FASEB J. FASEB
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published as doi: 10.1096/fj.07-8102com.
 (The FASEB Journal. 2008;22:19-29.)
© 2008 FASEB
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow All Versions of this Article:
fj.07-8102comv1
22/1/19    most recent
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 PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Breit, J. F.
Right arrow Articles by Gillespie, M. N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Breit, J. F.
Right arrow Articles by Gillespie, M. N.
(The FASEB Journal. 2008;22:19-29.)
© 2008 FASEB

Nuclear protein-induced bending and flexing of the hypoxic response element of the rat vascular endothelial growth factor promoter

Jeffrey F. Breit*, Katheryn Ault-Ziel*, Abu-Bakr Al-Mehdi* and Mark N. Gillespie*

* Department of Pharmacology and Center for Lung Biology, University of South Alabama College of Medicine, Mobile, Alabama, USA

1Correspondence: Dept. of Pharmacology, University of South Alabama College of Medicine, MSB 3366, Mobile, AL 36688-0002, USA. E-mail: mgillesp{at}jaguar1.usouthal.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Bending and flexing of DNA may contribute to transcriptional regulation. Because hypoxia and other physiological signals induce formation of an abasic site at a key base within the hypoxic response element (HRE) of the vascular endothelial growth factor (VEGF) gene (FASEB J. 19, 387–394, 2005) and because abasic sites can introduce flexibility in model DNA sequences, in the present study we used a fluorescence resonance energy transfer–based reporter system to assess topological changes in a wild-type (WT) sequence of the HRE of the rat VEGF gene and in a sequence harboring a single abasic site mimicking the effect of hypoxia. Binding of the hypoxia-inducible transcriptional complex present in hypoxic pulmonary artery endothelial cell nuclear extract to the WT sequence failed to alter sequence topology whereas nuclear protein binding to the modified HRE engendered considerable sequence flexibility. Topological effects of nuclear proteins on the modified VEGF HRE were dependent on the transcription factor hypoxia-inducible factor-1 and on formation of a single-strand break at the abasic site mediated by the coactivator, Ref-1/Ape1. These observations suggest that oxidative base modifications in the VEGF HRE evoked by physiological signals could be a precursor to single-strand break formation that has an impact on gene expression by modulating sequence flexibility.—Breit, J. F., Ault-Ziel, K., Al-Mehdi, A.-B., Gillespie, M. N. Nuclear protein-induced bending and flexing of the hypoxic response element of the rat vascular endothelial growth factor promoter.


Key Words: DNA topography • VEGF expression


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
HYPOXIA IS A FUNDAMENTAL STIMULUS in mammalian biology, playing significant roles in normal development and physiology as well as in numerous pathophysiologic states. Given its importance, considerable attention has been focused on understanding the mechanisms by which cells sense hypoxia and initiate adaptive gene expression. One of the key pathways activating transcription involves the hypoxia-inducible transcription factor, HIF-1, which forms a multiprotein complex with the cofactors, CBP/p300, ATF/CREB, Ref-1/Ape1, and probably others (1 2 3) on the hypoxic response element (HRE) to initiate expression of adaptive genes, including the angiogenic and cell survival vascular endothelial growth factor (VEGF) gene (3 , 4) .

Like many physiological stimuli, hypoxia uses reactive oxygen species as second messengers (5 6 7 8 9) . A surprising target of reactive oxygen species generated in hypoxia is the promoter of the VEGF gene (10 , 11) . In particular, the 3' guanine of the HIF-1 DNA recognition sequence is frequently converted to an alkali-labile abasic site, probably initiated by formation of the common base oxidation product 8-oxoguanine. Oxidative modifications at this nucleotide may be biologically significant; introduction of a model abasic site at the hypoxia-modified guanine alters composition of the hypoxia-inducible transcriptional complex binding to the HRE, increasing the abundance of HIF-1 along with the coactivator and DNA repair enzyme, Ref-1/Ape1, and engenders more robust reporter gene expression (11) .

Along with modulating the composition of the transcriptional complex, another mechanism whereby oxidative modifications targeted to specific DNA bases in promoter sequences may govern gene expression relates to changes in sequence flexibility. In this context, traditional concepts highlight the involvement of transcription factor-induced DNA bending in activation of gene expression (12) . More recent studies have emphasized the importance of sequence flexibility as a determinant of promoter DNA-nucleosome apposition and its role in governing DNA access to transcriptional regulatory proteins and chromatin remodeling enzymes (13 , 14) . Interestingly, base oxidation products, including abasic sites, are known to increase the torsional flexibility of model DNA sequences (15 16 17) . Single-strand DNA breaks, created by cleavage at abasic sites by the DNA repair enzyme and transcriptional coactivator, Ref-1/Ape1 (18) , also modulate conformational flexibility (19) . In light of these considerations, in the present study we tested the idea that introduction of an abasic site at a key nucleotide in the VEGF HRE, mimicking the oxidative modification that occurs during hypoxic signaling, alters the effect of transcriptional protein binding on DNA sequence flexibility.

To address this issue, we obtained two duplex oligonucleotides corresponding to the VEGF gene HRE, one without (wild type, "WT") and another with introduction of a tetrahydrofuran at the 3' guanine of the HIF-1 binding region to mimic the hypoxic-induced formation of an abasic site ("abasic"), and labeled these sequences at internal positions with two fluorophores, fluorescein and Texas red, suitable for detection of fluorescence resonance energy transfer (FRET). Using analytical strategies similar to those reported previously (20 21 22 23 24) , we performed steady-state FRET and lifetime fluorescence analyses in the absence and presence of normoxic and hypoxic nuclear protein extracts to provide insight into whether nuclear protein binding altered the apposition of the two fluorophores. In addition, we explored the roles of HIF-1 and Ref-1/Ape1, two important constituents of the hypoxia-inducible transcriptional complex, in inducing flexibility in the model VEGF HREs. We found that although the WT oligonucleotide was stiff and not perturbed in response to either normoxic or hypoxic nuclear protein binding, substantial sequence flexibility was induced by nuclear protein binding to the oligonucleotide harboring a single abasic site at the hypoxia-modified guanine. The structural effects of nuclear protein binding to the abasically modified sequence required the presence of HIF-1 and appeared to be dependent on the Ref-1/Ape1-mediated production of a single-strand break at the abasic site.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Rat main pulmonary artery endothelial cell (PAEC) cultures
Rat main PAECs were harvested and cultured as described previously (2 , 10 , 11) . In brief, main pulmonary arteries were isolated from 300-g Sprague-Dawley rats killed with an overdose of Nembutal. Isolated arteries were opened, and the intimal lining was gently scraped with a scalpel. The harvested cells were then placed into a T25 Corning flask in a 1:1 mixture of F-12 Nutrient Mixture and Dulbecco’s modified Eagle’s medium supplemented with 10% FBS, 100 U/ml penicillin, and 0.1 mg/ml streptomycin (Gibco BRL, Grand Island, NY, USA). Culture medium was changed once a week, and after reaching confluence, the cells were harvested using a 0.05% solution of trypsin (Gibco BRL) and passed up to 15 times. The endothelial cell phenotype was confirmed by acetylated LDL uptake, Factor VIII immunocytochemical staining and the lack of immunostaining with smooth muscle cell {alpha}-actin antibodies (Sigma-Aldrich Corp., St. Louis, MO, USA). When cells reached confluence, the plates were placed in incubators (Thermo Scientific, Inc., Marietta, OH, USA) purged with gas mixtures containing 21% (designated as "normoxic") or 2% O2 (termed "hypoxic"), 5% CO2, and the balance in nitrogen. Media Po2 values, determined using a model ABL30 blood-gas analyzer (Radiometer, Copenhagen, Denmark), were >120 torr for the control cultures and 25–35 torr for cultures designated as hypoxic.

