Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (179K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (19)
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Rai, P.
Right arrow Articles by Linn, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Rai, P.
Right arrow Articles by Linn, S.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Nucleic Acids Research, 2003, Vol. 31, No. 9 2323-2332
© 2003 Oxford University Press

Steady-state and time-resolved fluorescence studies indicate an unusual conformation of 2-aminopurine within ATAT and TATA duplex DNA sequences

Priyamvada Rai, Timothy David Cole, Elizabeth Thompson1, David P. Millar1 and Stuart Linn2

Biophysics Graduate Group, University of California, Berkeley, CA 94720, USA, 1 Department of Molecular Biology, Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037, USA and 2 Division of Biochemistry and Molecular Biology, Barker Hall, University of California, Berkeley, CA 94720-3202, USA

Timothy David Cole, Telik Inc., 750 Gateway Boulevard, San Francisco, CA 94080, USA

Received January 27, 2003; Revised February 24, 2003; Accepted March 8, 2003


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
2-Aminopurine (2-AP), a fluorescent analog of adenine, has been widely used as a probe for local DNA conformation, since excitation and emission characteristics and fluoresence lifetimes of 2-AP vary in a sequence-dependent manner within DNA. Using steady-state and time-resolved fluorescence techniques, we report that 2-AP appears to be unusually stacked in the internal positions of ATAT and TATA in duplex DNA. The excitation wavelength maxima for 2-AP within these contexts were red shifted, indicating reduced solvent exposure for the fluorophore. Furthermore, in these contexts, 2-AP fluorescence was resistant to acrylamide-dependent collisional quenching, suggesting that the fluorophore is protected by its stacked position within the duplex. This conclusion was further reinforced by the presence of a secondary peak at 275 nm in the fluorescence excitation spectra that is indicative of efficient excitation energy transfer from nearby non-fluorescent DNA bases. Fluorescence anisotropy decay and internal angular ‘wobbling’ motion measurements of 2-AP within these alternating AT contexts were also consistent with the fluorophore being highly constrained and immobile within the base stack. When these fluorescence characteristics are compared with those of 2-AP within other duplex DNA sequence contexts, they are unique.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Sequence-dependent local structure variation in DNA plays an important role in their recognition and binding by cognate proteins and synthetic ligands. Although subtle differences in local DNA structure such as altered base stacking or conformational flexibility may be extremely important, the nature of these differences is often not amenable to characterization by traditional structural techniques such as NMR or X-ray crystallographic analysis.

The fluorescent adenine analog, 2-aminopurine (2-AP), has frequently been used to study such variations in DNA structure because its emission properties are highly sensitive to local conformational changes and because it can substitute for adenine by Watson–Crick base pairing with thymine, minimally perturbing the helix conformation (Fig. 1) (1,2). Steady-state fluorescence spectroscopy can be used to compare differences in the local environment of 2-AP through observations of conformation-dependent internal quenching and solute-mediated collisional quenching, shifts in excitation wavelength maxima and excitation energy transfer (3–7). Differences in these spectral parameters can be correlated both with the strength of stacking interactions of 2-AP with its neighboring bases and with the extent of hydrogen bonding to thymine (8–10).



View larger version (15K):
[in this window]
[in a new window]
  		Figure 1. Schematic of Watson–Crick adenine-thymine (A-T) and 2-aminopurine-thymine (2-AP-T) base pairs.

 
While equilibrium fluorescence spectroscopy reflects an average of conformations of the fluorophore, it cannot be used to observe individual short-lived conformations. However, time-resolved fluorescence anisotropy decay can provide direct information about the internal rotational motions of 2-AP that occur in solution on the picosecond time scale (11). The dynamic fluctuations in base motion detected by this technique include those occurring through helical twisting, propeller twisting, base tilting and base rolling. The extent of these motions is affected by the local environment around the fluorescent base, such as strength of base pairing or stacking interactions with neighboring bases. The anisotropy decay behavior can be used to analyze the restricted angular range of 2-AP motion or ‘wobbling’ within the duplex (11,12).

In this report, we use both steady-state and time-resolved fluorescence spectroscopy to study the variation in 2-AP characteristics at various positions within two related 16 bp duplexes. We have found that 2-AP in the internal positions of ATAT and TATA is extremely well stacked and conformationally constrained relative to 2-AP in other contexts. Consequently, it is an unusually effective excitation energy acceptor from nearby adenines located on the opposite strand.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Sample preparation
The DNA oligonucleotides used in this study are described in Table 1. When 2-AP is substituted for adenine at position n, the single-stranded oligonucleotide is designated ss.Pn. When single-stranded DNA is annealed with its complementary strand, the duplex DNA is designated ds.Pn. All abbreviations are defined in Table 1.


View this table:
[in this window]
[in a new window]
  		Table 1. Sequences of oligonucleotides used in the study
 
DNA oligonucleotides were obtained from Operon Technologies (Alameda, CA) and were purified by reverse-phase HPLC through a 10 mm x 25 cm C18 column (Econosphere; Alltech, Deerfield, IL), using a 5–15% acetonitrile (HPLC grade) linear gradient in 50 mM TEAA buffer, pH 7.3. Fractions containing the desired oligonucleotides were combined, dried in vacuo in 14 M NH4OH to remove triethylamine, suspended in 1 M NaCl and purified on a Waters C18 Plus Sep-pak cartridge by washing with 1 M NaCl (99.999% pure; Sigma-Aldrich) followed by ddH2O and eluting with 50% HPLC-grade methanol. The oligonucleotides were then dried in vacuo and redissolved in ddH2O. Concentrations were determined from A260 measurements and complementary strands were annealed at a 1:1 molar ratio. All samples were in unbuffered 130 mM NaCl (99.999% pure; Sigma-Aldrich) at pH 6.60–6.75, determined with a Mettler-Toledo pH electrode. Sample concentrations ranged from 2.40 to 5.52 µM (duplex or single-stranded oligonucleotides).

