ATP–stimulated DNA–mediated Redox Signaling by XPD, a DNA
Repair and Transcription Helicase
Timothy P. Mui
#
,
Jill O. Fuss
†
,
Justin P. Ishida
†
,
John A. Tainer
†,+
, and
Jacqueline K.
Barton
#,*
#
Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena,
California 91125, United States
†
Life Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, California 94720,
United States
+
Department of Molecular Biology, Skaggs Institute of Chemical Biology, The Scripps Research
Institute, La Jolla, California 92037, United States
Abstract
Using DNA-modified electrodes, we show DNA-mediated signaling by XPD, a helicase that
contains a [4Fe-4S] cluster and is critical for nucleotide excision repair and transcription. The
DNA-mediated redox signal resembles that of base excision repair proteins, with a DNA-bound
redox potential of ~80 mV versus NHE. Significantly, this signal increases with ATP hydrolysis.
Moreover, the redox signal is substrate-dependent, reports on the DNA conformational changes
associated with enzymatic function, and may reflect a general biological role for DNA charge
transport.
To protect the genome, a variety of proteins with various functions must act in concert.
1–3
One such protein, XPD is a super family 2 helicase critical to nucleotide excision repair
(NER) and important to transcription.
4–6
Helicases are responsible for unwinding DNA in
an ATP-dependent fashion in order to access individual bases to allow the other proteins to
repair DNA damage and to both replicate and transcribe DNA. In humans, XPD is part of
the TFIIH machinery, with single site mutations leading to human diseases with increased
cancer risk or premature aging: Xeroderma Pigmentosum (XP), Cockayne Syndrome (CS),
Trichothiodystrophy (TTD) or combinations thereof.
4,5,7
Recent chemical analyses and
crystal structures of archaeal XPD homologues, which have ~22% sequence identity with
the human homologue, reveal the presence of a [4Fe-4S] cluster.
8–10
Furthermore, combined
structural, biochemical and mutational analyses show that the catalytic core of XPD is
conserved from archaea to humans and has functional relevance for understanding human
disease.
5
Mutational analyses of [4Fe-4S] coordinating cysteines have established the
importance of the [4Fe-4S] cluster in DNA unwinding activity,
10–12
yet a role for XPD as a
redox-active protein remains to be established.
DNA charge transport (CT), where electrons are transferred between proteins bound to DNA
in a path through the DNA bases, has been proposed as a first step in localizing a family of
base excision repair (BER) proteins containing [4Fe-4S] clusters in the vicinity of damage.
13
DNA CT chemistry facilitates electron transfer over long molecular distances through the
Corresponding Author: jkbarton@caltech.edu.
Supporting Information. Experimental procedures and supporting figures. This material is available free of charge via the Internet at
http://pubs.acs.org.
NIH Public Access
Author Manuscript
J Am Chem Soc
. Author manuscript; available in PMC 2012 October 19.
Published in final edited form as:
J Am Chem Soc
. 2011 October 19; 133(41): 16378–16381. doi:10.1021/ja207222t.
NIH-PA Author Manuscript
NIH-PA Author Manuscript
NIH-PA Author Manuscript
DNA duplex
14,15
but is remarkably sensitive to perturbations in base pair stacking, as, for
example, arise with damage.
16,17
DNA-mediated CT signaling by proteins was first explored in studies of a class of
E. coli
BER glycosylases that contain [4Fe-4S] clusters.
18,19
Electrochemistry on DNA-modified
electrodes showed that DNA binding shifts the cluster potential to ~80 mV, well within the
physiological range for redox signaling. Biophysical experiments were then used to examine
the redistribution of BER enzymes in the vicinity of damage, and genetic experiments, to
explore cooperative signaling between two BER enzymes, endonuclease (Endo) III and
MutY.
20,21
These results coupled with DNA electrochemistry linked the ability of Endo III
(i) to relocalize near a DNA mismatch, (ii) to cooperate in helping MutY repair lesions in
vivo, and (iii) to carry out DNA CT. If CT is generally important for DNA repair, we
hypothesized it should be detected in repair pathways besides BER, including NER, which is
the major pathway for chemically modified bases that disrupt the DNA double helix.
To test if DNA CT might occur with XPD, we first determined the DNA-bound redox
potential of an archael XPD on DNA-modified gold electrodes (Scheme 1, see Supporting
Information for Methods). We find that XPD from the thermophile
Sulfolobus
acidocaldarius
(SaXPD) has a DNA-bound redox potential of ~82 ± 10 mV versus NHE
(Figure 1). This potential, like those found in BER proteins, reflects physiological redox
activity and is not sufficient to damage DNA.
18
In the absence of DNA, the potential of the
cluster is expected to be significantly more positive and outside of the window of
physiological redox activity.
19
To confirm that the measured potential is DNA-mediated and
hence reflects the DNA-bound potential, rather than that of the protein directly interacting
with the surface without DNA, we compared the signal to that found on a surface with the
DNA containing a mismatch close to the surface; with this intervening mismatch, the
potential is unchanged yet the redox signal is significantly attenuated, consistent with the
protein electrochemistry signal being DNA-mediated (Figure 1).
16
We also observe that the
signal intensity exhibits a linear dependence on the square root of the scan rate, which
implies that the protein is binding to DNA in a diffusion-limited process.
22
In addition, we
observe an electron transfer rate of approximately 1.4 s
−
1
based on Laviron analysis, which
is similar to previously published rates that indicate that the rate of electron transfer is
limited by tunneling through the carbon linker.
23
Together these data establish that this
DNA-mediated signal corresponds to the one-electron redox couple of the [4Fe-4S] cluster
of SaXPD bound to DNA and that this redox couple can be physiologically active.
Unlike BER glycosylases, the principal activity of the XPD helicase is ATP-dependent, and
as such, we could also investigate the effect of ATP on the DNA-bound signal. After the
protein was allowed to equilibrate on the DNA-modified surface, various concentrations
were added of ATP or ATP-
γ
-S, a markedly more slowly hydrolysable ATP analogue
(Figure 2A).
24
Interestingly, as ATP was titrated onto the surface, a noticeable ATP-
dependent increase in the current was observed. No shift in potential was evident, indicating
that the cluster is neither degrading nor markedly changing in its environment; instead DNA
coupling appears to increase. In contrast, the slowly hydrolysable analogue shows little
effect on the electrochemical signal of the protein. Thus, the electrochemical signal reports
on the ATPase activity of XPD. This sensitivity in signal to ATP hydrolysis is remarkable
given that, based on the crystal structure of the protein without DNA bound, the distance
between the cluster and the ATP binding site is 30 Å.
10
The XPD structure revealed that the
[4Fe-4S] cluster domain is tightly linked to the ATP binding site by
β
-sheets that could
provide a mechanism for mechanical coupling of the motions of the cluster domain to those
in the ATP site as a result of hydrolysis. This increase in electrochemical signal must be
reporting on motions at the protein/DNA interface as the protein carries out ATP hydrolysis.
Mui et al.
Page 2
J Am Chem Soc
. Author manuscript; available in PMC 2012 October 19.
NIH-PA Author Manuscript
NIH-PA Author Manuscript
NIH-PA Author Manuscript
We next examined the effect of DNA substrate (Figure 2). XPD has shown a preference as a
5
′
-3
′
helicase;
25
modeled on a surface, the 5
′
-3
′
helicase would be expected, therefore, to
move downward towards the surface. Since the protein concentration is well above the
dissociation constant (66 nM),
10
the protein should bind DNAs with either 3
′
- or 5
′
-
overhangs, as well as duplex DNA. Indeed, we see differences in the DNA electrochemistry
based on substrate. When protein is placed on the surface modified with DNA with a 3
′
ssDNA overhang, with each addition of ATP, the signal increases temporarily, likely
reflecting increased coupling associated with reaction, but then, with each additional ATP
jump, decreases with time. This decay may result from the protein sliding off the small
segment of DNA bound to the surface. Notably, the absolute signal could not be compared
among the surfaces because of the variability in surface coverage using different substrates.
On the fully duplexed surface, which has no directionality bias, we see after the ATP
addition, the signal is mainly flat, as expected for binding by XPD in both orientations.
