also been shown that structural features w i l l allow the refinement of thermodynamic parameters that can account for those characteristics. Further testing of a larger set of A-tract sequences w i l l also serve to test the quality of new thermodynamic rules and additional secondary structure prediction algorithms. A t present, this work may provide a starting point for the appropriate modification and refinement of existing structural models for A-tract D N A oligomers. Base-stacking interactions in A-tracts is a dominant factor in the stabilization of B*-form. Hydrogen bonds, electrostatic interactions of the charged phosphate groups, and thermodynamic effects (entropy of internal motions and solvent molecules) also contribute to helix stability. Since the stability of A-tract containing D N A duplexes is a sum of all these factors, another interesting avenue to consider is the quantification of the relative contributions from each of these interactions to the total stability of B * -80 conformation. This is an ambitious and somewhat difficult task that has not been approached because of the non-linear additive effects of many different stabilization factors 1 6 4. The B to A transition, known to regulate protein-DNA recognition, has been studied in detail through mixed ethanol water solvent experiments and molecular dynamics (MD) simulations 1 6 8 . The B to A transition is not a spontaneous process because a significant activation barrier exists between the two helical states, whereas the A to B transition occurs spontaneously on the nanosecond time scale. In a similar fashion, well-characterized structural studies on B * versus B-conformation wi l l provide a basis for more detailed M D simulations on the B * <-» B helix transition. Since we know the optimal range of chain lengths that drives the different extent of formation of B*-form in A-tracts, it is possible to analyze the B * <-> B equilibrium owing to large changes of the 168 C D and of the U V absorbance spectra . The dynamics of the B * <-> B - D N A transition w i l l be extremely valuable to further understand how this transition is mediated by A -tracts. For instance, what are the intermediates that exist, i f any, for this particular transition? If so, can their existence affect the preference of formation of one conformation over the other? Can dynamics be responsible for some characteristics of the current experimental observations? The role of D N A thermodynamics in molecular biology applications is well-documented and has been emphasized in the introductory chapter of this thesis. A brief overview on the rules and regulations for probe and primer design was mentioned, but was not listed in details because it is not the focus our work. However, many aspects of 81 the findings in this work do overlap the goals of perfecting those guidelines for practical applications, such as consideration of template secondary structure. The presence or formation of secondary structure by target or probe D N A is known to inhibit probe/primer hybridization, leading to poor or no yield of the product. The formation of unfavorable secondary structures of a primer can reduce the binding of primers to the template, depending on factors such as temperature and base compositions. This decreases the availability of primers to the reaction and adversely affects the product yield of P C R amplification. Hence, i f an unwanted type of secondary structure is known and predicted to form, careful measures can be taken to avoid them in order to optimally take advantage of the time and effort put into those applications. 82 References 1. Watson, J. D . and Crick, F. H . C. Molecular structure Of nucleic acids. Nature 171,737-738 (1953). 2. Watson, J. D . and Crick, F. H . C. Genetical implications of the structure of deoxyribonucleic acid. Nature 171, 964-967 (1953). 3. Saenger, W . Principles of Nucleic Acid Structure (Springer, N e w York, 1984). 4. Voet, D . , Voet, J. G . , and Pratt, C . V . Fundamentals of Biochemistry. (John Wiley and Sons, Inc., New York, 1999) 5. Marky, L . A . and Breslauer, K . J. Origins of netropsin binding affinity and specificity: correlations of thermodynamic and structural data. Proceedings of the National Academy of Science USA 84, 4359-4363 (1987). 6. Vorlickova, M . Conformational transitions of alternating purine-pyrimidine D N A s in perchlorate ethanol solutions. Biophysical Journal 69, 2033-2043 (1995). 7. Davies, D . R. and Baldwin, R. L . X-ray studies on two synthetic D N A copolymers. Journal of Molecular Biology 6, 251 - 255 (1963). 8. Waring, M . Facilitating structural transitions in D N A . Proceedings of the. National Academy of Science USA 97 ,11685- 11687 (2000). 9. Maniatis, T., Venable, J. M . , and Lerman, L .S . The structure of *F D N A . Journal of Molecular Biology 84, 37 - 64 (1974). 10. Fuller, W. , Wilkins, M . H . F. , Wilson, H . R., and Hamilton, L . D . The molecular configuration of deoxyribonucleic acid. IV. X-ray diffraction study of the A-form. Journal of Molecular Biology 12, 60 - 76 (1965). . 