Steady-state fluorescence spectroscopy
A Hitachi model F-2000 spectrofluorometer was used for all steady-state fluorescence experiments. Standard fluorescence cuvettes (12 x 12 x 45 mm, no. 14-385-918A; Fisher Scientific) were used for 1 ml samples. Starna microcuvettes were used for 0.5 ml samples. The sample temperature was maintained at 30 ± 0.1°C with a circulating IsoTemp water bath. Samples were allowed to equilibrate in temperature within the fluorometer for a minimum of 5 min before acquisition of spectra. Spectra were acquired at a scan speed of 60 nm min–1, with a 2 s response time, a bandpass of 10 nm for both excitation and emission spectra and a voltage of 400 V applied to the photomultiplier. The instrument sensitivity was checked using the signal-to-noise ratio of a Raman spectrum of distilled water and maximized through manual lamp adjustment. Excitation spectra were recorded from 200 to 350 nm while monitoring emission at 370 nm. Emission spectra were obtained while scanning from 320 to 500 nm with an excitation wavelength of 303 nm. Fluorescence wavelength maxima and corresponding intensity values were determined automatically by the fluorometer software after each spectral acquisition.

Calculation of energy transfer efficiency
To observe energy transfer from normal DNA bases to 2-AP, oligonucleotides were excited at 275 nm and fluorescence of 2-AP was monitored at 370 nm. There is little direct excitation of 2-AP at this wavelength. Following Xu and Nordlund (6), the energy transfer efficiency was calculated according to equation 1.

{eta}t({lambda}ex) = [Aa({lambda}ex)/Ad({lambda}ex)]{[F({lambda}ex)/Fa({lambda}ex)] – 1}1

where Ad({lambda}ex)and Aa({lambda}ex) are the absorbances of the donor (normal DNA bases) and acceptor (2-AP), respectively, at the excitation wavelength ({lambda}ex), F({lambda}ex) is the measured fluorescence intensity of 2-AP in the oligonucleotide and Fa({lambda}ex) is the fluorescence intensity arising from direct excitation of 2-AP. To obtain Fa({lambda}ex), the fluorescence intensity of the free nucleotide was measured under the same experimental conditions and then multiplied by the ratio {phi}(oligo)/{phi}(nucleotide), where {phi} denotes the fluorescence quantum yield of 2-AP in each sample. This correction is necessary to account for the difference in the intrinsic quantum yields of the two samples. The absorbance of 2-AP at the excitation wavelength was also obtained from measurements of the free nucleotide.

The energy transfer efficiency defined in equation 1 is the average transfer efficiency per normal DNA base. The average transfer efficiency for ds.P7, ds.P26 and ds.P27 duplexes was decomposed into contributions from individual bases, as in equation 2.

{eta}t({lambda}ex) = [{epsilon}T{eta}T(–1) + {epsilon}T{eta}T(+1) + 2{epsilon}A{eta}A(cross)]/{sum}X{epsilon}X2

where {epsilon}T is the extinction coefficient of thymine, {eta}T(–1) is the transfer efficiency from the thymine on the 5' side of 2-AP, {eta}T(+1) is the transfer efficiency from the 3' thymine, {epsilon}A is the extinction coefficient of adenine and {eta}A(cross) denotes the transfer efficiency from each of the adjacent adenines on the opposite strand to 2-AP. All extinction coefficients are referenced to the excitation wavelength, {lambda}ex. The denominator in equation 2 represents the total extinction coefficient of the duplex (at {lambda}ex), obtained by summation over all normal bases in each strand. Equation 2 assumes that donors located two or more bases from 2-AP in either strand of the duplex make a negligible contribution to the energy transfer, which has been verified experimentally (6). The value of {eta}T(–1) was taken from Xu and Nordlund (6) and was assumed to be equal to {eta}T(+1). When the experimentally determined value of {eta}t is substituted on the left-hand side, equation 2 can be solved for {eta}A(cross).

Acrylamide titrations
Acrylamide titrations were done at 30.0 ± 0.1°C, by addition of a 10 M acrylamide (molecular biology grade; Aldrich) stock solution in ddH2O, to give a final concentration of 50 mM per increment, up to a total concentration of 300 mM. Samples were allowed to equilibrate with acrylamide for a minimum of 10 min before the acquisition of spectra.

Stern–Volmer plots were obtained from the intensity data by plotting F0/F as a function of acrylamide concentration, where F0 is the maximum emission intensity in the absence of acrylamide and F is the corresponding intensity at a particular acrylamide concentration. At the higher acrylamide concentrations at which the intensity was below the automatic detection threshold of the instrument, intensities were manually determined by the cursor function. The resulting data were fitted to the Stern–Volmer equation for analysis of collisional quenching:

F0/F = 1 + kq{tau}0[Q]3

where [Q] is the acrylamide concentration, kq is the bimolecular quenching rate constant and {tau}0 is the fluorescence lifetime in the absence of quencher. When more than one exponential term was required to fit the fluorescence intensity decay (2-AP in oligonucleotides), {tau}0 was replaced by the intensity-averaged fluorescence lifetime, {tau}int, defined in equation 5 (below). Quenching data were fitted to equation 3 using Kaleidagraph 3.0 software.

