Structural Features of the l-Argininamide-Binding DNA Aptamer

Sep 15, 2006 - ... of Harada and Frankel (Harada, K.; Frankel, A. D. EMBO J. 1995, 14, ... Yoshihiko Nagai , Jorge Dulanto Carbajal , John H. White , ...
2 downloads 0 Views 191KB Size
Anal. Chem. 2006, 78, 7259-7266

Structural Features of the L-Argininamide-Binding DNA Aptamer Studied with ESI-FTMS Xinhua Guo,†,‡ Zhiqiang Liu,† Shuying Liu,*,† Catherine M. Bentzley,‡ and Michael F. Bruist*,‡

Green Chemistry and Process Laboratory, Changchun Center of Mass Spectrometry, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, 5625 Renmin Street, Changchun 130022, P. R. China, and Department of Chemistry & Biochemistry, University of the Sciences in Philadelphia, 600 South 43rd Street, Philadelphia, Pennsylvania 19104

The 24-mer DNA aptamer of Harada and Frankel (Harada, K.; Frankel, A. D. EMBO J. 1995, 14, 5798-5811) that binds L-argininamide (L-Arm) was studied by electrospray ionization Fourier transform mass spectrometry (ESI-FTMS). This DNA folds into a stem and loop such that the loop is able to engulf L-Arm. As controls, two derivatives of the same base composition, one with the same stem but a scrambled loop and the other with no ability to form a secondary structure, were studied. The two DNAs that could fold into stem-loop structures showed a more negatively charged distribution of ions than the linear control. This tendency was preserved in the presence of ligand; complexes expected to have more secondary structure had ions with more negative charges. Distinct species corresponding to no, one, and two bound L-Arm molecules were observed for each DNA. The fractional peak intensities were fit to a straightforward binding model and binding constants were obtained. Thus, ESI-FTMS can provide both qualitative and quantitative data regarding the structure of DNA and its interactions with noncovalent ligands. Aptamers are nucleic acids that fold into a specific conformation and recognize ligands. They function as regulators of replication, transcription, and translation; numerous potential applications of these molecules are expected in medicine and technology.1-4 For example, TAR RNA is a stem-loop hairpin aptamer at the 5′-end of HIV-1 mRNA that specifically recognizes an arginine of the HIV Tat protein. The interaction between the TAR RNA and the Tat protein is a critical step to HIV gene expression.5-7 * To whom correspondence should be addressed. E-mail: [email protected]. Tel.: 0086-431-5262613. Fax: 0086-431-5262236. E-mail: [email protected]. Tel.: (215)596-8530. Fax: (215) 596-8543. † Changchun Institute of Applied Chemistry. ‡ University of the Sciences in Philadelphia. (1) Carothers, J. M.; Oestreich, S. C.; Davis, J. H.; Szostak, J. W. J. Am. Chem. Soc. 2004, 126, 5130-5137. (2) Breaker, R. R. Curr. Opin. Chem. Biol. 1997, 1, 26-31. (3) Bock, L. C.; Griffin, L. C.; Latham, J. A.; Vermaas, E. H.; Toole, J. J. Nature 1992, 355, 564-566. (4) Nickens, D. G.; Patterson, J. T.; Burke, D. H. RNA 2003, 9, 1029-1033. (5) Knight, R. D.; Landweber, L. F. Chem. Biol. 1998, 5, R215-220. (6) Calnan, B. J.; Tidor, B.; Biancalana, S.; Hudson, D.; Frankel, A. D. Science 1991, 252, 1167-1171. 10.1021/ac060606r CCC: $33.50 Published on Web 09/15/2006