11. Arnott, S., Fuller, W. , Hodgson, A . , and Prutton, I. Molecular conformations and structure transitions of R N A complementary helices and their possible biological significance. Nature. 220, 561 - 564 (1968). 12. Arnott, S. The geometry of nucleic acids. Progress in Biophysics and Molecular Biology 21, 265-319 (1970). 13. Wang, A . H . J., Quigley, G . J., Kolpak, F.J. , Crawford, J. L . , van Boom, J. H . , Vander Marel, G . , and Rich, A . Molecular structure of a left-handed double helical D N A fragment at atomic resolution. Nature 282, 680-686 (1979). 14. Rao, S. N . and Kollman, P. On the Role of Uniform and M i x e d Uniform Sugar Puckers in D N A Double Helical Structures. Journal of the American Chemical Society 107, 1611-1617(1985). 15. Dickerson, R. E . and N g , H . L . D N A structure from A to B . Proceedings of the National Academy of Science USA 98, 6986-6988 (2001). . 16. Yoon, C , Prive, G . G . , Goodsell, D . S. and Dickerson, R. E . Structure of an alternating B - D N A helix and its relationship to A-tract D N A . Proceedings of the National Academy of Science USA 85, 6332-6336 (1988). 17. Dickerson, R. E . D N A structure from A to Z . Methods in Enzymology 211, 67-127(1992). 18. Hud, N . V . and Plavec, J. A unified model for the origin of D N A sequence-directed curvature. Biopolymers 69, 144-159 (2003). 83 19. Tolstorukov, M . Y . , Ivanov, V . I., Malenkov, G . G . , Jernigan, R. L . , and Zhurkin, V . B . Sequence-dependent B <-> A transition in D N A evaluated with dimeric and trimeric scales. Biophysical Journal 81, 3409-3421 (2001). 20. Wang, A . H . J., Quigley, G . J., Kolpak, F. J., Crawford, J. L . , van Boom, J. H . , van der Marel, G . , and Rich, A . Molecular structure of a left handed double helical D N A fragment at atomic resolution. Nature 282, 680-686 (1979). 21. Callahan L . , Han F. S., Watt, W. , Duchamp, D. , Kezdy, F. J., and Agarwal, K . B - to Z - D N A transition probed by oligonucleotides containing methylphosphonates. Biochemistry 83, 1617-1621 (1986). • 22. Peck, L . J . and Wang, J. C. Energetics of B-to-Z transition in D N A . Proceedings of the National Academy of Sciences USA 80, 6206-6210 (1983). 23. Arnott, S. and Seising, E . Structures for the polynucleotide complexes poly(dA)-poly(dT) and poly(dT)-poly(dA)-poly(dT). Journal of Molecular Biology 88,509-521 (1974). 24. Nelson, H . C. M . , Finch, J. T., Luis i , B . F., and Klug , A . The structure of an oligo(dA)-oligo(dT) tract and its biological implications. Nature 330, 221-226 (1987). 25. Arnott, S., Chandrasekaran, R., Hukins, D . W . L . , Smith, P. J. C , and Watts, L . Structural details of a double helix observed for D N A s containing alternating purine and pyrimidine sequences. Journal of Molecular Biology 88, 523-533 (1974). 26. Bolshoy, A . , McNamara, P., Harrington, R. E . , and Trifonov, E . N . Curved D N A without A - A : Experimental estimation of all 16 D N A wedge angles. Proceedings of the National Academy of Science USA 88, 2312-2316 (1991). 27. Chuprina, V . P. Anomalous structure and properties of poly (dA)-poly(dT). Computer simulation of the polynucleotide structure with the spine of hydration in the minor groove. Nucleic Acids Research 15, 293-311 (1987). 28. Dickerson, R. E . , and Drew, H . R . Structure of a B - D N A dodecamer. II. Influence of base sequence on helix structure. Journal of Molecular Biology 149, 761-786(1981). 29. Lesser, D . R., Kurpiewski, M . R., Waters, T., Connolly, B . A . , and Jen-Jacobson, L . Facilitated distortion of the D N A site enhances EcoRI endonuclease-DNA recognition. Proceedings of the National Academy of Science USA 90, 7548-7552 (1993). 30. Burkhoff, A . M . and Tullius, T. D . The unusual conformation adopted by the adenine tracts in kinetoplast D N A . Cell 48, 935-943 (1987). 31. MacDonald, D . , Herbert, K . , Zhang, X . , Pologruto, T., L u , P. Solution structure of an A-tract D N A bend. Journal of Molecular Biology 306, 1081-1098 (2001). 32. Nadeau, J. G . and Crothers, D . M . Structural basis for D N A bending. Proceedings of the National Academy of Science USA 86, 2622-2626 (1989). 33. Strahs, D . and Schlick, T. A-Tract bending: Insights into experimental structures by computational models. Journal of Molecular Biology 301, 643-663 (2000). 34. Drak, J. and Crothers, D . M . Helical repeat and chirality effects on D N A gel electrophoretic mobility. Proceedings of the National Academy of Science USA 88, 3074-3078 (1991). 84 35. Stellwagen, E . , L u , Y . , and Stellwagen, N . C. Curved D N A molecules migrate anomalously slowly in free solution. Nucleic Acids Research 33, 4425-4432 (2005). 36. Madhumalar, A . and Bansal, M . Structural insights into the effect of hydration and ions on A-Tract D N A : A molecular dynamics study. Biophysical Journal 85, 1805-1816(2003). 37. McConnel l , K . J. and Beveridge, D . L . Molecular dynamics simulations of B ' -D N A : Sequence effects on A-tract-induced bending and flexibility. Journal of . Molecular Biology 314, 23-40 (2001). 38. Chen, Y . Z . and Prohofsky, E . W . The role of a minor groove spine of hydration in stabilizing poly(dA)-poly(dT) against fuctuational interbase H-bond disruption in the premelting temperature regime. Nucleic Acids Research 20, 415-419 (1992). . < 39. Hagerman, P. J. Sequence directed curvature of D N A . Annual Reviews of Biochemistry 59, 755-781 (1990). 