Time-resolved fluorescence spectroscopy
Time-resolved fluorescence measurements were performed using the time-correlated single photon counting method. Samples were placed in a temperature-controlled housing at 30°C and were repetitively excited by short pulses (<5 ps width, 1.87 MHz repetition rate) of vertically polarized laser light. The laser pulses were generated by a synchronously mode-locked DCM dye laser (Coherent 702) operating at 636 nm. The light from the dye laser was passed through a frequency doubler to generate 318 nm pulses for excitation. The resulting emission was collected at 90° to the excitation beam, collimated by a lens and then passed through a polarizer mounted on a motor driven rotary stage under computer control. Fluorescence emission was monitored at 370 nm using a monochromator with a spectral bandpass of 16 nm and photons were detected by a microchannel plate photomultiplier (Hamamatsu R2809U-01). A polarization scrambler placed after the emission polarizer was used to eliminate the polarization bias of the monochromator. The photomultiplier output was processed with standard time-correlated single photon counting electronics. Fluorescence decay curves were collected in 4096 channels with a time increment of 10 ps per channel. For fluorescence lifetime measurements, the emission polarizer was oriented at the magic angle (54.7° from vertical). For time-resolved fluorescence anisotropy measurements, two separate decays were recorded: one with the emission polarizer parallel to the excitation polarization and the other with a perpendicular orientation. The polarizer was switched between these two directions every 30 s and the two decays were accumulated in separate portions of the multichannel analyzer memory. Data collection was terminated when 40 000 counts were in the peak channel of the magic angle or vertical polarization decay curves. The instrument response function was collected at 318 nm using light scattered from a dilute suspension of non-dairy coffee creamer.

Time-resolved fluorescence data were analyzed by a standard reconvolution procedure (13) using non-linear least squares regression (14). The fluorescence intensity decay measured at the magic angle was fitted to a sum of exponentials,

where {alpha}i are the amplitudes of each component and {tau}i are the corresponding fluorescence lifetimes. Decay curves were represented with the minimum number of components (N) required for best fit, as judged by the reduced {chi}2 criterion.

For samples exhibiting multi-exponential fluorescence decay behavior, the intensity-averaged fluorescence lifetime, {tau}int, was calculated according to:

The lifetime parameters recovered from the fits of the magic angle decay were applied to the anisotropy decay data collected for the same samples. Polarized intensity decays were fitted using the following expressions:

I||(t) = 1/3[1 + 2r(t)]K(t)

I{perp}(t) = 1/3[1 – r(t)]K(t)6

where I||(t) and I{perp}(t) are the intensity decays measured with the emission polarizer parallel or perpendicular to the excitation polarization, respectively, and r(t) is the time-dependent fluorescence anisotropy. The latter was represented as a sum of exponential decays:

where ßk is the limiting anisotropy of component k, {phi}k is the corresponding rotational correlation time and M is the number of components. I||(t) and I{perp}(t) were simultaneously fitted by adjusting the values of ßk and {phi}k in equation 7 for best fit while keeping the parameters in K(t) fixed at the values recovered from the analysis of the corresponding magic angle decay. The minimum number of anisotropy decay components required for best fit was determined using the reduced {chi}2 criterion. The limiting anisotropy at time zero, r0, was obtained by summing the individual ßi values.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Equilibrium fluorescence characteristics of 2-AP in ATAT- and TATA-containing duplex DNA sequences
In order to assess sequence-dependent differences in steady-state fluorescence characteristics of 2-AP, excitation and emission properties of single substitutions of 2-AP for adenine in a series of related 16 bp duplexes (Table 1) at 5 µM were compared (Table 2). The position of 2-AP within single-stranded oligonucleotides did not affect excitation or emission wavelength maxima. The excitation maxima in these single-stranded oligonucleotides had a uniform range of 299–301 nm, while the emission maximum ranged between 370 and 373 nm, regardless of the 2-AP sequence context.


View this table:
[in this window]
[in a new window]
  		Table 2. Steady-state fluorescence parameters of oligonucleotides used in the study
 
The fluorescence quantum yields reflect the different environments of 2-AP in free base, single-stranded or double-stranded forms. Free 2-AP base experiences no quenching in the excited state from base stacking interactions and has the highest emission intensity of any of the species examined here (defined in Table 2 as a relative quantum yield of 1.0). The quantum yield of 2-AP is reduced by more than an order of magnitude in single-stranded oligonucleotides (Table 2), presumably because of base–base stacking interactions that quench the fluorescence (4). DNA-induced quenching of 2-AP is even more pronounced in the duplex oligonucleotides, resulting in a further 10-fold reduction in the fluorescence quantum yield (Table 2).

In contrast to 2-AP within the single-stranded oligonucleotides, the excitation wavelength maxima for 2-AP among the duplexed oligonucleotides showed considerable variation (Table 2). For a majority of the sequence positions, the wavelength maxima were between 301 and 304 nm, as observed previously for 2-AP-substituted DNA duplexes of a similar size (9). However, direct excitation of 2-AP at positions P7 and P27 in the ATAT-containing parent duplex and at position P26 in the TATA-containing parent duplex showed a pronounced red shift with wavelength maxima between 310 and 313 nm. Each of these positions corresponds to an internal adenine in an alternating AT context. A red-shifted excitation peak is associated with decreased exposure of 2-AP to aqueous solution (6).

Another unique spectral characteristic of 2-AP in ds.P7 and ds.P27 of the ATAT-containing duplex and ds.P26 of the TATA-containing duplex was the presence of a secondary excitation band at ~275 nm (Table 2 and Fig. 2A–C). Since 2-AP is only weakly excited at this wavelength, this band reflects excitation energy transfer to 2-AP from nearby non-fluorescent DNA bases. The most efficient energy donor for such a transfer has been shown to be adenine (6) and in each of the duplexes exhibiting an energy transfer peak, the 2-AP has two adjacent adenines on the opposite strand (Table 1). However, although Xu and Nordlund (6) found that measurable inter-strand energy transfer could be detected between adenine and 2-AP that are on opposite strands, they detected such transfer at very low efficiency (~7%). In this study, energy transfer from each of the adenines to 2-AP on the opposite strand was highly efficient (Table 3).