Nuclear protein isolation
Nuclear proteins were isolated from rat PAECs cultured for 3 h in normoxia or hypoxia as described previously (2 , 11) . Rat PAECs exposed to normoxic or hypoxic conditions for 3 h were washed and scraped into cold 1x PBS. Cells were centrifuged at 2000 g for 10 min at 4°C. Supernatant was removed, and the pellet was washed with 5 packed cell volumes (PCV) of buffer A [10 mM Tris-HCl (pH 7.5), 1.5 mM MgCl2, 10 mM KCl supplemented with 1M dithiothreitol, 0.2 M PMSF, 1 mg/ml leupeptin, 1 mg/ml aprotinin, 1 mg/ml pepstatin, and 0.5 M Na3VO4], resuspended in 4 PCV of buffer A and incubated on ice for 10 min. The cell suspension was homogenized, and nuclei were pelleted by centrifugation at 10,000 g for 10 min at 4°C, after which they were resuspended in 3 PCV of buffer C [20 mM Tris-HCl (pH 7.5), 0.42 M KCl, 1.5 mM MgCl2, and 20% glycerol] and rotated for 30 min at 4°C. The suspension was centrifuged at 15,000g for 30 min at 4°C and then dialyzed against three changes of buffer D [Tris-HCl (pH 7.5), 0.2 mM EDTA, 0.1 M KCl, and 20% glycerol] freshly supplemented with dithiothreitol and Na3VO4 for 4 h at 4°C. The dialysate was centrifuged at 15,000 g for 10 min at 4°C, divided into 25-µl aliquots, flash frozen in liquid N2, and stored at –80°C. The nuclear protein concentration was determined using the Bradford method (Bio-Rad, Hercules, CA, USA).

DNA affinity precipitation
DNA affinity precipitation analyses were performed as described previously (1 , 2 , 11) . Ten picomoles of gel-purified, biotin-labeled 64-mer oligonucleotide corresponding to a minimal functionally-active segment of the hypoxic response element of the VEGF gene (Fig. 1 ) were incubated with 125 µg of Dynabeads M-280 streptavidin (10 mg/ml; Dynal A.S., Oslo, Norway) for 20 min at 25°C. The biotinylated oligonucleotide bound to the streptavidin-conjugated beads was retrieved using a magnetic particle concentrator (Dynal A.S.). Bead-associated oligonucleotide was washed two times with Tris-EDTA and once with buffer D [25 mM Tris- HCl (pH 7.5), 0.2 mM EDTA, 0.1 M KCl, and 20% glycerol, supplemented with 100 mM DTT, 100 mM Na3VO4, and 500 mM PMSF]. Normoxic or hypoxic nuclear proteins (25 µg) were then added to the oligonucleotide and incubated on ice for 20 min. The protein-oligonucleotide-bead complexes were immobilized with the magnetic particle concentrator and washed twice with buffer D to remove unbound proteins. Protein-DNA complexes were resuspended in 2x sample buffer [20% glycerol, 0.2% bromphenol blue, 4% sodium dodecyl sulfate (SDS), and 100 mM Tris-HCl (pH 6.8)] and boiled for 2 min to elute bound proteins. Proteins eluted from the oligonucleotide were then subjected to Western blot analysis to identify specific components of the hypoxic transcriptional complex or enumerated on silver-stained gels.


Figure 1
View larger version (8K):
[in this window]
[in a new window]

  		Figure 1. Sequences of the WT and abasic site-containing fluorophore tethered VEGF HRE probes used for FRET analysis. F indicates the position in the abasic probe where a tetrahydrofuran was inserted to mimic the hypoxia-induced formation of an abasic site. FL and TR refer to positions where fluorescein and Texas red, respectively, were tethered to oligonucleotide probes to enable detection of oligonucleotide bending by steady-state and time-resolved fluorescence resonance energy transfer. See text for details.

Immunodepletion of Ref-1/Ape1 and HIF-1
To elucidate the role of Ref-1/Ape1 and HIF-1 relative to bending and flexing of the VEGF HRE, the proteins were discretely immunodepleted from normoxic and hypoxic nuclear extracts before DNA affinity precipitation and FRET analyses. Nuclear extracts (50 µg) were incubated on ice for 1 h with 7 µg of either Ref-1/Ape1 mouse monoclonal antibody (NB 100–116; Novus Biologicals, Inc., Littleton, CO, USA) or HIF-1{alpha} mouse monoclonal antibody (ab1; Novus Biologicals). After 1 h, 5 µl of protein A/G plus-agarose conjugated beads (0.5 ml of agarose/2.0 ml) (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) were added to the nuclear extract/antibody mixtures, and the samples were incubated overnight at 4°C with rocking. To remove the protein-antibody-A/G agarose complexes from the nuclear extracts, samples were centrifuged at 7500 g for 10 min at 4°C. The supernatant containing the remaining nuclear proteins was validated by Western blot analysis, transferred to a clean microcentrifuge tube and stored at –80°C until subsequent DNA affinity precipitation analysis.

Suppression of HIF-1 DNA binding
As another means of inhibiting HIF-1 binding to DNA, we used a "decoy" oligonucleotide consisting of the HIF-1 DNA recognition sequence, TACGTGGG (25) . The decoy serves as a sink, binding HIF-1 and preventing its incorporation into the transcriptional complex forming on the 64-bp HRE sequence.