40. Ulanovsky, L . E . and Trifonov, E . N . Estimation of wedge components in curved D N A . Nature 326, 720-722 (1987). 41. Ganunis, R. M . , Guo, H . , and Tullius, T. D . Effect of the crystallizing agent 2-methyl-2, 4-pentanediol on the structure of adenine tract D N A in solution. Biochemistry 35, 13729-13732 (1996). 42. • Drak, J. and Crothers, D . M . Helical repeat and chirality effects on D N A gel electrophoretic mobility.. Proceedings of the National Academy of Science USA 88,3074-3078 (1991). 43. Dlakic, M . and Harrington, R . E . The effects of sequence context on D N A curvature. Proceedings of the National Academy of Science USA 93, 3847-3852 (1996). 44. Katahira, M . , Sugeta, H . , and Kyogoku,Y. A new model for the bending of D N A s containing the oligo(dA) tracts based on N M R observations. Nucleic Acids Research 18, 613-618 (1990). 45. Nejedly, K . , Sykorova, E . , Diekmann, S., and Palecek, E . Analysis of a curved D N A constructed from alternating dA n:dT n-tracts in linear and supercoiled form by high resolution chemical probing. Biophysical Chemistry 73, 205-216 (1998). 46. . Koo , H . S., W u , H . M . , and Crothers, D . M . D N A bending at adenine-thymine tracts. Nature 320, 501-506(1986). 47. Haran, T. E . , Kahn, J. A . , and Crothers, D . M . Sequence elements responsible for D N A curvature. Journal of Molecular Biology 244, 135-143 (1994). 48. Shatzky-Schwatz, M . , Arbuckle, N . , Eisenstein, M . , Rabinovich, D . , Bareket-Samish, A . , Haran, T. E . , Luis i , B . F., and Shakked, Z . X-ray and solution studies of D N A oligomers and implications for the structural basis of A-tract dependent curvature. Journal of Molecular Biology 267,595-623 (1997). 49. Wu , H . M . and Crothers, D . M . The locus of sequence-directed and protein-induced D N A bending. Nature 308, 509-513 (1984). 50. Zinkel , S. S and Crothers, D . M . D N A bend direction by phase sensitive detection. Nature 328, 178-181 (1987). 85 51. Trifonov, E . N . and Sussman, J. L . The pitch of chromatin D N A is reflected in its nucleotide sequence. Proceedings of the National Academy of Science USA 77, 3816-3820(1980). 52. Trifonov, E . N . Sequence-dependent deformational anisotropy of chromatin D N A . Nucleic Acids Research 8, 4041-4053 (1980). 53. Struhl, K . Naturally occurring poly(dA-dT) sequences are upstream promoter elements for constitutive transcription in yeast. Proceedings of the National Academy of Sciences USA 82, 8419-8432 (1985). 54. Roque, A . , Orrego, M . , Ponte, I., and Suau, P. The preferential binding of histone H I to D N A scaffold-associated regions is determined by its C-terminal domain. Nucleic Acids Research 32, 6111-6119 (2004). 55. Beloin, C , Jeusset, J., Re'vet B . , Mirambeau, G . , Le He'garatt, F., and Le Cam, E . Contribution of D N A conformation and topology in right-handed D N A wrapping by the Bacillus subtilis L rpC Protein. The Journal of Biological Chemistry 278, 5333-5342 (2003). 56. Skoog, D . A . , Holler, F. J., and Nieman, T. A . Principles of instrumental analysis. (Harcourt Brace and Company, Philadelphia, 1998). 57. Hammes, G . G . Spectroscopy for the biological sciences. (Wiley-Interscience, N e w Jersey, 2005). 58. Rich, A . and Kasha, M . The n —» TC* transition in nucleic acids and . polynucleotides. Journal of the American Chemical Society 82, 6197-6199 (1960). 59. Haque, K . A . , Pfeiffer, R. M . , Beerman, M . B . , Struewing, J. P., Chanock, S. J., and Bergen, A . W . Performance of high-throughput D N A quantification methods. BMC Biotechnology3,20-29 (2003). 60. Yakovchuk, P., Protozanova, E . , and Frank-Kamenetskii, M . D . Base-stacking and base-pairing contributions into thermal stability of the D N A double helix. Nucleic Acids Research 34, 564-574 (2006). 61. Falk, M . . The U V spectra of native and denatured D N A . Journal of the American . Chemical Society 86, 1226-1228(1964). 62. Tinoco, I. Jr. Hypochromism in polynucleotides. Journal of the American Chemical Society 82, 4785-4790 (1960). 63. von Ashen, N . , Oellerich, M . , Armstrong, V . W. , and Schutz, E . Application of a thermodynamic nearest-neighbor model to estimate nucleic acid stability and optimize probe design: Prediction of melting points of multiple mutations of apolipoprotein B-3500 and Factor V with a hybridization probe genotyping assay on the LightCycler. Clinical Chemistry 45, 2094-2101 (1999) 64. Ihtshamul, H . , Chowdhry, B . Z . , and Chaires, J. B . Singular value decomposition of 3-D D N A melting curves reveals complexity in the melting process. European Biophysics Journal 26, 419-426 (1997). 65. Movileanu, L . , Benevides, J. M . , and Thomas, G . J. Determination of base and backbone contributions to the thermodynamics of premelting and melting transitions in B - D N A . Nucleic Acids Research 30,3*/'61'-3777 (2002). 66. Albergo, D . D . , Marky, L . A . , Breslauer, K . J., and Turner, D . H . Thermodynamics of (dG-dC )3 double-helix formation in water and deuterium oxide. Biochemistry 20, 1409-1413 (1981). 86 67. Johnson, W . C. Jr. Circular dichroism and its empirical application to biopolymers. Methods of Biochemical Analysis 31, 62-125 (1985). 68. http://en.wikipedia.org/wiki/Circular_Dichroism 69. Service, R. F. D N A chips survey an entire genome. Science 281, 1122 (1998). 70. ' Binder, H . and Preibisch, S. Specific and nonspecific hybridization of oligonucleotide probes on microarrays. Biophysical Journal 89, 337-352 (2005). 71. L i , X . , He, Z . , and Zhou, J. Selection of optimal oligonucleotide probes for microarrays using multiple criteria, global alignment and parameter estimation. Nucleic Acids Research 33, 6114-6123 (2005). 72. Reymond, N . , Charles, H . , Duret, L . , Calevro, F., Beslon, G . , and Fayard, J - M . R O S O : optimizing oligonucleotide probes for microarrays: Bioinformatics 20 271-273 (2004). 73. Li l ley, D . M . J. and Dahlberg, J. E . D N A Structures Part A : Synthesis and Physical Analysis of D N A ^ Methods in Enzymology'-211, 533-567 (1992). 74. SantaLucia, J. Jr., A l l a w i , H . T., and Seneviratne, P. A . Improved nearest-neighbor parameters for predicting D N A duplex stability. Biochemistry 35, 3555-3562(1996). 75. Erlich, H . A . Gelfand, D . , Sninsky, J. J. Recent advances in polymerase chain reaction. Science 252, 1643-1651 (1991). x 76. Steger, G . Thermal denaturation of double-stranded nucleic acids: prediction of temperatures critical for gradient gel electrophoresis and polymerase chain reaction. Nucleic Acids Research 22, 2760-2768 (1994). 77. Chakrabarti, M . C. and Schwarz, F. P. Thermal stability of P N A / D N A and D N A / D N A duplexes by differential scanning calorimetry. Nucleic Acids Research 27, 4801-4806 (1999). 78. Wittung, P., Nielsen, P. E . , Buchardt, O., Egholm, M . , and Norden, B . Nature 368 ,561-563 (1994). 79. McTigue, P. M . , Peterson, R. J., and Kahn J. D . Sequence-dependent thermodynamic parameters for locked nucleic acid ( L N A ) - D N A duplex formation. Biochemistry 43, 5388 (2004). 80. Tolstrup, N . , Nielsen, P. S., Kolberg, J. G . , Frankel, A . M . , Vissing, H . , and Kauppinen, S. OligoDesign: optimal design of L N A (locked nucleic acid) oligonucleotide capture probes for gene expression profiling. Nucleic Acids Research 31, 3758-3762 (2003). 81. Chou, L . S., Meadows, C , Wittwer, C. T., and Lyon, E . Unlabeled oligonucleotide probes modified with locked nucleic acids for improved mismatch discrimination in genotyping by melting analysis. Biotechniques 39, 644-650 (2005). 82. Wahlestedt, C. et al. Potent and nontoxic antisense oligonucleotides containing locked nucleic acids. Proceedings of the National Academy of Science USA 97 5633-5638 (2000). 83. Oerum, H . and Wengel, J. Locked nucleic acids: a promising molecular family for gene-function analysis and antisense drug development. Current Opinion in Molecular Therapeutics 3, 239-243 (2001). 84. Atkins, P. Physical Chemistry. (Freeman, N e w York, 1998). 87 85. Wu , P., Nakano, S., and Sugimoto, N . Temperature dependence of thermodynamic properties for D N A / D N A and R N A / D N A duplex formation. European Journal of Biochemistry 269, 2821-2830 (2002). 86. Marky, L . A . and Breslauer, K . J. Calculating thermodynamic data for transitions of any molecularity from equilibrium melting curves. Biopolymers 26, 1601-1620 (1987). 87. Rouzina, I. and Bloomfield, V . A . Heat capacity effects on the melting of D N A . 1. General aspects. Biophysical Journal 11, 3242-3251 (1999). 88. Rouzina, I. and Bloomfield, V . A . Heat capacity effects on the melting of D N A . 2. Analysis of nearest-neighbor base pair effects. Biophysical Journal 11, 3252-3255 (1999). 89. Naghibi, H . , Tamura, A . , and Sturtevant, J. M . Significant discrepancies between van't Hoff and calorimetric enthalpies. Proceedings of the National Academy of Sciences USA 92, 5597-5599 (1995). 90. Weber, G . van't Hoff Revisited: Enthalpy of association of protein subunits. Journal of Physical Chemistry 99, 1052-1059 (1995). 91. Holbrook, J. A . , Capp, M . W. , Saecker, R. M . , and Record, M . T. Jr. Enthalpy and heat capcity changes for formation of an oligomeric D N A duplex: interpretation in terms of coupled processes of formation and association of single-stranded helices. Biochemistry 38, 8409-8422 (1999). 92. L i u , Y . and Sturtevant, J. M . Significant discrepancies between van't Hoff and calorimetric enthalpies. Biophysical Chemistry 64, 121-126 (1997). 93. Wu , P. and Sugimoto, N . Transition characteristics and thermodynamic analysis of D N A duplex formation: a quantitative consideration for the extent of duplex association. Nucleic Acids Research 28, 4762-4768 (2000). 94. Chaires, J. B . Possible origin of differences between van't Hoff and calorimetric enthalpy estimates. Biophysical Chemistry 64, 15-23 (1997). 95. Horn, J. R., Russell, D . , Lewis, E . A . , and Murphy, K . P. van't Hoff and calorimetric enthalpies from isothermal titration calorimetry: are there significant discrepancies? Biochemistry 40,1774-1778 (2001). 96. Borer, P. N . , Dengler, B . , Tinoco, I. Jr., and Uhlenbeck, O. C. Stability of ribonucleic acid double-stranded helices. Journal of Molecular Biology 86, 843-853 (1974). 97. Breslauer, K . J., Frank, R., Blocker, H . , and Marky, L . A . Predicting D N A duplex stability from the base sequence. Proceedings of the National Academy of Sciences USA 83, 3746-3750 (1986). 98. Wu , P., Nakano, S., and Sugimoto, N . Temperature dependence of thermodynamic properties for D N A / D N A and R N A / D N A duplex formation. European Journal of Biochemistry 269, 2821 -2830 (2002). 99. Santa Lucia, J. Jr. A unified view of polymer, dumbbell, and oligonucleotide D N A nearest-neighbor thermodynamics. Proceedings of the National Academy of Sciences USA 95, 1460-1465 (1998). 100. Santa Lucia, J. Jr. and Hicks, D . The thermodynamics of D N A structural motifs. Annual Review of Biophysics & Biomolecular Structure 33, 415^440 (2004). 88 101. Dlakic, M . and Harrington, R. E . The effects of sequence context on D N A curvature. Proceedings of the National Academy of Sciences USA 93, 3847-3852 (1996). 102. Dlakic, M . and Harrington, R. E . Bending and torsional flexibility of G/C-r ich sequences as determined by cyclization assays. The Journal of Biological Chemistry 270, 29945-29952 (1995). 103. Goobes, R. and Minsky, A . Contextual equilibrium effects in D N A molecules. Journal of Biological Chemistry 276, 16155-16160 (2001). 104. Doktycz, M . J., Morris, M . D . , Dormady, S. J., Beattie, K . L . , and Jacobson, K . B . Optical melting of 128 cctamer D N A duplexes: effects of base pair location and nearest neighbors on thermal stability. Journal of Biological Chemistry 270, 8439-8445 (1995). 105. Paiva, A . M . and Sheardy, R. D . Influence of sequence context and length on the structure and stability of triplet repeat D N A oligomers. Biochemistry 43, 14218-14227(2004). 106. Peyret, N . , Seneviratne, P. A . , A l l a w i , H . T., and SantaLucia, J. Jr. Nearest-neighbor thermodynamics and N M R of D N A sequences with internal A - A , C C , G G , and T T mismatches. Biochemistry 38, 3468-3477 (1999). 107. Minetti, C . A . S. A . , Remeta, D . P., Mil ler , H . , Gelfand, C. A . , Plum, G . E . , Grollman, A . P., and Breslauer, K . J. The thermodynamics of template-directed D N A synthesis: base insertion and extension enthalpies. Proceedings of the National Academy of Sciences USA 100, 14719-14724 (2003). 108. Sugimoto, N . , Satoh, N . , Yasuda, K . , and Nakano, S. Stabilization factors affecting duplex formation of peptide nucleic acid with D N A . Biochemistry 40, 8444-8451(2001). 109. Owczarzy, R., Vallone, P. M . , Goldstein, R. F., and Benight, A . S. Studies of D N A dumbbells VII: Evaluation of the next-nearest-neighbor sequence-dependent interactions in Duplex D N A . Biopolymers 52, 29-56 (1999). 110. Mergny, J -L, L i , J., Lacroix, L . , Amrane, S., and Chaires, J .B. Thermal difference spectra: A specific signature for nucleic acid structures. Nucleic Acids Research^, e l38 (2005). 111. Lankas, F., Cheatham, T. E . , Spackova, N . , Hobza, P.; Langowski, J. , and Sponer, J. Critical effect of the N 2 amino group on structure, dynamics, and elasticity of D N A polypurine tracts. Biophysical Journal 82, 2592-2609 (2002). 112. Fasman, G . D . Circular Dichroism and the Conformational Analysis of Biomolecules (Plenum Press, N e w York, 1996). 113. Herrera, J. E . and Chaires, J. B . A premelting conformational transition in poly(dA)-poly(dT) coupled to daunomycin binding. Biochemistry 28, 1993-2000 (1989). 114. Greve, J., Maestre, M . F. , and Levin, A . Circular dichroism of adenine and thymine containing synthetic polynucleotides. Biopolymers 16, 1489-1504 (1977). 115. Arnott, S., Chandrasekaran, R., Ha l l , I. H . , and Puigjaner, L . C. Heteronomous D N A . Nucleic Acids Research 11, 4141-4155 (1983). 89 116. Alexeev, D . G . , Lipanov, A . A . , and Skuratovskii, I. Y . Poly(dA)-poly(dT) is a re-type double helix with a distinctively narrow minor groove. Nature 325 , 821-823 (1987). 117. Chan, S. S., Breslauer, K . J., Hogan, M . E . , and Kessler, D . J. Physical studies of D N A premelting equilibria in duplexes with and without homo dA-dT tracts: ' Correlations with D N A bending. Biochemistry 29 , 6161 -6171 (1990). 118. Marky, L . A . and Macgregor, R. B . Jr. Hydration of dA-dT polymers: role of H 2 O in the thermodynamics of ethidium and propidium intercalation. Biochemistry 29 , 4805-4811 (1990). 119. Bhattacharyya, D . , Kundu, S., Thakur, A . R., Majumdar, R. Journal of Biomolecular and Structural Dynamics 17',289-300 (1999). 120. Chan, S. S., Breslauer, K . J., Austin, R. H . , Hogan, M . E . Thermodynamics and premelting conformational changes of phased (dA)s tracts. Biochemistry 32 , 11776-1 1784(1993). 121. Jean, J. M . and Hal l , K . B . Stacking-unstacking dynamics of oligodeoxynucleotide trimers. Biochemistry 43 10277-10284 (2004). 122. Basham, B . , Schroth, G . P., and Ho, P. S. A n A - D N A triplet code: Thermodynamic rules for predicting A - and B - D N A . Proceedings of the National Academy of Science USA 92 , 6464-6468 (1995). 123. Scarlett, G . P., Elgar, S. J. , Cary, P. D . , Noble, A . M . , Orford, R. L . , Kneale, G . G. , and Guille, M . J. Intact RNA-binding domains are necessary for structure-specific D N A binding and transcription control by C B T F during Xenopus development. Journal of Biological Chemistry 279 , 52447-52455 (2004). 124. Edmondson, S. P., and Johnson, W . C. Base tilt of poly (dA)-poly(dT) and poly (dAT)-poly(dAT) in solution determined by linear dichroism. Biopolymers 24 , 825-841 (1985). 125. Stefl, R., Wu , H . , Ravindranathan, S., Sklena, V . , and Feigon, J. D N A A-tract bending in three dimenstions: solving the d A 4 T 4 vs. d T 4 A 4 conundrum. Proceedings of the National Academy of Sciences USA 101 , 1177-1182 (2004). 126. Gudibande, S. R., Jayasena, S. D . and Behe, M . J. C D studies of double-stranded polydeoxynucleotides composed of repeating units of contiguous homopurine residues. Biopolymers 21, 1905-1915 (1988). 127. Brahms, S. and Brahms, J. G . D N A with adenine tracts contains poly(dA) -poly(dT) conformational features in solution. Nucleic Acids Research 18 ,1559-1564(1990). 128. Mollegaard, N . E . , Bail ly, C , Waring, M . J., and Nielsen, P. E . Effects of diaminopurine and inosine substitutions on A-tract induced D N A curvature. Importance of the 3'-A-tract junction. Nucleic Acids Research 25 , 3497-3502 (1997). 129. Kopka, M . L . , Goodsell, D . S., Han, G . W. , Chiu, T. K . , Lown, J. W. , and Dickerson, R. E . Defining GC-specificity in the minor groove: side-by-side binding of the di-imidazole lexitropsin to C - A - T - G - G - C - C - A - T - G . Structure 5, 1033-1046(1997). 130. White, S., Szewczyk, J. W. , Turner, J. M . , Baird, E . E . , and Dervan, P. B . Recognition of the four Watson-Crick base pairs in the D N A minor groove by -synthetic ligands. Nature 3 9 1 , 468-471 (1998). 90 131. Dixi t , S. B . , Pi t ici , F., and Beveridge, D . L . Structure and axis curvature in two dA6'dT6 D N A oligonucleotides: Comparison of molecular dynamics simulations with results from crystallography and N M R spectroscopy. Biopolymers 7 5 , 468-479 (2004). 132. Breslauer, K J . , Frank, R., Blocker, H . , and Marky, L . A . Predicting D N A duplex stability from the base sequence. Proceedings of the National Academy of Sciences USA 8 3 , 3746-3750 (1986). 133. Digabriele, A . D. , Sanderson, M . R., Steitz, T. A . Crystal lattice packing is important in determining the bend of a D N A dodecamer containing an adenine tract. Proceedings of the National Academy of Sciences USA 8 6 , 1816-1820 (1989). 134. Blake, R. D . and Delcourt, D . G . Thermal stability of D N A . Nucleic Acids Research 2 6 , 3323-3332 (1998). 135. Paiva, A . M . and Sheardy, R. D . Influence of sequence context and length on the structure and stability of triplet repeat D N A oligomers. Biochemistry 4 3 , 14218-14227 (2004). 136. Opalinska, J. B . and Gewirtz, A . M . Therapeutic potential of antisense nucleic acid molecules. Science STKE 2 0 6 , 1-5 (2003). 137. Crooke, S. T. Progress in antisense technology. Annual Review of Medicine 5 5 , 61-95 (2004). 138. Kane, M . D . , Jatkoe, T. A . , Stumpf, C. R., L u , J., Thomas, J. D . , and Madore, S. J. Assessment of the sensitivity and specificity of oligonucleotide (50mer) microarrays. Nucleic Acids Research 2 8 , 4552-4557 (2000). 139. Nielsen, H . B . , Wernersson, R., and Knudsen, S. Design of oligonucleotides for microarrays and perspectives for design of multi-transcriptome arrays. Nucleic Acids Research 3 1 , 3491-3496 (2003). 140. Kaur, H . , Arora, A . , Wengel, J., and Mai t i , S. Thermodynamic, counterion, and hydration effects for the incorporation of locked nucleic acid nucleotides into D N A duplexes. Biochemistry 4 5 , 7347-7355 (2006). 141. Nguyen, A . , Zhao, C , Dorris, D . , and Mazumder, A . Quantitative assessment of the use of modified nucleoside triphosphates in expression profiling: differential effects on signal intensities and impacts on expression ratios. BMC Biotechnology 2 , 4642-4648 (2002). 142. Matray, T., Gamsey, S., Pongracz, K . , and Gryaznov, S. A remarkable stabilization of complexes formed by 2,6-diaminopurine oligonucleotide N3 ' —> P5' phophoramidates. Nucleosides Nucleotides Nucleic Acids 19 ,1553-1567 (2000). 143. Baker, E . S., Hong, J. W. , Gaylord, B . S., Bazan, G . C , and Bowers, M . T. P N A / d s D N A Complexes: Site Specific Binding and d s D N A Biosensor Applications. Journal of the American Chemical Society 1 2 8 , 8484-8492 (2006). 144. Giesen, U . , Kleider, W. , Berding, C , Geiger, A . , Orum, H . , and Nielsen, P. E . A formula for thermal stability (T m ) prediction of P N A / D N A duplexes. Nucleic Acids Research 2 6 , 5004-5006 (1998). 145. L i u , J., Tiefenauer, L . , Tian, S., Nielsen, P. E . , and K n o l l , W . P N A - D N A Hybridization Study Using Labeled Streptavidin by Voltammetry and Surface Plasmon Fluorescence Spectroscopy. Analytical Chemistry 7 8 , 470-476 (2006). 91 146. Gao, W . Y . , Han, F. S., Storm, C , Egan, W. , and Cheng, Y . C. Phosphorothioate oligonucleotides are inhibitors of human D N A polymerases and RNase H : implications for antisense technology. Molecular Pharmacology 41, 223-229 (1992). 147. Lysik, M . A . , Wu-Pong, S. Innovations in oligonucleotide drug delivery. Journal of Pharmceutical Sciences 92, 1559-1573 (2003). 148. Koshkin, A . A . , Singh, S. K . , Nielsen, P., Rajwanshi, V . K . , Kumar, R., Meldgaard, M . , Olsen, C E . and Wengel, J. L N A (Locked Nucleic Acids): Synthesis of the Adenine, Cytosine, Guanine, 5-Methylcytosine, Thymine and Uraci l Bicyclonucleoside Monomers, Oligomerisation, and Unprecedented Nucleic A c i d Recognition. Tetrahedron 54, 3607-3630 (1998). 149. Wengel, J. Synthesis of 3'-C- and 4'-C-branched oligodeoxynucleotides and the development of locked nucleic acid ( L N A ) . Accounts of Chemical Research 32, 301-310(1998). 150. www.sigmaldrich.com 151. Latorra, D. , Campbell, K . , Wolter, A . , and Hurley, J. M . Enhanced allele-specific P C R discrimination in SNP genotyping using 30 locked nucleic acid ( L N A ) primers. Human Mutation 22, 79-85 (2003). 152. Hertoghs, K . M . L . , El l is , J. H . , and Catchpole, I. R. Use of locked nucleic acid oligonucleotides to add functionality to plasmid D N A . Nucleic Acids Research 31,5817-5830(2003). 153. Koshkin, A . A . , Nielsen, P., Meldgaard, M . , Rajwanshi, V . K . , Singh, S. K . and Wengel, J. L N A (locked nucleic acid): an R N A mimic forming exceedingly stable L N A : L N A duplexes. Journal of the American Chemical Society 120, 13252-13260(1998). 154. Silahtaroglu, A . , Pfundheller, H . , Koshkin, A . , Tommerup, N . , and Kauppinen, S. LNA-modif ied oligonucleotides are highly efficient as F ISH probes. Cytogenetic and Genome Research 107, 32-37 (2004). 155. Peterson, M., Nielsen, C. B . , Nielsen, K . E . , Jensen, G . A . , Bondensgaard, K . , Singh, S. K . , Rajwanshi, V . K . , Koshkin, A . A . , Dahl, B . M . Wengel, J., and Jacobsen, J. P. The conformations of locked nucleic acids ( L N A ) . Journal of Molecular Recognition 13, 44-53 (2000). 156. Simeonov, A . and Nikiforav, T. T. Single nucleotide polymorphism genotyping using short, fluorescently labeled locked nucleic acid ( L N A ) probes and fluorescence polarization detection. Nucleic Acids Research, 30, e31,. (2002). 157. Kvaerno, L . and Wengel, J. Investigation of restricted backbone conformations as an explanation for the exceptional thermal stabilities of duplexes involving L N A (Locked Nucleic Acid) : synthesis and evaluation of abasic L N A . Chemical Communications 1, 657-658 (1999). 158. Mouritzen, P., Nielsen, A . T., Pfundheller, H . M . ; Choleva, Y . , Kongsbak, L . , and Moller , S. Single nucleotide polymorphism genotyping using locked nucleic acid ( L N A ) . Expert Review of Molecular Diagnostics 3, 27-38 (2003). 159. http://lna-tm.com 160. Marin, V . , Hansen, H . F., Koch, T., Armitage, B . A . Effect of L N A modifications on small molecule binding to nucleic acids. Journal of Biomolecular Structure and Dynamics 21, 841-850 (2004) 92 161. Searle, M . S. and Will iams, D . H . On the stability of nucleic acid structures in solution: enthalpy-entropy compensations, internal rotations and reversibility. Nucleic Acids Research 21, 2051 -2056 (1993). 162. Davis, T. M . , McFail-Isom, L . , Keane, E . , and Will iams, L . D . Melting of a D N A hairpin without hyperchromism. Biochemistry 37, 6975-6978 (1998). 163. Sacca, B . , Lacroix, L . , and Mergny, J-L. The effect of chemical modifications on the thermal stability of different G-quadruplex-forming oligonucleotides. Nucleic Acids Research 33, 1182-1192 (2005). 164. Digabriele, A . D . and Steitz, T. A . A D N A dodecamer containing an adenine tract crystallizes in a unique lattice and exhibits a new bond. Journal of Molecular Biology 231, 1024-1039 (1993). 165. C o l l , M . , Frederick, C. A . , Wang, A . H . J., and Rich , A . A bifurcated hydrogen-bonded conformation in the d(AT) base pairs of the D N A dodecamer d ( C G C A A A T T T G C G ) and its.complex with distamycin. Proceedings of the National Academy of Science USA 84, 8385-8389 (1987). 166. Jose, D . and Porschke, D . Dynamics of the B - A transition of D N A double helices. Nucleic Acids Research 32, 2251-2258 (2004). 167. Merling, A . , Sagaydakova, N . , and Haran, T. E . A-Tract polarity dominate the curvature in flanking sequences. Biochemistry, 42, 4978 - 4984 (2003). 168. N g , H . L . , Kopka, M . L . , and Dickerson, R. E . The structure of a stable intermediate in the A <->• B D N A helix transition. Proceedings of the National Academy of Sciences USA 91, 2035-2039 (2000). 93 Appendix 1.0 Circular Dichroism Measurement Procedures 1.1 S Y S T E M START-UP 1.1.1 Turn on the nitrogen supply. For measurements in the normal wavelength region (more than 190 nm), the proper flow rate of nitrogen gas is 3 litres/minute. N O T E : A l l o w the nitrogen gas to purge the Xenon lamp compartment for about 10 minutes. The high intensity Xenon light source w i l l convert oxygen to ozone, which destroys the optics. 1.1.2 Turn on the water recirculator. Check to ensure that the level of distilled water in the water recirculator is full. N O T E : the level should be near the top of the reservoir - i f it needs to be topped up use distilled water to avoid mineral deposits. 