View larger version (19K):
[in this window]
[in a new window]
  		Figure 2. Representative fluorescence excitation spectra. Excitation profiles were taken at 30°C and automatically printed by the Hitachi-F2000 fluorometer. They were then scanned and traced over using the ‘autotrace’ function of Canvas 7.0 to improve the print quality of the plotter-generated spectra. The wavelength (in nm) is on the x-axis and excitation intensity (in absorbance units) is on the y-axis and is denoted by I. Excitation energy transfer peaks are indicated with arrows. Spectra are shown for: (A) ds.P7; (B) ds.P27; (C) ds.P26; (D) ds.P12; (E) ss.P8; (F) ds.P7_eh.

 

View this table:
[in this window]
[in a new window]
  		Table 3. Energy transfer efficiencies from normal DNA bases to 2-AP in duplex oligonucleotides
 
The energy transfer peaks evident in the uncorrected excitation spectra shown in Figure 2A–C actually underestimate the true extent of energy transfer because of the significant background absorbance of DNA at 275 nm, which attenuates the intensity of the exciting light reaching the sample. The relative intensities of the energy transfer bands listed in Table 2 have been corrected for this inner filter effect. Excitation energy transfer is thought to be facilitated by stacking of 2-AP with neighboring bases as well as by duplex stability around 2-AP (6). Consistent with these ideas, the energy transfer band was not seen for ss.P7, ss.P27 or ss.P26, nor was it seen for 2-AP in any of the other duplex contexts (representative data are shown for ds.P12 and ss.P8 in Fig. 2D and E, respectively). It should be noted that the peak at ~320 nm seen for ds.P12 (Fig. 2D) is not an energy transfer peak, but a Raman scattering peak from solvent water that is often seen in excitation spectra when the intensity of the primary excitation maximum is low.

As noted above, the pronounced energy transfer peaks indicate a significant degree of electronic coupling between 2-AP and nearby adenine donors, which are located on the opposite strands in ds.P7, ds.P27 and ds.P26. Moreover, the red-shifted excitation wavelength maximum reflects significant shielding of 2-AP from aqueous solvent within these particular sequence contexts. These observations are further supported by comparison of the fluorescence characteristics of 2-AP in the duplexes ds.P7 and ds.P7_eh, an oligonucleotide in which A25 has been deleted so that T8, which flanks P7, is now unpaired (Table 2, and Fig. 2A versus F). NMR studies have shown that in DNA duplexes between 25 and 30°C, unpaired thymines are predominantly extrahelical when flanked by adenines (15), but mainly intrahelical when flanked by guanines (16,17). In a mixed context, such as that in ds.P7_eh, in which the unpaired thymine is flanked by 2-AP and G, the unpaired T is likely to have significant populations in both intrahelical and extrahelical states, resulting, on average, in reduced stacking of the adjacent 2-AP. Indeed there is a blue-shifted excitation wavelength maximum (304 nm) relative to 2-AP in ds.P7 (310 nm) (Table 2). Moreover, there remains a slight energy transfer peak, presumably from A27, in the excitation spectrum of ds.P7_eh (Fig. 2B), as expected if the 2-AP is in a relatively stacked environment for some fraction of the time.

In summation, the red-shifted excitation spectra and the presence of strong energy transfer excitation peaks suggest that 2-AP in positions P27 and P7 in the ATAT-containing duplex and P26 in the TATA-containing duplex are particularly well stacked and constrained within the duplexes compared to 2-AP in their single-stranded counterparts or in GPC, GPT, CPT, CPC and TPG contexts within a DNA duplex. Specifically, the TATA and ATAT motifs appear to adopt conformations that shield the purine bases from aqueous solution and facilitate cross-strand energy transfer from adenine donors to 2-AP.

Assessment of solvent exposure of 2-AP in various sequence contexts by monitoring fluorescence quenching by acrylamide
In order to directly assess the solvent accessibility of 2-AP in various sequence contexts, fluorescence emission intensities were measured in the presence of varying concentrations of acrylamide. Acrylamide quenches fluorescence through a dynamic collisional mechanism with the fluorophore and this quenching can therefore provide an indication of the solvent exposure of 2-AP. To assess the degree of acrylamide accessibility of 2-AP within the different oligonucleotides, Stern–Volmer plots of F0/F versus acrylamide concentration were utilized, where F0 is the fluorescence emission intensity in the absence of acrylamide and F is the intensity at a given acrylamide concentration. The slopes of such plots are equal to kq{tau}0 and therefore reflect both the accessibility of 2-AP to the quenching agent and the lifetime of the excited state (18) (see Materials and Methods). A smaller slope is indicative of a smaller acrylamide-dependent change in fluorescence intensity and therefore a lesser degree of solvent exposure of the fluorophore.

Significant differences in acrylamide quenching for 2-AP in different environments can be seen from such plots for ss.P26 and ds.P26 (Fig. 3A). The free 2-AP base is extremely sensitive to acrylamide-mediated fluorescence quenching, showing an upward curvature of the Stern–Volmer plot which is indicative of a contribution of static quenching at high quencher concentrations (19) due to formation of a weak complex between the acrylamide and the 2-AP base in its electronic ground state (data not shown). The single-stranded oligonucleotide is also very sensitive to acrylamide quenching, but although there is some scatter in the linear Stern–Volmer plot, no curvature is observed (Fig. 3A), allowing for an accurate estimation of the bimolecular quenching rate constant. The large value of the bimolecular quenching rate constant (kq = 6.0 x 109 M–1 s–1) is similar in magnitude to the values observed for acrylamide quenching of small molecules or surface-exposed tryptophan residues in proteins (19), indicating that the 2-AP base is highly solvent exposed in the single-stranded oligonucleotide. In contrast, in the corresponding duplex, 2-AP fluorescence shows very weak quenching by acrylamide and the bimolecular quenching rate constant (kq = 1.1 x 108 M–1 s–1) is reduced by almost two orders of magnitude compared with the single strand. Thus, the 2-AP base is strongly protected from collisional quenching in this particular duplex.




View larger version (73K):
[in this window]
[in a new window]
  		Figure 3. Effect of acrylamide upon the fluorescence of 2-AP in different contexts. Acrylamide was added at 30°C to give a final concentration of 50 mM per increment as described in Materials and Methods. (A) Stern–Volmer plots for single-stranded (ss.P26) and duplexed (ds.P26) oligonucleotides. (B) Bimolecular quenching rate constants obtained from analysis of Stern–Volmer plots according to equations 3 and 5 (Materials and Methods).

 
Looking at quenching at other positions of substitution, the collisional quenching rate for ss.P8 was similar to that of ss.P26, as expected (data not shown). However, in the duplex contexts, the bimolecular quenching rates, while lower than in their single-stranded counterparts, show considerable sequence-dependent variability (Fig. 3B). Those for ds.P7, ds.P27 and ds.P26 are roughly an order of magnitude smaller than the corresponding rate constants determined for ds.P5, ds.P25, ds.P10, ds.P12 and ds.P8. These results verify that 2-AP in each of the aforementioned three positions is much more protected from collisional quenching and is less solvent exposed relative to 2-AP in the other duplex contexts. This result is in close agreement with the equilibrium fluorescence data, which suggest that 2-AP internal to ATAT or TATA is well stacked within the duplex.

Fluorescence anisotropy decay measurements to assess the dynamics of 2-AP motion in various contexts
In order to directly assess the mobility of 2-AP within different sequence contexts, time-resolved fluorescence anisotropy decay measurements were performed as described in Materials and Methods. For most duplexes, the anisotropy was fitted with two exponential decay components (Table 4). The longer anisotropy decay time ({phi}1) is attributed to overall tumbling of the duplex molecules, whereas the shorter decay time ({phi}2) arises from internal rotation of the 2-AP base (11). The angular range of 2-AP motion is estimated by modeling the base rotation as diffusion within a cone (12,20). The cone semi-angle is given by {theta}0, where


View this table:
[in this window]
[in a new window]
  		Table 4. Time-resolved fluorescence anisotropy decay parameters, order parameter and cone angle for 2-AP in various sequence contexts
 
{theta}0 = cos–1 {(0.5)[(1 + 8S)1/2 – 1]}

and S is the generalized order parameter,

S = [ß1/(ß1 + ß2)]1/2.

The magnitude of the cone semi-angle provides an indication of the motional restriction of 2-AP within the duplexes. The absence of a short anisotropy decay time (S = 1, {theta}0 = 0) would suggest that the base is conformationally restrained within the duplex.

From the data in Table 4, it is evident that 2-AP is rotationally constrained in ds.P7, ds.P27 and ds.P26, the same positions in which 2-AP is well stacked and solvent protected (Table 2). At these sites, a single nanosecond scale motion is detected for 2-AP, a time that is too long to be attributed to motion of the base itself, but instead is likely to be due to rotation of the duplex as a whole. The failure to detect a rapid local motion of 2-AP at these sites is not simply a consequence of limited time resolution, because the initial value of the anisotropy (r0) is similar to that seen at the other fluorophore positions (Table 4). The low mobility of 2-AP in these contexts is also consistent with the presence of a significant excitation energy transfer peak seen at room temperature (Fig. 2A–C and Table 2). This type of transfer is dependent (among other factors) upon reduced mobility of 2-AP relative to the nearest adenine that is acting as the energy donor (6).

In contrast, the 2-AP base exhibits high rotational mobility in ds.P5, ds.P10 and ds.P25, as indicated by the large cone angles describing the angular range of motion of the base (Table 4). The cone angles at these positions are typical of the values normally detected for DNA duplexes containing 2-AP (11). In ds.P12, the cone angle is smaller, indicating that the local motion of 2-AP is somewhat more restricted at this position, although the base is clearly not as restricted as in ds.P7, ds.P27 or ds.P26. Thus, it appears that the degree of motional restriction at position P12 is intermediate between the highly restricted (ds.P7, ds.P27 and ds.P26) and highly mobile (ds.P5, ds.P10 and ds.P25) sites. This may be the result of 2-AP in ds.P12 being in a CAC context, which puts it in a similar (but not identical) base stacking conformation as the highly restricted 2-AP positions, all of which lie within TAT contexts.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
The positions that gave rise to anomalous structural parameters of 2-AP in duplex DNA are summarized in Figure 4. The variations among the oligonucleotide positions in excitation wavelengths and profiles, slopes of Stern–Volmer plots and fluorescence anisotropy decay behavior each point to the same position contexts having conformational irregularities. In each case, the 2-AP lies within a staggered TAT duplex context in which 2-AP is flanked by two thymines and adjacent to two adenines on the opposite strand.



View larger version (27K):
[in this window]
[in a new window]
  		Figure 4. Summary of sequence-dependent anomalies in 2-AP conformation within the ATAT-containing and TATA-containing duplexes. The positions of the various 2-AP substitutions studied within the two duplexes are underlined. The positions at which 2-AP exhibited red-shifted excitation maxima and energy transfer peaks at 275 nm as well as low rotational motion are shown (outlined) within the ATAT- and TATA-containing duplexes. In each instance, 2-AP falls within a staggered duplexed TAT context.

 
The internal sites of 2-AP substitution in these contexts (ds.P7, ds.P27 and ds.P26) are quite different in conformation than external sites of 2-AP substitution (ds.P5 and ds.P8) (Fig. 4). These differences are a function of the duplexed contexts since they do not occur among the corresponding 2-AP-containing single-stranded oligonucleotides. There are several reported fluorescence studies of 2-AP in d(TnpPpTn) or d(AnpPpAn) that have observed a red-shifted direct excitation spectrum for 2-AP (3) and unusual dynamic behavior (5). However, to our knowledge, there are no reports of the properties of 2-AP fluorescence substituted at internal versus external positions of alternating AT sequences having been investigated.

Variations in 2-AP fluorescence among conformations that differ fundamentally in secondary structure have been previously observed. However, the differences observed in this study are significant in that they reflect subtleties of sequence-dependent local conformations within very specific short sequence contexts. A red shift in 2-AP excitation wavelength maxima by ~10 nm has been attributed to reduced solvent exposure of the fluorophore (9). A recent study involving quantum mechanical calculations of the (2-AP)T dinucleotide suggests that this shift arises from a perturbed excited state electronic structure of 2-AP resulting from a stacked conformation (21). Additionally, the red shift has been shown to occur upon base pairing between the N1 of 2-AP and the imino proton on thymine (22). The absence of a protonated form of N1, as in a melted duplex or in the free 2-AP base under neutral solution conditions, causes the excitation wavelength maximum to be blue shifted by ~10 nm (8,9,23). In the three locations noted, the red-shifted excitation wavelength maximum therefore indicates that the fluorophore spends a larger fraction of time as a fully base paired, intrahelical entity than when present at other positions in the parent duplexes. This situation was altered in ds.P7 by deleting A25 to form ds.P7_eh so as to cause the flanking 3' thymine to spend part of the time in an extrahelical conformation. In this case, the excitation wavelength of 2-AP was blue shifted into the range seen for the majority of duplex 2-AP contexts, indicating weakened base pairing at P7 and reduced base stacking.

Another feature that distinguished the excitation fluorescence spectra of 2-AP in ds.P7, ds.P27 and ds.P26 was the presence of the energy transfer peak at ~275 nm. This peak is thought to originate as a result of excitation energy transfer from nearby adenines that are efficient energy donors and is dependent on proximity to the donor adenine and a stacked and immobile conformation for the fluorophore (6,7). The energy transfer peaks observed at 30°C in this study are much more pronounced examples of interstrand energy transfer than previously reported (6). The 2-AP in these duplexes has two adjacent adenines on the opposite strand (Fig. 4) that are partly responsible for the strong energy transfer peaks. In the case of 2-AP in ds.P5, ds.P25 and ds.P8, only one adenine is present on the opposite strand and a transfer peak is not present. However, the intrinsic energy transfer efficiency from each of the adenines opposite 2-AP in ds.P7, ds.P27 and ds.P26 is much higher than previously observed for cross-strand transfer from A to 2-AP (6). Therefore, the alternating AT sequence context of 2-AP in ds.P7, ds.P27 and ds.P26 may be contributing to a structural conformation that aids energy transfer. The anisotropy decay data for ds.P7, ds.P27 and ds.P26 support this proposal as they can be fitted to a single correlation time corresponding only to overall rotation of the duplex, strongly suggesting that the 2-AP base has little, if any, mobility at these positions.

Interestingly, there is a slight but perceptible energy transfer peak in the excitation spectrum of 2-AP in ds.P7_eh, in which 2-AP is 5' to a thymine that lacks an adenine on the opposite strand. This energy transfer peak becomes more pronounced at lowered temperatures (data not shown), indicating that it is not a spectral artifact. The presence of this visible energy transfer peak suggests that the local conformation of 2-AP at P7 may be partially preserved and therefore still able to facilitate some energy transfer from adenine on the opposite strand.

NMR studies of the ATAT-containing parent duplex suggest an unusual conformation of ATGA in which there is a kink at the TG step, leading to increased helical curvature towards the major groove at this site (24,25). Since P7 lies within this sequence and P27 flanks it, we wanted to ensure that the anomalous characteristics of 2-AP at these positions were indeed a consequence of the alternating AT stretch as opposed to the conformational anomaly at ATGA. By replacing the ATAT stretch with a TATA stretch to form the second parent duplex set, the staggered TAT duplex context was maintained while the ATGA sequence was eliminated. However, 2-AP in position P26 in the TATA-containing duplex had fluoresence characteristics similar to those of 2-AP in ds.P7 and ds.P27.

The acrylamide quenching data (Fig. 3B) and fluorescence anisotropy decay data (Table 4) with the two 16mer duplexes used in this study suggest that the 2-AP conformational stability in duplex DNA shows the following context dependence: TPT (ds.P7, ds.P27 and ds.P26) >> CPC (ds.P12) > GPT (ds.P5) = TPG (ds.P8). [The data for ds.P25 (TPC) and ds.P10 (GPC) cannot be generalized because these sites lie within the RTGR context, which itself is structurally anomalous (23–25).]

Although purines are not thought to stack well between pyrimidines, Escherichia coli DNA polymerase I in the absence of template exclusively produces poly(dA·dT) even with all four dNTPs present (26), suggesting that alternating AT stretches are somehow energetically stabilized. Tetranucleotide conformational maps indicate that ATAT and TATA are both relatively flexible with respect to slide of the central dinucleotide step (27). However, positive values of slide would cause adenines on opposite strands to become spatially closer and partially stacked, which may account for the efficient cross-strand energy transfer and reduced rotational motion observed in this study. Cyclization kinetic studies of TATA-containing sequences find that the direction of flexibility at TATA can be explained by a high value of roll on either side of the internal TA base pair which is abolished when TATA is replaced with TACA (28). This configuration of TATA would also serve to bring the internal adenines in the duplexed sequence into closer proximity. A crystal structure of a decamer containing ATATAT exhibits an alternation of twist angles at the AT tracts with a low twist angle between AT steps and a high twist angle between TA steps (29). The effect of this pattern is to strengthen the intrinsically favorable overlap at AT steps at the expense of the already poor overlap at TA steps. Furthermore, interior segments of alternating poly(dA·dT) tracts are able to adopt a number of different conformations and have variable minor groove widths, depending on the flanking sequences (29,30).

