Advertisement
JBC

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


     

Originally published In Press as doi:10.1074/jbc.M313348200 on January 20, 2004

J. Biol. Chem., Vol. 279, Issue 15, 14694-14702, April 9, 2004
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/15/14694    most recent
M313348200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
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 arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Surguladze, N.
Right arrow Articles by Fried, M. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Surguladze, N.
Right arrow Articles by Fried, M. G.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Interactions and Reactions of Ferritin with DNA*

Nodar Surguladze{ddagger}, Khristy M. Thompson{ddagger}, John L. Beard§, James R. Connor{ddagger}, and Michael G. Fried¶||

From the Departments of {ddagger}Neural and Behavioral Science and Biochemistry and Molecular Biology, the Pennsylvania State University College of Medicine, Hershey, Pennsylvania 17033 and the §Department of Nutritional Sciences, the Pennsylvania State University, University Park, Pennsylvania 16802

Received for publication, January 12, 2004 , and in revised form, January 12, 2004.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 APPENDIX
 REFERENCES
 
Ferritin, normally considered a cytoplasmic iron-storage protein, is also found in the nuclei of some cells. There is no current agreement about its function(s) in this environment. Proposals include DNA protection, provision of iron to nuclear enzymes, and regulation of transcription initiation, but evidence for these functions is scanty. We have shown previously that H-ferritin subunits can be cross-linked to chromosomal DNA in vivo (Thompson, K. J., Fried, M. G., Ye, Z., Boyer, P., and Connor, J. R. (2002) J. Cell Sci. 115, 2165–2177). Here we describe systematic analyses of DNA binding and the covalent stability of DNA in the presence of ferritins from several different sources. Our data show that the H-subunit of human ferritin binds DNA, whereas neither the L-subunit nor the ferroxidase-deficient 222-mutant of the H-subunit has detectable binding activity. DNA binding is without significant preference for base composition, sequence, or the nature of DNA ends. H- and L-ferritins and ferritins of mixed subunit composition stimulate the conversion of superhelical plasmid DNA to the relaxed form. The sensitivity of this conversion to glycerol suggests that DNA is nicked by a free radical mechanism. The rate of nicking correlates with the iron content of the ferritin and is strongly inhibited by chelators. Ferritin-dependent nicking is characterized by a kinetic lag that is not seen in control reactions containing free iron species. These results suggest that the release of iron from ferritin is an important part of the nicking mechanism. The potential role of ferritin as a protector of the genome is discussed in the context of these results.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 APPENDIX
 REFERENCES
 
Ferritin is a cytoplasmic protein that binds and sequesters iron. Intranuclear iron-positive inclusions discovered to be ferritin have been found in hepatocytes following iron overload conditions in mice (1), rats (2), and baboons (3). Intranuclear ferritin-like particles have been described in human liver cells, bone marrow macrophages, reticular cells, and muscle and nerve cells in a number of pathological conditions such as pigmentary cirrhosis and chronic progressive motor disturbance (4, 5). We have shown that ferritin is present in neuronal but not glial nuclei during postnatal development, following hypoxic ischemic insult (6), and in brain tumor cells (7). We have also demonstrated that the amount of ferritin in the nucleus can be altered by changes in cellular iron status and/or oxidative stress (7).

Although there is increasing acceptance of the concept that ferritin is present in cell nuclei, there is little agreement about its function in this environment. Ferritin can offer protection from UV-induced DNA damage in nuclei of avian corneal epithelial cells (8–10), but other studies implicate it as a source of hydroxyl radical production, either through the oxidation of iron following sequestration or after iron release from the molecule (11–13). In vivo cross-linking indicates that H-ferritin subunits have access to chromosomal DNA (7), whereas in vitro results demonstrate that some ferritins and ferritin-like proteins bind directly to DNA (14–16). Some of these studies indicate that ferritin can bind sequences regulating the expression of globin genes; this has given rise to the proposal that ferritin binding regulates globin transcription (14, 17, 18).

In addition to binding DNA, ferritin could serve to deliver iron to the nucleus for iron-dependent enzyme or transcription factor activities (19, 21, 22). On the other hand, DNA binds iron with relatively high affinity (23), and there is substantial evidence that free radical species produced in the redox reactions of iron are capable of DNA damage (cf. Ref. 4). The metabolic demand for iron and the accompanying need to avoid the toxic consequences of its release may be met by proteins like ferritin, which bind iron reversibly and retain it in a relatively unreactive form. To test features of this model, we have examined the DNA interactions of several ferritins, and their abilities to protect DNA from iron-induced damage over prolonged periods. The results indicate that long term DNA protection depends sensitively on the solution conditions, among which are the oxidation state of the dominant iron species and the presence of species capable of oxidizing Fe(II) or reducing Fe(III). Together these data have the potential to reconcile conflicts in earlier results that bear on the ability of ferritin to protect the genome against iron-mediated damage.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 APPENDIX
 REFERENCES
 
Ferritins—The human recombinant ferritins used in this study (recombinant heavy chain ferritin (rH),1 recombinant light chain ferritin (rL), and the H-chain 222-mutant) were the generous gift of Paolo Arosio (Milan, Italy). The rH and rL proteins have been shown to have proper assembly, folding, and functional properties, whereas the mutant 222 protein has only 6.3% the ferroxidase activity of the rH protein (24). Crystalline bovine serum albumin, chymotrypsin, horse spleen apoferritin, horse spleen ferritin, and ovalbumin were obtained from Sigma.

Measurement of Iron Content of Ferritin Samples—Iron was measured by atomic absorption spectrophotometry after acid digestion of samples at 75 °C for 24 h (25). All samples were analyzed in triplicate.

DNA Samples—Oligodeoxyribonucleotides with sequences 5'-GATCCAACTCCTAAGCCAGTGCCAGAAGAGCCAAGGACAGGTG-3' and 5'-TCGTCACCTGTCCTTGGCTCTTCTGGCACTGGCTTAGGAGTT-3' were synthesized by the Pennsylvania State College of Medicine Macromolecular Core Facility. These DNAs were annealed to form a 43-bp duplex with a sequence identical to that of base pairs –164 to –128 in the human {beta}-globin promoter (17). This DNA was 3'-end-labeled with [32P]dCTP as described (26), and unincorporated products were removed with GD-25 spin columns (5 Prime -> 3 Prime, Inc., Boulder, CO). Binding competition analyses were performed with a mixture of HinfI, RsaI, and SinI fragments of plasmid pGEM-3, obtained from Promega (Madison, WI). The DNAs of this mixture were labeled with 32P at 5' termini as described previously (27). Plasmid pUC19 (lot 28) was purchased from New England Biolabs.