XPD mutations in humans are associated with several often fatal diseases: XP, CS and
TTD.
4,5,7
One such mutant, G34R, shows attenuated ATPase and helicase activity relative to
wild type (WT) protein in biochemical assays (Table 10).
10
Interestingly, this SaXPD
mutant exhibits a redox signal comparable to that of WT in the absence of ATP (Table 1).
However, the rate of electronic signal increase with ATP for the G34R mutant is
significantly lower compared to WT, further demonstrating the sensitivity of our assay to
ATP hydrolysis (Figure 3). While there is certainly not a simple linear relationship between
activity measured electrically and biochemically, the electronic signal appears to be a
sensitive reporter of changes in protein/DNA coupling that result from ATP hydrolysis. This
assay complements well a fluorescence helicase assay seen earlier, but without the need for
DNA labeling.
26
XPD, a critical helicase for NER, thus contains a redox-active [4Fe-4S] cluster that is sensed
electronically as a reporter of activity. The DNA-bound redox potential is similar to
previously reported BER proteins. Here, DNA electrochemistry is seen to provide a
sensitive means for detecting ATP-dependent signaling that may be generally useful in
screening the activity of DNA-binding proteins containing redox centers. The activity can be
distinguished between the WT protein and ATP-deficient activity of the G34R mutant as
well as between the native and non-native DNA substrates. Additionally, these results
prompt the question as to how this electronic signaling of XPD activity might be utilized
in
vivo
, and further, to which proteins XPD may be signaling inside the cell. Various repair and
replication proteins, including FancJ and Dna2, which act in maintaining genomic stability,
also have an associated [4Fe-4S] cluster.
27,28
We suggest that it will be important to test
other proteins acting at the interface of repair with major DNA processes of replication and
transcription for their CT ability.
Supplementary Material
Refer to Web version on PubMed Central for supplementary material.
Acknowledgments
This research was supported by the NIH (GM49216 to J.K.B. and CA112093 to J.A.T.) and the DOE (ENIGMA
program under Contract No. DE-AC02-05CH11231 to J.A.T.) T.P.M. also thanks the NSF for a graduate
fellowship.
References
1. Eisen JA, Hanawalt PC. Mutat Res. 1999; 435:171. [PubMed: 10606811]
2. Fry RC, Begley TJ, Samson LD. Annu Rev Microbiol. 2005; 59:357. [PubMed: 16153173]
Mui et al.
Page 3
J Am Chem Soc
. Author manuscript; available in PMC 2012 October 19.
NIH-PA Author Manuscript
NIH-PA Author Manuscript
NIH-PA Author Manuscript
3. David SS, O’Shea VL, Kundu S. Nature. 2007; 447:941. [PubMed: 17581577]
4. Sancar A. Annu Rev Biochem. 1996; 65:43. [PubMed: 8811174]
5. Fuss JO, Tainer JA. DNA Repair (Amst). 2011; 10:697. [PubMed: 21571596]
6. de Laat WL, Jaspers NG, Hoeijmakers JH. Genes Dev. 1999; 13:768. [PubMed: 10197977]
7. Lehmann AH. Genes Dev. 2001; 15:15. [PubMed: 11156600]
8. Liu H, Rudolf J, Johnson KA, McMahon SA, Oke M, Carter L, McRobbie AM, Brown SE,
Naismith JH, White MF. Cell. 2008; 133:801. [PubMed: 18510925]
9. Wolski SC, Kuper J, Hanzelmann P, Truglio JJ, Croteau DL, Van Houten B, Kisker C. PLoS Biol.
2008; 6:e149. [PubMed: 18578568]
10. Fan L, Fuss JO, Cheng QJ, Arvai AS, Hammel M, Roberts VA, Cooper PK, Tainer JA. Cell. 2008;
133:789. [PubMed: 18510924]
11. Rudolf J, Makrantoni V, Ingledew WJ, Stark MJ, White MF. Mol Cell. 2006; 23:801. [PubMed:
16973432]
12. Pugh RA, Honda M, Leesley H, Thomas A, Lin Y, Nilges MJ, Cann IK, Spies M. J Biol Chem.
2008; 283:1732. [PubMed: 18029358]
13. Genereux JC, Boal AK, Barton JK. J Am Chem Soc. 2010; 132:891. [PubMed: 20047321]
14. Genereux JC, Barton JK. Chem Rev. 2010; 110:1642. [PubMed: 20214403]
15. Slinker JD, Muren NB, Renfrew SE, Barton JK. Nat Chem. 2011; 3:230.
16. Boon EM, Ceres DM, Drummond TG, Hill MG, Barton JK. Nat Biotechnol. 2000; 18:1096.
[PubMed: 11017050]
17. Boal AK, Barton JK. Bioconjugate Chem. 2005; 16:312.
18. Boal AK, Yavin E, Lukianova OA, O’Shea VL, David SS, Barton JK. Biochemistry. 2005;
44:8397. [PubMed: 15938629]
19. Gorodetsky AA, Boal AK, Barton JK. J Am Chem Soc. 2006; 128:12082. [PubMed: 16967954]
20. Boal AK, Genereux JC, Sontz PA, Gralnick JA, Newman DK, Barton JK. Proc Natl Acad Sci
USA. 2009; 106:15237. [PubMed: 19720997]
21. Romano CA, Sontz PA, Barton JK. Biochemistry. 2011; 50:6133. [PubMed: 21651304]
22. Bard, AJ.; Faulkner, LR. Electrochemical Methods – Fundamental and Application. 2. John Wiley
& Sons; New York: 2000. p. 580-631.
23. Drummond TG, Hill MG, Barton JK. J Am Chem Soc. 2004; 126:15010. [PubMed: 15547981]
24. Eckstein F. Angew Chem, Int Ed. 1983; 22:423.
25. Singleton MR, Dillingham MS, Wigley DB. Annu Rev Biochem. 2007; 76:23. [PubMed:
17506634]
26. Ha T, Rasnik I, Cheng W, Babcock HP, Gauss GH, Lohman TM, Chu S. Nature. 2002; 419:638.
[PubMed: 12374984]
27. Wu Y, Suhasini AN, Brosh RM Jr. Cell Mol Life Sci. 2009; 66:1209. [PubMed: 19099189]
28. Vaithiyalingam S, Warren EM, Eichman BF, Chazin WJ. Proc Natl Acad Sci USA. 2010;
107:13684. [PubMed: 20643958]
Mui et al.
Page 4
J Am Chem Soc
. Author manuscript; available in PMC 2012 October 19.
NIH-PA Author Manuscript
NIH-PA Author Manuscript
NIH-PA Author Manuscript
Figure 1.
Electrochemistry of DNA-bound WT SaXPD. (Top) Cyclic voltammogram (CV) of SaXPD
[120
μ
M] on a well-matched DNA-modified electrode (red) and on DNA with a CA
mismatch located near the gold surface as in Scheme 1 (blue). (Ag/AgCl reference
electrodes; Pt auxiliary electrode, 50 mV/s scan rate, NHE = normal hydrogen electrode).
(Bottom) Plot of current versus
ν
1/2
(square root of scan rate). The data indicate that our
signal is obtained through a diffusion-limited process.
Mui et al.
Page 5
J Am Chem Soc
. Author manuscript; available in PMC 2012 October 19.
NIH-PA Author Manuscript
NIH-PA Author Manuscript
NIH-PA Author Manuscript
Figure 2.
ATP-dependent Electrochemistry of SaXPD on different DNA substrates. Time points
shown are every 6 minutes. (Ag/AgCl reference electrode; Pt auxiliary electrode, 50 mV/s
scan rate). % Difference in Current for SaXPD [9
μ
M] on a 5
′
-ssDNA overhang
(A)
, 3
′
-
ssDNA overhang
(B)
, fully duplexed DNA
(C)
. The signal is seen to be ATP-dependent and
sensitive to substrate. Note at the high concentration addition [5 mM ATP-
γ
-S], some ATP
still remains; once depleted, the signal levels off.
Mui et al.
Page 6
J Am Chem Soc
. Author manuscript; available in PMC 2012 October 19.
NIH-PA Author Manuscript
NIH-PA Author Manuscript
NIH-PA Author Manuscript