The duplex TATA sequence is of particular interest since it is an essential element of the ‘TATA’ box (TATAAAAG) that is involved in transcription initiation. The TATA-binding protein (TBP) binds the TATA box through the minor groove and bends the central 6 bp (ATAAAA) by ~90° (31). Studies with TATA box sequence variants indicate that sequences that abolish the alternating TA segment lead to lower deformability (32) and also result in the most reduced TBP binding (33). The crystal structure of human TBP bound to the TATA box shows that the TATA region of the bound DNA exhibits the low twist angle characteristic of AT steps at AT followed by a higher twist angle at the adjacent TA step (31). This suggests that TBP might recognize and exaggerate the unique structural features of duplex TATA that were observed in this study.

It should be pointed out that both the ATAT and TATA sequences studied here were flanked by GC base pairs (Fig. 4) and the unusual properties observed for the internal sites of 2-AP substitutions might be perturbed if these sequences were part of a longer alternating AT sequence. Given that alternating AT regions have complex sequence context-dependent patterns of base and base pair conformational flexibility, any one structural technique might not expose each of the features of this sequence. Therefore, multiple approaches are necessary to better understand the structural role of alternating AT sequences in processes such as DNA replication and transcription initiation.


   ACKNOWLEDGEMENTS
 
We thank Prof. David E. Wemmer for critical reading of the manuscript, Prof. Daniel Portnoy for the use of his fluorometer and Prof. Howard Schachman for the use of his temperature control unit. This research was supported by NIH grants RO1GM19020 (S.L.), P3OES01896 (S.L.) and RO1GM44060 (D.M.).


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Sowers,L.C., Fazakerly,G.V., Eritja,R., Kaplan,B.E. and Goodman,M.F. (1986) Base pairing and mutagenesis: observation of a protonated base pair between 2-aminopurine and cytosine in an oligonucleotide by proton NMR. Proc. Natl Acad. Sci. USA, 83, 5434–5438.[Abstract/Free Full Text]

  2. Eritja,R., Kaplan,B.E., Mhaskar,D., Sowers,L.C., Petruska,J. and Goodman,M.F. (1986) Synthesis and properties of defined DNA oligomers containing base mispairs involving 2-aminopurine. Nucleic Acids Res., 14, 5869–5884.[Abstract/Free Full Text]

  3. Stivers,J.T. (1998) 2-Aminopurine fluorescence studies of base stacking interactions at abasic sites in DNA: metal-ion and base sequence effects. Nucleic Acids Res., 26, 3837–3844.[Abstract/Free Full Text]

  4. Rachofsky,E.L., Osman,R. and Ross,J.B.A. (2001) Probing structure and dynamics of DNA with 2-aminopurine: effects of local environment on fluorescence. Biochemistry, 40, 946–956.[CrossRef][Medline]

  5. Rachofsky,E.L., Seibert,E., Stivers,J.T., Osman,R. and Ross,J.B.A. (2001) Conformation and dynamics of abasic sites in DNA investigated by time-resolved fluorescence of 2-aminopurine. Biochemistry, 40, 957–967.[CrossRef][Medline]

  6. Xu,D.-G. and Nordlund,T.M. (2000) Sequence dependence of energy transfer in DNA oligonucleotides. Biophys. J., 78, 1042–1058.[Abstract/Free Full Text]

  7. Nordlund,T.M., Xu,D. and Evans,K.O. (1993) Excitation energy transfer in DNA. Duplex melting and transfer from normal bases to 2-aminopurine. Biochemistry, 32, 12090–12095.[CrossRef][Medline]

  8. Xu,D.-G., Evans,K.O. and Nordlund,T.M. (1994) Melting and pre-melting transitions of an oligomer measured by DNA base fluorescence and absorption. Biochemistry, 33, 9592–9599.[CrossRef][Medline]

  9. Evans,K., Xu,D.-G., Kim,Y.-S. and Nordlund,T.M. (1992) 2-Aminopurine optical spectra: solvent, pentose ring and DNA helix melting dependence. J. Fluorescence, 2, 209–219.[CrossRef]

  10. Nordlund,T.M., Anderson,S., Nisson,L., Rigler,R., Graslund,A. and McLaughlin,L.W. (1989) Structure and dynamics of a fluorescent DNA oligomer containing the EcoRI recognition sequence: fluorescence, molecular dynamics and NMR studies. Biochemistry, 28, 9095–9103.[CrossRef][Medline]

  11. Guest,C.R., Hochstrasser,R.A., Sowers,L.C. and Millar,D.P. (1991) Dynamics of mismatched base pairs in DNA. Biochemistry, 30, 3271–3279.[CrossRef][Medline]

  12. Lipari,G. and Szabo,A. (1980) Effect of librational motion on fluorescence depolarization and nuclear magnetic resonance relaxation in macromolecules and membranes. Biophys. J., 30, 489–506.[Abstract/Free Full Text]

  13. Grinvald,A. and Steinberg,I.Z. (1974) On analysis of fluorescence decay kinetics by the method of least-squares. Anal. Biochem., 59, 583–598.[CrossRef][ISI][Medline]

  14. Bevington,P.R. (1969) Data Reduction and Error Analysis for the Physical Sciences. McGraw-Hill, New York, NY, pp. 134–136.