© 2006 American Chemical Society

Currently, the most common method for studying the details of the structure and binding features of aptamers has been NMR spectroscopy8 and X-ray crystallography.9 Electrospray ionization mass spectrometry (ESI-MS) now provides a fast and powerful technique for the analysis of noncovalent complexes of oligonucleotides.10,11 DNA duplexes,12-14 triplexes,15 and quadruplexes,15-17 and specific DNA-ligand complexes17-20 have been detected by ESI-MS. Recently, Yu and Fabris21 reported how direct mass spectrometric analysis in combination with chemical modification provided insights into the structure of the Ψ-recognition element of HIV-1 RNA. Vairamani and Gross22 detected the formation of a thrombin-binding G-quadruplex aptamer by measuring the formation of metal adducts and H/D exchange using precise ESI-MS determinations. In addition to providing explicit stoichiometries of complexes, ESI-MS offers macromolecular structural information in the envelope of multiply charged ions that it creates. The ESI-MS charged ion envelope or charge-state distribution of proteins has been well studied and is used to determine the distributions of folding states of proteins in solution.23,24 Recently, the Bentzley laboratory has expanded this observation to nucleic acids: folded (7) Zeffman, A.; Hassard, S.; Varani, G.; Lever, A. J. Mol. Biol. 2000, 297, 877893. (8) Patel, D. J. Curr. Opin. Chem. Biol. 1997, 1, 32-46. (9) Convery, M. A.; Rowsell, S.; Stonehouse, M. J.; Ellington, A. D.; Hirao I.; Murray, J. B.; Peabody, D. S.; Phillips, S. E.; Stockley, P. G. Nat. Struct. Biol. 1998, 5, 133-139. (10) Hofstadler, S. A.; Griffey, R. H. Chem. Rev. 2001, 101, 377-390. (11) Hofstadler, S. A.; Sannes-Lowery, K. A.; Hannis, J. C. Mass Spectrom. Rev. 2005, 24, 265-285. (12) Light-Wahl, K. J.; Springer, D. L.; Winger, B. E.; Edmonds, C. G.; Camp II, D. G.; Thrall, B. D.; Smith, R. D. J. Am. Chem. Soc. 1993, 115, 803-804. (13) Doktycz, M. J.; Habibi-Goudarzi, S.; McLuckey, S. A. Anal. Chem. 1994, 66, 3416-3422. (14) Wan, K. X.; Shibue, T.; Gross, M. L. J. Am. Chem. Soc. 2000, 122, 300307 (15) Rosu, F.; Gabelica, V.; Houssier, C.; Colson, P.; Pauw, E. D. Rapid Commun. Mass Spectrom. 2002, 16, 1729-1736. (16) Goodlett, D. R.; Camp, D. G. ??.; Hardin, C. C.; Corregan, M.; Smith, R. D. Biol. Mass Spectrom. 1993, 22, 181-183. (17) Gale, D. C.; Goodlett, D. R.; Light-Wahl, K. J.; Smith, R. D. J. Am. Chem. Soc. 1994, 116, 6027. (18) Sannes-Lowery, K. A.; Hu, P.; Mack, D. P.; Mei, H.-Y.; Loo, J. A. Anal. Chem. 1997, 69, 5130-5135. (19) Rosu, F.; Gabelica, V.; Houssier, C.; De Pauw, E. Nucleic Acids Res. 2002, 30, e82. (20) Hofstadler, S. A.; Griffey, R. H. Chem. Rev. 2001, 101, 377-390. (21) Yu, E.; Fabris, D. J. Mol. Biol. 2003, 330, 211-223. (22) Vairamani, M.; Gross, M. L. J. Am. Chem. Soc. 2003, 125, 42-43.

Analytical Chemistry, Vol. 78, No. 20, October 15, 2006 7259

Table 1. Experimental Oligonucleotide Strands strand Ap1 Ap2 Lin

sequence 5′GATCGAAACGTAGCGCCTTCGATC3′ 5′GATCGAACGTCACCGGATTCGATC3′ 5′GATCGAAACGTAGTCTCCCGCTGA3′

conformation hairpin hairpin linear

DNAs create an ion envelope with species that are more negatively charged than unfolded ones.25 Here, we compare the ions created by ESI-Fourier transform (FT)MS from a DNA aptamer that binds L-arginine with two derivatives each with reduced binding and one lacking the ability to fold. The structural differences measured in the gas phase are supported by binding experiments and UV spectroscopy in solution. The goal is to enhance the application of the charge-state distributions of negative oligonucleotide ions to structural studies. We chose to investigate the DNA aptamer (PDB ID 1OLD) that binds L-argininamide (L-Arm). This aptamer has been well studied by NMR and circular dichroism (CD).26-28 It comprises a 7-base-pair stem extending from a 10-residue loop. Chemical modification and binding interference experiments show that the aptamer binds both specifically and nonspecifically to L-Arm. The sequences in the loop and the stem base pairs at the loop are important in creating the specific L-Arm binding site: upon the binding of L-Arm, the aptamer changes conformation to form a binding pocket in the loop that envelops one L-Arm molecule.26,28 The nonspecific binding arises from the general electrostatic interactions between the polyanionic DNA and L-Arm, which is a dication at neutral pH. Takeda et al.29 determined that DNA nonspecifically binds one putrescine (1,4-diaminobutane) per seven nucleotides with a dissociation constant of ∼1 mM. Here we demonstrate how the ability of ESI to show the charge envelopes for different stoichiometries of aptamer and ligand aids the analysis to these two forms of binding. We choose this aptamer as a model for four reasons: (1) The aptamer is sufficiently short (24-mer) to be analyze by ESI-MS (many instruments have an upper limit of 2000 m/z). (2) With a 42 °C melting temperature, the hairpin folds at room temperature and should remain folded in the gas phase.25 (3) The 100 µM dissociation constant26 of the aptamer-L-Arm complex is sufficiently tight for MS detection. (4) The interaction between the DNA aptamer and the L-Arm may reflect a common recognition between DNA and arginine in other proteins. In the present work, three DNA strands (Ap1, Ap2, Lin) of the same composition but different sequences were studied (Table 1). Ap1 is the original L-Arm binding aptamer with a stem-loop hairpin conformation.26 Ap2 is a hairpin with a same stem as the Ap1, but the sequence in the loop region is scrambled. Harada (23) Chowdhury, S. K.; Katta, V.; Chait, B. T. J. Am. Chem. Soc. 1990, 112, 9012-9013. (24) Lee, V. W. S.; Chen, Y.-L.; Konermann, L. Anal. Chem. 1999, 71, 41544159. (25) Guo, X.; Bruist, M. F.; Davis, D. L.; Bentzley, C. M. Nucleic Acids Res. 2005, 33, 3659-3666. (26) Harada, K.; Frankel, A. D. EMBO J. 1995, 14, 5798-5811 (27) Lin, C. H.; Patel, D. J. Nat. Struct. Mol. Biol. 1996, 3, 1046-1050. (28) Robertson, S. A.; Harada, K.; Frankel, A. D.; Wemmer, D. E. Biochemistry 2000, 39, 946-954. (29) Takeda, Y.; Nara, H.; Iwahashi, K.; Mitsui, Y.; IItaka, Y. J. Biochem. 1983, 94, 275-282.

7260

Analytical Chemistry, Vol. 78, No. 20, October 15, 2006

and Frankel26 showed that such a change disrupts the specific L-Arm binding site. Strand Lin lacks self-complementarity and should remain linear and have no specific affinity for L-Arm. Using the above strands, we investigated their structure and structurebased binding of L-Arm. EXPERIMENTAL SECTION Materials. Table 1 presents the sequences of DNA oligonucleotides used in this work. Strand Lin was designed and analyzed for minimal self-interactions using the SEQUIN30 and mfold31 (http://www.bioinfo.rpi.edu/∼zukerm/rna/) programs. Synthetic oligonucleotide strands from TaKaRa (TaKaRa Biotechnology Co., Ltd., Dalian, China) and Midland (Midland Certified Reagent Co., Midland, TX) were used, respectively, for ESI-FTMS and UV spectroscopy experiments. DNAs were prepared using a SpinDialyzer Turbo dialyzer with regenerated cellulose membranes of molecular weight cutoff 1000 (Harvard Bioscience, Holliston, MA) in 20 mM ammonium acetate. Concentrations of the purified strands were calculated based on their UV absorbance at 260-nm wavelength, using the extinction coefficient calculated base on the nearest-neighbor method (http://scitools.idtdna.com/scitools/ Application/OligoAnalyzer). Aqueous stocks of L-argininamide dihydrochloride (Sigma Chemical Co., St. Louis, MO) were prepared at 500 µM and adjusted to pH 7.0 with ammonia. UV Spectral Analysis. For melting analysis by UV absorbance, an Agilent 8453 UV-visible spectrophotometer with a Peltier thermostated cell holder (Palo Alto, CA) was used in the thermal denaturation mode. Sample stock solutions were diluted to ∼2 µM in aqueous 2 mM ammonium acetate. The absorbance of each sample was measured at 260 nm, while the temperature of the solution changed from 20 to 70 °C at 1 °C increments, with holding times of 1 min. The melting temperatures were determined from an absorbance versus temperature derivative curve. Mass Spectrometry. For ESI determination of DNA chargestate distribution, 30 µM DNA was prepared in 20/80 (v/v) methanol/20 mM ammonium solution. For ligand binding studies, 50 µM DNA in 20/80 methanol/30 mM ammonium acetate containing 50-200 µM L-Arm was used. The inclusion of 20% methanol stabilized the spray, and it was added just before injection. Mass spectrometric analysis was performed on an IonSpec HiRes Fourier transform ion cyclotron resonance mass spectrometer (FT-ICR-MS) (Lake Forest, CA) in the negative ion detection mode equipped with a 7-T shielded superconducting magnet and a Micromass Z-spray electrospray source. A DNA duplex formed from complementary 10-mer strands was employed to tune instrumental parameters to produce the most abundant duplex ions. Samples were directly infused into the source region at a rate of 5 µL/min. The spray voltage was 2300 V. The source and the probe temperature were 80 and 100 °C, respectively. The sample cone voltage was -25 V. Ions were accumulated in the external hexapole for 1500 ms and then were collected with 512K data points at an ADC rate of 1 MHz. All spectra were single acquisition scanned from 400 to 2500 m/z. All experiments were performed in duplicate. Relative peak intensities were determined by integration with InoSpec99 soft(30) Seeman, N. C. J. Biomol. Struct. Dyn. 1990, 8, 573-581. (31) Zuker, M. Nucleic Acids Res. 2003, 31, 3406-34015.

Table 2. Melting Temperature of the Ap1, Ap2, and Lin before/after Being Mixed with L-Arm

Tm (°C)

Ap1

Ap2

Lin

Ap1-L-Arm

Ap2-L-Arm

Lin-L-Arm

42

37

none

57

53

none

ware. For binding quantification, the relative intensities of ions differing by an adduct of ammonia, sodium, or potassium were summed together. Gabelica et al.32 demonstrated that the ESI response factor of DNA duplexes is changed little by the addition of minor complexing or binding agents, since the mass and conformation are altered only to a minor extent. Thus, we have assumed that members of adduct groups of ions with same charge, such as DNA, DNA-L-Arm and DNA-(L-Arm)3, all have the same response factors. THEORETICAL BASIS Binding Model. Analysis of the binding of L-Arm to the aptamer assumed two independent binding sites, one tighter than the other. There are four species possible: empty DNA (D); DNA with L-Arm in the tight site (AD); DNA with L-Arm in the weak site (DA); DNA with both tight and weak sites occupied (ADA). In mass spectrometry, AD and DA cannot be distinguished, and one should not assume that DA can be ignored. Hence, we define the mass spectroscopy observable L1 as

the fraction of the ions represented by the empty DNA, singly occupied DNA, and double-occupied DNA. These are related to the dissociation constants and [A] by:

νjL0 )

D ) Dtot

[D] [D][A] [D][A] [D][A]2 [D] + + + K1 K2 K1K2

)

K1K2 K1K2 + [A]K1 + [A]K2 + [A]2 νjL1 )

νjL2 )

[A]K1 + [A]K2 K1K2 + [A]K1 + [A]K2 + [A]2 [A]2 K1K2 + [A]K1 + [A]K2 + [A]2

(1)

(2)

(3)

For aptamers Ap2 and Lin, the two sites were modeled as equivalent, and K1 ) K2. Note that the mass spectroscopy data cannot distinguish the case in which the two sites are not independent from the independent case described here. If the binding affinity for one site is altered by the occupancy of the other, the following more complex model can be described.

L1 ) [AD] + [DA] The concentrations of a particular DNA complex M was determined by multiplying the total DNA concentration Ctotal DNA by the fractional intensity of the peaks for M. This assumes that all species observed have the same response factor, an assumption that is addressed later.

This model has three independent variables since K1′K2 ) K2′K1. However, since DA and AD cannot be distinguished, only

IM [M] ) Ctotal DNA IfreeDNA + IDNA•L-Arm + IDNA•(L-Arm)2 Since L-Arm ions could not be detected, the free L-Arm concentration was determined by subtracting the concentration of bound L-Arm from the total L-Arm concentration.

[A] ) [L-Arm]free ) Ctotal L-Arm (CDNA•L-Arm + 2CDNA•(L-Arm)2 + 3CDNA•(L-Arm)3) The two independent dissociation constants are defined as follows: K1

AD [\] D + A

K2

DA [\] D + A

K1

ADA [\] DA + A K1)

[D][A] [DA][A] ) [AD] [ADA]

K2 )

[D][A] [AD][A] ) [DA] [ADA]

K2

ADA [\] AD + A

From the mass spectrum we get three fractions, νjL0, νjL1, and νjL2,

can be modeled, which has only two independent variables. Without the ability to distinguish AD and DA, we cannot determine whether the sites are independent. RESULTS AND DISCUSSION Structural and Binding Features by UV Spectroscopy. This work studies the folding properties of three DNAs, the L-Armbinding aptamer Ap1, its scrambled-loop variant Ap2, and the nonhairpin of the same base composition, Lin. We confirmed the folding behavior of these molecules by UV spectroscopy. The melting temperature of each DNA was determined in 2 mM ammonium acetate by the increase in absorbance at 260 nm associated with the disruption of double helices. Each DNA strand (2 µM) was observed in the absence and presence of 200 µM L-Arm. The results are given in Table 2. As expected only Ap1 and Ap2 show melting transitions, supporting the assumption that the Lin DNA does not fold into a specific stable structure. The melting temperature of Ap1 is 5 °C higher than Ap2, indicating a contribution of the L-Arm-specificloop bases to the folded structure of Ap1 that is not present in (32) Gabelica, V.; et al. J. Mass Spectrom. 2003, 38, 491-501.)

Analytical Chemistry, Vol. 78, No. 20, October 15, 2006

7261

Figure 1. Mass spectra of three strands of DNA. (A) Ap1, an L-Arm-binding DNA aptamer, (B) Ap2, the loop of Ap1 has been scrambled in order to disrupt L-Arm binding, and (C) DNA Lin has the same base composition as Ap1 and Ap2, but the sequence no longer supports secondary structure formation. All spectra were acquired in 20/80 methanol/20 mM ammonium acetate.

the loop of Ap2. L-Arm increases the melting temperature of both Ap1 and Ap2 by ∼15 °C but does not induce any meltable structure 7262 Analytical Chemistry, Vol. 78, No. 20, October 15, 2006

in strand Lin. The increased melting temperature is most likely due to nonspecific stabilization of double-helical DNA by the L-Arm

Figure 2. Ratio R for peaks (areas of -6 peaks to -5 peaks) versus the concentrations of ammonium acetate. Ap1 black squares; Ap2 gray circles; Lin open diamonds. Each point represents the average of six spectra. Error bars indicate the standard deviation.

dication. The spacing of the two positive charges of L-Arm is similar to that in putrescine. The addition of putrescine to nonspecific Escherichia coli DNA in 15 mM NaCl increases the UV melting transition by 10 °C.29 The magnitude of this shift becomes even greater at lower ionic strengths.33 Structure Characterization by Charge-State Distributions. Our previous work demonstrated that, on an ESI single quadrupole mass spectrometer, DNA strands that can fold into a hairpin produce ions with an average greater negative charge than strands of the same base composition that cannot fold into a hairpin.25 DNA strands related to the L-Arm aptamer show similar behavior with ESI-FTMS at 10-40 mM ammonium acetate. Figure 1 shows mass spectra of 30 µM Ap1 (A), Ap2 (B), and Lin (C) in 20 mM ammonium acetate. All give rise to major peaks for ions with charges of -6 and -5 and a minor peak for -7. The -6 peak is strongest for Ap1, while the -5 peak is strongest for Lin; the Ap2 spectrum is intermediate, with a slightly higher intensity for the -5 peak. The -5 and -6 peaks dominated all spectra for all three strands from 5 to 30 mM ammonium acetate. Signals decrease as the concentration of ammonium acetate increases; strongly suppressed signals are observed at 40 mM ammonium acetate. The relative intensity ratio R of peak -6 to peak -5 was computed in order to quantitatively describe the charge-state distributions of the three strands. R for each DNA strand is plotted against ammonium acetate concentration in Figure 2. R values for the three strands show a consistent order, Ap1 > Ap2 > Lin, for ammonium acetate from 10 to 30 mM. This order follows the extent of interactions of residues within each strand. The Lin strand is designed to have minimal intrastrand interactions and shows the lowest charge-state distribution. In contrast, Ap1 and Ap2 are designed to form hairpins and produces ions with higher negative charges. We observe that Ap1 forms negative ions with charge values higher than the Ap2, which suggests that the native Ap1 aptamer has additional interactions of residues within the loop, which do not occur in the scrambled Ap2 loop. This agrees with the higher melting temperature observed for Ap1 by UV spectroscopy. (33) Hora´cˇek, P.; Cernohorsky´, I. J. Biochem. Biophys. Res. Commun. 1968, 32, 956-962.

We do not know the basis for the correlation of higher charge with more secondary structure; its cause might be thermodynamic or kinetic. There is extensive evidence that nucleic acid helices are preserved in the gas phase.34,35 Such helices give good separation of the phosphates that carry the negative charges. Molecular modeling and collision cross-section measurements indicate that nonhelical DNA takes on a collapsed globular form.34-36 It is possible that the backbones in these structures do not give such good separation of the phosphates and these structures would prefer lower charged states to reduce phosphatephosphate repulsions. Another consideration as to why helical DNA prefers the more negatively charged arises from the necessity to include ammonia or other nitrogenous bases to reduce the charge and number of species in the envelope.25,37 Ammonium ions most likely donate the protons that neutralize the phosphates during gas-phase ion formation; they have a lower gas-phase proton affinity than the nucleobases of mononucleotides.38 The propensity of ammonium ions to donate protons to the backbone is limited not only the by kinetics of the transfer reaction but also by the density of ammonium sites on the backbone. Molecular dynamics simulations by Korolev et al.39 show that a bound cation on the DNA backbone precludes the binding of another cation within ∼7.5 Å, even at 2 M counterion concentrations. Thus, helical and collapsed nucleic acids may have different affinities for ammonium ions, giving different rates of protonation. Binding Properties. Noncovalent complexes of the DNA aptamers and L-Arm survive in the gas phase and are detected by ESI-FTMS. This binding reaction was studied by quantitating the observed peaks for 50 µM DNA with 50, 100, 150, and 200 µM L-Arm. Ion envelopes corresponding to empty DNAs, 1:1 DNAL-Arm complexes, and at high L-Arm concentrations, DNA-(LArm)2 complexes were detected (Figure 3). Ions of L-Arm should be positive and therefore were not detected. Ap1 gave the highest occurrence of DNA-L-Arm complexes, but significant binding was observed for Ap2 and Lin. At the highest L-Arm concentration, all three DNAs complexed two L-Arm molecules. At this highest L-Arm concentration, a slight amount of an Ap1-(L-Arm)3 was seen (Figure 3A). This variety in stoichiometry indicates that both specific and nonspecific binding is preserved in the gas phase. The ratio R followed the trend described above and is displayed in Table 3. For uncomplexed DNAs, Ap1 and Ap2 have much higher R values than Lin. As L-Arm concentrations increased, this trend is maintained as R values decrease. This decrease may reflect an ability of L-Arm to promote proton transfers in the gas phase and emphasizes that ion distributions may not be at equilibrium. Ion distributions will still reflect the structures present in solution at the time of ionization. The Ap1-L-Arm complex, which should be dominated by specific binding, had R values (34) Gidden, J.; Ferzoco, A.; Baker, E. S.; Bowers, M. T. J. Am. Chem. Soc. 2004, 126, 15132-15140. (35) Gidden, J.; Baker, E. S.; Ferzoco, A.; and Bowers, M. T. Int. J. Mass Spectrom. 2005, 240, 183-193. (36) Schnier, P. D.; Klassen, J. S.; Strittmatter, E. F.; Williams, E. R. J. Am. Chem. Soc. 1998, 120, 9605-9613. (37) Muddiman, D. C.; Cheng, X.; Udseth, H. R.; Smith, R. D. J. Am. Soc. Mass Spectrom. 1996, 7, 697-706. (38) Green-Church, K. B.; Limbach, P. A. J. Am. Soc. Mass Spectrom. 2000, 11, 24-32.32. (39) Korolev, N.; Lyubartsev, A. P.; Laaksonen, A.; Nordenskiold, L. Nucleic Acids Res. 2003, 31, 5971-81.

Analytical Chemistry, Vol. 78, No. 20, October 15, 2006

7263

Figure 3. Mass spectra of the three DNAs in the presence of L-Arm. Each spectrum contains 50 µM DNA in 20/80 methanol/30 mM ammonium acetate containing 200 µM L-Arm. (A) Ap1, (B) Ap2, and (C) Lin. The peak marked with / is an impurity. 7264 Analytical Chemistry, Vol. 78, No. 20, October 15, 2006

Table 3. R Valuesafor DNAs Complexed with Different Stoichiometries of L-Armb DNA Ap1

Ap2

Lin

DNA-L-Arm [L-Arm] (µM)

1:0

1:1

1:0

1:1

1:0

1:1

0 50 100 150 200

1.11 ( 0.07c 0.39 ( 0.03 0.35 ( 0.04 0.22 ( 0.02 0.15 ( 0.02

0.51 ( 0.05 0.47 ( 0.06 0.27 ( 0.02 0.18 ( 0.03

0.86 ( 0.06 0.34 ( 0.04 0.26 ( 0.02 0.16 ( 0.01 0.13 ( 0.03