SDS-PAGE and Western blot analysis
Samples were suspended in 2x SDS sample buffer, boiled for 2 min, and loaded onto a precast 4–15% polyacrylamide gradient Tris-HCl gel (Bio-Rad). Proteins were separated at 100 V for 1.5 h and transferred to a nitrocellulose membrane at 250 mA for 45 min using a protein transfer unit (Bio-Rad). The gel was removed from the transfer unit, and proteins were visualized by silver staining. In other studies, the membranes were subjected to Western immunoblot analysis to detect some of the known components of the hypoxic transcriptional complex. In this instance, the membrane was blocked with Blotto (5% nonfat dry milk in 1x PBS with 0.05% Tween) for 1 h. The blocking solution was removed, and the primary antibody was diluted in 5 ml of Blotto and applied to the nitrocellulose membrane. Antibody concentrations and incubation conditions varied, depending on the specific requirements of each antibody as per the manufacturer’s recommendations: HIF-{alpha} mouse monoclonal antibody (ab1) was diluted 1:500 and incubated overnight at 4°C. Ref-1/Ape1 mouse monoclonal antibody (NB 100–116) was diluted to 2 µg/ml in 5 ml of Blotto and incubated for 1 h at 25°C. ATF/CREB mouse monoclonal antibody (25C10G; Santa Cruz Biotechnology, Inc.) was diluted 1:100 in 5 ml of Blotto and incubated at 25°C for 1.5 h. p300 mouse monoclonal antibody (NM11; NeoMarkers, Inc., Fremont, CA, USA) was diluted to 1 µg/ml in 5 ml of Blotto and incubated at 25°C for 2 h. For all of the above antibodies, goat anti-mouse horseradish peroxidase-conjugated secondary antibody (Bio-Rad) was diluted 1:5000 in Blotto, and membranes were incubated at 25°C for 45 min. Antibody was removed, and membranes were washed three times for 5 min in 1x PBS + 0.05% Tween and one time for 5 min in 1x PBS. Proteins were visualized with a Luminol (Santa Cruz Biotechnology, Inc.) -enhanced chemiluminescence detection system using Hyperfilm ECL (Amersham Pharmacia Biotech, Buckinghamshire, England).

Steady-state FRET analysis
Sixty-four-mer oligonucleotides corresponding to the VEGF HRE were labeled with the donor fluorophore, fluorescein, alone or in the presence of the acceptor fluorophore, Texas red at the positions indicated in Fig. 1 . Also shown in Fig. 1 is the position of a tetrahydrofuran introduced into the oligonucleotide to mimic the abasic site formed during hypoxic signaling. Steady-state fluorescence intensity measurements of the acceptor fluorophore were performed in a cuvette containing 1 ml of N-morpholino propanesulfonic acid (MOPS) buffer [1 M MOPS, 100 mM NaCl, and 5% glycerol (pH 7.5)] and 2 µg of the 64-mer oligonucleotide (Oligos Etc, Portland, OR, USA). The cuvette was placed in a fixed temperature (37°C) photon-counting spectrofluorometer (PTI Inc., Lawrenceville, NJ, USA), using a band pass of 3 nm on both the excitation and emission monochromators. FRET analyses were performed by excitation of fluorescein at 490 nm and detection of Texas red emission at 613 nm. Nuclear proteins from normoxic or hypoxic PAECs were added at the indicated concentrations to the cuvettes containing one of the two oligonucleotide sequences, and the cuvette was vortexed and incubated for 5 min before FRET detection. FRET signals, expressed as arbitrary units, were calculated as described previously (20) : the baseline Texas red emission intensity was subtracted from the increase in intensity after exposure to proteins, and this number was divided by the emission intensity of fluorescein and multiplied by 1000.

Lifetime FRET analysis
Fluorescence intensity decays were measured at 37°C in the time domain using 490 nm as excitation light from a LED (model 05B; IBH) light source with a 1-ns pulse duration and 100 kHz repetition rate. Fluorescence intensity decays of the donor, fluorescein, in the absence and presence of the acceptor, Texas red, were monitored at 520 nm and collected into 1024 channels of a multichannel analyzer (PCA3) under identical experimental conditions and used to calculate the means and distributions of the interfluorophore distances using an analytical approach described previously (26 27 28) . In brief, each fluorescence decay curve was fit to a single or multiexponential equation and resolved into components, and the decays of each component used to calculate interfluorophore distances using the Forster equation. Preliminary experiments demonstrated that anisotropies of the fluorophore-labeled WT and abasically modified oligonucleotides were identical, and the {kappa} value was assumed to be 2/3. Calculated interfluorophore distances were then fit to a Gaussian distribution to generate probability–interfluorophore distance curves in the absence of nuclear proteins and in the presence of nuclear proteins from normoxic and hypoxic PAECs. From these curves, the value of the "half-width" reflecting the degree of oligonucleotide flexibility was determined.

It should be noted that the precise intersite distances reported herein cannot be determined from our analysis. As indicated above, although the value of {kappa} is unknown, preliminary experiments demonstrated that anisotropies of the fluorophore-labeled WT and abasically modified oligonucleotides were identical. Thus, comparisons of calculated intersite distances between the two oligonucleotides is valid, but the interfluorophore distances calculated from the Forster equation must be considered as estimates until the actual value of {kappa} is determined.

Oligonucleotide cleavage assay
Endonuclease activity present in normoxic and hypoxic PAEC nuclear extracts was determined using an adaptation of methods described previously (29) . In brief, nuclear extracts from normoxic PAECs and PAECs cultured in hypoxia for 3 h were prepared as described above and applied to duplex WT and abasically modified oligonucleotides corresponding to the VEGF HRE. Either the WT, abasically modified, or complementary strands were end-labeled with 32P using T4 kinase. In some experiments, Ref-1/Ape1 was immunodepleted from the nuclear extracts using the protocol described above before their addition to the 32P-labeled oligonucleotides. After incubation with nuclear protein extracts for 5 min at 37°C, samples were loaded on a 20% acrylamide-8 M urea gel and electrophoresed to separate intact, single-strand oligonucleotides from cleavage fragments.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Nuclear protein binding to the fluorophore-labeled VEGF HRE
We and others reported that a functionally active 64-bp sequence of the VEGF HRE associates with the transcription factor, HIF-1, along with the coactivator proteins p300/CBP, ATF/CREB, and Ref-1/Ape1 (1 2 3) . Because we now wished to determine whether these and potentially other nuclear proteins altered the topology of the VEGF HRE, it was first necessary to confirm that the presence of the two fluorophores required for FRET analyses (Fig. 1) did not alter nuclear protein-DNA interactions. Accordingly, DNA affinity precipitation was applied to magnetic bead-tagged oligonucleotide sequences of the WT and abasically modified VEGF HRE, either with or without the two fluorophores ligated to the sequences at the indicated positions. The proteins eluted from the oligonucleotide probes were then either enumerated from silver-stained gels, or Western immunoblot analyses were used to identify selected components of the transcriptional complex. As shown in Fig. 2 , inspection of the silver-stained gel reveals that a similar number, ~25, and pattern of proteins present in either normoxic or hypoxic PAEC nuclear extract are associated with the nonlabeled and fluorescently labeled oligonucleotide probes. The Western immunoblot analysis, also shown in Fig. 2 , indicates that both the fluorescently labeled and nonlabeled oligonucleotide probes associated with ATF/CREB, p300, and Ref-1/Ape1 and, in hypoxic nuclear extract, HIF-1. Differences in the association of these proteins between the nonlabeled and fluorophore-labeled probes were generally modest. It is noteworthy that, as previously reported (11) , the presence of an abasic site at the hypoxia-modified guanine was accompanied by increased binding of HIF-1 and that this occurred with both the WT and fluorophore-labeled oligonucleotide probes.