Electrophoretic Analyses—Submarine-format agarose gels (1.5% w/v) were cast and run in 40 mM Tris acetate, 1 mM EDTA (pH 8.0). Polyacrylamide gels (nominal concentration 4% w/v) were cast as described (28). Both gel and electrophoresis buffers contained 20 mM Tris acetate and 0.5 mM EDTA. Gels with non-radioactive DNA samples were stained with ethidium bromide (0.5 µg/ml) and photographed using a digital image capture system (Stratagene). Gels containing 32P-labeled DNA were visualized by autoradiography using Kodak XAR-5 film, exposed at 4 °C. Film densitometry was performed using a digital scanner, as described previously (29).

Supercoil Relaxation Assay—DNA backbone breakage was detected using a superhelical DNA relaxation assay (30). Supercoiled plasmid pUC19 DNA (31) was used as substrate. Reaction mixtures (20–30 µl) contained DNA (0.5 or 1 µg) and were dissolved in 10 mM Hepes (pH 7.5), 50 mM NaCl, 2.5 mM MgCl2, 2.5 mM DTT. Where indicated, reactions were carried out in buffer consisting of 10 mM Tris (pH 7.4 at 21 °C), 100 mM KCl. Ferritin, FeCl3, FeSO4, glycerol, and chelators (EDTA and EGTA) were added as needed. Reactions were quenched by addition of 10 µl of 50% glycerol, 50 mM EDTA, and 0.1% bromphenol blue. Following electrophoresis as described above, the mole fractions of supercoiled and relaxed forms were measured by integration of digitized gel images. DNA relaxation reactions following pseudo first-order kinetics were analyzed as shown in Equation 1,

(Eq. 1)
where A(t) is the concentration of supercoiled DNA at time t; A0 is its initial concentration; and k is the rate constant for the reaction.

Electrophoretic Mobility Shift Assays—DNA samples (2 nM of 43-bp {beta}-globin promoter fragment or 60 pM of 5'-ends when the pGEM fragment mixture was used) were dissolved in binding buffer (10 mM Hepes (pH 7.4 at 20 °C), 5 mM spermidine, 2.5 mM DTT, 50 mM NaCl, 50 µg/ml bovine serum albumin, 0.05 mg/ml poly(dI-dC), 0.2% Nonidet P-40). Recombinant H-, L-, or 222-ferritins were added to obtain desired final concentrations, and samples were equilibrated for 60 min at 21 ± 1 °C before resolution by PAGE (29).

Error Analysis—The estimation of errors of fitted parameters was performed by the method of Brodersen et al. (32). The parameters representing the least squares "best" fit to the data were multiplied by random variables with a mean value of 1.0, to generate 100–200 sets of parameters that gave fits to the experimental data that were acceptable within a probability limit of 0.95, according to the F test (33). Minimum and maximum values of parameters in these acceptable sets were taken to represent the 95% confidence limits of a given parameter.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 APPENDIX
 REFERENCES
 
A Subunit-specific Ferritin-DNA Interaction—To test directly whether ferritin binds DNA, electrophoresis mobility shift assays (28)) were carried out using a 43-bp DNA duplex. This DNA has a sequence identical to residues –164 to –128 of the human {beta}-globin promoter (17). Titration of this DNA with recombinant human H-ferritin (rH-ferritin) results in the concentration-dependent formation of a discrete mobility-shifted band (Fig. 1, lanes b–f). Titration of pre-formed complexes with unlabeled competing DNA (of identical sequence) decreased retention of the 32P-labeled DNA, indicating that the ferritin-DNA interaction is reversible (result not shown). In contrast, binding was not detected with the recombinant human L chain ferritin under the same solution conditions (Fig. 1, compare lanes e and i). Since H- and L-ferritins differ in feroxidase activities, the observed difference in their DNA binding prompted investigation of the 222-mutant of human H-ferritin. The 222-mutant forms native-like 24-subunit assemblies but has a three-amino acid substitution that diminishes the ferroxidase activity normally associated with the H-subunit (34). As a result, it is only 6.3% as active in iron uptake as is the recombinant H homopolymer (35). The mutant 222-ferritin did not possess detectable DNA binding activity over the range of protein concentrations that gave detectable DNA binding with rH-ferritin (Fig. 1, lanes j–m). Although proof is still lacking, the absence of detectable mobility shift with L-ferritin and the 222-mutant H-ferritin raises the intriguing possibility that ferroxidase activity might be required for DNA binding.



View larger version (63K):
[in this window]
[in a new window]
  		FIG. 1.
Subunit-specific binding of ferritin to the 43-bp {beta}-globin fragment, detected by mobility shift assay. Comparison of recombinant H-, L-, and 222-mutant ferritins. All samples contained 32P-labeled 43-bp duplex DNA (2 nM). Samples shown in lanes b–e contained, in addition, recombinant H-ferritin at final concentrations of 0.5, 1.0, 2.5, and 5.0 µM, respectively. Samples shown in lanes f–i contained, in addition, recombinant L-ferritin at final concentrations of 0.5, 1.0, 2.5, and 5.0 µM, respectively. Samples shown in lanes j–m contained, in addition, 222-ferritin at final concentrations of 0.5, 1.0, 2.5, and 5.0 µM, respectively. Binding and electrophoresis conditions are described under "Materials and Methods."

 
Binding Is Independent of DNA Sequence, Base Composition, or Type of Molecular End—Binding competition assays were performed to determine whether the rH-ferritin-DNA interaction was dependent on base composition, sequence, or the structures present at the ends of restriction fragments. A population of fragments derived from plasmid pGEM-3 (36) was titrated with rH-ferritin (Fig. 2A). Three of these fragments (of 1605, 1198, and 222 bp) contain a CAGTGC sequence motif (36). This sequence has been reported to bind ferritin (17). Because the many free DNA bands interfere with visualization of mobility-shifted ferritin-DNA complexes, binding is most easily monitored by following the disappearance of the free DNA bands, as originally described by Garner and Revzin (37). In addition, this assay differs from the binary competition assays described previously (28, 29) in that here the protein partitions between 15 discrete DNA fragments, ranging in size from 36 to 2645 bp. As discussed in the "Appendix," the ratio of protein affinities for a target fragment ({alpha}) and a reference fragment (ref) is given by Equation 2.

(Eq. 2)
Band intensities are represented by I{alpha},1 and Iref,1, whereas the input mole fractions of the corresponding DNA fragments are f({alpha}) and f(ref). For this analysis, we have selected the 65-bp fragment as the reference species. Representative graphs of the dependence of the normalized binding ratio (I{alpha},1·f(ref)/Iref,1·f({alpha})), on ferritin concentration for the 460, 126, and 36-bp fragments, are shown in Fig. 2B, upper panel. For every fragment tested, the normalized binding ratio appeared to depend linearly on [ferritin], over the entire experimental concentration range, and linear extrapolation to [ferritin] = 0 was straightforward. As shown in Table I, values of K{alpha},1/Kref,1 for all fragments lie within a narrow range (0.42 ± 0.15 <= K{alpha},1/Kref,1 <= 2.71 ± 0.17). This result is inconsistent with a strong binding preference for any sequence present in the population, including those fragments containing the CAGTGC sequence. Similarly, a graph of normalized binding ratios as a function of the percent of G + C residues in each fragment reveals no systematic dependence of relative affinity on base composition (Fig. 2B, center panel). Intriguingly, a graph of normalized binding ratios as a function of fragment length (Fig. 2B, bottom panel) shows the absence of a systematic dependence of binding affinity on DNA length. This result is inconsistent with binding models in which the number of potential binding sites increases with DNA length (such as that represented by Equation 8 under the "Appendix"), and it justifies our choice of the fragment length-independent binding model, represented by Equation 7.



View larger version (24K):
[in this window]
[in a new window]
  		FIG. 2.
Binding competition analysis. A, titration of 32P-labeled restriction fragments derived from plasmid pGEM-3 with recombinant H-ferritin. All samples contained 32P-labeled restriction fragments (5'- end concentration 60 pM). Samples shown in lanes b–g contained, in addition, rH-ferritin at final concentrations of 0.1, 0.25, 0.5, 1.0, 1.5, and 2.5 µM, respectively. Fragment sizes are given in units of base pairs. B, analysis of binding competition data. Top, graph of the normalized binding ratio as a function of the ferritin concentration. Data are from the experiment shown in A. The normalized binding ratio, equal to K{alpha},1/Kref,1, was determined according to Equation 7 in the "Appendix," using the 65-bp DNA fragment as the reference ligand. The data presented are for the 36-bp fragment ({circ}), the 126-bp fragment ({blacktriangleup}), and the 460-bp fragment ({square}). Center, graph of the normalized binding ratio as a function of restriction fragment base composition. The values plotted are those obtained in the limit of low [ferritin], extrapolated graphically as shown in the top panel. The error bars represent 95% confidence limits. The horizontal line represents a binding affinity equal to that of the 65-bp reference fragment. Bottom, dependence of normalized binding ratio on restriction fragment length. Fragment lengths are normalized to that of the 65-bp reference fragment. The values plotted are those obtained in the limit of low [ferritin], extrapolated graphically as shown in the top panel. The error bars represent 95% confidence limits. The horizontal line represents a binding affinity equal to that of the 65-bp reference fragment.

 


View this table:
[in this window]
[in a new window]
  		TABLE I
Relative affinities of ferritin for restriction fragments from pGEM-3

Ratios K{alpha},1/Kref,1 were determined using Equation 7, given under the "Appendix." Error ranges are 95% confidence limits, estimated as described under "Materials and Methods."

 
The observation that binding affinity is independent of DNA length suggests that binding sites might be within structures that are present at the same relative concentration for all DNA fragments. DNA ends are the most obvious class of structures with this property. However, the values obtained with blunt-ended DNA fragments K{alpha},1/Kref,1 = 1.1 ± 0.5 (n = 4) and with fragments possessing single-stranded 5' overhangs K{alpha},1/Kref,1 = 0.98 ± 0.71 (n = 10) are not significantly different (Table I). Although this does not rule out DNA ends as binding sites for ferritin, it indicates that the differences in DNA end structures that we have chosen to test do not greatly influence the strength of the interaction.

DNA Nicking in the Presence of Ferritin—Addition of ferritin to solutions containing iron (initially present as either FeCl3 or FeSO4) dramatically reduces the extent of nicking experienced by plasmid DNA added at a later time (7). Presumably, this is a consequence of iron sequestration by the ferritin because the magnitude of the effect correlates with the iron uptake ferroxidase activity of the proteins tested (recombinant human H-ferritin > 222-mutant H-ferritin > horse spleen apoferritin). However, the long term protection of DNA requires both iron uptake and iron retention. Here we test the notion that the release of iron from ferritin can be a source of DNA damage.

Supercoil-relaxation assays (30) were performed with plasmid pUC19 DNA incubated with ferritin (Fig. 3). Under the electrophoretic conditions chosen, individual topoisomers and the relaxed circle form were resolved (Fig. 3A, lanes i and q). After addition of ferritin, the mole fraction of the superhelical form decreased and that of the relaxed form increased with time (Fig. 3A, lanes b–h and j–p). At long reaction times, linear form DNA can also be detected.2 Comparison of the time course of reactions run with rH-ferritin, 222-mutant H-ferritin, liver ferritin, and spleen ferritin (Fig. 3B) shows that all reactions have a characteristic lag at early times, followed by an interval of increasing nicking activity. Such a lag phase is not seen when ferrous or ferric iron was added to ferritin-free control reactions (Fig. 3C). The duration of lag phase was similar for all ferritins tested, as were the rates of DNA nicking during lag phase (initial nicking rates average = 4.8 ± 1.1 x 10–5 s–1 within this sample population). Because the ferroxidase activities of rH, 222, spleen, and liver ferritins differ markedly (40), these results are inconsistent with the idea that the lag reflects a function of the iron-transport ferroxidase. In addition, because there are large differences in the iron contents of our ferritin samples (Fig. 4), the uniform duration and nicking rates of the lag phase are unlikely to be a function of the degree of iron loading of ferritin. These issues will be discussed in more detail below.



View larger version (28K):
[in this window]
[in a new window]
  		FIG. 3.
Time course of DNA-nicking in the presence of rH-, 222-, liver, and spleen-ferritins. A, electrophoretic profiles of pUC19 DNA incubated in the presence of recombinant H-ferritin (upper panel) and horse spleen (lower panel) ferritin. Reactions were initiated by the addition of ferritin to samples containing 0.5 µg of supercoiled pUC19 DNA. The reaction buffer was 10 mM Hepes (pH 7.5), 50 mM NaCl, 2.5 mM MgCl2, 2.5 mM DTT. Reaction times for samples shown in lanes b–h and lanes j–p were 15, 30, 45, 60, 75, 90, and 120 min, respectively. Lane a contains a sample of pUC19 DNA linearized with endonuclease BamHI; lanes i and q contain samples of superhelical pUC19 DNA that was not exposed to ferritin. Reactions were quenched, and electrophoresis performed as described under "Materials and Methods." Band designations: R, relaxed form, L, linear form, SC, supercoiled form. B, time course of the conversion of superhelical DNA to the relaxed form. The mole fraction at each time point (F) has been normalized to its initial value prior to the start of the reaction (Fo). Symbols are as follows: {circ}, rH-ferritin; {blacksquare}, 222-ferritin; {diamond}, liver-ferritin; , spleen ferritin. Averaged over all samples, the initial reaction rate (lag phase) was 4.8 ± 1.1 x 10–5 s–1. Maximum rates, estimated about the inflection points of each reaction profile, are as follows: rH-ferritin, 1.18 ± 0.5 x 10–4 s–1; 222-ferritin, 1.26 ± 0.16 x 10–4 s–1; liver ferritin, 2.77 ± 0.36 x 10–4 s–1; spleen ferritin, 3.50 ± 0.11 x 10–4 s–1. C, comparison of nicking rates of liver and spleen ferritins with matched concentrations of FeCl3. Samples containing liver ferritin ({blacksquare}) had [iron]total = 24 µM. Samples containing spleen ferritin () had [iron]total = 40 µM. Control reactions ({square} and {circ}) had initial concentrations of FeCl3 = 24 and 40 µM and gave pseudo-first order DNA-nicking rates of 5.69 ± 0.61 x 10–4 and 8.47 ± 0.41 x 10–4 s–1, respectively. D, comparison of nicking rates of liver and spleen ferritins with matched concentrations of FeSO4. Samples containing liver ferritin ({blacksquare}) had [iron]total = 24 µM. Samples containing spleen ferritin () had [iron]total = 40 µM. Control reactions ({square} and {circ}) had initial concentrations of FeSO4 = 24 and 40 µM and gave pseudo-first order DNA nicking rates of 3.74 ± 0.74 x 10–4 and 4.97 ± 0.39 x 10–4 s–1, respectively.

 



View larger version (20K):
[in this window]
[in a new window]
  		FIG. 4.
Iron contents of the ferritin samples used in this study. Samples were hydrolyzed in acid for 24 h at 75 °C, as described under "Materials and Methods." The iron concentrations in the resulting solutions were determined by atomic absorption spectrophotometry. Determinations were made in triplicate; mean values are reported here.

 
Following the lag phase, the rate of DNA nicking accelerates (Fig. 3B). The nicking rates increase in the order spleen ferritin > liver ferritin > (rH, 222-ferritins). This order is the same as that of the iron content in our samples (Fig. 4), suggesting that the maximal rate of DNA nicking in each reaction may be related to the amount of ferritin-iron that is potentially available within the system. On the other hand, the small difference in nicking rate observed with rH and 222-mutant ferritins that differ significantly in ferroxidase activity suggests that the maximal nicking rate observed is not determined by the ferroxidase activity of the preparation. Finally, at long reaction times, the rate of detectable DNA nicking declines as the supercoiled substrate is depleted.

Relationship of DNA Relaxation Rate and Superhelical Density—Does ferritin-mediated nicking of DNA depend on superhelical density? Different topoisomers in our DNA population differ in torsional stress and compaction because of supercoiling and possibly secondary structure (41). These differences could influence the susceptibility of DNA to attack by reactive species. To determine whether this is the case, superhelical pUC19 DNA was exposed to ferritin for varying times as described above, but the electrophoretic step was carried out longer (24 h) to improve resolution of the topoisomers in the population. As shown in Fig. 5, the time course of nicking appears similar for all topoisomers in the population.



View larger version (58K):
[in this window]
[in a new window]
  		FIG. 5.
Time course of relaxation of individual topoisomers in the presence of rH-ferritin. A, electrophoretic profiles of reaction products as a function of time. Superhelical pUC19 DNA was incubated with rH-ferritin as described under "Materials and Methods." The reaction buffer was 10 mM Hepes (pH 7.5), 50 mM NaCl, 2.5 mM MgCl2, 2.5 mM DTT. Reactions were quenched after reactions of 15, 30, 45, 60, 75, 90, and 120 min (samples a–g, respectively). An unreacted control sample is shown in lane h. The positions of relaxed (R), linear (L), and superhelical DNAs are indicated. The bands corresponding to topoisomers that have been analyzed quantitatively are numbered 1–6. B, time course of relaxation reactions for individual topoisomers. The mole fractions of topoisomers 1 ({blacktriangleup}), 2 (), 3 ({triangleup}), 4 ({blacksquare}), 5 ({circ}), and 6 ({square}) (identified in A, above) are plotted as a function of time. All gave similar lag phase duration and nicking rates and similar maximum nicking rates.

 
Iron Release and the Formation of Free Radicals Appear to Be Necessary for DNA Nicking—The kinetic lag in ferritin-mediated DNA nicking suggests that one or more reaction steps precede the one in which DNA is nicked. We hypothesize that the lag corresponds to the interval required for iron release by ferritin. Following iron release, the Fenton reaction or related pathways should result in the formation of reactive iron species and/or hydroxyl radicals, which react with DNA (42). If free iron is an intermediate in this pathway, DNA nicking should be inhibited by chelators. Consistent with this prediction, the addition of EDTA and EGTA to the reaction mixture strongly inhibits nicking (Fig. 6A). If free radicals are necessary for DNA nicking, the reaction should be inhibited by radical scavengers such as glycerol. The dependence of the nicking rate on [glycerol] is consistent with this prediction (Fig. 6A, inset). Finally, the production of free radicals at locations distant from the DNA should result in random nicking of each DNA strand. Only very heavy damage of this kind should result in double-strand cleavage. This is the pattern that is seen when DNA is attacked by hydroxyl radicals produced by untethered iron-EDTA complexes (Fig. 6B, upper panel). On the other hand, the production of radicals by ferrous and ferric ions that are bound at the DNA surface should result in spatially correlated damage to both strands and a high probability of double-strand scission. The rapid accumulation of double-strand breaks is a feature of the reactions that take place in ferritin-DNA solutions (Fig. 6B, lower panel). Together, these results are consistent with a view of DNA nicking in the presence of ferritin in which iron release from ferritin and the production of free radicals at points close to both DNA strands are important features.