1.1.3 Open the sample compartment and confirm that there are no samples in the sample holder and that there are no obstructions to the light path. 1.1.4 Turn on the power to the computer, J-810 Spectropolarimeter, and the Peltier device. 1.1.5 Warm up the instrument for 30 minutes until it is stabilized after starting it. 1.1.6 Double click.on the Spectra Manager Software icon on the desktop of the P C . 1.1.7 From the Spectra Manager window select the Spectrum Measurement application for scanning. 1.1.8 Checking of CD-value stability: Check often the CD-value stability with the sample while warming up the instrument. Pour an aqueous solution of 0.06% (w/v) ammonium d-10-camphorsulfonate into a 1 cm cell and set the cell in the instrument to make measurements with a wavelength of 290.5 nm for approximately 2 hours in the T-scan mode. If the stabilized C D value (value after base correction) remains within +190.4 mdeg (± 1%), it is normal. 1.2 SCANNING SET-UP 1.2.1 Ensure the scan parameters are set up as follows: 1.2.2 Select the Control Tab: 1.2.2.1 Bandwidth: set to 1 nm 1.2.2.2 Response: set to 4 seconds 1.2.2.3 Sensitivity: select Standard 1.2.2.4 Data Pitch: set to 1 nm 1.2.2.5 Scan Speed: set to 50 nm/min 1.2.2.6 Accumulations: set to 5 1.3 B A S E L I N E C O R R E C T I O N 1.3.1 C l ick on the Measurement tab and select Baseline Correct. 1.3.2 Zero the spectropolarimeter with the same buffer used to prepare the test sample. 1.3.3 C l ick on the Start button to acquire the buffer spectrum. 1.3.4 Save the buffer spectrum to use as a baseline file. 94 1.4 MEASURING A SAMPLE 1.4.1 Load the correct baseline file. 1.4.2 Inspect the cuvette for particulates and ensure the cuvette is clean and free from scratches. 1.4.3 Place the sample in the sample holder. 1.4.4 C l ick on the Start button to acquire the spectrum of the sample and save the spectrum after acquisition. 1.4.5 Perform all readings within an assay using the same type of cuvette. 1.5 ZOOMING IN A SPECTRUM, AUTO SCALE and FULL SCREEN DISPLAY 1.5.1 To zoom in a spectrum using the mouse, click and drag the mouse so that a rectangle surrounds the area to be zoomed. The rectangle can be moved with keeping the same size. Once the rectangle is on the area to be zoomed, click the mouse again on the selected area. 1.5.2 To Autoscale the spectra move the cursor on the spectrum window and right click, select autoscale in the menu (changes the vertical axis to a proper size). 1.5.3 To return to full scale after zooming in, right click with the mouse on the spectrum window, select full from the menu. 1.6 SPECTRA OVERLAY 1.6.1 Select display (by right clicking on the file in the tree view window and select display) for all spectra to be overlaid. Select window and then jo in visible from the pull-down menu. To return each spectrum to an original window select split from the pull-down menu under window. To change the display style of overlaid spectra, select window and multi spectra from the pull down window, select a style from the submenu. 1.7 SYSTEM SHUT-DOWN 1.7.1 Confirm that all necessary C D data has been saved. 1.7.2 Close all the Spectra Manager Applications this w i l l close down the connection to the J-810. 1.7.3 Remove all samples from the sample compartment. 1.7.4 Turn off the Water Recirculator power switch which is on the top left of the unit. 1.7.5 Turn off the Peltier temperature unit. 1.7.6 Turn off the J-810 the power switch. 1.7.7 Turn off the nitrogen supply. 1.7.8 Enter all activities in the instrument logbook. 9 5 2.0 U V Absorbance Measurement Procedures 2.1 S Y S T E M S T A R T UP 2.1.1 Turn on the main power unit and allow the instrument to warm up for at least 30 minutes. 2.1.2 From the Start menu, click on Programs —» Cary W i n U V —*• Scan. 2.1.3 On the top toolbar options, click on the Setup tab. 2.1.3.1 Under the X - M o d e of the Cary tab, type in the start (340 nm) and end (200 nm) parameters of the wavelength region. Cl ick O K . 2.1.3.2 Under the Options tab, select Fixed S B W 2.1.3.3 Under the Baseline tab, select Baseline Correction. 2.1.4 Use a Kimwipe to wipe down the sides of the cuvette and be careful to avoid touching the sides of the cuvette when putting it in the instrument. 2.1.5 Place the blank sample (buffer) in cell 1. Close the l id and click on the Start button. 2.1.6 The box in the upper left-hand corner w i l l show that the instrument has been zeroed by displaying an absorbance of 0.000. 2.1.7 Remove the blank sample and place the oligomer sample into the same cell compartment. Ensure that the cuvette faces the same direction to ensure the same pathlength at all times. 2.1.8 C l ick Start to obtain a scan of the sample spectrum. 2.2 SAVING AND R E T R I E V I N G D A T A 2.2.1 To save data, go to "F i l e " —> "Save A s " . Enter a filename and click "Save" to save .the file in the *.bsw format. 2.2.2 To open saved data, go to "F i l e" —> "Open". Select the desired file. 2.3 S Y S T E M S H U T - D O W N 2.3.1 Save all data,files and exit out of the program. 2.3.2 Turn off the main power unit. 2.3.3 Enter all activities in the instrument logbook. 96 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0060162/manifest

Comment

Related Items