Figure 2
View larger version (17K):
[in this window]
[in a new window]

  		Figure 2. DNA affinity precipitation analyses of nuclear proteins derived from normoxic and hypoxic PAECs binding to WT and fluorophore-labeled 64-mer oligonucleotides corresponding to the hypoxic response element of the rat VEGF gene. Proteins eluting from the oligonucleotide were enumerated from silver-stained gels (A), whereas specific transcription activators and coactivators were detected by Western immunoblot (B). Results are representative of at least 4 experiments.

Steady-state FRET analyses
The baseline steady-state FRET values, expressed in arbitrary units, for the WT and abasically modified probes incubated in the absence of nuclear proteins were 10 ± 1.5 and 15 ± 1, respectively. Subsequent studies were performed to determine the impact of nuclear proteins derived from PAECs under normoxic and hypoxic conditions on the steady-state FRET response of the fluorophore-tagged VEGF HRE probes. FRET intensity of the WT probe increased slightly when exposed to nuclear extracts isolated from both normoxic and hypoxic cells (Fig. 3 A). The abasically modified FRET probe exhibited an increased FRET response in comparison to the WT sequence when exposed to either normoxic or hypoxic nuclear extracts (Fig. 3B ). In addition, the FRET response of the abasically modified probe to the nuclear proteins from normoxic PAECs was much greater than that evoked by nuclear proteins from hypoxic cells, thus suggesting that when associating with the abasically modified probe, the DNA binding proteins present in hypoxic nuclear extract are associated with a different change in global topology than that evoked by DNA binding proteins from normoxic nuclear extract.


Figure 3
View larger version (11K):
[in this window]
[in a new window]

  		Figure 3. Nuclear extracts from normoxic and hypoxic PAECs elicit different FRET responses in the WT (A) and abasic site-containing (B) VEGF HREs (n=5). Differences between the effects of normoxic and hypoxic nuclear extract on steady-state FRET responses of both the WT and abasically-modified HREs were significant at P < 0.05. A.U. = arbitrary units.

A critical transcription factor driving VEGF expression in hypoxia is HIF-1 (1 , 30) . Nevertheless, even in normoxia where HIF-1 abundance is low or undetectable in most cells, nuclear proteins bind to the HIF-1-regulated erythropoietin enhancer element in vivo (31) and to an oligonucleotide sequence corresponding to the VEGF HRE in vitro (1 , 2) . To determine whether HIF-1 is responsible for the FRET responses observed in WT and abasically modified VEGF HREs, the transcription factor was depleted from nuclear extract using a HIF-1-specific antibody or its binding to the model HREs was suppressed using a decoy oligonucleotide. Results with these two HIF-1 inhibitory strategies were identical, and only the outcomes of the immunodepletion experiments are shown. The DNA affinity precipitation analysis depicted in Fig. 4 verified that antibody-mediated HIF-1 depletion both suppressed HIF-1 binding to the VEGF HRE while preserving associations between other DNA binding proteins and the sequence. The outcome of steady-state FRET experiments, also shown in Fig. 4 , indicated that when HIF-1 is immunodepleted from normoxic PAEC nuclear extract, the curves depicting the steady-state FRET responses as a function of nuclear protein concentrations were shifted slightly to the right for both the WT and abasically modified VEGF HRE oligonucleotide. However, when hypoxic PAEC nuclear extract was used, HIF-1 depletion nearly abolished nuclear protein-dependent FRET on both of the oligonucleotide probes.


Figure 4
View larger version (16K):
[in this window]
[in a new window]

  		Figure 4. A) Western immunoblot analysis, representative of at least 4 experiments, showing that in comparison to control nuclear extracts, immunodepletion of HIF-1 (–HIF1) from normoxic (N) or hypoxic (H) PAEC nuclear extracts suppressed incorporation of HIF-1 in the hypoxia-inducible transcription complex forming on both the WT and abasically modified (abasic) VEGF HRE. Other components of the transcriptional complex, including p300, CREB, ATF, and Ref-1/Ape1, were largely unaffected. Summary data (n=5) showing that FRET responses induced in WT (B) and abasically modified (C) VEGF HREs by normoxic and hypoxic nuclear extracts are significantly (P<0.05) right-shifted or depressed as a result of HIF-1 immunodepletion. A.U. = arbitrary units.

The bifunctional transcriptional coactivator and DNA repair enzyme, Ref-1/Ape1, is critical to the formation of the hypoxia-inducible transcriptional complex on the VEGF HRE, possibly because it functions in part to preassemble components of the transcriptional complex before association with the HRE (2 , 3) . We sought to determine whether Ref-1/Ape1 also played an important role in FRET responses of the WT VEGF HRE and especially in the abasically modified sequence, where Ref-1/Ape1 interactions with the tetrahydrofuran might be expected to distort oligonucleotide structure and/or cleave the sequence at the abasic site. Ref-1/Ape1 was immunodepleted from both normoxic and hypoxic nuclear extracts and steady-state FRET studies were performed. As shown in Fig. 5 , immunodepletion of Ref-1/Ape1 prevented interactions between the probe and all of the selected protein components of hypoxic transcriptional complex, with the exception of a modest amount of Ref-1/Ape1 binding to both the WT and abasically modified probes. In addition, Fig. 5 also shows that Ref-1/Ape1 immunodepletion suppressed FRET responses of both the WT and abasically modified probes in the presence of either normoxic or hypoxic nuclear proteins.


Figure 5
View larger version (14K):
[in this window]
[in a new window]

  		Figure 5. A) Western immunoblot analyses, representative of at least 4 experiments, showing that in comparison to control nuclear extracts, immunodepletion of Ref-1/APE1 (–Ref-1/Ape1) from normoxic (N) or hypoxic (H) PAEC nuclear extracts inhibited transcription complex formation on both the WT and abasically-modified (abasic) VEGF HRE. Summary data (n=5) showing that FRET responses induced in WT (B) and abasically modified (C) VEGF HREs by normoxic and hypoxic nuclear extracts are significantly (P<0.05) depressed as a result of Ref-1/Ape1 immunodepletion. A.U. = arbitrary units.

Fluorescence lifetime analyses
The FRET responses noted above could be attributed to two general mechanisms; nuclear proteins could induce bending or flexibility of the oligonucleotide, resulting in changes in the intersite distances, or they could alter the fluorophore microenvironment, leading to a change in dipole moment and fluorescence responses unrelated to oligonucleotide bending. To discriminate between these possibilities and thus provide unambiguous evidence for bending or flexing of the VEGF HRE, fluorescence lifetime studies were performed.

In these experiments, the donor fluorophore, fluorescein, was excited with consecutive 1.0-ns pulses in the absence and presence of the acceptor fluorophore, Texas Red, and in the absence and presence of either normoxic or hypoxic PAEC nuclear extracts. Time courses of fluorescence after each excitatory pulse were then recorded and the rates of fluorescence decay of each of the individual components of the aggregate decay curves were used to calculate the energy transfer efficiencies from which distances between fluorophores were determined using the Forster equation. Individual interfluorophore distances determined after each excitatory pulse were then fit to a Gaussian distribution to derive the probabilities at which the donor and fluorophores were a given distance apart. These probability-fluorophore distance curves for the WT probe, displayed in Fig. 6 , show that in the absence of nuclear proteins, this sequence is characterized by a most probable interfluorophore distance of 59.46 Å with a half-width of 0.566 Å, reflecting relatively little oligonucleotide conformational variation (Table 1 ). Neither the general shape of the curve, the most probable interfluorophore distance, nor the half-width was altered by addition of nuclear proteins from either normoxic or hypoxic cells. The WT sequence is thus stiff and its association with nuclear proteins from either normoxic or hypoxic PAECs altered neither its flexibility nor the apposition between fluorophores. In the absence of nuclear proteins, the abasically modified VEGF HRE behaved similarly, displaying a most probable interfluorophore distance of 59.46 Å and minimal flexibility, characterized by a half-width of 0.662 Å (Fig. 6 , Table 1 ). However, the addition of nuclear proteins engendered considerable differences. The most probable interfluorophore distance was decreased slightly in the abasically modified HRE from 59.46 to 56.78 and 58.84 Å by normoxic and hypoxic nuclear protein extracts, respectively. However marked increases in sequence flexibility also were induced in the abasically modified VEGF HRE by normoxic and hypoxic nuclear protein extracts as evidenced by an increase in half-width to 9.9 Å.


Figure 6
View larger version (21K):
[in this window]
[in a new window]

  		Figure 6. Probability-distance curves describing the impact of nuclear proteins from normoxic (Norm.) and hypoxic (Hyp.) PAECs on intersite distances between donor and acceptor fluorophores tethered to WT and abasically modified VEGF HRE oligonucleotide (Oligo.). Note the similarities between WT and abasically modified (Abasic) oligonucleotides in the absence of nuclear proteins, and prominent dispersions of distances induced by nuclear proteins selectively in the abasically modified oligonucleotide. See text for details.


View this table:
[in this window]
[in a new window]

  		Table 1. Intersite distances and flexibility (half-width) in WT and abasic VEGF HREs in the absence and presence of nuclear proteins from normoxic and hypoxic PAECs

Ref-1/Ape1-dependent DNA strand cleavage
Because abasic sites are believed to introduce torsional flexibility into model DNA sequences (15 16 17) , we were surprised to observe that the abasically modified oligonucleotide was relatively stiff in the absence of nuclear proteins, adopting essentially the same conformational profile as the WT oligonucleotide (Fig. 6) . However, the substantial increase in flexibility induced in the abasically modified oligonucleotide by addition of nuclear proteins suggested that one or more of the bound proteins may modulate flexibility by altering integrity of the DNA strand. In this regard, one possibility is that the abasic site is converted to a single-strand break by the action of Ref-1/Ape1, and this in turn is responsible for the marked increase in conformational instability caused by nuclear protein binding. To address this idea, we performed endonuclease cleavage assays using the WT and abasically modified duplex oligonucleotides 5'-labeled with 32P. In initial experiments, electrophoresis on an acrylamide/urea gel indicated that the both WT and abasically modified oligonucleotides migrated as single, intact bands, thus indicating that there was no degradation of either probe in solution (data not shown). In addition and as shown in Fig. 7 , incubation of the WT oligonucleotide for 5 min with nuclear extract from normoxic or hypoxic PAECs failed to alter its integrity. However, there was substantial cleavage of the oligonucleotide harboring the abasic site, with slightly more product formed when the oligonucleotide was incubated with hypoxic nuclear extract relative to normoxic extract. Importantly, formation of the cleavage product was suppressed when Ref-1/Ape1 was immunodepleted from the nuclear extract. The ability of Ref-1/Ape1 immunodepletion to attenuate single-strand break formation was similar to the inhibitory effect of this immunodepletion strategy on steady-state FRET responses to nuclear extracts (see Fig. 5 ). To verify that a single-strand and not a double-strand break was created, identical experiments were performed in which the complementary DNA strand was 32P-labeled. Under these conditions, no cleavage product was detected regardless of whether the duplex oligonucleotide was incubated with normoxic or hypoxic PAEC nuclear extract (data not shown).


Figure 7
View larger version (12K):
[in this window]
[in a new window]

  		Figure 7. Autoradiogram, representative of 4 experiments, showing results of an oligonucleotide cleavage assay to detect single-strand breaks in the model VEGF HREs. The substrate consisted of the 64-mer oligonucleotide probes shown in Fig. 1 , end-labeled with 32P. After incubation for 5 min with nuclear extract from normoxic (N) or hypoxic (H) PAECs, cleavage products are detected only when the abasically modified oligonucleotide (abasic), but not the WT oligonucleotide, was the substrate. Immunodepletion of Ref-1/Ape1 before application to the 32P-labeled oligonucleotide (–Ref-1/Ape1) attenuated formation of the cleavage product.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
The intrinsic flexibility of promoter DNA sequences may impact transcriptional regulation in at least two general ways. First, transcription factors may induce bending of the promoter as an early event in transcriptional initiation, perhaps as a means of bringing regulatory proteins bound to nearby or remote sequences in closer proximity (12) . Second, more recent studies by Ju and coworkers have shown that regulated transcription requires a topoisomerase IIβ-mediated formation of a double-strand DNA break that may serve to relax the close association between promoter DNA and the nucleosome core particle, thereby facilitating access to the promoter by chromatin remodeling enzymes (13 , 14) . Against this background, previous findings in PAECs that hypoxia and other signals using reactive oxygen species as second messengers cause reversible, nucleotide-specific oxidative modifications in functionally relevant promoter sequences (10 , 11) suggest another mechanism by which such critical changes in DNA flexibility could be dynamically introduced into otherwise stiff DNA sequences. The intent of the present studies was to provide proof-of-concept for this hypothesis using a duplex oligonucleotide corresponding to the HRE of the VEGF promoter harboring a model abasic site at a key, hypoxic-modified guanine in the HIF-1 DNA recognition sequence.

Minimal impact of fluorophore labeling on nuclear protein binding
Before we used the FRET method to determine the effects of nuclear proteins on sequence flexibility, it was necessary to verify that the presence of the two fluorophores did not influence nuclear protein binding to the labeled oligonucleotides. We found that the two fluorophores, ligated into the WT and abasically modified DNA duplexes at positions 3' to the HIF-1 DNA recognition sequence, did not qualitatively alter nuclear protein association with the oligonucleotide. Silver-stained gels revealed that ~25 proteins precipitated with both oligonucleotides after incubation with either normoxic or hypoxic nuclear extract, with no differences in the position or intensities of any of the protein bands. Western blot analyses indicated that the fluorophore-labeled oligonucleotides associated with the same transcription factor, HIF-1 and coactivators, p300, ATF/CREB, and Ref-1/Ape1, as the unlabeled oligonucleotide probes, with minor changes in the relative abundances of ATF/CREB and Ref-1/Ape1. Importantly and similar to our previous finding, the abasically modified probes associated with more HIF-1 and more Ref-1/Ape1 than the WT despite the presence of the two fluorophores. Viewed collectively, these findings demonstrate that introduction of fluorescein and Texas red required for FRET detection of oligonucleotide bending and flexing does not markedly distort nuclear protein binding to the probes.

Critical roles of HIF-1 and Ref-1/Ape1 in steady-state FRET responses
Application of nuclear proteins to the WT oligonucleotide resulted in modest protein concentration-related increases in FRET that did not differ substantially between normoxic and hypoxic nuclear extract. In contrast, the abasically modified oligonucleotide displayed large increases in steady-state FRET when incubated with normoxic nuclear protein extract. Increases associated with hypoxic nuclear proteins were lesser in magnitude, but still somewhat greater than those occurring on the WT oligonucleotide.

HIF-1 seemed to play a key role in steady-state FRET responses mediated by PAEC nuclear extract binding to the oligonucleotide probes. The two methods used to suppress HIF-1 DNA binding—immunodepletion of HIF-1 from nuclear extract and a HIF-1 decoy oligonucleotide—both caused relatively selective reductions in HIF-1 association with the oligonucleotide probes without modifying probe associations with the other transcriptional coactivators, p300, ATF/CREB, and Ref-1/Ape1. Suppression of HIF-1 binding to the probes caused slight rightward shifts in the nuclear protein concentration–steady-state FRET curves when normoxic nuclear extracts were used and nearly abolished FRET responses to hypoxic PAEC nuclear extract. The impact of HIF-1 inhibition was similar between the WT and abasically modified model HREs. Differences in the magnitude of the HIF-1 inhibitory effects observed between normoxic and hypoxic nuclear extract probably reflect the fact that HIF-1 abundance is low in normoxia and markedly elevated in hypoxia. More importantly, these observations indicate that steady-state FRET responses of both the WT and abasically modified oligonucleotide probes appear to have a specific requirement for DNA binding of the key transcription factor governing gene expression in hypoxia, HIF-1.

Because of its involvement in assembly of the hypoxic transcriptional complex (2 , 3) , we were interested in the contribution of Ref-1/Ape1 to PAEC nuclear protein-induced FRET responses in the model HREs. The Ref-1 domain of this bifunctional protein is responsible for redox activation of a number of transcription factors, including HIF-1, whereas the Ape1 domain is crucial for the second step in the base excision pathway repairing oxidative DNA damage (18) . Exactly how Ape1 "surveys" DNA for the presence of abasic sites remains somewhat speculative, but it is believed to search for regions of increased flexibility (32) . Direct, persistent DNA binding does not appear to be critical to this survey function. Once an abasic site is detected, the Ape1 domain of Ref-1/Ape1 inserts loops into the DNA base stack, creating an approximate 35° bend in the sequence, and stabilizes the DNA such that the site is "flipped" or projected into a sequestered enzyme pocket before strand cleavage (33) . Because of the different mechanisms by which the two activities of Ref-1/Ape1 interact with undamaged and abasically modified DNA, we were alert for the possibility that the contribution of Ref-1/Ape1 to FRET responses of the two oligonucleotides might differ. This does not appear to be the case. Immunodepletion of Ref-1/Ape1 from nuclear extract before its application to the oligonucleotide probe eliminated or suppressed association with the known components of the transcriptional complex and attenuated steady-state FRET responses of both the WT and abasically modified model HREs to nuclear protein extract from both normoxic and hypoxic PAECs. This observation supports the idea that Ref-1/Ape1 and the transcriptional proteins with which it interacts are important for steady-state FRET responses of the fluorophore-labeled oligonucleotides, but it leaves open the question of how the survey function and the DNA binding and repair activities of the Ape1 domain contribute to oligonucleotide structure.

Nuclear protein-induced changes in sequence flexibility
Steady-state FRET responses of the WT and abasically modified oligonucleotides evoked by PAEC nuclear extracts could be caused by changes in fluorophore apposition or by factors unrelated to oligonucleotide topology. The outcome of fluorescence lifetime analysis of the WT probe showed that neither the most probable interfluorophore distance nor the range of interfluorophore distances—the half-widths—were altered by nuclear protein binding. The WT sequence is thus stiff, and the steady state FRET responses observed herein cannot be ascribed to distortion of oligonucleotide probe topology. That such a short oligonucleotide sequence would exhibit such behavior is consistent with classic studies holding that double-stranded DNA behaves as a semiflexible polymer characterized by a bending-persistence length P = 50 nm (~150 bp). Thus, whereas DNA sequences longer than P would be expected to be flexible and require little force to bend more dramatically, sequences shorter than P would require substantial force to induce formation of multiple conformational states (34 35 36 37) . Evidently, binding of the hypoxia-inducible transcriptional complex to the WT HRE is incapable of overcoming the intrinsic stiffness of this short DNA sequence to distort its local topology.

In the absence of nuclear proteins, the abasically modified and WT oligonucleotides displayed similar interfluorophore distances and half-widths. Thus, despite the presence of a model abasic site believed to introduce torsional flexibility in DNA sequences, the WT and abasically modified oligonucleotides adopt similarly constrained conformations. It is known, however, that the extent to which an abasic lesion modifies sequence flexibility depends on the nature of the abasic site and the surrounding base sequences (15) . In this regard, modeling bendability of the VEGF HRE (11) revealed that the guanine oxidatively modified in hypoxia resides in the stiffest region of the HRE. Thus, local sequence context may minimize the impact of the model abasic site on sequence flexibility in the resting state.

In marked contrast to its behavior in the absence of nuclear proteins, the abasically modified oligonucleotide displayed large increases in local sequence flexibility when incubated with nuclear proteins from either normoxic or hypoxic PAECs, with increases in the half-width from ~0.5 to almost 10 Å. Nuclear protein binding to the abasically modified probe also may have caused slight alterations in interfluorophore distances, decreasing the most probable intersite distance by a modest extent when normoxic nuclear proteins bound and widening the interfluorophore distance toward the protein-free conditions when hypoxic nuclear extract was applied. These findings indicate that increases in steady-state FRET responses in the abasically modified oligonucleotide probe evoked by normoxic and hypoxic PAEC nuclear proteins are caused by changes in local sequence topology.