  15. Morden,K.M., Gunn,B.M. and Maskos,K. (1990) NMR studies of a deoxyribodecanucleotide containing an extrahelical thymidine surrounded by an oligo(dA).oligo(dT) tract. Biochemistry, 29, 8835–8845.[CrossRef][Medline]

  16. Coppel,Y., Berthet,N., Coulombeau,C., Coulombeau,C., Garcia,J. and Lhomme,J. (1997) Solution conformation of an abasic DNA undecamer duplex d(CGCACXCGC).d(GCGTGTGTGCG): the unpaired thymine stacks inside the helix. Biochemistry, 36, 4817–4830.[CrossRef][Medline]

  17. Kalnik,M.W., Norman,D.G., Li,B.F., Swann,P.F. and Patel,D.J. (1990) Conformational transitions in thymidine bulge-containing deoxytridecanucleotide duplexes. Role of flanking sequence and temperature in modulating the equilibrium between looped out and stacked thymidine bulge states. J. Biol. Chem., 265, 636–647.[Abstract/Free Full Text]

  18. Bell,J.E. (1981) Fluoresence: Solution studies. In Bell,J.E. (ed.), Spectroscopy in Biochemistry, Vol. I. CRC Press, Boca Raton, FL, pp. 183–188.

  19. Eftink,M.R. (1991) Fluorescence quenching: theory and applications. In Lakowicz,J.R. (ed.), Topics in Fluorescence Spectroscopy, Vol. 2. Plenum Press, New York, NY, pp. 53–126.

  20. Kinoshita,K.,Jr, Kawato,S. and Ikegami,A. (1977) A theory of polarization decay in membranes. Biophys. J., 20, 289–305.[Abstract/Free Full Text]

  21. Jean,J.M. and Hall,K.B. (2001) 2-Aminopurine fluorescence quenching and lifetimes: role of base-stacking. Proc. Natl Acad. Sci. USA, 98, 37–41.[Abstract/Free Full Text]

  22. Goodman,M.F. and Ratcliff,R.M. (1983) Evidence of 2-aminopurine-cytosine base mispairs involving two hydrogen bonds. J. Biol. Chem., 258, 12842–12846.[Abstract/Free Full Text]

  23. Dornberger,U., Flemming,J. and Fritzsche,H. (1998) Structure determination and analysis of helix parameters in the DNA decamer d(CATGGCATG)2. Comparison of results from NMR and crystallography. J. Mol. Biol., 284, 1453–1463.[CrossRef][ISI][Medline]

  24. Rai,P., Cole,T.D., Wemmer,D.E. and Linn,S. (2001) Localization of Fe2+ at an RTGR within duplex DNA explains preferential cleavage by Fe2+ and H2O2. J. Mol. Biol., 312, 1089–1101.[CrossRef][ISI][Medline]

  25. Tonelli,M., Ragg,E., Bianucci,A.M., Lesiak,K. and James,T.L. (1998) Nuclear magnetic resonance structure of d(GCATATGATAG).d(CTATCATATGC): a consensus sequence for promoters recognized by {sigma}k RNA polymerase. Biochemistry, 37, 11745–11761.[CrossRef][Medline]

  26. Schachman,H., Adler,J., Radding,C.M., Lehman,I.R. and Kornberg,A. (1960) Enzymatic synthesis of deoxyribonucleotide acid. J. Biol. Chem., 235, 3242–3249.[Free Full Text]

  27. Packer,M.J., Dauncey,M.P. and Hunter,C.A. (2000) Sequence-dependent DNA structure: tetranucleotide conformational maps. J. Mol. Biol., 295, 85–103.[CrossRef][ISI][Medline]

  28. Davis,N.A., Majee,S.S. and Kahn,J.D. (1999) TATA box DNA deformation with and without the TATA box-binding protein. J. Mol. Biol., 291, 249–265.[CrossRef][ISI][Medline]

  29. Yuan,H., Quintana,J. and Dickerson,R.E. (1992) Alternative structures for alternating poly(dA-dT) tracts: the structure of the B-DNA decamer C-G-A-T-A-T-A-T-C-G. Biochemistry, 31, 8009–8021.[CrossRef][Medline]

  30. Bareket-Samish,A., Cohen,I. and Haran,T.E. (2000) Signals for TBP/TATA box recognition. J. Mol. Biol., 299, 965–977.[CrossRef][ISI][Medline]

  31. Nikolov,D.B., Chen,H., Halay,E.D., Hoffmann,A., Roeder,R.G. and Burley,S.K. (1996) Crystal structure of a human TATA box-binding protein/TATA element complex. Proc. Natl Acad. Sci. USA, 93, 4862–4867.[Abstract/Free Full Text]

  32. de Souza,O.M. and Ornstein,R.L. (1998) Inherent DNA curvature and flexibility correlate with TATA box functionality. Biopolymers, 46, 402–415.

  33. Wu,J., Parkhurst,K.M., Powell,R.M., Brenowitz,M. and Parkhurst,L.J. (2001) DNA bends in TATA-binding protein-TATA complexes in solution are DNA sequence-dependent. J. Biol. Chem., 276, 14614–14622.[Abstract/Free Full Text]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Nucleic Acids ResHome page
S. V. Avilov, E. Piemont, V. Shvadchak, H. de Rocquigny, and Y. Mely
Probing dynamics of HIV-1 nucleocapsid protein/target hexanucleotide complexes by 2-aminopurine
Nucleic Acids Res., February 11, 2008; 36(3): 885 - 896.
[Abstract] [Full Text] [PDF]

Home page
 Nucleic Acids ResHome page
C. Yuan, X. W. Lou, E. Rhoades, H. Chen, and L. A. Archer
T4 DNA ligase is more than an effective trap of cyclized dsDNA
Nucleic Acids Res., August 13, 2007; 35(16): 5294 - 5302.
[Abstract] [Full Text] [PDF]

Home page
 Nucleic Acids ResHome page
A. A. Marti, S. Jockusch, Z. Li, J. Ju, and N. J. Turro
Molecular beacons with intrinsically fluorescent nucleotides.
Nucleic Acids Res., January 1, 2006; 34(6): e50 - e50.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (179K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (19)
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Rai, P.
Right arrow Articles by Linn, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Rai, P.
Right arrow Articles by Linn, S.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?