Loop Sequence Context Influences the Formation ... - ACS Publications

Jul 20, 2016 - Mikeal McKim, Alexander Buxton, Courtney Johnson, Amanda Metz, and Richard D. Sheardy*. Department of Chemistry and Biochemistry, ...
0 downloads 0 Views 1MB Size
Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article

Loop Sequence Context Influences the Formation and Stability of the i-Motif Formed from DNA Oligomers of Sequence (CCCXXX) 4

Mikeal McKim, Alexander Buxton, Courtney C. Johnson, Amanda Metz, and Richard D. Sheardy J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.6b04561 • Publication Date (Web): 20 Jul 2016 Downloaded from http://pubs.acs.org on July 22, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry B is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Loop Sequence Context Influences the Formation and Stability of the i-Motif for DNA Oligomers of Sequence (CCCXXX)4, where X = A and/or T, under Slightly Acidic Conditions

Mikeal McKim, Alexander Buxton, Courtney Johnson, Amanda Metz, and Richard D. Sheardy1 Department of Chemistry and Biochemistry Texas Woman’s University PO Box 425859 Denton, TX 76204

1

Address correspondence to this author

Phone: 940-898-2551 Email: [email protected]

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract The structure and stability of DNA is highly dependent upon the sequence context of the bases (A, G, C, and T) and the environment under which the DNA is prepared (e.g., buffer, temperature, pH, ionic strength). Understanding the factors that influence structure and stability of the i-motif conformation can lead to the design of DNA sequences with highly tunable properties. We have been investigating the influence of pH and temperature on the conformations and stabilities for all permutations of the DNA sequence (CCCXXX)4, where X = A and/or T, using spectroscopic approaches. All oligomers undergo transitions from single stranded structures at pH 7.0 to i-motif conformations at pH 5.0 as evidenced by circular dichroism (CD) studies. These folded structures possess stacked C:CH+ base pairs joined by loops of 5’-XXX-3’. Although the pH at the midpoint of the transition (pHmp) varies slightly with loop sequence, the linkage between pH and log K for the proton induced transition is highly loop sequence dependent. All oligomers also undergo the thermally induced i-motif to single strand transition at pH 5.0 as the temperature is increased from 25 oC to 95 oC. The temperature at the midpoint of this transition (Tm) is also highly dependent on loop sequence context effects. For seven of eight possible permutations, the pH induced and thermally induced transitions appear to be highly cooperative and two state. Analysis of the CD optical melting profiles via a van’t Hoff approach reveals sequence dependent thermodynamic parameters for the unfolding as well. Together, these data reveal that the i-motif conformation exhibits exquisite sensitivity to loop sequence context with respect to formation and stability.

ACS Paragon Plus Environment

Page 2 of 28

Page 3 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Introduction The protonation of N3 of cytosine (C) has been extensively studied.1-6 Protonation of N3 of C changes the Crick face from an acceptor-acceptor-donor (C2O-N3-C4NH2) motif to an acceptor-donor-donor motif leading to the formation of non Watson-Crick base pairing schemes such as Hoogsteen G:CH+ base pairs2,3,5 and C:CH+ base pairs.1,4,6 Such base pairing schemes are associated with non canonical conformations of DNA such as multi-stranded structures,1 duplexes,2 triplexes,3 quadruplexes,4 left handed DNA,5 and, of particular interest, the DNA imotif.6 The N3 of free C in solution has a pKa of around 4.2,7 however, the pKa of C embedded in an oligonucleotide can vary dramatically from 4.5 to nearly 7.0. For example, the pKa of C is 7.4 in poly[dC] and 6.2 in poly[dCT].1 For triplexes, the pKa of C has been determined experimentally to be 5.8 for [CTTCCTCCTCT],8 6.15 for [TTCTTCTTC],9 5.5 for a (dCT) containing oligonucleotide10 and found computationally to be 4.1 for C in [dTC]n:[AG]n:[TC+]n. In this latter case, the intrinsic pKa of 6.7 is lowered to 4.1 due to influences of site-site interactions on the DNA dielectric constant.3 Protonation of the N3 of C is essential for the formation of C:CH+ base pairs found in the so called DNA i-motif (Figure 1). The first i-motif to be studied and characterized was formed from [dTCCCCC].11, 12 Since then, the conformational properties of a number of C rich DNA oligonucleotide model systems have been investigated from both structural and thermodynamic perspectives.6,13-21 Mergny et al6 investigated oligonucleotide sequences possessing repeats of combinations of T and C bases. Depending upon the actual sequence context, these oligomers would either form a bimolecular i-motif, a unimolecular i-motif or remain single stranded. They report pKa values for C N3 of 4.8 at low ionic strength and 4.55 in 100 mM NaCl.6 Recent investigations using DNA oligomers of general sequence (CCCTx)3CCC (where x = 1, 3, 5, 10, 15 or 20) reported pKa values for C N3 ranging from 5.98 to 6.68. These values were obtained from the pH at the midpoint of the single strand to i-motif transition.21 A separate study reported that i-motif stability is highly related to sequence context and that loop length also plays a role in conformational stability. With regard to the latter, i-motif conformations with shorter loops were more stable than those conformations with longer loops.20 Our group has been investigating the conformational properties of DNA oligomers with sequences related to the human telomere TTAGGG repeat.22-25 For these studies, we investigated the structure and stability of DNA sequences (XXXGGG)4, where X = A and /or T. In the presence of K+, these oligomers fold into intramolecular hybrid like quadruplexes possessing three G tetrads with loops in specific orientations in the folded structure. The results of spectroscopic studies indicated that changing the sequence context of the loops, even if just one base in any one loop, affected the structure and stability of the folded conformation.23,24

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

To complement these studies, we set out to investigate the conformational properties of all permutations of (CCCXXX)4, where X = A and/or T, to determine the influence of loop sequence on the structure and stability of the i-motif conformations formed under slightly acidic conditions. A schematic of the folded structure is depicted in Figure 1. This structure is stabilized by the formation of C:CH+ base pairs that interdigitate in the folded structure. This folded structure has three 5’-XXX -3’ loops and a 3’-XXX tail. Here, DNA sequences were prepared in standard phosphate buffer with pH ranging from 7.0 to 5.0. The CD spectra of each sequence under different conditions of pH at 25 oC were determined and transition profiles constructed. Further, upon formation of the i-motif at pH 5.0 and 25 oC, each oligomer was slowly heated to 95 oC to investigate the thermal unfolding of the i-motif. Materials and Methods Preparation of buffers Standard potassium phosphate buffers (10 mM phosphate, 0.1 mM EDTA, 21.4 mM K+, pH 7.0 and 10 mM phosphate, 0.1 mM EDTA, 34.2 mM K+, pH 5.0) were prepared using KH2PO4 (VWR International), K2HPO4 (VWR International), and EDTA (EMD Chemicals). By adjusting the ratio of H2PO4- to HPO4-2, the two different pHs were obtained. The final concentration of K+ was brought to 115 mM by adding KCl. The buffers were then filtered through a 0.45 µm Millipore filter and degassed before storing them for use. DNA oligomers All DNA oligomers were purchased from Biosynthesis Inc. (Lewisville, TX) and used without further purification. Each sequence was reconstituted in 1 mL of buffer at either pH 7.0 or pH 5.0, heated to 95°C, and slowly cooled to room temperature and stored at 4°C. Extinction coefficients, provided by the supplier, are shown in Table 1. DNA samples of different pH values were prepared by mixing different quantities of the samples prepared at 7.0 and 5.0. Circular Dichroism An Olis RMS 1000 CD spectrophotometer (Olis, Inc. Athens, GA) was used to carry out all circular dichroism studies. A 1 mm circular quartz cuvette was used to run all samples including the baseline. Initially, the CD spectra of a particular DNA sequence in buffer of pH 7.0 and 5.0 were determined. Different ratios of the DNA samples in pH 7.0 and 5.0 were mixed to obtain samples of different pH values. Continuing in this fashion, DNA solutions in buffers of pH ranging from 7.0 to 5.0 at about 0.1 to 0.2 pH unit increments were generated. Since the volumes of these solutions were typically around 400 µL, a micro pH electrode was used to measure the exact pH of each solution before placing that sample in the CD spectrometer. CD spectra were determined at 25 oC from 320 nm to 220 nm with 1 nm intervals. Upon completion of the scan, an aliquot of the DNA solution was diluted for UV/Vis for determination (see below)

ACS Paragon Plus Environment

Page 4 of 28

Page 5 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

of its concentration. For those samples at pH 5.0, CD spectra were also obtained from 320 nm to 220 nm every 5 oC from 25 oC to 95 oC to generate a three dimensional optical melt (i.e., ellipticity vs wavelength vs temperature). The raw CD data were smoothed using the Olis software and then transported to SigmaPlot for graphing and analysis. Each spectral set was baseline corrected and divided by the DNA concentration in bases for that sample to generate Molar Ellipticities as a function of wavelength with units of 106 mdeg M-1 cm-1 using the User Defined Transform macro of SigmaPlot. For each sequence that displayed an apparent isoelliptic point for the thermalinduced melt, the average value of the molar ellipticities at that wavelength for each temperature was used to normalize the CD spectra. Hence, for a particular temperature scan, the ellipticity at each wavelength was multiplied by the ratio of the average ellipticity at the isoelliptic point to the unnormalized ellipticity at the isoelliptic point which gives rise, naturally, to a very sharp isoelliptic point. UV/Vis Spectroscopy All UV/Vis spectra were determined using a Varian Cary 100 Bio model (Varian Associates, Palo Alto, CA). Once the appropriate dilutions were made on each CD sample, the baseline was subtracted and UV/Vis spectra from 320 nm to 220 nm were determined using 10 mm square quartz cuvettes at 25 oC and 95 °C. Concentrations of the DNA samples were then calculated using the appropriate extension coefficient and the base line corrected absorbance at 260 nm and 95 oC. Data Analysis As noted above, we examined the CD spectra of the various DNA sequences as a function of both temperature and pH. To generate transition profiles of fraction of i-motif vs pH or fraction of i-motif vs temperature we used the molar ellipticity at 292 nm: Fraction i-Motif = (θ292, pH - θ292, pH 7.0)/ (θ292, pH 5.0 - θ292, pH 7.0)

(1)

where θ292, pH is the molar ellipticity at 292 nm for any pH solution, θ292, pH 7.0 is the molar ellipticity at 292 nm at pH 7.0 and θ292, pH 5.0 is the molar ellipticity at 292 nm at pH 5.0; or Fraction i-motif = (θ292, T - θ292, 95C)/ (θ292, 25 C - θ292, 95C)

(2)

where θ292, T is the molar ellipticity at 292 nm at any temperature T, θ292, 25C is the molar ellipticity at 292 nm at 25 oC and θ292, 95C is the molar ellipticity at 292 nm at 95 oC. From these plots, we can evaluate the equilibrium constant, K, as a function of pH or temperature: K = [i-motif]/[single strands] = (Fraction i-Motif)/(1 – Fraction i-Motif)

ACS Paragon Plus Environment

(3)

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Investigators who study pH conformational changes in proteins use a Tanford-Wyman approach to analyze the data.26-28

δ∆Go/δpH = 2.303RT∆Q

(4); or

δ(log K)/δ(pH) = - ∆Q

(5)

Thus, a plot of log K vs pH reveals information about the protonation where ∆Q (the linkage), the slope of the resultant linear lines the number of protons gained or lost during the transition. Using a van’t Hoff approach (a plot of log K vs 1/T), the enthalpic (∆Ho) and entropic (∆So) contributions to the total free energy (∆Go) of i-motif formation can also be determined: log K = -∆Ho/RT + ∆So/R ∆Go = ∆Ho – T∆So

(6) (7)

Results and Discussion All Oligomers Studied Form the i-Motif under Slightly Acidic Conditions at 25 oC Spectroscopic studies of DNA oligomers of general sequence (CCCXXX)4, where X = A and/or T, were carried out to investigate their structures and stabilities under different pH and temperature conditions. Each DNA sequence was prepared in standard 10 mM phosphate buffer with 115 mM K+ and different pHs as described in the experimental section. Figure 2 shows the CD spectra of (CCCTAA)4 at different pH and temperature values. At pH 7.0, the oligomer is single-stranded at both 25 oC and 95 oC (upper left hand panel of Figure 2). Further, the oligomer is single-stranded at 95 oC at either pH 7.0 or 5.0 (lower right hand panel of Figure 2). All spectra in these two panels have slight peaks at around 279 nm and shallow troughs at around 240 nm – typical of the spectral characteristics of single stranded DNA. However, the CD spectrum of the oligomer at 25 oC and pH 5.0 indicates that the oligomer has undergone a conformational transition to the i-motif (lower left hand panel of Figure 2), with a well pronounced peak at 292 nm and a deeper trough at 240 nm – typical of the spectral characteristics of the i-motif. This i-motif conformation unfolds to the single strand at pH 5.0 and 95 oC (upper right hand panel of Figure 2). All oligomers studied assume the i-motif conformation as determined by their CD spectra at 25 C and pH 5.0 (Figure 3). The UV/Vis and CD spectral characteristics are found in Table o

ACS Paragon Plus Environment

Page 6 of 28

Page 7 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

1. As can be seen from Figure 3 and Table 1, there are slight differences in λmax and extinction coefficients in the UV/Vis spectra due to the sequence context variations. At pH 7.0, all oligomers have a CD peak in the range of 276 to 281 nm and molar ellipticities ranging from 1.89 to 3.77 x 10 6 mdeg M-1 cm-1. At pH 5.0, the peak shifts to 289-292 nm with much higher molar ellipticities ranging from 5.04 to 8.87 x 106 mdeg M-1 cm-1. Further, the T rich sequences have, in general, higher ellipticities at λmax than the A rich sequences. The variations in peak locations and molar ellipticities are due to variations in loop sequence context. Panel A of Figure 4 shows the overlay of the CD spectra of (CCCTAA)4 at 25 oC as a function of pH. These data demonstrate the pH-induced transition from the single strand at pH 7.0 to the i-motif at pH 5.0. The presence of the isoelliptic point at 278 nm is consistent with a two state transition and the sigmoidal nature of the titration curve (Panel B) suggests a highly cooperative transition. To simplify the analysis, we start with the assumption that the single strand to i-motif transition is indeed cooperative and two state. Thus, θλ, total = cssθλ,ss + ci-motifθλ, i-motif

(8)

where θλ, total is the total observed molar ellipticity at wavelength λ, css and ci-motif are the concentrations of single-stranded structure and i-motif, respectively, and θλ,ss and θλ, i-motif are the molar ellipticities of the single-stranded structure and i-motif, respectively, at wavelength λ. Thus, the molar ellipticity at a particular wavelength and pH is simply the concentration weighted average of the molar ellipticities of the pure single stranded or pure i-motif. For all oligomers, no further spectral changes were observed below pH 5.2 to pH 4.9 indicating that the transition was complete. The pH titration curve in Panel B (Fraction i-Motif vs pH) was obtained using equation (1) to evaluate the fraction of i-motif at every pH studied. Assuming a two state transition, the midpoint of the transition, pHmp, represents the point in the titration when there are equal concentrations of both the single stranded species and the i-motif species, i.e. css = ci-motif. For this particular oligomer, pHmp occurs at pH = 5.86 + 0.03. The two state assumption dictates that 50% of the C bases in a single molecule have been protonated and that this molecule has assumed the i-motif conformation. Since each sequence has 12 C bases, the protonation of six would allow i-motif formation. This point will be addressed in more detail below. All Loop Variant i-Motifs Thermally Denature at pH 5.0 Panel C of Figure 4 shows the overlay of the CD spectra of (CCCTAA)4 at pH 5.0 as a function of temperature. These data demonstrate the thermally induced transition from the folded i-motif at 25 oC to the unfolded single stranded structure at 95 oC. The presence of an isoelliptic point in the CD spectra and the sigmoidal shape of the melting profile (Panel D) suggest a

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

cooperative, two state transition for the thermally induced denaturation. The optical melting curve in Panel D was obtained using the molar ellipticity at 292 nm and equation 2 to evaluate the fraction of i-motif at any temperature studied. The midpoint of the transition, Tm, represents the point in the transition where there are equal concentrations of both the single stranded species and the i-motif species – assuming a two state transition. The value of the temperature at the midpoint of the transition should be related to the stability of the i-motif conformation. For this particular oligomer, Tm occurs at T = 66.2 + 0.3 oC. All oligomers studied, with the exception of (CCCTTA)4, gave similar pH and temperature dependent CD spectra, and pH titration and optical melting curves (see Supporting Information) albeit with different pHmp and Tm values (Table 2). As noted above, the spectral overlays presented in Figure 4 show sharp isoelliptic points at 278 nm for both the pH titration and the thermal melt. The initial raw CD spectra were corrected for DNA concentration allowing the plot of molar ellipticity as a function of wavelength. Examination of these temperature dependent spectra revealed fairly sharp isoelliptic points at a wavelength particular to a specific sequence in all but one of the DNA oligomers studies, (CCCTTA)4. Hence, for each DNA sequence displaying an isoelliptic point, the average value of the molar ellipticities at that particular wavelength was used to normalize all spectra which resulted in the sharp isoelliptic points observed. It should be noted that all resultant normalized ellipticities at the isoelliptic points were within + 2% of the averaged values used to generate the plots. The concentration corrected ellipticities varied due to noise in the acquisition not to differences in DNA concentration. Hence, this normalization process “smoothed” out some of that noise. The molar ellipticity at the isoelliptic point was then used to normalize the pH titration spectra. Since the characteristics of the CD plot for any of these oligomers are dependent upon both pH and [DNA], minimizing error in the [DNA] contribution can be achieved through this normalization process. Table 1 shows all isoelliptic points and ellipticity values for these oligomers. There is a striking similarity in the CD spectra for the pH titration and for the thermal melt – as there should be since the initial states (e.g., folded) and final states (e.g., unfolded) are similar for both transitions. The normalization process worked very well for all but one of the DNA oligomers, namely (CCCTTA)4. The lack of a fairly well defined isoelliptic point in the overlays of the concentration corrected spectra for the optical melting is quite puzzling. The experiment was repeated five times with five new samples and similar results were obtained all five times: as the temperature increases from 25 oC to 55 oC, the molar ellipticity at 292 nm increases and then decreases rapidly as the temperature continues to increase to 95 oC (Figure 5). It appears as if there is a non denaturational conformational change between 35 and 55 oC - before the unfolding event (panels A and B of Figure 5). Typically, molar ellipticities increase with base stacking. If this is the case, the increased molar ellipticity may be due to the transient formation of an A:T base pair that can stack with the first or last C:CH+ base pair in the folded structure. Hence, the

ACS Paragon Plus Environment

Page 8 of 28

Page 9 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

temperature induced unfolding of the i-motif conformation for (CCCTTA)4 is not two state. None the less, the pH induced folding does appear to be two state (panel C of Figure 5). Loop Sequence Context Affects both pHmp and Tm The data in Table 2 allow a comparison of the pHmp and Tm values obtained for all eight oligomers. Although there is not much variation in the pHmp values (ranging from 5.82 to 6.35), there are some correlations related to sequence context effects of the XXX linkers (Table 2): 1) those oligomers with more A bases than T bases in the XXX linker have nearly identical pHmp values ranging between 5.82 and 5.86; and, 2) the oligomers with more T bases than A bases in the XXX linker have higher pHmp values with a broader range from 6.07 to 6.35. In addition, for loop sequences A-X-T and T-X-A, the pHmp is considerably higher when X = T rather than A. Other than these observations, one cannot correlate sequence effects to pHmp since the differences in pHmp are rather small. Examination of Tm values listed in Table 2 indicates a wide range of values – from 56.5 C to over 80 oC. Clearly, the loop sequence context of the XXX segments is influencing the stability of the folded i-motif conformation. It is interesting to note that sequences that have an A in the first and third positions have the lowest Tm values while sequences with a T in the third position have the highest Tm values. Further, a T in the second position stabilizes more than an A in the second position. Inspection of the schematic for the i-motif in Figure 1 indicates that the folded structure has an XXX tail on the 3’ terminus and three XXX loops. Base stacking plays a significant role in the stability of DNA secondary structures29 so the question pertains to what the bases in the loops and tail are doing: Are they conformationally unconstrained or do they have some structure as well? Clearly, the bases have some sort of structure, either stacking with themselves or with the C:CH+ base pairs. In addition, an A in the first position with a T in the third position (or a T in the first position with an A in the third) can also result in an A:T base pair adjacent to a C:CH+ base pair. Hence the differences in Tm values observed are due to sequence context effects whereby different sequences have slightly different initial and final structures, stacking interactions and possible base pairing interactions and therefore, different stabilities. o

Loop Sequence Context Influences the Proton Induced Folding One of the initial aims of this project was to determine if and how loop sequence context would influence both the protonation induced folding and the thermally induced unfolding. Although differences in the observed pHmp values are small, examination of the linkage (∆Q) of log K and pH is informative. Plots of log K vs pH were generated (Supporting Information) and the slopes (-∆Q) of the resultant least squares regression lines determined. Figure 6 dramatically demonstrates the variation in linkage as a result of sequence context. If we assume strictly a two

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

state transition, ∆Q should approach six for all sequences since six cytosines per strand must be protonated to form the fully folded hemiprotonated i-motif, the final state of our proposed two state model. Yet the values shown in Figure 6 range from 1.20 to 3.98. Based on the observation of an isoelliptic point in the CD spectra for the for the proton induced i-motif formation, we are going to initially assume a two state transition since we can readily observe the initial and final states with CD spectroscopy. The initial state is the totally unfolded, totally unprotonated single strand; the final state is the fully folded, hemiprotonated imotif. Further, both the single stranded structure and the i-motif have associated waters and counterions (i.e., K+ and, for the final state, H+). The folding process will release water as basepair formation and stacking is hydrophobically driven. Finally, the folding process will require an uptake of K+ since the folded structure will have a higher charge density. The net reaction can best be described as: Initial State: fully unfolded, fully unprotonated single strand

(CCCTAA)4 (H2O)x (K+)y + 6H+ + mK+

+ + (CCCTAA)4 (H2O)x-n (K )y+m (H )6 + nH2O

Final State: fully folded, hemiprotonated i-motif Clearly the equilibrium above will depend on pH, temperature, ionic strength and the activity of water. In this model, once the requisite number of C bases in any oligomer are protonated and the first C:CH+ base pairs formed, additional protonation and C:CH+ base pair formation takes place in a very cooperative fashion to form the ultimate i-motif structure. However, once the requisite number of C bases are protonated, the structure must also fold to allow the formation of hydrogen bonds to unprotonated C bases at other locations in the sequence. Hence, base pair formation requires folding. Although formation of any hydrogen bond is enthalpically favorable, the folding process itself has an unfavorable conformational entropy. The release of water and the uptake of K+ will also influence the thermodynamics of folding. Although the transition appears to be two state, the proton induced folding of these C rich strands can best be described by multiple equilibria as discussed below. For these experiments, all DNA oligomers were dissolved in standard phosphate buffer of 10 mM phosphate, 115 mK

ACS Paragon Plus Environment

Page 10 of 28

Page 11 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

K+, 0.1 mM EDTA and pH ranging from 7.0 to 5.0. Not considered below are the multiple equilibria involving H2PO4-2, HPO4-1, H2O and H3O+. It is these equilibria that determine which specie is the primary proton donor. Thus, at pH 7.0, the [H2PO4-] is 6.1 mM, the [HPO4-2] is 3.9 mM and the [H3O+] is 1.0 x 10-4 mM and at pH 5.0, those concentrations are 9.9 mM, 1.0 x 10-2 mM and 10-5 mM, respectively. Further, the total concentration of C bases is 0.12 mM. As the pH decreases with the addition of H3O+, the concentration of H2PO4-1 increases. Hence, at any pH between 7.0 and 5.0, the major proton donor is most likely H2PO4-1. In the scheme below, we will designate the proton source as the generic H+. As far as the proton acceptor is concerned, the pKa values of other ionizable groups in the DNA sequence (e.g., phosphate back bone, and the other bases) are well outside of the pH values used for this study, so their protonation by added H+ is negligible. Hence, the primary site of protonation by added H+ is N3 of C. Ignoring the loss of water and the uptake of K+, the first step is protonation of the single strand: (CCCTAA)4 + zH+ Fully Unprotonated Single Strand

[(CCCTAA)4 (H+)z] Partially Protonated Single Strand

The first equilibrium above is the protonation of N3 of z C bases in the strand. Here, z represents the minimum number of hydrogen bonds that must be present in the partially protonated, partially folded structure that forms in the next step (see below). Protonation would also lead to the loss of some associated K+ (not shown). The equilibrium constant for the protonation will depend upon the pKas of the protonated C bases. The fully unprotonated single strand has 12 C bases and hence 12 potential, but not necessarily equivalent, protonation sites. Since the pKa of cytosine is especially sensitive to environmental conditions, one of the goals of this project was to investigate the effect of sequence context of bases flanking the protonatable cytosine bases on the pH of the midpoint of the proton induced transition, i.e., can the sequence of the flanking bases influence the apparent pKa (and hence the observable pHmp) of the protonatable cytosines? Recent publications 20, 21 have used the pH at the midpoint of the transition to determine the pKa of the C bases, since half of the C bases in one strand must be protonated to fully form the i-motif. The pKa of free C is 4.2.7 The pHmp values reported in Table 2 are much higher than 4.2 indicating that the apparent pKa of N3 is higher than it’s intrinsic pKa. For triplex formation, the pKa has been measured to range from 5.8 to 6.2.3 This range is very similar to what is observed here (5.82 to 6.35). In the formation of C+:C Hoogstein base pairs, the pKa has been determined by NMR to be between 7.2 and 6.7.2 For all these systems, the apparent pKas are higher than that of the free base indicating that embedding a C in a polynucleotide affects the pKa of N3 very significantly.

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 28

Once the first C is protonated, what is the pKa of an adjacent unprotonated C and what is the pKa of an unprotonated C adjacent to an already formed C:CH+ base pair? While the dielectric constant for the exterior of DNA is considered to be close to 78.4, the dielectric for the interior on DNA has been calculated to be anywhere from 10 to 50.3 The pKa of an N3 on a C base immediately adjacent to a C:CH+ base pair will experience a lower dielectric constant due to the exclusion of water and an higher electrostatic potential due to base pairing for the folded oligomer. Both effects will raise the apparent pKa of N3 of the neighboring N3. It is likely that the actual pKa for any N3 depends upon several factors in addition to flanking sequence effects. Thus, the pHmp values listed in Table 2 reflect the apparent pKa values for the entire particular oligomer not necessarily the pKa of any one C base. Perhaps what is more informative than pHmp is the value of the ∆Q (linking number) for each sequence. The second equilibrium, shown below, is the formation of the partially protonated, partially folded structure. This is likely the rate determining step due to the unfavorable conformational entropy. Our interpretation of the linking numbers is that they represent z in the equilibrium below, i.e., ∆Q is the minimum number of protons that must be added to the single strand to allow folding. Although not shown, this step would also lead to release of water and uptake of K+.

[(CCCTAA)4 (H+)z] Partially Protonated Single Strand

+ [(CCCTAA)4 (H )z]

Partially Protonated, Partially Folded Structure

Once partially folded, the structure becomes a “proton sponge” and immediately acquires the total number of protons needed for the i-motif (i.e., 6): + + [(CCCTAA)4 (H )z] + (6 -z)H

Partially Protonated, Partially Folded Structure

+

[(CCCTAA)4 (H )6] Hemiprotonated, Fully Folded i-Motif

According to the schematic in Figure 1, the six base pairs formed in the final i-motif structure involve: C1:C13, C2:C14; C3:C15, C7:C19, C8:C20 and C9:C21. Since the base pairs formed in the initial partially protonated, partially folded structure may involve other combinations, some reorganization of base pairing may also have to occur in this fast step. However, the mobility of protons in the folded structure should be enough to enable protonation

ACS Paragon Plus Environment

Page 13 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

at multiple sites. Even though we can describe the folding in three steps, the kinetics of the third step must be such that the lifetime of the partially protonated, partially folded structure is so short that all we observe is the final state – hence the appearance of a two state transition. While loop sequence context has little influence on the pH at the midpoint of the transition, it highly influences the number of protons needed to fold. In other words, some sequences are more predisposed to folding than others. It is not unreasonable to assume that the single strands have some structure even at 25 oC and pH 7.0. Further, the populations of conformational states available to the single strands are most likely sequence context dependent. Thus, some single strands may be more conformationally poised to fold resulting in a less unfavorable conformational entropy change. Loops Sequence Context Influences the Thermally Induced Unfolding The data presented in Table 2 demonstrate a sequence context effect on the stability of these i-motif conformations. Using equations (6) and (7), we calculated van’t Hoff enthalpies, entropies and free energies (Supporting Information). The results are shown graphically in the right panel of Figure 7. With the exception of (CCCTTT)4, there is generally good correlation between Tm values and the respective ∆Go values, i. e., those with lower melting temperatures have lower free energies of i-motif formation. It is interesting to note that those oligomers with at least two T bases have the highest Tm values, and, in general, the highest ∆Go values. Further, for these oligomers where one of the two T bases is in the middle of the loop sequence, the free energies are much higher than for the other oligomers. Since the thermal denaturation of (CCCTTT)4 is incomplete at 95 oC, the data in Table 2 and Figure 6 for this sequence may not reflect the actual thermodynamic parameters but does reveal relative stability. Finally, the Tm and ∆Go values (55.6 to > 75 oC and 2.90 to 12.8 kcal/mol) are in the range of values reported recently for intramolecular i-motif forming sequences with T loops of various length and location within that sequence.21 The graphical representation of the data in the right panel of Figure 6 reveals the enthalpy-entropy compensation resulting in a small range of free energies. Although the van’t Hoff approach is subject to up to 10% error, particularly for unimolecular transitions, the general trends in Tm and calculated free energies support a sequence context effect on stability. Kaushik et al14 studied the unfolding of (CCCTAA)4 in cacodylate buffer at various pH values using differential scanning calorimetry (DSC). Their findings revealed a biphasic thermal transition which they attribute to the coexistence of two different complexes with different molecularities. For the i-motif complex (molecularity of 1), the thermodynamic parameters were determined with a Tm of 39.9 oC and ∆Go of 2.4 kcal/mol at pH = 5.2. Comparison to our values of 66.2 oC and 5.1 kcal/mol, respectively, and which were obtained through van’t Hoff analysis of the CD optical melting profile in phosphate buffer at pH 5.0, reveals a significantly higher

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

melting temperature and higher free energy for our studies. Further, our spectroscopic data are more consistent with a two state transition for (CCCTAA)4 for both the pH titration at 25 oC and thermal melt at pH 5.0. The difference in free energies, melting temperature and number of states in the transition may simply be due to differences in sample preparation or experimental techniques (CD vs DSC) employed. The above discussion focused on the step by step formation of the i-motif. Now consider the net reactions:

+H+ Single Strand

+q i-Motif

+

-H

-q

Single Strand

The equilibrium on the left depicts the protonation-induced folding to the i-motif while the equilibrium on the right is the thermally-induced unfolding back to the single strand. Although we can measure the heat (q) required to carry out the thermally induced unfolding we cannot necessarily use it in a simple thermodynamic cycle since the single strand on the far left (at pH 7.0 and 25 oC) is really not the same thermodynamically and conformationally as the single strand on the right (at pH 5.0 and 95 oC). In other words, the enthalpy for the thermally induced i-motif to single strand transition at pH 5.0 may not equal to the enthalpy for the deprotonation induced unfolding of the i-motif to single strand at 25 oC. We will be initiating some isothermal titration calorimetry approaches to address this question.

Conclusions DNA oligomers of general sequence (CCCXXX)4, where X = A and/or T, undergo the single strand to i-motif transition when the pH decreases from 7.0 to 5.0 at 25 oC. Further, at pH 5.0, all oligomers undergo the i-motif to single strand transition as the temperature is increased from 25 oC to 95 oC. For seven of the eight sequence permutations, these transitions appear to be highly cooperative and two state. The number of protons needed for the protonation induced transitions and the temperature at the midpoints of the thermally induced transitions are sequence context dependent. Further, the thermodynamic parameters for the thermally induced transitions are also sequence context dependent.

Supporting Information Available Supporting information includes CD spectra as a function of pH and T as well as the transition plots for all other oligomers studied, Wyman and van’t Hfof plots and tables of associated line parameters.

ACS Paragon Plus Environment

Page 14 of 28

Page 15 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Acknowledgements The authors thank the National Science Foundation for the acquisition of the CD spectrometer, and the Robert A. Welch Foundation and Texas Woman’s University for their generous support of this project.

References 1. Edwards, E. L.; Patrick, M. H.; Ratliff, R. L.; Gray D. M. A:T and C:C+ Base Pairs Can Form Simultaneously in a Novel Multistranded DNA Complex. Biochemistry 1990, 29, 828-836. 2. Nikolava, E. N.; Goh, G. B.; Brooks III, C. L.; Al-Hashami, H. M. 2013. Characterizing the Protonation State of Cytosine in Transient G:C Hoogsteen Base Pairs in DNA. J Am Chem Soc 2013, 135, 6766-6769. 3. Pack, G. R.; Wong, L.; Lamm, G. pKa of Cytosine on the Third Strand of Triplex DNA: Preliminary Poisson-Boltzmann Calculations. Inter J Quantum Chem 1998, 70, 1177-1184. 4. Hardin, C. C.; Corregan, M.; Brown II, B. A.;. Frederick , L. N. Cytosine-Cytosine+ Base Pairing Stabilizes DNA Quadruplexes and Cytosine Methylation Greatly Enhances the Effect. Biochemistry 1993, 32, 5870-5880. 5. Segers-Nolten, G. M. J.; Sijtsema, N. M.; Otto, C. Evidence for Hoogsteen GC Base Pairs in the Proton-Induced Transition from Right-Handed to Left-Handed Poly(dGdC):poly(dG-dC). Biochemistry 1997, 36, 13241-13247. 6. Mergny, J.-L.; Lacroix, L.; Han, X.; Leroy, J.-L.; Helene, C. Intramolecular Folding of Pyrimidine Oligodeoxynucleotides into an i-DNA Motif. J Am Chem Soc 1995, 117, 8887-8898. 7. Sanger, W. Principles of Nucleic Acid Structure; Springer-Verlag, New York. 1984. 8. Xodo. L. E.; Manzini, G.; Quadrifoglio, F.; van der Marel; G. A.; van Boom, J. H. Effect of 5-Methylcytosine on the Stability of Triple-stranded DNA – A Thermodynamic Study. Nucl Acids Res 1991, 19, 5625-5631.

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

9. Husler, P. L; Klump, H. H. Thermodynamic Characterization of a Triple-helical Three-way Junction Containing a Hoogsteen Branch Point. Arch Biochem Biophys 1995, 322, 149-166. 10. Singleton, S. F.; Dervan, P. B. Influence of pH on the Equilibrium Association Constants for Oligodeoxyribonucleotide-directed Triple Helix Formation at Single DNA sites. Biochemistry 1992, 31, 10995-101003. 11. Gehring, K.; Leroy, J.L.; Gueron, M. A Tetrameric DNA Structure with Protonated Cytosine-Cytosine Base Pairs. Nature 1993, 363, 561-564. 12. Leroy, J. L.; Gehring, K.; Kettani, A.; Gueron, M. Acid Multimers of Oligodeoxycytidine Strands: Stoichiometry, Base-pair Characterization, and Proton Exchange Properties. Biochemistry 1993, 32, 6019-6031. 13. Esmaili, N.; Leroy, J. L. i-Motif Solution Structure and Dynamics of the d(AACCC) and d(CCCCAA) Tetrahymena Telomeric Repeats. Nucl Acids Res 2005, 33, 213-224. 14. Kaushik, M.; Suehl, N.; Marky, L. A. Calorimetric Unfolding of the Bimolecular and i-Motif Complexes of the Human Telomere Complementary Strand, d(C3TA2)4. Biophys Chem 2007, 126, 154-164. 15. Leroy, J. L. The Formation Pathway of i-Motif Tetramers. Nucl Acids Res 2009, 37, 4127-4234. 16. Kaushik, M.; Prasad, M.; Kaushik, S.; Singh, A.; Kukreti, S. Structural Transition for Dimeric to Tetrameric i-Motif, Caused by the Presence of TAAQQ at the 3’-End of Human Telomeric C-Rich Sequence. Biopolymers 2009, 93, 150-160. 17. Dhakal, S.; Schonhoft, J. D.; Koirala, D.; Yu, Z.; Basu, S.; Mao, M. Coexistence of an ILPR i-Motif and a Partially Folded Structure with Comparable Mechanical Stability Revealed at the Single-Molecule Level. J Am Chem Soc 2010, 132, 8991-8997. 18. Yang, B.; Rodgers, M. T. 2013. Base-Pairing Energies of Proton-Bound Heterodimers of Cytosine and Modified Cytosines: Implications for the Stability of DNA i-Motif Conformations. J Am Chem Soc 2013, 136, 282-290. 19. Kim, S. E.; Lee, I.-B.; Hyeon, C.; Hong, S.-C. Destabilization of i-Motif by Submolar Concentrations of a Monovalent Cation. J Phys Chem 2014, 118, 4753-4760.

ACS Paragon Plus Environment

Page 16 of 28

Page 17 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

20. Gurung, S. P.; Schwarz, C.; Hall, J. P.; Cardin, C. J.; Brazier, J. A. The Importance of

Loop Length on the Stability of i-Motif Structures. Chem Comm 2015, 51, 5630-5632 21. Reilly, S. M.; Morgan, R. K.; Brooks, T. A.; Wadkins, R. M. Effect of Interior Loop

Length on the Thermal Stability and pKa of i-Motif DNA. Biochemistry 2015, 54, 1364-1370. 22. Sharma, V. R.; Sheardy, R. D. The Human Telomere Sequence, (TTAGGG)4, in the Absence and Presence of Cosolutes: A Spectroscopic Investigation. Molecules 2014, 19, 595-608. 23. Yadav, D.; Sheardy, R, D. 2012. A Single Base Permutation in Any loop of a Folded Intramolecular Quadruplex Influences Its Structure and Stability. J Biophys Chem 2012, 3, 341-347. 24. Tucker, B. A.; Gabriel, S.; Sheardy, R. D. A CD Spectroscopic Investigation of Interand Intramolecular DNA Quadruplexes. Frontiers in Nucleic Acids, edited by R.D. Sheardy and S. A. Winkle, ACS Symposium Books, Washington, D. C. 2011 25. Antonacci, C.; Chaires, J. B.; Sheardy, R. D. Biophysical Characterization of the Human Telomeric Repeat (TTAGGG)4 in Potassium Solution. Biochemistry 2007, 47, 4654-4660. 26. Meng, W.; Raleigh, D. P. Analysis of Electrostatic Interactions in the Denatured State Ensemble of the N-terminal Domain of L9 under Native Conditions. Proteins 2011, 79, 3500-3510. 27. Tanford, C. Protein Denaturation. C. Theoretical Models for the Mechanism of Denaturation. Adv. Protein Chem 1970, 24, 1-95. 28. Di Cera, E.; Gill, S. J.; Wyman, J. Canonical Formulation of Linkage Thermodynamics. Proc. Natl. Acad. Sci. USA 1988, 85, 5077-5081.

29. Sponer, J.; Sponer, J. E.; Mladek, A.; Jurecka, P.; Banas, P.; Olyepka, M. Nature and Magnitude of Aromatic Base Stacking in DNA and RNA: Quantum Chemistry, Molecular Mechanics, and Experiment. Biopolymers 2013, 99, 978-988.

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure Legends Figure 1. A. A C:CH+ base pair. Since the N3 of cytosine has a pKa of 4.67 (free), C:CH+ base pairs form under acidic conditions. B. A schematic of the intramolecular i-motif conformation formed from a general sequence 5’-(CCCXXX)4-3’ (where X = A and /or T) when under acidic conditions (pH < 6). In this schematic, the red spheres are Cs, the tail and the loops have the XXX sequence (where X = A and/or T) and the dashed and wedged lines represent hydrogen bonds. The C:CH+ base pairs interdigitate in the folded structure. Figure 2. CD spectra for (CCCTAA)4 in 10 mM phosphate buffer, 115 mM K+, as a function of pH and temperature. Figure 3. CD spectra of (CCCXXX)4, where X = A and/or T, in 10 mM phosphate buffer, 115 mM Na+, pH 5.0 at 25 oC. Each oligomer under these conditions has spectral characteristics consistent with the i-motif conformation. The left hand panel features the A rich sequences while the right hand panel features the T rich sequences. Figure 4. A. CD spectra of (CCCTAA)4 in 10 mM phosphate buffer, 115 mM Na+, 25 oC as a function of pH of the solution. B. A plot of the fraction of i-motif as a function of pH for the spectra in panel A. C. CD spectra of (CCCTAA)4 in 10 mM phosphate buffer, 115 mM Na+, pH 5.0 as a function of temperature of the solution. D. A plot of the fraction of i-motif as a function of T for the spectra in panel C. Figure 5. A. CD spectra of (CCCTTA)4 in 10 mM phosphate buffer, 115 mM Na+, pH 5.0 as a function of temperature of the solution. B. A plot of the fraction of i-motif as a function of T for the spectra in panel A. C. A plot of the fraction of i-motif as a function of pH for the protonation induced transition. Figure 6. The variation of ∆Q as a function of loop sequence context. Figure 7. The left panel displays a graphical representation of the van’t Hoff thermodynamic parameters as determine using equations (5) and (6). The right panel demonstrates the enthalpyentropy compensation.

ACS Paragon Plus Environment

Page 18 of 28

Page 19 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

The Journal of Physical Chemistry

Table 1. Spectral properties of the i-motif forming oligomers at 25 oC. Sequence

UV/Vis λmax

ε260

(nm)

(M-1 cm-1)

CD pH 7.0

pH 5.0

Isoelliptic Point

λmax

θmax

λmax

θmax

λip

θip

(nm)

(106 mdeg M-1 cm-1)

(nm)

(106 mdeg M-1 cm-1)

(nm)

(106 mdeg M-1 cm-1)

(CCCAAA)4

259

233,600

278

1.99

292

5.72

280

1.96

(CCCAAT)4

262

220,300

276

2.16

291

5.04

279

2.11

(CCCATA)4

262

226,000

277

2.53

291

6.14

279

2.45

(CCCATT)4

264

204,700

278

2.66

290

6.42

278

2.66

(CCCTAA)4

260

220,400

279

1.89

291

6.35

278

1.88

(CCCTAT)4

265

207,100

279

2.39

290

6.86

278

2.40

(CCCTTA)4

265

204,800

278

3.77

289

6.98

NO

NO

(CCCTTT)4

267

204,700

281

2.56

289

8.87

278

2.48

NO= not observed.

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 28

Table 2. Loop sequence effects on the i-motif:single strand equilibrium.1

Tm (oC)

pHmp

Loop

∆Tm (oC)

∆pHmp Sequence

X=A

X= T

X=A

X=T

A-X-A

5.82

5.84

0.02

55.6

59.2

3.6

T-X-T

6.12

6.15

0.01

65.9

> 75

>9

A-X-T

5.85

6.35

0.50

62.2

75.6

13.4

T-X-A

5.86

6.07

0.21

66.2

67.1

0.9

1

pHmp is the pH at the midpoint of the proton induced single strand to i-motif transition determined from the Fraction i-Motif vs pH plot (Figure 4B) and Tm is T at the midpoint of the heat induced i-motif to single strand transition determined from the Fraction i-Motif vs T plot (Figure 4D). The delta values are calculated as: (X = T) – (X = A).

ACS Paragon Plus Environment

Page 21 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 1. A. A C:CH+ base pair. Since the N3 of cytosine has a pKa of 4.67 (free), C:CH+ base pairs form under acidic conditions. B. A schematic of the intramolecular i-motif conformation formed from a general sequence 5’-(CCCXXX)4-3’ (where X = A and /or T) when under acidic conditions (pH < 6). In this schematic, the red spheres are Cs, the tail and the loops have the XXX sequence (where X = A and/or T) and the dashed and wedged lines represent hydrogen bonds. The C:CH+ base pairs interdigitate in the folded structure. Figure 1 338x190mm (96 x 96 DPI)

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2. CD spectra for (CCCTAA)4 in 10 mM phosphate buffer, 115 mM K+, as a function of pH and temperature. Figure 2 279x362mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 22 of 28

Page 23 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 3. CD spectra of (CCCXXX)4, where X = A and/or T, in 10 mM phosphate buffer, 115 mM Na+, pH 5.0 at 25 oC. Each oligomer under these conditions has spectral characteristics consistent with the i-motif conformation. The left hand panel features the A rich sequences while the right hand panel features the T rich sequences. Figure 3 279x362mm (300 x 300 DPI)

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4. A. CD spectra of (CCCTAA)4 in 10 mM phosphate buffer, 115 mM Na+, 25 oC as a function of pH of the solution. B. A plot of the fraction of i-motif as a function of pH for the spectra in panel A. C. CD spectra of (CCCTAA)4 in 10 mM phosphate buffer, 115 mM Na+, pH 5.0 as a function of temperature of the solution. D. A plot of the fraction of i-motif as a function of T for the spectra in panel C. Figure 4 215x279mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 24 of 28

Page 25 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 5. A. CD spectra of (CCCTTA)4 in 10 mM phosphate buffer, 115 mM Na+, pH 5.0 as a function of temperature of the solution. B. A plot of the fraction of i-motif as a function of T for the spectra in panel A. C. A plot of the fraction of i-motif as a function of pH for the protonation induced transition. Figure 5 279x362mm (300 x 300 DPI)

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 6. The variation of ∆Q as a function of loop sequence context. Figure 6 81x90mm (150 x 150 DPI)

ACS Paragon Plus Environment

Page 26 of 28

Page 27 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 7. The left panel displays a graphical representation of the van’t Hoff thermodynamic parameters as determine using equations (5) and (6). The right panel demonstrates the enthalpy-entropy compensation. Figure 7 279x362mm (300 x 300 DPI)

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

TOC Graphic 215x279mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 28 of 28