Why does the abasically modified probe display such impressive nuclear protein-induced increases in flexibility? It is not a property of the abasic site per se because, in the absence of nuclear proteins, the WT and abasically modified probes adopt the same, stiff conformational profiles. The flexibility of the abasic probe also cannot be ascribed to the proteins that associate with it, as they are qualitatively the same as those associating with the WT probe. Our data suggest that the marked increase in nuclear protein-induced flexibility of the abasically modified HRE can be attributed to the Ref-1/Ape1-mediated formation of a single-strand break at the abasic site, a documented action of the Ape1 domain that plays a critical role in the base excision pathway repairing oxidative DNA damage (18) . Importantly, formation of the single-strand break does not prevent DNA-protein interactions; rather, binding of HIF-1, believed to occur in the immediate vicinity of the Ref-1/Ape1-mediated break at the abasic site, appears to be increased. It is also interesting to note that the donor and acceptor fluorophores are both positioned 3' to the probable break site at the 3' end of the HIF-1 DNA recognition sequence (see Fig. 1 ), thus indicating that a single-strand break at this key, hypoxia-modified nucleotide impacts the flexibility of downstream sequences of the functional HRE.

Biological implications of this study
We previously found that hypoxia and other stimuli using reactive oxygen species as second messengers caused oxidative modifications at specific nucleotides in the VEGF promoter (10 , 11) . The formation and elimination of these oxidative modifications coincided with a sharp hypoxia-induced increase followed by a sustained, lower-level elevation in VEGF mRNA expression. The most frequently modified nucleotide was the 3' guanine of the HIF-1 DNA recognition sequence in the VEGF HRE. Because the modification was revealed by treatment of the DNA with alkali, we speculated that the detected base oxidation product was an abasic site formed after loss or removal of the precursor lesion, 8-oxoguanine. The biological consequences of such an oxidant attack on a nucleotide within the VEGF HRE remain speculative. However, we found that incorporation of the abasic site analog, tetrahydrofuran, at the hypoxia-modified guanine in an oligonucleotide corresponding to the VEGF HRE increased the abundance of HIF-1 and Ref-1/Ape1 in the transcriptional complex and increased reporter gene expression. These observations underscored the intriguing possibility that introduction of a base modification product at this location in the VEGF promoter, which occurs during hypoxia, may play a role in governing VEGF expression. To this general model, the present findings suggest that an abasic site at the modified guanine may be a substrate for the formation of Ref-1/Ape1-dependent single-strand DNA break which, in turn, causes marked increases in local sequence flexibility. The increased sequence flexibility could have at least two important and interrelated consequences for regulation of gene expression (38) . First, the increased sequence flexibility could reduce the force needed by the transcriptional complex to overcome the intrinsic stiffness of the hypoxic response element, thereby facilitating alterations in local sequence topology necessary for transcriptional initiation. Second, and similar to the model proposed for the estrogen response element wherein a topoisomerase II-mediated double-strand DNA break is required for regulated transcription (13 , 14) , the enhanced local sequence flexibility may serve to unwrap or relax promoter DNA from its normally compacted state in the nucleosome, thereby reducing steric obstruction to transcription factor binding. Determining exactly how oxidative base modifications within promoter sequences impact gene expression will require additional study and could be facilitated by the FRET-based reporter system reported herein.


   ACKNOWLEDGMENTS
 
The contribution of Drs. H. Cheung and W. Dong of the University of Alabama at Birmingham to the fluorescence lifetime studies reported herein is gratefully appreciated. In addition, the expert technical assistance of Ms. Gina C. Capley is acknowledged. This investigation was supported in part by grants from the National Institutes of Health (RO1 HL058234, RO1 HL073244, R21 HL084521, and PO1 HL066299).

Received for publication January 22, 2007. Accepted for publication August 2, 2007.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Ebert, B. L., Bunn, H. F. (1998) Regulation of transcription by hypoxia requires a multiprotein complex that includes hypoxia-inducible factor 1, an adjacent transcription factor, and p300/CREB binding protein. Mol. Cell. Biol. 18,4089-4096[Abstract/Free Full Text]
  2. Ziel, K. A., Campbell, C. C., Wilson, G. L., Gillespie, M. N. (2004) Ref-1/Ape is critical for formation of the hypoxia-inducible transcriptional complex on the hypoxic response element of the rat pulmonary artery endothelial cell VEGF gene. FASEB J. 18,986-988[Abstract/Free Full Text]
  3. Gray, M. J., Zhang, J., Ellis, L. M., Semenza, G. L., Evans, D. B., Watowich, S. S., Gallick, G. E. (2005) HIF-1{alpha}, STAT3, CBP/p300 and Ref-1/APE are components of a transcriptional complex that regulates Src-dependent hypoxia-induced expression of VEGF in pancreatic and prostate carcinomas. Oncogene 24,3110-3120[CrossRef][Medline]
  4. Levy, A. P., Levy, N. S., Wegner, S., Goldberg, M. A. (1995) Transcriptional regulation of the rat vascular endothelial growth factor gene by hypoxia. J. Biol. Chem. 270,13333-13340[Abstract/Free Full Text]
  5. Killilea, D. W., Hester, R., Balczon, R., Babal, P., Gillespie, M. N. (2000) Free radical production in hypoxic pulmonary artery smooth muscle cells. Am. J. Physiol. 279,L408-L412
  6. Chandel, N. S., Maltepe, E., Goldwasser, E., Mathieu, C. E., Simon, M. C., Schumacker, P. T. (1998) Mitochondrial reactive oxygen species trigger hypoxia-induced transcription. Proc. Natl. Acad. Sci. U. S. A. 95,11715-11720[Abstract/Free Full Text]
  7. Liu, J. Q., Zelko, I. N., Erbynn, E. M., Sham, J. S., Folz, R. J. (2006) Hypoxic pulmonary hypertension: role of superoxide and NADPH oxidase (gp91phox). Am. J. Physiol. 290,L2-L10
  8. Lin, M. J., Leung, G. P., Zhang, W. M., Yang, X. R., Yip, K. P., Tse, C. M., Sham, J. S. (2004) Chronic hypoxia-induced upregulation of store-operated and receptor-operated Ca2+ channels in pulmonary arterial smooth muscle cells: a novel mechanism of hypoxic pulmonary hypertension. Circ. Res. 95,496-505[Abstract/Free Full Text]
  9. Liu, J. Q., Sham, J. S., Shimoda, L. A., Kuppusamy, P., Sylvester, J. T. (2003) Hypoxic constriction and reactive oxygen species in porcine distal pulmonary arteries. Am. J. Physiol. 285,L322-L333
  10. Grishko, V., Solomon, M., Breit, J. F., Killilea, D. W., Ledoux, S. P., Wilson, G. L., Gillespie, M. N. (2001) Hypoxia promotes oxidative base modifications in the pulmonary artery endothelial cell VEGF gene. FASEB J. 15,1267-1269[Free Full Text]
  11. Ziel, K. A., Grishko, V., Campbell, C. C., Breit, J. F., Wilson, G. L., Gillespie, M. N. (2005) Oxidants in signal transduction: impact on DNA integrity and gene expression. FASEB J. 19,387-394[Abstract/Free Full Text]
  12. Ptashne, M. (1986) Gene regulation by proteins acting nearby and at a distance. Nature 322,697-701[CrossRef][Medline]
  13. Haince, J. F., Rouleau, M., Poirier, G. G. (2006) Transcription: gene expression needs a break to unwind before carrying on. Science 312,1752-1753[Abstract/Free Full Text]
  14. Ju, B. G., Lunyak, V. V., Perissi, V., Garcia-Bassets, I., Rose, D. W., Glass, C. K., Rosenfeld, M. G. (2006) A topoisomerase IIβ-mediated dsDNA break required for regulated transcription. Science 312,1798-1802[Abstract/Free Full Text]
  15. Ayadi, L., Coulombeau, C., Lavery, R. (2000) The impact of abasic sites on DNA flexibility. J. Biomol. Struct. Dyn. 17,645-653[Medline]
  16. Barsky, D., Foloppe, N., Ahmadia, S., Wilson, D. M., 3rd, MacKerell, A. D., Jr (2000) New insights into the structure of abasic DNA from molecular dynamics simulations. Nucleic Acids Res. 28,2613-2626[Abstract/Free Full Text]
  17. Cheng, X., Kelso, C., Hornak, V., de los Santos, C., Grollman, A. P., Simmerling, C. (2005) Dynamic behavior of DNA base pairs containing 8-oxoguanine. J. Am. Chem. Soc. 127,13906-13918[CrossRef][Medline]
  18. Evans, A. R., Limp-Foster, M., Kelley, M. R. (2000) Going APE over ref-1. Mutat. Res. 461,83-108[Medline]
  19. Hays, J. B., Zimm, B. H. (1970) Flexibility and stiffness in nicked DNA. J. Mol. Biol. 48,297-317[CrossRef][Medline]
  20. Berney, C., Danuser, G. (2003) FRET or no FRET: a quantitative comparison. Biophys. J. 84,3992-4010[Medline]
  21. Parkhurst, L. J. (2004) Distance parameters derived from time-resolved Forster resonance energy transfer measurements and their use in structural interpretations of thermodynamic quantities associated with protein-DNA interactions. Methods Enzymol. 379,235-262[CrossRef][Medline]
  22. Williams, S. L., Parkhurst, L. K., Parkhurst, L. J. (2006) Changes in DNA bending and flexing due to tethered cations detected by fluorescence resonance energy transfer. Nucleic Acids Res. 34,1028-1035[Abstract/Free Full Text]
  23. Hardwidge, P. R., Parkhurst, K. M., Parkhurst, L. J., Maher, L. J., 3rd (2003) Reflections on apparent DNA bending by charge variants of bZIP proteins. Biopolymers 69,110-117[CrossRef][Medline]
  24. Parkhurst, L. J., Parkhurst, K. M., Powell, R., Wu, J., Williams, S. (2001) Time-resolved fluorescence resonance energy transfer studies of DNA bending in double-stranded oligonucleotides and in DNA-protein complexes. Biopolymers 61,180-200[Medline]
  25. Yang, D. I., Chen, S. D., Yang, Y. T., Ju, T. C., Xu, J. M., Hsu, C. Y. (2004) Carbamoylating chemoresistance induced by cobalt pretreatment in C6 glioma cells: putative roles of hypoxia-inducible factor-1. Br. J. Pharmacol. 141,988-996[CrossRef][Medline]
  26. Robinson, J. M., Dong, W. J., Cheung, H. C. (2003) Can Forster resonance energy transfer measurements uniquely position troponin residues on the actin filament? A case study in multiple-acceptor FRET. J. Mol. Biol. 329,371-380[CrossRef][Medline]
  27. Dong, W. J., Robinson, J. M., Stagg, S., Xing, J., Cheung, H. C. (2003) Ca2+-induced conformational transition in the inhibitory and regulatory regions of cardiac troponin I. J. Biol. Chem. 278,8686-8692[Abstract/Free Full Text]
  28. She, M., Dong, W. J., Umeda, P. K., Cheung, H. C. (1998) Tryptophan mutants of troponin C from skeletal muscle–an optical probe of the regulatory domain. Eur. J. Biochem. 252,600-607[Medline]
  29. Ruchko, M., Gorodnya, O., LeDoux, S. P., Alexeyev, M. F., Al-Mehdi, A. B., Gillespie, M. N. (2005) Mitochondrial DNA damage triggers mitochondrial dysfunction and apoptosis in oxidant-challenged lung endothelial cells. Am. J. Physiol. 288,L530-L535
  30. Semenza, G. L. (1998) Hypoxia-inducible factor 1: master regulator of O2 homeostasis. Curr. Opin. Genet. Dev. 8,588-594[CrossRef][Medline]
  31. Hu, B., Wright, E., Campbell, L., Blanchard, K. L. (1997) In vivo analysis of DNA-protein interactions on the human erythropoietin enhancer. Mol. Cell. Biol. 17,851-856[Abstract/Free Full Text]
  32. Wilson, D. M., 3rd, Barsky, D. (2001) The major human abasic endonuclease: formation, consequences and repair of abasic lesions in DNA. Mutat. Res. 485,283-307[Medline]
  33. Mol, C. D., Hosfield, D. J., Tainer, J. A. (2000) Abasic site recognition by two apurinic/apyrimidinic endonuclease families in DNA base excision repair: the 3' ends justify the means. Mutat. Res. 460,211-229[Medline]
  34. Zhang, Y., Crothers, D. M. (2003) Statistical mechanics of sequence-dependent circular DNA and its application for DNA cyclization. Biophys. J. 84,136-153[Medline]
  35. Crothers, D. M., Drak, J., Kahn, J. D., Levene, S. D. (1992) DNA bending, flexibility, and helical repeat by cyclization kinetics. Methods Enzymol. 212,3-29[Medline]
  36. Vologodskaia, M., Vologodskii, A. (2002) Contribution of the intrinsic curvature to measured DNA persistence length. J. Mol. Biol. 317,205-213[CrossRef][Medline]
  37. Hagerman, P. J., Ramadevi, V. A. (1990) Application of the method of phage T4 DNA ligase-catalyzed ring-closure to the study of DNA structure. I. Computational analysis. J. Mol. Biol. 212,351-362[CrossRef][Medline]
  38. Gillespie, M. N., Wilson, G. L. (2007) Bending and breaking the code: dynamic changes in promoter integrity may underlie a new mechanism regulating gene expression. Am. J. Physiol. 292,L1-L3




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow All Versions of this Article:
fj.07-8102comv1
22/1/19    most recent
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 PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Breit, J. F.
Right arrow Articles by Gillespie, M. N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Breit, J. F.
Right arrow Articles by Gillespie, M. N.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS