DNA Complexes Containing Joined Triplex and Duplex Motifs: Melting

Nov 23, 2009 - Our laboratory is interested in predicting the thermal stability and melting behavior of nucleic acids from knowledge of their sequence...
0 downloads 0 Views 533KB Size
Download PDF
J. Phys. Chem. B 2010, 114, 541–548

541

DNA Complexes Containing Joined Triplex and Duplex Motifs: Melting Behavior of Intramolecular and Bimolecular Complexes with Similar Sequences Hui-Ting Lee,† Irine Khutsishvili,† and Luis A. Marky*,†,‡,§ Department of Pharmaceutical Sciences, Department of Biochemistry and Molecular Biology, and Eppley Institute for Cancer Research, UniVersity of Nebraska Medical Center, 986025 Nebraska Medical Center, Omaha, Nebraska 68198-6025 ReceiVed: August 31, 2009; ReVised Manuscript ReceiVed: October 20, 2009

Our laboratory is interested in predicting the thermal stability and melting behavior of nucleic acids from knowledge of their sequence. One focus is to understand how sequence, duplex and triplex stabilities, and solution conditions affect the melting behavior of complex DNA structures, such as intramolecular DNA complexes containing triplex and duplex motifs. Nucleic acid oligonucleotides (ODNs), as drugs, present an exquisite selectivity and affinity that can be used in antigene and antisense strategies for the control of gene expression. In this work, we try to answer the following question: How does the molecularity of a DNA complex affect its overall stability and melting behavior? We used a combination of temperature-dependent UV spectroscopy and calorimetric (DSC) techniques to investigate the melting behavior of DNA complexes with a similar helical stem sequence, TC+TC+TC+T/AGAGAGACGCG/CGCGTCTCTCT, but formed with different strand molecularity. We determined standard thermodynamic profiles, and the differential binding of protons and counterions accompanying their unfolding. The formation of a DNA complex is accompanied by a favorable free energy term resulting from the typical compensation of favorable enthalpy-unfavorable entropy contributions. As expected, acidic pH stabilized each complex by allowing protonation of the cytosines in the third strand; however, the percentage of protonation increases as the molecularity decreases. The results help in the design of oligonucleotide sequences as targeting reagents that could effectively react with DNA or RNA sequences involved in human diseases, thereby increasing the feasibility of using the antigene and antisense strategies, respectively, for therapeutic purposes. Introduction The use of oligonucleotides (ODNs) as modulators of gene expression has been exploited through two main approaches: the antisense and the antigene strategies.1,2 In the antisense strategy, an ODN binds mRNA and inhibits the translation of the corresponding protein.3 In the antigene strategy, the ODN binds to the major groove of a DNA double helix, resulting in the formation of a triple helix and inhibiting the transcription of its target gene.4 In both cases, formation of stable duplex or triplex molecules is needed, which is dictated by the energy provided by the formation of base-pair and base-triplet stacks, respectively. Triplex formation is sequence specific; recognition of different DNA sequences is through Hoogsteen base pairing.5,6 The inhibition of gene expression via triple helix formation is feasible due to the natural abundance of homopurine-homopyrimidine tracts in genomes.7-11 In general, the affinity in the stable formation of a triplex is provided by base stacking contributions, hydrogen bonding between the third strand bases and purine bases of the duplex, and electrostatic and hydration contributions.12-14 For instance, the formation of a very stable triplex (with ∆G° of nearly -20 kcal mol-1) can potentially inhibit transcription by competing with the binding of proteins that activate the transcriptional machinery.15 The most common triplex motif is the “Pyrimidine” motif in which a pyrimidine* Corresponding author. Tel.: (402) 559-4628. Fax: (402) 559-9543. E-mail: [email protected]. † Department of Pharmaceutical Sciences. ‡ Department of Biochemistry and Molecular Biology. § Eppley Institute for Cancer Research.

rich strand binds parallel to the purine strand of a duplex, forming T · A*T and C · G*C+ base triplets (“ · ” and “*” represent Watson-Crick and Hoogsteen base pairing, respectively). The stability of Pyrimidine triplexes containing exclusively TAT base triplets is pH independent and salt dependent, while those with exclusively C+GC, or combinations of these base triplets, are pH dependent and salt independent.16,17 It has been demonstrated that the presence of hairpin loops in DNA may play an important role in biological processes, such as the palindromic sequences of plasmids forming cruciform structures in response to a topological stress,18-21 providing considerable interest in both the structure and the overall physical properties of nucleic acid hairpin loops. Oligonucleotide hairpin loops form intramolecularly, and their unfolding resembles the pseudomonomolecular melting of a nucleic acid polymer. Their stable formation is accompanied by a lower entropy penalty, and their unfolding takes place at convenient transition temperatures.17,22-24 Our laboratory has investigated a variety of intramolecular DNA triplexes;17,22,25,26 their design involves the inclusion of two hairpin loops that will render them slightly more hydrophobic and compact. Substitution of their duplex closing loop with a hairpin loop containing four base pairs in their stem generates intramolecular DNA complexes containing triplex and duplex motifs with an extended duplex stem. A bimolecular complex with similar sequence can be generated by combining a hairpin loop with a homopurine dangling end and a single strand with pyrimidine ends that are complementary to the dangling end. These DNA complexes were investigated to

10.1021/jp9084074  2010 American Chemical Society Published on Web 11/23/2009

542

J. Phys. Chem. B, Vol. 114, No. 1, 2010

SCHEME 1: Cartoon of the Structures and Designations of DNA Complexes, Control Triplexes, and Hairpin Loopsa

a The sequences of the two strands of BiComp and BiTrip are in black and gray.

further our understanding of the melting behavior of DNA and to construct nanoscale two-dimensional shapes (DNA Origami).27 Another important application includes the targeting of the transient loops of DNA, or the loops of RNA structures, with complementary strands.28 In this work, we used a combination of UV and circular dichroism spectroscopies and calorimetric techniques to investigate the temperature-induced unfolding of an intramolecular DNA complex with joined triplex-duplex motifs and a bimolecular complex with similar helical stem sequence. We obtain a complete thermodynamic description of their melting behavior as a function of pH and salt concentration. The results clearly show how sequence, loops, ion, and proton binding control the stability and melting behavior of a complex nucleic acid molecule. Materials and Methods Materials. All oligonucleotides were synthesized by the Core Synthetic Facility of the Eppley Research Institute at the University of Nebraska Medical Center, HPLC purified, and desalted by column chromatography. The concentration of each oligonucleotide in solution was determined at 260 nm and 90 °C using the following molar extinction coefficients (in mM-1 cm-1 of strands): d(AGAGAGACGCGT5CGCGTCTCTCTC5TCTCTCT), 339; d(AGAGAGAC5TCTCTCTC5TCTCTCT), 259; d(TCTCTCTC5TCTCTCT), 145; d(AGAGAGACGCGT5CGCG), 185; d(AGAGAGA), 92; and d(CGCGC5CGCG), 112. These values were calculated by extrapolation of the tabulated values of the dimer and monomer bases at 25 °C to high temperatures, using procedures reported earlier.26,29 All experiments were done using a 10 mM sodium phosphate buffer solution at pH 6.2 or pH 7.0, and adjusted to the appropriate sodium concentration with NaCl. All other chemicals, from Sigma Chemicals, were of reagent grade and used without further purification. Design of DNA Complexes Containing Joined Triplex and Duplex Motifs. A cartoon of the putative structure of the two complexes and control molecules is shown in Scheme 1. One complex is composed of a single strand that folds intramolecularly with two loops, while the second complex is formed bimolecularly by combining a hairpin loop containing a homopurine dangling end and a pyrimidine single strand with complementary ends to the dangling end forming a triplex motif of seven base triplets. Both complexes have a common helical stem with sequence, TC+TC+TC+T/AGAGAGACGCG/CGCGTCTCTCT, with joined triplex and duplex motifs, and two

Lee et al. loops. These complexes differed from previously investigated intramolecular triplexes in that their stem is elongated with a duplex motif; that is, compare IntraComp with IntraTrip of Scheme 1. Temperature-Dependent UV Spectroscopy. Absorbance versus temperature profiles (UV melts) for each complex and their control molecules were measured at 260 and 275 nm, with a thermoelectrically controlled Aviv 14-DS or a Perkin-Elmer Lambda-10 spectrophotometer, as a function of strand, pH, and salt concentration. The temperature was scanned at a heating rate of ∼0.6 °C/min. The analysis of the shape of these melting curves allows us to measure TM’s, which are the midpoint temperatures of their order-disorder transition, and van’t Hoff enthalpies, ∆HvH. We used the following relationship and procedures that have been described earlier to determine these ∆HvH’s:30 ∆HvH ) (2n + 2)RTM2(∂R/∂T); where “R” is the universal gas constant, “R” is the fraction of single strands in the helical state, and (∂R/∂T) is the slope of the R versus T curve measured around the TM; “n” is the molecularity of the transition, equal to 1 or 2 for monomolecular and bimolecular transitions, respectively. The TM’s of multiphasic melting curves were obtained from the peaks of the corresponding differential melting curves. Circular Dichroism Spectroscopy (CD). We use a thermoelectrically controlled Aviv circular dichroism spectrometer model 202SF (Lakewood, NJ) to measure the CD spectrum of each oligonucleotide. The analysis of these spectra yielded the conformation adopted by the helical state of each ODN. Typically, we prepared an ODN solution with an absorbance of 1 in appropriate buffered solutions, and the CD spectrum is measured from 320 to 200 nm every 1 nm, using a strained free quartz cuvette with a path length of 1 cm, and at temperatures that the ODN is 100% in the helical state. The reported spectra correspond to an average of at least two scans. Differential Scanning Calorimetry (DSC). To thermodynamically investigate the helix-coil transition of each molecule, excess heat capacity functions (DSC melts) were measured with a Microcal VP-DSC differential scanning microcalorimeter (Northampton, MA). Two cells, the sample cell containing 0.7 mL of oligonucleotide solution and the reference cell filled with the same volume of buffer solution, were heated differentially from 1 to 110 °C at a heating rate of 0.75 °C/min. Analysis of the resulting thermograms yielded standard thermodynamic profiles (∆Hcal, ∆Scal, and ∆G°T). These profiles were obtained from the following relationships, using procedures described previously:30 ∆Hcal ) ∫∆Cpa dT and ∆Scal ) ∫(∆Cpa/T) dT, where ∆Cpa represents the anomalous heat capacity during the unfolding process. The free energy change at any temperature, ∆G°T, is obtained from the Gibbs equation: ∆G°T ) ∆Hcal T∆Scal. The ∆HvH’s are determined from analysis of the shape of the experimental DSC curves using procedures described earlier.30 The nature of each transition, two-state versus non twostate, was simply estimated by the ∆HvH/∆Hcal ratio; for a twostate transition, ∆HvH ) ∆Hcal, while for a non-two-state, ∆HvH * ∆Hcal.30 These ratios are obtained using the average ∆HvH values obtained from UV and DSC melts. Differential Thermodynamic Binding of Protons and Counterions. UV melting curves were carried out as a function of pH and salt concentration to measure the differential thermodynamic binding of protons, ∆nH+, and counterions, ∆nNa+, between the folded and unfolded states, respectively. These two linking numbers were determined with the following relationships:31 ∆nH+ ) -0.43(∆Hcal/RTM2) [dTM/dpH], and ∆nNa+ ) 1.1(∆Hcal/RTM2) [dTM/d ln[Na+]]. The first term in

Unfolding of DNA Complexes

J. Phys. Chem. B, Vol. 114, No. 1, 2010 543

Figure 1. Typical UV melting curves of complexes in 10 mM sodium phosphate buffer at pH 6.2 (solid) or 7.0 (open), 0.2 M NaCl, and at two different wavelengths. Insets: TM-dependences on strand concentration of the first (circles) and second (squares) transitions at pH 6.2 (solid) or 7.0 (open).

parentheses in these two equations was measured directly from DSC experiments, while the second term in brackets was measured from the slopes of the dependences of the TM on pH and salt, respectively, 0.43 is the factor that converts natural logarithms into decimal logarithms, while 1.1 is a factor for converting solution activities into concentrations.

TABLE 1: TM’s and ∆HvH’s for the UV Unfolding of Complexes at 260 and 275 nm pH 7 TM 260 nm 275 nm

Results and Discussion Overall Experimental Protocol. The molecularity of each complex is determined initially with UV melts by investigating their temperature-induced transitions as a function of strand concentration. This is determined from the dependence of their TM on strand concentration; if the TM remains constant with the increase in strand concentration the complex is formed intramolecularly, and if the TM increases the complex forms with higher molecularity, that is, involving two or more strands. CD spectroscopy is then used to determine the overall conformation of each complex by simple inspection of its CD spectrum. Next, DSC is used to determine unfolding thermodynamic profiles from analysis of their heat capacity functions and the Gibbs equation.30 To determine the ∆nH+ and ∆nNa+ parameters for each complex, additional UV melts were obtained to measure their TM-dependences on pH and salt concentration, respectively. These TM-dependences together with their corresponding unfolding DSC enthalpies yield the ∆nH+ and ∆nNa+ for each complex. Optical Unfolding of Complexes (UV Melting Curves). UV melting experiments as a function of strand and salt concentrations and pH were used to characterize the helix-coil transition of all molecules shown in Scheme 1. Figure 1 shows typical UV melts for each complex at the pH’s of 6.2 and 7 and at wavelengths of 260 and 275 nm. All transitions shown in these curves follow the characteristic sigmoidal behavior for the unfolding of a nucleic acid helical structure. The UV melts of both complexes exhibit clear biphasic transitions; the IntraComp shows a broad first transition followed by a sharp second transition, while the bimolecular complex shows a somewhat sharper first transition followed by a broad second transition. However, the overall melting behavior of each molecule changes with the pH. Their first transition is shifted to higher temper-

260 nm 275 nm

pH 6.2 TM

∆HvH

44 74 36 79

45.4 82.4 46.1 82.7

41 77 39 75

68 37 58 37

30.0 72.0 31.4 72.8

72 38 88 39

∆HvH

first second first second

IntraComp 25.0 82.4 25.8 82.9

first second first second

BiComp 18.5 72.0 19.2 72.7

a All parameters measured in a 10 mM sodium phosphate buffer, 0.2 M NaCl, at pH 7 or 6.2. Experimental errors are shown in parentheses: TM ((0.5 °C) and ∆HvH ((10%). The TM’s correspond to a total ODN concentration of 4 µM.

atures at pH 6.2, by 20.4 °C (IntraComp) and 11.9 °C (BiComp), while no temperature shifts are observed for their second transition. The comparison of the melting curves at the two wavelengths of measurement showed hyperchomicity differences, which were consistent with the wavelength of measurement; that is, 260 nm monitors the unfolding of sequences containing AT and GC base pairs and TAT and C+GC base triplets, while the 275 nm wavelength is more sensitive to the unfolding of CG base pairs and C+GC base triplets. For instance, the second transition of each complex corresponds more or less to the unfolding of four CG base pairs; therefore, higher hyperchromicities were observed at 275 nm than at 260 nm. The main observation is that the analysis of all transitions yielded similar TM’s; see Table 1. UV melts of each complex were obtained over a total strand concentration range of 1-20 µM to determine their transition molecularities; the TM-dependences on strand concentration are shown in the insets of Figure 1. At the two pH’s, the TM’s of the two transitions of IntraComp and the second transition of the BiComp do not depend on strand concentration, while the TM’s of the first transition of BiComp do depend on strand concentration. This confirms IntraComp forms intramolecularly,

544

J. Phys. Chem. B, Vol. 114, No. 1, 2010

Lee et al.

Figure 2. Circular dichroism spectra of IntraComp (solid line) and BiComp (b) in 10 mM sodium phosphate buffer, 0.2 M NaCl, at pH 6.2.

while BiComp forms intermolecularly. Shape analysis of the UV melts yielded ∆HvH’s for each transition of both complexes. At the two wavelengths of measurement (Table 1), we obtained similar ∆HvH’s, within experimental error. For the first transition(s) of each complex, we obtained a ∆HvH of 40 kcal mol-1 at both pH’s (IntraComp), 63 and 80 kcal mol-1 at pH 7 and 6.2, respectively (BiComp); and for their second transition at both pH’s, 76 kcal mol-1 (IntraComp) and 38 kcal mol-1 (BiComp). The total ∆HvH’s for the overall unfolding of the two complexes at pH 6.2 range from 116 to 118 kcal mol-1, assuming additivity of their component hairpins. These values are higher than the estimated value of 80-85 kcal mol-1 for the unfolding enthalpy of their entire stem duplex.32,33 These enthalpy differences of 35 kcal mol-1 confirm the formation of a triplex motif in each complex. The comparison of the TM and ∆HvH values of each complex with those of their duplex component, together with their TMdependences on strand concentration and observed number of transitions, allow us to explain their melting behavior, as follows. The biphasic unfolding of IntraComp corresponds to the sequential intramolecular removal of the third strand followed by unfolding of its whole stem-loop component, while the biphasic unfolding of BiComp corresponds to the unfolding of the whole triplex motif followed by the unfolding of the hairpin loop with its dangling end. Overall, each complex unfolds according to their design. CD Spectroscopy and Complex Conformation. The CD spectra of both complexes are shown in Figure 2. The spectra consist of a positive band centered at 285 nm, a crossover at 265-270 nm, and a negative band centered at ∼246 nm. The positive band has a slightly larger area than the negative band due to the contribution of the two constrained loops of each complex, which also affects the standard 260 nm crossover by shifting it to longer wavelengths. Furthermore, each complex exhibits a broad negative band at 210 nm, which has been observed in previous reports,14,17,34 and it has been considered as a characteristic feature of the formation of a triplex. This analysis indicates that both complexes exhibit the typical CD spectra of a nucleic acid helix in the “B” conformation and that binding of the third strand does not impose major distortions in the geometry of the duplex. Furthermore, the magnitude of both bands, especially the negative band, is smaller for BiComp, suggesting that this complex is less compact; that is, the extent of base-pair and base-triplet stacking is somewhat lowered. Calorimetric Unfolding of Complexes (DSC Melting Curves). Figure 3 shows the calorimetric unfolding of each complex at pH 7 and 6.2. The unfolding of each complex takes place mainly through biphasic transitions with TM’s that follow the same trend as the TM’s of the UV melts. The exception is IntraComp at pH 6.2 that shows a broad and small additional

Figure 3. DSC melting curves of complexes IntraComp and BiComp, in 10 mM sodium phosphate buffer, 0.2 M NaCl, at pH 6.2 (b) or 7.0 (solid lines).

transition at 18 °C. Each thermogram show similar pre- and post-transition baselines, indicating that complex unfolding takes place with negligible heat capacity effects. Integration of each transition of these DSC melts yielded endothermic enthalpies, which are shown as enthalpies of formation in Table 2. We obtained a ∆Hcal of 46.2 kcal mol-1 for the first transition of IntraComp at pH 7 that increases to 70.8 kcal mol-1 (sum of the two transitions below 70 °C) when the pH is decreased to 6.2, and a similar ∆Hcal of ∼83.3 kcal mol-1 is obtained for the last transition of this complex at both pH’s. The ∆Hcal of the first transition of BiComp increases from 77.6 to 94.3 kcal mol-1 when the pH is decreased from 7 to 6.2, while the enthalpy of the second transition remains similar (∆Hcal ) 34.5 kcal mol-1) with this pH change. The main observation is that the total unfolding enthalpy increases by 20.2 kcal mol-1 (IntraComp) and 16.5 kcal mol-1 (BiComp) with this decrease in pH (Table 2), which is consistent with a similar trend observed with the control triplexes, IntraTrip and BiTrip (Table 3). This confirms the presence of a triplex motif in each complex. At pH 6.2, there is an optimization of the base-triplet stacks of both complexes, due to a more effective protonation of cytosines. This optimization is higher with IntraComp, suggesting a more compact complex, which is qualitatively in agreement with the observed differences in their CD spectra. Table 2 also shows van’t Hoff enthalpies determined from the shape of the DSC curves, which are similar to the ones obtained from UV melting curves within experimental error. We have averaged the optical and calorimetric ∆HvH’s for each transition to determine the ∆HvH/∆Hcal ratios. We obtained ∆HvH/∆Hcal ratios in the range of 0.9-1.1, indicating that all transitions are two-state without the presence of intermediate states; one exception is the second transition of IntraComp at pH 6.2 that has a ∆HvH/∆Hcal ratio of 0.7, which indicates a non-two-state transition with the presence of intermediate states. Similar analysis for the control molecules yielded ∆HvH/∆Hcal ratios in the range of 0.9-1.1, indicating that these transitions are two-state; the exception was the second transition of IntraTrip at pH 6.2 that had a ∆HvH/∆Hcal ratio of 0.4, which indicates a non-two-state transition. Relative to IntraComp, the lower ∆HvH/∆Hcal ratio of IntraTrip indicates a lower size of its cooperative unit melt.

Unfolding of DNA Complexes

J. Phys. Chem. B, Vol. 114, No. 1, 2010 545

TABLE 2: Folding Thermodynamic Profiles of DNA Complexesa pH 7.0 6.2

7.0 6.2

TM (°C) first second total first second third total

24.7 84.0

first second total first second total

30.8 73.5

17.4 46.0 84.2

42.6 72.5

∆Go(5) (kcal/mol)

∆Hcal (kcal/mol)

–3.1 –18.9 –22.0 –0.6 –7.4 –18.0 –26.0

–46.2 –85.5 –131.7 –12.9 –57.9 –81.1 –151.9

IntraComp –43.1 –66.6 –109.7 –12.3 –50.5 –63.1 –125.9

–6.6 –6.8 –13.4 –11.2 –6.7 –17.9

–77.6 –34.6 –112.2 –94.3 –34.4 –128.7

BiComp –71.0 –27.8 –98.8 –83.1 –27.7 –110.8

T∆Scal (kcal/mol)

∆HvH (kcal/mol)

∆HvHavg/∆Hcal

–41 –76

0.9 0.9

∆nH+ (molH+/mol)

∆nNa+ (molNa+/mol)

–2.7

0.9 –2.4 –1.5

–2.7 –42 –74

0.7 0.9

–3.0 –3.0

–74 –41

0.9 1.1

–89 –40

0.9 1.1

–1.4 –1.4 –1.8 –1.8

0.7 –2.3 –1.6 –1.9 –0.9 –2.8 –1.2 –0.7 –1.9

a All parameters measured in a 10 mM sodium phosphate buffer, 0.2 M NaCl, at pH 7 or 6.2. Experimental errors are shown in parentheses: TM ((0.5 °C), ∆HvH ((10%), ∆Hcal ((5%), T∆Scal ((5%), and ∆Go(5) ((7%), ∆nNa+ and ∆nNa+ ((20%). The TM’s correspond to a total ODN concentration of 80 µM.

TABLE 3: Folding Thermodynamic Profiles of Control Triplexes and Hairpin Loopsa pH 7.0

TM (°C)

∆Go(5) (kcal/mol)

∆Hcal (kcal/mol)

–4.6 –7.2 –11.8 –1.2 –16.0 –17.2

–50.5 –44.2 –94.7 –13.6 –110.0 –123.6

IntraTrip –45.9 –37.0 –82.9 –12.4 –94.0 –106.4 BiTrip –82.8 –94.3

first second total first second total

33.1 58.9

7.0 6.2

single single

31.3 39.6

–7.8 –11.7

–90.6 –106.0

7.0 6.2

single single

72.6 73.1

–7.0 –6.9

–35.7 –35.0

7.0 6.2

single single

69.6 69.5

–6.2 –6.1

–33.1 –32.6

6.2

31.0 52.5

T∆Scal (kcal/mol)

∆HvH (kcal/mol)

∆HvHavg/∆Hcal

–51 –42

1.0 0.9

∆nH+ (molH+/mol)

∆nNa+ (molNa+/mol)

–2.9

–41

0.4

–3.3 –3.3

0.9 –1.3 –0.4 0.1 –1.5 –1.4

–85 –95

0.9 0.9

–1.9 –2.1

–0.4 –0.5

(AG)3A(CG)2Hp –28.7 –41 –28.1 –39

1.1 1.1

–0.5 –0.5

1.1 1.1

–0.3 –0.3

(CG)2Hp –26.9 –26.5

–2.9

–34 –36

a All parameters measured in a 10 mM sodium phosphate buffer, 0.2 M NaCl, at pH 7 or 6.2. Experimental errors are shown in parentheses: TM ((0.5 °C), ∆HvH ((10%), ∆Hcal ((5%), T∆Scal ((5%), and ∆Go(5) ((7%), ∆nNa+ and ∆nNa+ ((20%). The TM’s correspond to a total ODN concentration of 80 µM.

Furthermore, IntraComp is much more stable (∆∆G° ) -8.4) than BiComp in this pH range, due to a more favorable enthalpy contribution (∆∆Hcal ) -21.4 kcal mol-1); see Table 2. This indicates more robust base-triplet stacks, confirming that IntraComp forms a much stronger and compact complex. Melting Behavior of Complexes Relative to Control Molecules. The calorimetric unfolding of the two complexes at pH 6.2, and their corresponding control triplex and hairpin loop molecules at this pH, is shown in Figure 4. Standard thermodynamic profiles for the folding of each complex are presented in Table 2, while those of the control molecules are presented in Table 3. The temperature-unfolding of IntraComp is triphasic, while the unfolding of its control molecules, IntraTrip and (CG)2Hp, is monophasic; however, IntraTrip shows a shoulder at lower temperatures. The first two transitions of IntraComp have TM’s of ∼10 °C below the TM’s of the shoulder and the main transition of IntraTrip, while its third transition shows a higher TM, by 14 °C, than the TM of its control hairpin loop (see Figure 4 and Figure S1). However, the total unfolding enthalpy (151.9 kcal mol-1) of IntraComp is 4.3 kcal

Figure 4. DSC curves of complexes and control molecules in 10 mM sodium phosphate buffer, 0.2 M NaCl at pH 6.2. Top panel: IntraComp (solid line), IntraTrip (O), and (CG)2Hp (0). Bottom panel: BiComp (38 µM, solid line), BiTrip (98 µM, O), and (AG)3A(CG)2Hp (0).

546

J. Phys. Chem. B, Vol. 114, No. 1, 2010

mol-1 lower than the sum of the enthalpies of its component molecules (Table 2), which may be explained by an enthalpy compensation in the formation of the triplex-duplex junction, that is, removal of the stacked bases in the triplex loop with the formation of an additional base-pair stack upon the inclusion of the hairpin loop. Clearly, the first two transitions of IntraComp have a lower enthalpy of 52.8 kcal mol-1 than the total enthalpy of IntraTrip, while its third transition has a much higher unfolding enthalpy, by 48.5 kcal mol-1, than (CG)2Hp. IntraComp consists of a triplex motif of seven base triplets and a duplex motif of four base pairs (Scheme 1); its melting behavior is therefore explained as follows: the first two transitions correspond to the cooperative removal of the third strand and disruption of 3-4 AG/TC and GA/CT base-pair stacks followed by the cooperative disruption of the remaining 8-9 base-pair stacks of the duplex domain. This indicates that the elongation of the duplex domain of this complex, by replacing the right-hand side loop of the control triplex with its control hairpin loop, forces the removal of the third strand. On the other hand, the unfolding of BiComp is biphasic, while its two control molecules, IntraTrip and (AG)3A(CG)2Hp, undergo monophasic transitions. These transitions overlap very well with each transition of the control molecules (lower panel of Figure 4). The TM’s of the transitions of BiComp agree well (2-4 °C) with the TM’s of their control molecules; however, the total unfolding enthalpy (128.7 kcal mol-1) of BiComp is 12.3 kcal mol-1 lower than the sum of the enthalpies of its component molecules (Table 2). The biphasic melting behavior of BiComp corresponds to the disruption of the triplex motif, releasing (AG)3A(CG)2Hp and followed by the independent unfolding of this hairpin. This provides a good strategy for using a double complementary (duplex and triplex) pyrimidine-rich single strand to target homopurine single stranded ends attached to hairpin loops. The differences in the overall unfolding free energies between each complex and the sum of its control molecules at pH 6.2 are small (Tables 1 and 2), in kcal mol-1: -2.7 (IntraComp) and 0.7 (BiComp), due to a relatively small differential enthalpy-entropy compensation of an unfavorable enthalpy contribution, ∆∆H ) 4.3 (IntraComp) and 12.3 (BiComp), with a favorable entropy contribution, ∆(T∆S) ) 7.0 (IntraComp) and 11.6 (BiComp). In the following sections, we discuss the differential binding of protons and counterions between the helical and random coils states for each transition of each complex. These parameters contribute effectively to the entropy terms. UV Melts as a Function of pH and Differential Binding of Protons. UV melting curves of each complex in the pH range of 6.2-7.2 and at the total Na+ concentration of 216 mM are shown in Figure 5. IntraComp unfolds in a biphasic transition at pH 7, lowering the pH shifts the first transition (third strand removal) to higher temperatures while the second transition (duplex unfolding) remains the same; the first transition generates a shoulder at lower temperatures with the decrease in pH and the overall hyperchromicity at pH 6.2 is 16%. The melting behavior of InterComp is similar to its DSC melts (Figure 3). The UV melting curves of BiComp in this pH range show biphasic transitions, as both transitions are shifted to higher temperature with the decrease in pH; however, the second transition shifts are to a much lesser extent, and the overall complex hyperchromicity is 15% at pH 6.2 (Figure 5). Similar comparison of the control triplexes (data not shown) yielded hyperchromicities of 17% (IntraTrip) and 21% (BiTrip).

Lee et al.

Figure 5. UV melting curves of complexes in 10 mM sodium phosphate, 0.2 M NaCl, as a function of pH (6.2-7.2). Insets: TMdependences on pH of the first (solid) and second (open) transitions.

We used well-defined UV melting curves through the pH range of 6.2-7.2 to determine the slope of the TM-dependences on pH for each transition. These TM-dependences are shown in the insets of Figure 5, which together with the enthalpy factor (∆Hcal/RTM2) allowed us to measure the ∆nH+ linking number. The TM-dependence on pH yielded slopes of -24.2 and -14.8 °C for the first transition of these complexes, and nearly zero slopes for their second transition. As expected, the second transition of each complex and the control hairpin loops ((CG)2Hp and (AG)3A(CG)2Hp), shown in Figure S2, yielded nearly negligible ∆nH+ values; these values were actually nonmeasurable (Figure S2). This shows that the differential binding of protons in the folding of these complexes is associated with the protonation of the third strand cytosines, and to a smaller extent to protonation of the loop cytosines.35,36 Therefore, the ∆nH+ value of each complex has been determined using the TM and enthalpy for the association of the third strand. For IntraComp, we use the folding enthalpy of its first transition, while the folding enthalpy of BiComp is estimated by subtracting the folding enthalpy of its duplex stem from the overall triplex enthalpy, as used earlier.22 At pH 6.2 and 216 mM Na+, we obtained net uptake of protons; the ∆nH+ values were (in mol H+ per mol) -3.0 ( 0.2 (IntraComp) and -1.8 ( 0.2 (BiComp), see Table 1, whereas the control triplexes have ∆nH+’s of -3.3 ( 0.2 (IntraTrip) and -2.1 ( 0.2 (BiTrip), see Table 2. The expected protonation value for 100% formation of this triplehelical stem sequence is 3 (∆nH+ ) -3), which lowers in magnitude to 2.7 at pH 6.2 when the pKa of cytosine is considered,17,37 and increases in magnitude, by 0.10-0.15 per loop,17,35,36 if cytosine loops are involved. The 0.1 value was calculated using a pK value of 4.6 that corresponds to cytosines in the single-stranded state,35,36 while the 0.15 value has been estimated from the experimental ∆nH+ values of intramolecular triplexes and control duplexes with exclusively TAT base triplets

Unfolding of DNA Complexes

Figure 6. UV melting curves of complexes in 10 mM sodium phosphate buffer at pH 6.2, as a function of Na+ concentration (16-216 mM). Insets: TM-dependences on the total concentration of Na+ for the first (solid) and second (open) transitions.

and AT base pairs, respectively, with C5 loops.17 Overall, the closer is the ∆nH+ value to -3, the better is the formation of the triplex. Therefore, the proton uptake values of these triplex components showed that each complex has a similar number of base-triplets and base-triplet stacks as compared to its control triplexes, and that the intramolecular complex and triplex are stronger than their counterpart bimolecular ones, consistent with earlier reports.22 UV Melts as a Function of Salt Concentration and Differential Binding of Counterions. UV melting curves as a function of salt concentration at pH 6.2 are shown in Figure 6. In this range of salt concentration (16-216 mM Na+), IntraComp unfolds in biphasic transitions at low salt concentrations, changing to triphasic transitions at higher salt concentrations, while BiComp unfolds in clear biphasic transitions. All melting curves are shifted to higher temperatures with the increase in salt concentration; the exception is the first two transitions of IntraComp that are shifted to lower temperatures. The increase in salt yielded small increases in hyperchromicities of 1% (IntraComp) and 2% (BiComp). Despite the complicated melting behavior of IntraComp as a function of salt concentration, the increase in Na+ concentration has a small effect on the thermal stability of its triplex state and a very large effect on the unfolding of its duplex motif. However, the increase in salt concentration shifts the first transition of BiComp to higher temperatures to a larger extent than it does to the second transition (Figure 6). The TM-dependences on salt concentration for the complexes are shown in the insets of Figure 6, while those for the control molecules are shown in Figure S3, which together with its enthalpy factor allow us to measure the associated ∆nNa+ values, shown in Table 2. However, it should be noted that the measurement of ∆nNa+ is difficult because of the competition

J. Phys. Chem. B, Vol. 114, No. 1, 2010 547 between counterion exclusion by the triplex state and counterion binding by the duplex state within the triplex motif.17,22 The slope of the TM-dependences is negative for the first transition of IntraComp and positive for the second transition of IntraComp and for the transitions of BiComp (inset of Figure 6). Therefore, folding of IntraComp at pH 6.2 is accompanied by an overall counterion uptake, ∆nNa+ ) -1.6 ( 0.3 mol Na+ per mol (Table 1), resulting from the release of 0.7 ( 0.2 mol Na+ per mol (first transition) and uptake of 2.3 ( 0.1 mol Na+ per mol (second transition), while the folding of BiComp yielded a net ∆nNa+ value of -1.9 ( 0.2 mol Na+ per mol, resulting from the sum of the uptakes of counterions from both transitions. On the other hand, the folding of IntraTrip yielded a net uptake of counterions (∆nNa+ ) -1.4 ( 0.1 mol Na+ per mol), resulting from the release of 0.1 mol Na+ per mol (first transition) and uptake of 1.5 ( 0.1 mol Na+ per mol (second transition), while BiTrip has a net uptake of 0.5 mol Na+ per mol (Table 2 and Figure S3).38 The control hairpins have net uptakes of 0.3 mol Na+ per mol ((CG)2Hp) and 0.5 mol Na+ per mol ((AG)3A(CG)2Hp). In general, the uptake of counterions by each complex is higher than the sum of their control molecules (by a ∆∆nNa+ of -1 mol Na+ per mol) at both pH’s; however, this differential ∆nNa+ drops to -0.1 for IntraComp at pH 6.2. This is explained in terms of the more effective exclusion of counterions by the protonated cytosines of IntraComp. This counterion-proton competition is in excellent agreement with previous reports on the unfolding of intramolecular triplexes with C+GC base triplets.17 Complete Thermodynamic Profiles for the Folding of Complexes. Standard thermodynamic profiles for the folding of each DNA complex at 5 °C are summarized in Table 1; at this temperature, all molecules are in the helical state. Inspection of Table 1 indicates that the folding of each molecule is accompanied by a favorable free energy term, which results from the characteristic compensation of a favorable enthalpy and unfavorable entropy contributions.39 In general, favorable heat contributions correspond to the formation of base-pairs and basetriplet stacks, while unfavorable entropy contributions arise from the unfavorable ordering of strands and the putative uptake of counterions, protons, and perhaps water molecules. The main observation is that the favorable folding free energy of each complex is enthalpy driven, that is, exothermic enthalpy contributions. This is consistent with the overall observation that the folding of a helical structure is enthalpy driven.39 Specifically, the ∆G°(5) terms of Table 2 indicate IntraComp is more stable than BiComp, by -8.1 kcal mol-1 (pH 6.2) and -8.6 kcal mol-1 (pH 7), indicating the favorable formation of a complex increases with decreasing its molecularity. IntraComp also has a more favorable folding enthalpy, by -23.2 kcal mol-1 (pH 6.2) and -19.5 kcal mol-1 (pH 7). This shows that the magnitude of the enthalpy depends on the number and strength of the base-triplet stacks that formed within each triplex and stacking contribution from the loops, that is, the compactness of the triplex motif, and on the number of base pair-stacks of the duplex motif. This suggests that the inclusion of loops may have a stabilizing effect that improves base-triplet stacking, by preventing the exposure of base-triplets to the solvent. In terms of the entropy contribution, T∆Scal, IntraComp has a lower entropy penalty, by 15.1 kcal mol-1 (pH 6.2) and 10.9 kcal mol-1 (pH 7). From the point of view of triplex molecularity, the lower is the number of strands that formed a particular complex, the lower is the entropy. However, the above observation indicates that the magnitude of the unfavorable entropy contributions depends strongly on the extent of proton and

548

J. Phys. Chem. B, Vol. 114, No. 1, 2010

counterion binding by the triplex motif contained in these complexes; the lowering of pH stabilizes triplexes containing C+GC base triplets, while the increase in salt concentration is destabilizing. Moreover, triplexes and complexes with triplex motifs that formed intermolecularly will need high salt concentrations for their stable formation. Conclusion We have investigated the unfolding of a pair of DNA complexes containing joined triplex and double helical motifs as a function of their molecularity. Complete thermodynamic profiles, including the differential binding of protons and counterions, are reported for the observed transitions of each complex, and their overall thermodynamic profiles are compared to those of their control molecules. Overall, the temperatureinduced unfolding of each complex corresponds to the initial disruption of the triplex motif (removal of the third strand) followed by the partial or full unfolding of the duplex stem. Complex formation is accompanied by a favorable free energy term, resulting from the typical compensation of favorable enthalpy-unfavorable entropy contributions; that is, the folding of a particular complex is enthalpy driven. The magnitude of the favorable enthalpy contributions corresponds to the number and strength of the base-triplet stacks and base-pair stacks formed, which are helped by stacking contributions due to the incorporation of dangling ends or loops. The intramolecular complex (IntraComp) is more stable, consistent with its lower molecularity. As expected, acidic pH stabilized both triplexes by allowing protonation of the cytosines in the third strand; however, the percentage of protonation depended on triplex molecularity. The folding of these triplexes is consistent with a model reported earlier in which the triplex to duplex transition is accompanied by a release of protons and the uptake of sodium ions, while the duplex to random coil transition is accompanied by a release of sodium ions.17 The results help to choose adequate solution conditions for the study of triplexes containing different ratios of C+GC and TAT base-triplets. Moreover, the resulting data should improve our current picture of how sequence, loops, proton, and ion binding control the stability and melting behavior of nucleic acid molecules, and would supplement existing nearest-neighbor thermodynamic parameters to help in the prediction of secondary structure from a given sequence, as demonstrated earlier.32,33,40,41 Furthermore, the design of these complexes mimics portions of the complex structures presented by nucleic acid molecules with internal and end loops; therefore, the data presented should help in the optimization of oligonucleotide reagents for the targeting of specific transient structures that form in vivo,42 using antigene and antisense strategies. It should be noted that the use of DNA strands in the antisense strategy would yield DNA/RNA hybrid duplexes that will have a lower stability, by an average ∆G of 0.2 kcal mol-1, than the corresponding DNA/DNA duplexes.41,43 Acknowledgment. This work was supported by Grants MCB0315746 and MCB-0616005 from the National Science Foundation. We greatly appreciate the preliminary data obtained in our laboratory by Andre Seidel under a fellowship from the Student ExChange Program (SEP) of the International Students Federation (IPSF), during the Summer of 2006.

Lee et al. Supporting Information Available: Additional figures. This material is available free of charge via the Internet at http:// pubs.acs.org. References and Notes (1) Helene, C. Eur. J. Cancer 1991, 27, 1466. (2) Helene, C. Eur. J. Cancer 1994, 30A, 1721. (3) Dove, A. Nat. Biotechnol. 2002, 20, 121. (4) Duval-Valentin, G.; Thuong, N. T.; Helene, C. Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 504. (5) Soyfer, V. N.; Potaman, V. N. Triple-Helical Nucleic Acids; Springer-Verlag: New York, 1996. (6) Frank-Kamenetskii, M. D.; Mirkin, S. M. Annu. ReV. Biochem. 1995, 64, 65. (7) Larsen, A.; H., W. Cell 1982, 29, 609. (8) Yavachev, L. P.; Georgiev, O. I.; Braga, E. A.; Avdonina, T. A.; Bogomolova, A. E.; Zhurkin, V. B.; Nosikov, V. V.; Hadjiolov, A. A. Nucleic Acids Res. 1986, 14, 2799. (9) Behe, M. J. Biochemistry 1987, 26, 7870. (10) Behe, M. J. Nucleic Acids Res. 1995, 23, 689. (11) Kato, M.; Kudoh, J.; Shimizu, N. Biochem. J. 1990, 268, 175. (12) Cheng, Y. K.; Pettitt, B. M. Prog. Biophys. Mol. Biol. 1992, 58, 225. (13) Soto, A. M.; Marky, L. A. Biochemistry 2002, 41, 12475. (14) Shikiya, R.; Marky, L. A. J. Phys. Chem. B 2005, 109, 18177. (15) Maher, L. J., III; Wold, B.; Dervan, P. B. Science 1989, 245, 725. (16) Keppler, M. D.; Fox, K. R. Nucleic Acids Res. 1997, 25, 4644. (17) Soto, A. M.; Loo, J.; Marky, L. A. J. Am. Chem. Soc. 2002, 124, 14355. (18) Panayotatos, N.; Wells, R. D. Nature (London) 1981, 289, 466. (19) Lilley, D. M. J. Proc. Natl. Acad. Sci. U.S.A. 1980, 77, 6468. (20) Pollack, Y.; Stein, R.; Razin, A.; Cedar, H. Proc. Natl. Acad. Sci. U.S.A. 1980, 77, 6463. (21) Lilley, D. M. J. Nucleic Acids Res. 1981, 9, 1271. (22) Lee, H.-T.; Arciniegas, S.; Marky, L. A. J. Phys. Chem. 2008, 112, 4833. (23) Rentzeperis, D.; Alessi, K.; Marky, L. A. Nucleic Acids Res. 1993, 21, 2683. (24) Marky, L. A.; Maiti, S.; Olsen, C. M.; Shikiya, R.; Johnson, S. E.; Kaushik, M.; Khutsishvili, I. Building blocks of nucleic acid nanostructures: unfolding thermodynamics of intramolecular DNA complexes. In Biomedical Applications of Nanotechnology; Labhasetwar, V., Leslie-Pelecky, D., Eds.; John Wiley & Sons, Inc.: New York, 2007; p 191. (25) Rentzeperis, D.; Marky, L. A. J. Am. Chem. Soc. 1995, 117, 5423. (26) Marky, L. A.; Blumenfeld, K. S.; Kozlowski, S.; Breslauer, K. J. Biopolymers 1983, 22, 1247. (27) Rothemund, P. W. K. Nature (London) 2006, 440, 297. (28) Lee, H.-T.; Olsen, C. M.; Waters, L.; Sukup, H.; Marky, L. A. Biochimie 2008, 90, 1052. (29) Cantor, C. R.; Warshaw, M. M.; Shapiro, H. Biopolymers 1970, 9, 1059. (30) Marky, L. A.; Breslauer, K. J. Biopolymers 1987, 26, 1601. (31) Kaushik, M.; Suehl, N.; Marky, L. A. Biophys. Chem. 2006, 126, 154. (32) Breslauer, K. J.; Frank, R.; Bloecker, H.; Marky, L. A. Proc. Natl. Acad. Sci. U.S.A. 1986, 83, 3746. (33) SantaLucia, J., Jr.; Allawi, H. T.; Seneviratne, P. A. Biochemistry 1996, 35, 3555. (34) Escude, C.; Francois, J.-C.; Sun, J.-S.; Ott, G.; Sprinzl, M.; Garestier, T.; Helene, C. Nucleic Acids Res. 1993, 21, 5547. (35) Zimmer, C.; Luck, G.; Venner, H.; Fric, J. Biopolymers 1968, 6, 563. (36) Zimmer, C.; Venner, H. Biopolymers 1966, 4, 1073. (37) Leitner, D.; Schroeder, W.; Weisz, K. Biochemistry 2000, 39, 5886. (38) Rentzeperis, D.; Kharakoz, D. P.; Marky, L. A. Biochemistry 1991, 30, 6276. (39) Marky, L. A.; Kupke, D. W. Methods Enzymol. 2000, 323, 419. (40) Xia, T.; SantaLucia, J., Jr.; Burkard, M. E.; Kierzek, R.; Schroeder, S. J.; Jiao, X.; Cox, C.; Turner, D. H. Biochemistry 1998, 37, 14719. (41) Sugimoto, N.; Nakano, S.-i.; Katoh, M.; Matsumura, A.; Nakamuta, H.; Ohmichi, T.; Yoneyama, M.; Sasaki, M. Biochemistry 1995, 34, 11211. (42) Kankia, B. J. Phys. Chem. B 1999, 103, 8759. (43) Sugimoto, N.; Nakano, S.-i.; Yoneyama, M.; Honda, K.-i. Nucleic Acids Res. 1996, 24, 4501.

JP9084074