View larger version (57K):
[in this window]
[in a new window]
  		FIG. 6.
Effects of chelators and glycerol on DNA nicking in the presence of ferritin. A, time course of DNA nicking in the presence of rH-ferritin in the absence ({blacksquare}) and presence ({circ}) of 10 mM EDTA plus 5 mM EGTA. The reaction buffer was 10 mM Hepes (pH 7.5), 50 mM NaCl, 2.5 mM MgCl2, 2.5 mM DTT. The apparent maximum rate of DNA nicking, observed near the mid-point of the ferritin reaction, was 6.20 ± 2.8 x 10–4 s–1. The rate in the presence of EDTA + EGTA chelators was 2.36 ± 0.20 x 10–5 s–1. Inset, the effect of glycerol on apparent maximum nicking rate, k(app). B, the rate of double-strand breakage is slower with iron-EDTA (upper panel) than with equivalent concentration of ferritin iron (lower panel). Reaction mixtures (30 µl) contained supercoiled pUC19 DNA (0.5 µg), in 10 mM Hepes, 2.5 mM DTT, 50 mM NaCl, pH 7.4. Samples shown in lanes a–i contained, in addition 40 µM FeCl3, 40 mM EDTA and 2 mM EGTA; lanes l–r contained in addition 8 µg of spleen ferritin, respectively. Reaction times for samples a–i were 0, 2, 4, 8, 16, 32, 64, 96, and 128 min, respectively. Reaction times for samples l–r were 2, 4, 8, 16, 32, 64, and 128 min, respectively. All reactions were performed at 37 °C. Lanes j and s contain superhelical pUC19 DNA, and lane k contains BamHI-linearized pUC19 DNA as controls. Nicking reactions were quenched by the addition of 10 µl of a solution containing 50% glycerol, 50 mM EDTA, and 0.1% bromphenol blue. Aliquots were subjected to electrophoresis on 1.5% agarose gels at 40 mA for 12 h. Gels were stained with ethidium bromide (0.5 µg/ml). Band designations are as follows: R, relaxed form; L, linear form; Sc, supercoiled form.

 
Does Ferritin Protect DNA?—It is clear that DNA is nicked more rapidly in the presence of ferritin-iron complexes than in iron-free solutions. However, because iron is abundant in nuclei (43), a more appropriate comparison is whether ferritin-iron damages DNA more rapidly than an equivalent concentration of inorganic iron. Such a comparison is shown in Fig. 3, C and D. Under our standard reaction conditions (10 mM Hepes (pH 7.5 at 21 °C), 2.5 mM DTT, 50 mM NaCl), the initial DNA-nicking rates of free ferric iron (from FeCl3) are significantly faster than those of equivalent concentrations of ferritin-iron. However, following the lag phase, the maximal rates of DNA relaxation with ferritin are similar to those observed with free iron. Thus, whether ferritin protects DNA may depend on factors that influence the duration and amplitude of the kinetic lag phase.

The oxidation state of the free iron strongly influences the rate of DNA nicking (compare Fig. 3, C and D). For reactions carried out under our standard conditions (10 mM Hepes (pH 7.5 at 21 °C), 2.5 mM DTT, 50 mM NaCl), the initial DNA-nicking rates of free ferrous iron (from FeSO4) are similar to those of the ferritin lag phase and slower than the maximum rates obtained with ferritin at later reaction times. In these reactions the FeSO4 concentration has been adjusted to equal the bulk concentrations of ferritin-iron. Because under these conditions some iron is likely to remain associated with ferritin, the concentration of free iron in the ferritin reactions must be less than or equal to those in the reference FeSO4 reactions. It follows that the iron species released by ferritin are significantly more reactive toward DNA than equivalent (or slightly greater) concentrations of FeSO4 in the reference reaction.

A final complication is that the rates of DNA nicking by free iron species depend significantly on the buffer composition. When 10 mM Hepes (pH 7.5 at 21 °C), 50 mM NaCl, 2.5 mM MgCl2, 2.5 mM DTT buffer was used, 50 µM FeCl3 alone damaged the DNA after 15 min (Fig. 7, top panel), whereas the addition of 10 mM H2O2 accelerated the damage. In control experiments, H2O2 alone produced no detectable damage after 2 h under these conditions. In contrast, when the reaction was run in 10 mM Tris (pH 7.4 at 21 °C), 50 mM NaCl, 2.5 mM MgCl2, no DNA nicking was detectable after 2 h with 50 µM FeCl3 or 10 mM H2O2 alone. Only in combination did FeCl3 and 10 mM H2O2 produce significant cleavage (Fig. 7, middle panel). Finally, in parallel reactions carried out in 10 mM Tris (pH 7.4 at 21 °C), 50 mM NaCl, 2.5 mM MgCl2, 50 µM FeCl3 were as effective at nicking DNA in the presence of 2.5 mM DTT as in combination with 10 mM H2O2. Thus, whether ferritin is seen to damage DNA or to protect it depends on the conditions of the reference reaction performed with inorganic iron species. In particular, modest concentrations of DTT, a reagent commonly included in ferritin storage and dilution buffers, are highly effective in promoting DNA nicking by ferric iron. Together, these observations have the potential to account for much of the current controversy over whether ferritin protects DNA (44).



View larger version (23K):
[in this window]
[in a new window]
  		FIG. 7.
The rates of DNA nicking by free iron species depend on the buffer composition. Top panel, time course of reactions carried out in 10 mM Hepes (pH 7.5), 50 mM NaCl, 2.5 mM MgCl2, 2.5 mM DTT. Reaction promoted by 50 µM FeCl ({diamond}); apparent pseudo first-order rate constant kapp = 6.75 ± 0.38 x 10–4 s–1. Reaction promoted by 50 µM FeCl3 plus 10 mM H2O2 (); apparent pseudo first-order rate constant kapp = 8.19 ± 0.54 x 10–4 s–1. Reaction promoted by 10 mM H2O2 alone ({blacktriangleup}); apparent pseudo first-order rate constant kapp= 9.98 ± 1.67 x 10–6 s–1. Center panel, time course of reactions carried out in 10 mM Tris (pH 7.4 at 21 °C), 100 mM KCl. Reaction promoted by 50 µM FeCl3 ({blacktriangledown}); apparent pseudo first-order rate constant kapp = 2.92 ± 1.04 x 10–6 s–1. Reaction promoted by 50 µM FeCl3 plus 10 mM H2O2 ({diamondsuit}); apparent pseudo first-order rate constant kappr = 7.66 ± 0.34 x 10–4 s–1. Reaction promoted by 10 mM H2O2 alone ({square}); apparent pseudo first-order rate constant kapp = 1.98 ± 0.37 x 10–6 s–1. Bottom panel, time course of reactions carried out in 10 mM Tris (pH 7.4), 50 mM NaCl, 2.5 mM MgCl2. Reaction promoted by 50 µM FeCl3 ({triangleup}); apparent pseudo first-order rate constant kapp = 2.32 ± 0.24 x 10–6 s–1. Reaction promoted by 50 µM FeCl3 plus 10 mM H2O2 ({blacksquare}); apparent pseudo first-order rate constant kapp = 4.58 ± 0.25 x 10–4 s–1. Reaction promoted by 50 µM FeCl3 plus 2.5 mM DTT ({circ}); apparent pseudo first-order rate constant kapp = 6.75 ± 0.48 x 10–4 s–1.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
APPENDIX
 REFERENCES
 
The idea that ferritin can gain access to the nucleus of a cell in a regulated manner is gaining acceptance. However, the functions of ferritin within the nucleus remain to be discovered. Several possible, non-exclusive functions have been proposed, including (i) transcriptional regulation (14), (ii) iron sequestration to protect nuclear contents from iron-mediated damage (7), and (iii) delivery of iron to enzymes that require it (45).

In this paper we have demonstrated that the H-subunit of human ferritin forms a stable complex with DNA. There is, in fact, considerable precedent for such a function. Broyles and co-workers (14) reported that a ferritin-like protein from K562 cells binds to the sequence CAGTGC, and this finding has been reproduced and amplified by Pountney et al. (46). In addition, several bacterial ferritins and ferritin-related proteins bind DNA quite strongly (47, 48). Finally, we have shown that 125I-labeled rH-ferritin can be cross-linked, in vivo, to the chromosomal DNA of human SW1088 astrocytoma cells (7). In our initial DNA binding studies, we observed weak and reversible binding to the CAGTGC sequence using a 43-bp DNA fragment as binding substrate (Fig. 1). Under the same buffer conditions, a comparable complex was not detectable with the L-subunit. These findings make an intriguing parallel with the observation that the H-subunit is more efficiently translocated into cell nuclei than is the L-subunit (7). An intact ferroxidase center may be necessary for DNA binding, because the 222-mutant protein (a ferroxidase-defective mutant of H-ferritin) does not form stable complexes with DNA. Although it is not possible to accurately infer the molecular weight of the H-ferritin-DNA complex from the available electrophoresis mobility shift assays data, its relatively high electrophoretic mobility is incompatible with a very large structure. As a working hypothesis, we propose that ferritin monomers or dimers form the functional DNA-binding unit. Sedimentation equilibrium experiments to test this notion are currently underway.

Under our reaction conditions, the DNA binding of H-ferritin is apparently sequence- and base composition-independent. In addition, it does not depend on the DNA length or the structure of the DNA ends. The observed lack of sequence specificity is inconsistent with the function of ferritin as a conventional transcription regulatory protein, but it does not rule out such function if the DNA binding specificity were provided by a separate factor. On the other hand, nonspecific DNA binding may help to retain ferritin within the nucleus and in close proximity to the DNA, where it may perform other valuable functions.

The question of whether ferritin protects DNA from iron-mediated oxidative damage is a complex one. We have demonstrated previously (7) that incubation of iron with ferritin greatly reduces its ability to nick superhelical DNA that is added later. H-ferritin gave more efficient protection than the 222-mutant, suggesting that ferroxidase-dependent iron sequestration was part of the mechanism. However, these short duration (15 min) experiments did not address the longer term ability of ferritin to retain iron. In this article we have examined the stability of superhelical DNA in ferritin solutions during prolonged (>=2 h) incubations. Over this interval, DNA nicking does occur, although its onset is characterized by a lag of ~3000 s (50 min) during which nicking is slow. Kinetic lags like those seen here are characteristic of multistep sequential reactions in which the rate-limiting step is not the last one (49). Such a lag phase is not seen when ferrous or ferric iron is added to ferritin-free control reactions, suggesting that the lag is a property of the ferritin protein. The maximal rate of DNA nicking in ferritin solutions depends on the iron content of the protein, and nicking is inhibited by chelators and glycerol. These features suggest iron released from ferritin is the source of DNA damage and that free radicals participate in the reaction pathway. Finally, the DNA damage is characterized by a higher frequency of double-strand breaks than is observed when cleavage is mediated by iron-EDTA complexes that bind DNA weakly or not at all (50). Double-strand breaks require the scission of both strands within a short contour length. This spatially correlated cleavage implies a source of reactive species located at (or near) specific positions on the DNA surface. Together these results suggest a mechanism in which iron is released from ferritin and binds (or is located very near) the DNA before strand scission occurs.

The reactivity of ferritin-iron toward DNA is similar to equivalent bulk concentrations of FeCl3 and greater than that of equivalent bulk concentrations of FeSO4. This raises the possibility that ferric iron released from ferritin is the species responsible for DNA damage. If correct, this interpretation indicates that iron release is not simply the reverse of the ferroxidase-driven iron accumulation reaction, because this should liberate the less reactive ferrous form. This interpretation is supported by the recent observation (51) that there are two or more kinetically distinct pathways for iron to enter (and by implication to leave) the ferritin shell.

Is there a cellular role for ferritin-mediated DNA nicking? The relaxation of superhelical stress may be necessary for chromatin re-modeling (52) and is known to be important for transcriptional elongation and DNA replication (53). Observations of ferritin in the nuclei of tumor cells and those of cells in actively growing tissue (7) are consistent with its potential involvement in these processes.


   APPENDIX
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 APPENDIX
REFERENCES
 
At equilibrium, the molecules in an interacting protein-DNA system exist in a series of binding states, each differing in distribution of protein(s) among the available DNA sites. The probability with which a given state (j) of the system occurs is proportional to exp(–{Delta}Gj/RT), in which {Delta}Gj is the free energy change that accompanies the formation of state j from a reference state (54). If the state in which the DNA molecule has no protein bound is taken as the reference state, the term exp(–{Delta}Gj/RT) for the binding of a protein (P) to a particular DNA site is equal to the product of its association constant Kj and the free protein concentration, [P]. If all DNA sites are bound independently, the occupancy states of a DNA species {alpha} are enumerated by the polynomial shown in Equation 3,

(Eq. 3)
in which the numerical subscripts denote the 1st, 2nd, 3rd, etc., binding event, in order of relative affinity. In the limit of low protein concentration, this simplifies to F({alpha}) = 1 + K{alpha},1[P]. For the binding of protein to a mixture of DNA molecules, the binding polynomial for the entire mixture (the system partition function) is the product of polynomials for the individual DNAs in the system (here designated {alpha}, {beta}, {lambda}, {delta},...) (Equation 4).

(Eq. 4)

The probability, Pj with which state j occurs, is equal to the term in the partition function for the jth state, divided by Q. Because the samples employed in mobility shift assays contain large numbers of molecules, this probability will be very closely approximated by the relative frequency with which state j occurs. Thus, for samples at equilibrium, the mole fraction of the DNA that is present in a given binding state is a good measure of the formation probability of that state (28). We have described previously conditions under which the mole fractions of individual DNA species can be accurately measured as integrated intensities (I) of resolved electrophoretic bands (29).

If we wish to compare the affinities of a protein for DNA of species {alpha} with that of a reference species (for example {beta}), present at identical concentration, the ratio of binding affinities K{alpha},1/K{beta},1 is given by Equation 5.

(Eq. 5)
This expression is analogous to one previously derived for the distribution of a protein between two DNA species, in a system in which only two are available (see Equation 8 in Ref. 28). As before, this expression simplifies in the limit of low protein concentration as shown in Equation 6.

(Eq. 6)
To generalize this result, it is necessary to account for any differences in the concentrations of {alpha} and reference ({beta}) DNAs. This is done by dividing each band intensity term by the mole fraction of the DNA form to which it refers (f({alpha}) or f({beta})) as shown in Equation 7.

(Eq. 7)
This expression gives the ratio of binding affinities for the binding model in which every DNA fragment in the system contains the same number of protein-binding sites. If the DNA fragments in the system contain different numbers of protein-binding sites, this difference must be taken into account by dividing each band intensity term by the number of protein-binding sites present in the DNA to which it refers (ns({alpha}) or ns({beta})) (Equation 8).

(Eq. 8)
A similar approach can be used to solve more complicated distribution problems (39).


   FOOTNOTES
 
* This work was supported by National Institutes of Health Grant DK54289. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

|| To whom correspondence should be addressed. Tel.: 717-531-5250; Fax: 717-533-9592; E-mail: mfried{at}psu.edu.

1 The abbreviations used are: rH, recombinant heavy chain; DTT, dithiothreitol; rL, recombinant light chain. Back

2 Mobility-shifted species are not detected in these assays, presumably because they are unstable under the conditions of agarose gel electrophoresis. Many protein DNA complexes are less stable in agarose gels than in the more concentrated polyacrylamide matrices (38, 39). Back


   ACKNOWLEDGMENTS
 
We thank Dr. Paolo Arosio (Milan, Italy) for the recombinant H- and L-ferritin proteins.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 APPENDIX
 REFERENCES
 

  1. Haddow, A., and Horning, E. S. (1960) J. Natl. Cancer Inst. 24, 109–147[Medline] [Order article via Infotrieve]
  2. Smith, A. G., Carthew, P., Francis, J. E., Edwards, R. E., and Dinsdale, D. (1990) Hepatology 12, 1399–1405[Medline] [Order article via Infotrieve]
  3. Iancu, T. C., Rabinowitz, H., Brissot, P., Guillouzo, A., Deugnier, Y., and Bourel, M. (1985) J. Hepatol. 1, 261–275[CrossRef][Medline] [Order article via Infotrieve]
  4. Shires, T. K. (1982) Biochem. J. 205, 321–329[Medline] [Order article via Infotrieve]
  5. Vaca, C. E., and Harms-Ringdahl, M. (1989) Biochim. Biophys. Acta 1001, 35–43[Medline] [Order article via Infotrieve]
  6. Cheepsunthorn, P., Palmer, C., Menzies, S., Roberts, R. L., and Connor, J. R. (2001) J. Comp. Neurol. 431, 382–396[CrossRef][Medline] [Order article via Infotrieve]
  7. Thompson, K. J., Fried, M. G., Ye, Z., Boyer, P., and Connor, J. R. (2002) J. Cell Sci. 115, 2165–2177[Abstract/Free Full Text]
  8. Cai, C. X., Birk, D. E., and Linsenmayer, T. F. (1997) J. Biol. Chem. 272, 12831–12839[Abstract/Free Full Text]
  9. Cai, C. X., Birk, D. E., and Linsenmayer, T. F. (1998) Mol. Biol. Cell 9, 1037–1051[Abstract/Free Full Text]
  10. Cai, C. X., and Linsenmayer, T. F. (2001) J. Cell Sci. 114, 2327–2334[Medline] [Order article via Infotrieve]
  11. Reif, D. W., Schubert, J., and Aust, S. D. (1988) Arch. Biochem. Biophys. 264, 238–243[CrossRef][Medline] [Order article via Infotrieve]
  12. Samokyszyn, V. M., Thomas, C. E., Reif, D. W., Saito, M., and Aust, S. D. (1988) Drug Metab. Rev. 19, 283–303[Medline] [Order article via Infotrieve]
  13. Reif, D. W. (1992) Free Radic. Biol. Med. 12, 417–427[CrossRef][Medline] [Order article via Infotrieve]
  14. Broyles, R. H., Belegu, V., DeWitt, C. R., Shah, S. N., Stewart, C. A., Pye, Q. N., and Floyd, R. A. (2001) Proc. Natl. Acad. Sci. U. S. A. 98, 9145–9150[Abstract/Free Full Text]
  15. Bozzi, M., Mignogna, G., Stefanini, S., Barra, D., Longhi, C., Valenti, P., and Chiancone, E. (1997) J. Biol. Chem. 272, 3259–3265[Abstract/Free Full Text]
  16. Grant, R. A., Filman, D. J., Finkel, S. E., Kolter, R., and Hogle, J. M. (1998) Nat. Struct. Biol. 5, 294–303[CrossRef][Medline] [Order article via Infotrieve]
  17. Broyles, R. H., Blair, F. C., Kyker, K. D., Kurein, B. T., Stewart, D. R., Hala'sz, H., Berg, P. E., and Schechter, A. N. (1995) Colloq. INSERM 234, 43–51
  18. Wu, Y. J., and Noguchi, C. T. (1991) J. Biol. Chem. 266, 17566–17572[Abstract/Free Full Text]
  19. Prince, R. C., and Grossman, M. J. (1993) Trends Biochem. Sci. 18, 153–154[CrossRef][Medline] [Order article via Infotrieve]
  20. Hidalgo, E., and Demple, B. (1994) EMBO J. 13, 138–146[Medline] [Order article via Infotrieve]
  21. Khoroshilova, N., Beinert, H., and Kiley, P. J. (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 2499–2503[Abstract/Free Full Text]
  22. Hurta, R. A., and Wright, J. A. (1991) Biochem. Cell Biol. 69, 635–642[Medline] [Order article via Infotrieve]
  23. Andronikashvili, E. L., Mosulishvili, L. M., Belokobilski, A. I., Kharabadze, N. E., Tevzieva, T. K., and Efremova, E. Y. (1974) Cancer Res. 34, 271–274[Abstract/Free Full Text]
  24. Levi, S., Santambrogio, P., Cozzi, A., Rovida, E., Corsi, B., Tamborini, E., Spada, S., Albertini, A., and Arosio, P. (1994) J. Mol. Biol. 238, 649–654[CrossRef][Medline] [Order article via Infotrieve]
  25. Erikson, K. M., Pinero, D. J., Connor, J. R., and Beard, J. L. (1997) J. Nutr. 127, 2030–2038[Abstract/Free Full Text]
  26. Kurien, B. T., Scofield, R. H., and Broyles, R. H. (1997) Anal. Biochem. 245, 123–126[CrossRef][Medline] [Order article via Infotrieve]
  27. Maxam, A. M., and Gilbert, W. (1977) Proc. Natl. Acad. Sci. U. S. A. 74, 560–564[Abstract/Free Full Text]
  28. Fried, M., and Crothers, D. M. (1981) Nucleic Acids Res. 9, 6505–6525[Abstract/Free Full Text]
  29. Fried, M. G. (1989) Electrophoresis 10, 366–376[CrossRef][Medline] [Order article via Infotrieve]
  30. Tachon, P. (1989) Free Radic. Res. Commun. 7, 1–10[Medline] [Order article via Infotrieve]
  31. Yanisch-Perron, C., Vieira, J., and Messing, J. (1985) Gene (Amst.) 33, 103–119[CrossRef][Medline] [Order article via Infotrieve]
  32. Brodersen, R., Nielsen, F., Christiansen, J. C., and Andersen, K. (1987) Eur. J. Biochem. 169, 487–495[Medline] [Order article via Infotrieve]
  33. Sprague, E. D., Larrabee, C. E., Jr., and Halsall, H. B. (1980) Anal. Biochem. 101, 175–181[CrossRef][Medline] [Order article via Infotrieve]
  34. Lawson, D. M., Treffry, A., Artymiuk, P. J., Harrison, P. M., Yewdall, S. J., Luzzago, A., Cesareni, G., Levi, S., and Arosio, P. (1989) FEBS Lett. 254, 207–210[CrossRef][Medline] [Order article via Infotrieve]
  35. Chasteen, N. D., Sun, S., Levi, S., and Arosio, P. (1994) Adv. Exp. Med. Biol. 356, 23–30[Medline] [Order article via Infotrieve]
  36. Promega Corp. (2000) Promega Technical Bulletin 033: pGM-3Z Vector, Madison, WI
  37. Garner, M. M., and Revzin, A. (1981) Nucleic Acids Res. 9, 3047–3060[Abstract/Free Full Text]
  38. Fried, M. G., and Bromberg, J. L. (1997) Electrophoresis 18, 6–11[CrossRef][Medline] [Order article via Infotrieve]
  39. Fried, M. G., and Daugherty, M. A. (1998) Electrophoresis 19, 1247–1253[CrossRef][Medline] [Order article via Infotrieve]
  40. Harrison, P. M., and Arosio, P. (1996) Biochim. Biophys. Acta 1275, 161–203[Medline] [Order article via Infotrieve]
  41. Bauer, W. R. (1978) Annu. Rev. Biophys. Bioeng. 7, 287–313[CrossRef][Medline] [Order article via Infotrieve]
  42. Theil, E. C. (2003) J. Nutr. 133, S1549–S1553[Abstract/Free Full Text]
  43. Gurgueira, S. A., and Meneghini, R. (1996) J. Biol. Chem. 271, 13616–13620[Abstract/Free Full Text]
  44. Whiting, R. F., Wei, L., and Stich, H. F. (1981) Cancer Res. 41, 1628–1636[Abstract/Free Full Text]
  45. Romeo, A. M., Christen, L., Niles, E. G., and Kosman, D. J. (2001) J. Biol. Chem. 276, 24301–24308[Abstract/Free Full Text]
  46. Pountney, D., Trugnan, G., Bourgeois, M., and Beaumont, C. (1999) J. Cell Sci. 112, 825–831[Abstract]
  47. Theil, E. C. (1987) Annu. Rev. Biochem. 56, 289–315[CrossRef][Medline] [Order article via Infotrieve]
  48. Theil, E. C. (1990) Adv. Enzymol. Relat. Areas Mol. Biol. 63, 421–449[Medline] [Order article via Infotrieve]
  49. Gutfreund, H. (1995) Kinetics for the Life Sciences, pp. 103–137, Cambridge University Press, Cambridge, UK
  50. Tullius, T. D., Dombroski, B. A., Churchill, M. E., and Kam, L. (1987) Methods Enzymol. 155, 537–558[Medline] [Order article via Infotrieve]
  51. Orino, K., Kamura, S., Natsuhori, M., Yamamoto, S., and Watanabe, K. (2002) Biometals 15, 59–63[CrossRef][Medline] [Order article via Infotrieve]
  52. Wang, J. (1996) Annu. Rev. Biochem. 65, 635–692[CrossRef][Medline] [Order article via Infotrieve]
  53. Widom, J. (1998) Annu. Rev. Biophys. Biomol. Struct. 27, 285–327[CrossRef][Medline] [Order article via Infotrieve]
  54. Hill, T. L. (1986) An Introduction to Statistical Thermodynamics, pp. 59–73, Dover, New York

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 Nucleic Acids ResHome page
M. E. Beaudoin, V.-J. Poirel, and L. A. Krushel
Regulating amyloid precursor protein synthesis through an internal ribosomal entry site
Nucleic Acids Res., December 1, 2008; 36(21): 6835 - 6847.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
G. Bhattacharyya and A. Grove
The N-terminal Extensions of Deinococcus radiodurans Dps-1 Mediate DNA Major Groove Interactions as well as Assembly of the Dodecamer
J. Biol. Chem., April 20, 2007; 282(16): 11921 - 11930.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
R. Li, C. Luo, M. Mines, J. Zhang, and G.-H. Fan
Chemokine CXCL12 Induces Binding of Ferritin Heavy Chain to the Chemokine Receptor CXCR4, Alters CXCR4 Signaling, and Induces Phosphorylation and Nuclear Translocation of Ferritin Heavy Chain
J. Biol. Chem., December 8, 2006; 281(49): 37616 - 37627.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Cell. Biol.Home page
K. Iwasaki, E. L. MacKenzie, K. Hailemariam, K. Sakamoto, and Y. Tsuji
Hemin-Mediated Regulation of an Antioxidant-Responsive Element of the Human Ferritin H Gene and Role of Ref-1 during Erythroid Differentiation of K562 Cells.
Mol. Cell. Biol., April 1, 2006; 26(7): 2845 - 2856.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/15/14694    most recent
M313348200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
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 arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Surguladze, N.
Right arrow Articles by Fried, M. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Surguladze, N.
Right arrow Articles by Fried, M. G.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2004 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement