Chemistry & Biology

Article

A Small Molecule Screen Identifies an Inhibitor

of DNA Repair Inducing the Degradation of TFIIH
and the Chemosensitization of Tumor Cells to Platinum
Sergey Alekseev,1 Mériam Ayadi,2 Laurent Brino,1 Jean-Marc Egly,1 Annette K. Larsen,2 and Frédéric Coin1,*
1IGBMC, Department of Functional Genomics and Cancer, Equipe Labellisée Ligue 2014, CNRS/INSERM/Université de Strasbourg,

BP 163, 67404 Illkirch Cedex, CU Strasbourg, France

2Laboratory of Cancer Biology and Therapeutics, Centre de Recherche Saint-Antoine, INSERM and Université Pierre et Marie Curie, UPCM,

4 Place Jussieu, 75005 Paris, France

*Correspondence: [email protected]
http://dx.doi.org/10.1016/j.chembiol.2013.12.014

SUMMARY Helicases allow recruitment of the complex and unwinding of

the DNA around the lesion thereby providing a three-dimensional
Nucleotide excision repair (NER) removes DNA structure able to recruit XPA and RPA (Oksenych et al., 2009).
lesions resulting from exposure to UV irradiation The edges of the resulting DNA bubble are recognized by XPG
or chemical agents such as platinum-based drugs and ERCC1-XPF, two junction-specific endonucleases with
used as anticancer molecules. Pharmacological opposite single-strand polarities that generate 30 and 50 single
inhibition of NER is expected to enhance chemosen- DNA incisions relative to the damage, respectively (O’Donovan
et al., 1994; Sijbers et al., 1996). This dual DNA incision leads
sitivity but nontoxic NER inhibitors are rare. Using a
to the excision of a 24–32-mer damaged oligonucleotide and
drug repositioning approach, we identify spironolac-
to gap-filling DNA resynthesis (Ogi et al., 2010).
tone (SP), an antagonist of aldosterone, as a potent DNA lesions that are removed by NER are chemically and
NER inhibitor. We found that SP promotes a rapid structurally diverse, being caused by a wide variety of chemical
and reversible degradation of XPB, a subunit of agents and UV. Among these lesions, those derived from cell
transcription/repair factor TFIIH. Such degradation exposure to cisplatin and its derivative have attracted much
depends both on ubiquitin-activating enzyme and attention due to their widespread use in cancer chemotherapy.
on the 26S proteasome. Supplementation of extracts Cisplatin, as well as its derivatives (carboplatin, oxaliplatin), is
from SP-treated cells with purified TFIIH restored now largely employed for the treatment of a wide range of solid
TFIIH-dependent repair and transcription activities tumors, including testicular, bladder, ovarian, colorectal, lung,
in vitro, demonstrating the specific impact of SP and head and neck cancers (Galanski, 2006).
Platinum-based compounds exert their antineoplastic effect
on two fundamental functions of TFIIH. Finally, SP
through direct binding to DNA resulting in saturation of several
potentiated the cytotoxicity of platinum derivatives
DNA repair mechanisms, including NER. Consequently, plat-
toward tumor cells, making it a potential therapeutic inum blocks fundamental cellular processes such as transcrip-
and research tool. tion and replication ultimately resulting in apoptotic cell death.
Because platinum derivatives constitute the major therapeutic
option in many clinical settings, the development of chemo-
INTRODUCTION sensitization strategies constitutes a goal with important
clinical implications. Indeed, inhibition of DNA repair pathways
One of the most versatile mammalian repair pathways is nucleo- such as NER may increase the efficiency of current chemo-
tide excision repair (NER), in which two distinct modes of therapeutic regimes, thereby minimizing resistance and ulti-
damage recognition are operating through the sequential mately decreasing the possibility of the negative side effects
accumulation of NER factors at the damage site (de Boer and (Kelland, 2007). However, DNA repair inhibitors are generally
Hoeijmakers, 2000). Within transcription-coupled repair (TCR), toxic in the own right, whereas siRNA-directed strategies to
the removal of lesions located on the transcribed strand of knock down proteins have not proven readily applicable in a
a gene is rapid owing to their detection by the elongating clinical setting (Burnett and Rossi, 2012). Hence, we aimed
transcription machinery (Fousteri and Mullenders, 2008). The at identifying potent, nontoxic, and specific drug candidates
XPC-hHR23B complex locates injuries in the nontranscribed able to inhibit NER, thereby potentially improving existing
sequences to initiate the slower process of global genome repair cancer therapies (Barakat et al., 2012). We established a
(GGR) (Sugasawa et al., 1998). Both subpathways converge cell-based screening assay and applied a chemical library to
subsequently into a common process through the recruitment this screen. Among the 1,200 molecules tested, spironolactone
of the multiprotein complex TFIIH. This transcription/repair (17-hydroxy-7a-mercapto-3-oxo-17a-pregn-4-ene-21-carboxylic
factor contains several enzymatic activities including helicases acid g-lactone acetate) (SP) was selected owing to its capacity
(XPB and XPD) and a kinase (cdk7) (Compe and Egly, 2012). to induce a dramatic decrease in NER activity that was

398 Chemistry & Biology 21, 398–407, March 20, 2014 ª2014 Elsevier Ltd All rights reserved
Chemistry & Biology
DNA Repair Inhibitor Targets XPB Helicase

A B

C D

E F

Figure 1. SP Inhibits the Repair of UV-Induced DNA Damage

(A) Schematic representation of the protocol that was used for the identification of NER inhibitors. Following 12 hr incubation with the chemical library in 96-well
plates, cells were UVC irradiated (60 J/m2), postincubated for 3 hr, and fixed. Subsequently (6-4)PP lesions were labeled using an anti-(6-4)PP antibody and
detected by high throughput microscopy (HTM).
(B) The protocol presented in (A) was validated using shCTL and shXPA HeLa cells. The levels of remaining (6-4)PP lesions was determined using HTM. One
hundred percent corresponds to the (6-4)PP signal measured just after UV irradiation.
(C) Results of the screening performed with the 1,200 small molecules from the Prestwick library tested at 10 mM following the protocol described in (A). The
vertical axis represents the percent of (6-4)PP lesions relative to the level measured right after irradiation that is set to 100% in all experiments.
(D) (6-4)PP repair inhibitory activity of different SP concentrations as determined by HTM. Antazoline (AN) did not induce any (6-4)PP repair inhibitory activity in
our screening and thus was used as negative control. The vertical axis represents the percent of (6-4)PP lesions relative to the level measured just after irradiation
that is set to 100% in all experiments. The values represent the averages (±SD) from three independent experiments.
(E) Structure of Budesonide, Cortisone, Diflorasone, and Eplerenone, four compounds structurally close to SP (shown at the center of the figure).
(F) The relative NER inhibitory activity of Budesonide (1), Cortisone (2), Diflorasone (3), and Eplerenone (4) was determined and compared to SP (5) following
the protocol described in (A). The vertical axis represents the percent of (6-4)PP lesions relative to the level measured just after irradiation that is set to 100% in all
experiments. The values represent averages (±SD) from three independent experiments.
See also Figure S1.

associated with a rapid and reversible degradation of the RESULTS

XPB helicase subunit of TFIIH depending both on ubiquitin-
activating enzyme and on proteasome. Furthermore, we have SP Inhibits the Repair of UV-Induced DNA Lesions
shown that a combination of SP and platinum derivatives To set up a screening for NER inhibitors, we optimized an assay
produced significant increases in cytotoxicity toward human detecting UV-induced DNA damages in a 96-well plate and
colon and ovarian carcinoma cells, pointing to the reposition- automated the quantification of the nuclear signal for (6-4)PP
ing of this small nontoxic molecule as an adjuvant in plat- lesions, a major UV-induced lesion (de Laat et al., 1999), through
inum-based chemotherapy. high-throughput microscopy (HTM) (Figure 1A). During the

Chemistry & Biology 21, 398–407, March 20, 2014 ª2014 Elsevier Ltd All rights reserved 399
 Chemistry & Biology
DNA Repair Inhibitor Targets XPB Helicase

A D Figure 2. SP Increases the Sensitivity to

UVC and Cisplatin Treatment
(A and B) Quantitative UV (A) or cisplatin (B) sur-
vival analysis of HeLa cells treated with DMSO or
SP (10 mM). Results are expressed as percent of
cell survival relative to the nontreated controls
that is set to 100%. The values represent averages
(±SD) from three independent experiments;
p value is indicated (**p < 0.01; ***p < 0.001).
(C) Cytotoxic activity of SP(10 mM) or cisplatin
(2.5 mM) in HeLa cells treated 24 hr with the indi-
cated compound and stained with 7-AAD before
flow cytometry analysis. Results are expressed as
percent of dead cells (positive for 7-AAD staining)
B E in the total population. The values represent
averages (±SD) from three independent experi-
ments; p value is indicated.
(D) Upper panel: increasing amount (1 mg, 3 mg,
and 6 mg) of nuclear extracts from DMSO
(NEDMSO)-treated or SP (NESP)-treated HeLa cells
were analyzed for dual incision. NEs were incu-
bated with cis-platinum mono-damaged plasmid
(Riedl et al., 2003). Incision products are indicated
as radiolabeled 27 nt to 34 nt products. Several
incisions products are observed that depend on
the cutting site of XPG that varies from two to three
C nucleotides. As control, a reconstituted system
(Rec) with recombinant proteins was used (Riedl
et al., 2003). Lower panel: similar amount of NE as
tested above was analyzed by western blotting
with an anti-TBP antibody.
(E) DMSO-treated or SP-treated HeLa cells were
incubated in the presence (+) or absence () of
100 ng/ml NCS for 30 min. Cells were then cultured
without NCS and harvested at 6 or 24 hr. Cells
were lysed and analyzed by western blotting with
anti-gH2AX or anti-TBP antibodies. When indi-
cated, cells were treated with DNA-PK inhibitor
NU7026 during the recovery time.

screening, we exposed cells to the drugs for 12 hr, irradiated and showed almost complete inhibitory activity at 1 and 10 mM
them with UVC (60J/m2) and postincubated them for 3 hr in and a partial response (40% inhibition) at submicromolar con-
drug-free media before processing for (6-4)PP HTM immunoflu- centration (Figure 1D).
orescence, using anti-(6-4)PP antibody (Mori et al., 1991). As In the chemical library, some compounds were structurally
proof of principle, use of HeLa cells knocked down for the NER related to SP (Figure 1E). We tested several of these chemicals
factor XPA (Le May et al., 2010) was associated with 70% resid- individually and observed no potent inhibition of NER (Figure 1F).
ual (6-4)PP lesions 3 hr postirradiation, in marked contrast to the Eplerenone (Figure 1E) is a SP derivative designed to enhance
shRNA control cells that displayed 10% residual (6-4)PP signal the selective binding to the mineralocorticoid receptor (de Gas-
(Figure 1B). paro et al., 1987). However, Eplerenone did not significantly
Prestwick Chemical Library (http://www.prestwickchemical. inhibit NER in our assay (Figure 1F). We concluded from these
com/index.php?pa=26) containing 1,200 small molecules was experiments that the effect of SP in preventing the removal of
then used at a 10 mM single point concentration. More than DNA lesions by NER is an additional and highly specific function
90% of the chemicals had no effect on NER with a (6-4)PP signal of this molecule, independent from its aldosterone antagonist
<20% of the initial signal measured just after UV irradiation (Fig- function. We therefore focused on this compound for the subse-
ure 1C and Figure S1 available online). The compounds that quent analyses.
had resulted in a (6-4)PP signal higher than 80% were studied.
Six of these compounds were excluded because they showed SP Increases Cell Sensitivity to UV and Cisplatin
a high cellular toxicity following the pre-irradiation incubation, Treatment
which may explain the NER inhibition. We also excluded A hallmark of NER-deficient cells is their sensitivity to UV- or
Merbromin as a false positive because it has a major excitation cisplatin-induced DNA lesions (Furuta et al., 2002; Gillet and
peak at 498 nm, the wavelength used in our screening. Finally, Schärer, 2006). In a survival assay, SP-treated cells were more
only the SP, an aldosterone antagonist that binds to the sensitive to UVC irradiation or cisplatin treatment, compared
mineralocorticoid receptors, was retained as a potential NER to vehicle (DMSO)-treated control cells (Figures 2A and 2B,
inhibitor candidate. SP was tested at different concentrations respectively). Furthermore, viability values of cells treated with

400 Chemistry & Biology 21, 398–407, March 20, 2014 ª2014 Elsevier Ltd All rights reserved
Chemistry & Biology
DNA Repair Inhibitor Targets XPB Helicase

SP (10 mM) alone and determined using 7-AAD staining of dead We also analyzed the amount of endogenous XPB directly
cells showed no significant cytotoxic activity of this compound in vivo by immunofluorescence. The intensity of the nuclear
following 24 hr incubation (Figure 2C). staining of XPB was strongly reduced after 2 hr treatment with
We then prepared nuclear extracts (NE) from HeLa cells SP (Figures 3D, panels b–d, and S2D). In contrast, immunostain-
treated with SP (10 mM) for 2 hr and tested them in a dual incision ing of the same cells with anti-XPC antibody did not reveal sig-
of a single 1,3-intrastrand d(GpTpG) cisplatin-DNA crosslink nificant differences (Figures 3D, panels a–c, and S2D). Using
in vitro (Coin et al., 2006). NEs prepared from SP-treated cells live cell imaging, we also monitored the decrease in the nuclear
were highly deficient in dual incision assay (Figure 2D), indicating concentration of exogenous XPB by time-lapse microscopy
that SP affects one of the basal steps of NER. Finally, we following SP treatment of HeLa cells transfected with XPB-
analyzed whether the inhibition induced by SP was restricted GFP (Oksenych et al., 2009). The homogenous nuclear XPB
to NER or included other DNA repair pathways. For this purpose, signal progressively decreased within the time frame of the SP
we estimated the rate of double-strand DNA break (DSB) repair treatment (Figure 3E; Movie S1), without any modification of
by analyzing the persistence of phosphorylated histone H2AX XPB cellular localization (Movie S1). In total agreement with the
(gH2AX) taking place after treatment with the DSB-inducer neo- data on endogenous XPB (Figure S2B), we estimated the half-
carzinostatin (NCS). In unsynchronized cells, the DSB repair rate life of exogenous XPB to be 70 min in the presence of SP.
in SP-treated cells was not reduced in clear contrast to the These results indicate that SP induces a rapid decrease of the
persistence of phospho-H2AX induced by the treatment with XPB concentration in the nucleus without cellular relocalization
the DNA-PK inhibitor NU7026 (Figure 2E). These findings sug- of the protein.
gest that SP selectively inhibits NER over DSB repair.
SP Induced Ubiquitin-Activating Enzyme- and
SP Downregulates XPB Proteasome-Dependent Degradation of XPB
The above results indicate that SP hinders a crucial step of the We next investigated the mechanism of SP-induced XPB
NER reaction that occurs by sequential assembly of repair fac- downregulation. Two hour treatment with SP or DMSO was not
tors at the site of DNA damage (Volker et al., 2001). Therefore, associated with any detectable differences in the levels of XPB
we examined the distribution of NER factors 15 min after UV mRNA (Figure 4A) indicating a direct effect of SP at the protein
irradiation (100 J/m2) using HeLa cells covered with isopore pol- level. Indeed, addition of the proteasome inhibitor MG-132
ycarbonate filters with 5 mm diameter pores (Volker et al., 2001). during SP treatment preserved XPB protein levels (Figure 4B).
In control cells, immunostaining against the damage recognition We next used PYR-41, a small molecule chemical inhibitor of
factor XPC showed that it was located in discrete local spots of ubiquitin-activating enzymes (Yang et al., 2007). HeLa cells
the nucleus corresponding to irradiated areas (Figure 3A, panels pretreated with PYR-41 did not show any decrease of XPB
a, e, and i). In these cells, p62 (a subunit of TFIIH), XPA, and XPF following incubation with SP (Figure 4C). Finally, XPB interacts
colocalized with the XPC spots, indicating that they were effi- with Sug1, a component of the 26S proteasome (Weeda et al.,
ciently recruited to the damaged sites (Figure 3A, panels c, g, 1997). Accordingly, siRNA-mediated knock down of Sug1
and k). Surprisingly, while XPC continued to accumulate effi- partially restored XPB protein level in SP-treated cells (Figures
ciently on UV-irradiated spots of the SP-treated cells (Figure 3A, 4C and S2E). Together, these findings suggest a critical role
panels b, f, and j), p62, XPA, and XPF did not (Figure 3A, panels for ubiquitin-activating enzymes and the proteasome in the
d, h, and i). downregulation of XPB induced by SP.
We then initiated a systematic analysis of the relative amounts
of NER factors in nuclear extracts obtained from SP-treated SP-Induced XPB Degradation Impairs Transcription
cells lysed at different time points. While the amounts of the Besides its role in DNA repair, TFIIH is also a basal transcription
NER factors XPC, XPA, RPA, XPG, and XPF remained un- factor involved in the hormone-dependent transcriptional
changed along the time course, we noticed a rapid decrease response (Compe and Egly, 2012). As a consequence, the effect
of the concentration of XPB, the biggest TFIIH subunit, after of SP on gene expression was studied in HeLa cells by moni-
2 hr treatment with SP (Figure 3B). TFIIH is a ten-subunit factor toring the transcription of Cyp26 and RARb2, two retinoic acid
(Compe and Egly, 2012): all core subunits that we tested (p62, (t-RA)-responsive genes (Le May et al., 2012). We observed
p52, p44, or TTDA) were reduced by 6 hr after SP treatment that 2 hr of exposure to SP (10 mM) before incubation with t-RA
compared to DMSO vehicle control (Figure 3B). However, the over 6 hr resulted in dose-dependent inhibition of Cyp26 and
downregulation of these subunits clearly took place at a slower RARb2 gene expression (Figures 5A and 5B). In addition, in an
rate compared to XPB (Figure S2A). Using a shorter time in vitro transcription assay (Gerard et al., 1991), we noticed a
course, we determined the half-life of XPB to be 70 min reduced transcriptional activity of the NESP that can be restored
(Figure S2B). following the addition of purified TFIIH (Figure 5C, compare
The results above suggest that the SP-induced NER defi- lanes 7–9 with 10–12). These data demonstrated that the SP-
ciency is due to a defect in TFIIH. To test this hypothesis, we sup- induced degradation of XPB impairs both basal and activated
plemented NESP with highly purified TFIIH (Coin et al., 1998) mRNA transcription.
(Figure S2C) in a NER assay and observed a clear recovery of
dual incision (Figure 3C, compare lanes 7–9 to 10–12). In a Reversion of SP-Induced Degradation of XPB
normal situation, TFIIH is not a limiting factor because its addition We next tested whether the SP-induced degradation of XPB was
to NEDMSO did not increase DNA repair (Figure 3C, compare reversible. Following treatment of HeLa cells with SP (10 mM)
lanes 1–3 to 4–6). for 4 hr, the drug was removed and fresh medium was added

Chemistry & Biology 21, 398–407, March 20, 2014 ª2014 Elsevier Ltd All rights reserved 401
 Chemistry & Biology
DNA Repair Inhibitor Targets XPB Helicase

B C

D E

Figure 3. SP-Induced NER Deficiency Is due to the Downregulation of TFIIH

(A) Following incubation with DMSO or SP (10 mM) for 2 hr, HeLa cells were locally UV-irradiated (100 J/m2), fixed after 15 min, and costained with antibodies to
XPC (a, b, e, f, i, and j) and to either TFIIH(p62) (c and d), XPA (g and h), or XPF (k and l). Arrows indicate locally irradiated areas.
(B) HeLa cells were incubated with DMSO or SP (10 mM) for the indicated time before lysis. Extracts from DMSO-treated or SP-treated cells were resolved by
SDS-PAGE and subsequently analyzed by immunoblotting. The different proteins are indicated and black arrows point to the corresponding bands. CTL, control.
(C) HeLa cells were incubated with DMSO or SP (10 mM) for 2 hr. Increasing amounts (1 mg, 3 mg, and 6 mg) of nuclear extracts from DMSO (NEDMSO)-treated or SP
(NESP)-treated HeLa cells were analyzed for dual incision as described in Figure 2C. When indicated, purified TFIIH (100 ng) from HeLa was added. Optical density
was determined using a Typhoon PhosphorImager 8600 and expressed as relative activity (RA).
(D) HeLa cells were treated with DMSO or SP (10 mM) for 2 hr, fixed and incubated with anti-XPC (a–c) and anti-XPB antibodies (b–d).
(E) HeLa cells were transfected with XPB-GFP and treated with either DMSO or SP (10 mM). Fluorescence was determined by live imaging every 12 min during
120 min and plotted on the graph. The values are indicated relative to the expression level of XPB-GFP before treatment that is set to 100. Values represent
averages (±SD) from three independent experiments.
See also Figure S2 and Movie S1.

before cells were lysed at different recovery time points. We with the restoration of the XPB protein levels, we observed a
observed a progressive recovery of XPB protein level along the recovery of (6-4)PP removal that reached the levels of the
time course of the recovery time (Figure 6A), which was pre- DMSO-treated control cells, 6 hr after the removal of SP (Fig-
vented by addition of cycloheximide (CHX), an inhibitor of the ure 6C). These observations indicate that SP-induced degrada-
de novo protein synthesis (Figure 6B, lanes 3 and 6). Together tion of XPB is rapidly reversible.

402 Chemistry & Biology 21, 398–407, March 20, 2014 ª2014 Elsevier Ltd All rights reserved
Chemistry & Biology
DNA Repair Inhibitor Targets XPB Helicase

A SP Chemo-Sensitizes Human Cancer Cells to Platinum

Derivatives
To determine the influence of SP on the sensitivity of cancer
cells to platinum-based compounds, we examined the cytotoxic
effect of two platinum-derivatives, cisplatin and oxaliplatin,
toward human cancer cells. In the absence of SP, IC50 values
of cisplatin and oxaliplatin toward A2780 ovarian and HCT-116
colon carcinoma cells were determined to be 2.8 and 1.2 mM,
respectively. Cotreatment with SP significantly decreased the
IC50 values of cisplatin and oxaliplatin to 0.9 and 0.7 mM, respec-
tively (Figures 7A and 7B). Together with an increase in the
platinum-induced cytotoxicity, SP induced XPB degradation in
B these cells (Figure 7C). Taken together, our data indicate that
the presence of SP increases the sensitivity of different types
of human cancer cells to platinum-based anticancer agents.

DISCUSSION

In our study, we show that SP induced the degradation of the

XPB subunit of TFIIH, which leads to an impairment of the NER
C
pathway. Platinum-based agents overwhelm DNA repair mech-
anisms involved in their removal such as the NER pathway.
Consequently, the platinum-derived lesions impair various
DNA-associated processes like replication and transcription
leading to apoptosis of a large fraction of tumor cells. Although
the initial response to platinum is high, there are several limita-
tions to the use of platinum-based compounds in cancer therapy
D including the toxic side effects as well as development of
resistance (Köberle et al., 2010). NER is considered to be a vital
target to improve platinum-based cancer therapy and reduce the
resistance (Barakat et al., 2012). Consequently, discovery of
potent, nontoxic DNA repair inhibitors could be an important
step in modern cancer therapy aimed at improving existing
treatments. With this paradigm in mind, we decided to follow a
drug repositioning approach and search for NER inhibitors in
the Prestwick Chemical Library that contain a limited number
of highly diverse drug molecules for which bioavailability and
toxicity studies have already been performed. This characteristic
Figure 4. SP Induces Ubiquitin-Activating Enzymes and Protea- was a priority for us because the cytotoxicity of many DNA
some-Dependent Degradation of XPB repair inhibitors such as Wortmannin or caffeine precludes their
(A) HeLa cells were treated with DMSO or SP (10 mM) for 2 hr. Total RNA was clinical use.
extracted and XPB mRNA was quantified by RT-qPCR. Values are normalized Using a functional assay measuring the removal of the (6-4)PP
relative to GAPDH expression. Values represent averages (±SD) from three
lesions by HTM, we identified in a drug repositioning approach,
independent experiments.
SP as a potent inhibitor of NER. UV irradiation was chosen
(B) HeLa cells were incubated in the presence (+) or absence () of the pro-
teasome inhibitor MG132 (10 mM) for 1 hr before addition of DMSO or SP instead of platinum for technical reasons but additional experi-
(10 mM) for 4 hr. Cells were lysed and the extracts were resolved by SDS-PAGE ments performed in vitro and ex vivo demonstrated that similar
and analyzed by immunoblotting for XPB or TBP. inhibition of platinum-derived lesions was observed with SP
(C) HeLa cells were incubated in the presence (+) or absence () of the ubiq- (see Figure 2). Strikingly, none of the structurally related analogs
uitin-activating enzyme inhibitor PYR-41 (35 mM) for 2 hr before addition of of SP present in the Prestwick library showed potent NER
DMSO or SP (10 mM) for 4 hr. Cells were lysed and the extracts were resolved
inhibition indicating that a strict conservation of the structure of
by SDS-PAGE and analyzed by immunoblotting for XPB or TBP.
(D) HeLa cells were transfected with Sug1-pool siRNA or Control (CTL) smart- SP is required to obtain NER inhibition. This was further
pool siRNA (Dharmacon) for 36 hr following incubation with DMSO or SP confirmed by the use of eplerenone, an aldosterone antagonist
(10 mM) for 4 hr. Cells were lysed and extracts were resolved by SDS-PAGE that is as derivative of SP, which did not show any NER inhibi-
and analyzed by immunoblotting. The signals were quantified using Genetool tion activity. Because the antihypertensive action of eplerenone
(Syngene) and indicated below the immunoblot. and SP includes their binding to the mineralocorticoid receptors,
we conclude that SP impairs NER by a different pathway, unre-
lated to its function as an aldosterone antagonist. In our assay,
SP was used at 10 mM, a concentration that is widely used for
potent inhibitors of ATM or DNA-PKcs tested in cellular assays

Chemistry & Biology 21, 398–407, March 20, 2014 ª2014 Elsevier Ltd All rights reserved 403
 Chemistry & Biology
DNA Repair Inhibitor Targets XPB Helicase

A B Figure 5. SP-Induced Degradation of TFIIH

Impairs Transcription
(A and B) mRNA levels for Cyp26 (A) and RARb2 (B)
were analyzed in DMSO-treated or SP-treated
cells through the time after the addition of tRA
(1 mM). SP (10 mM) was added to the medium 2 hr
before addition of tRA and kept during the time
course. Expression was represented as the level of
the Cyp26 or RARb2 mRNA relative to GAPDH.
Values represent averages (±SD) from three inde-
pendent experiments.
(C) HeLa cells were incubated with DMSO or SP
(10 mM) for 2 hr. Increasing amounts (1 mg, 3 mg,
and 6 mg) of NEDMSO or NESP were tested in an
C in vitro transcription assay using AdML promoter
yielding a 309 nt transcript as indicated. When
indicated, purified TFIIH (100 ng) from HeLa
(Gerard et al., 1991) was added. The signals were
quantified using Genetool (Syngene) and indicated
below the autoradiogram. RA, relative activity.

(Hickson et al., 2004; Willmore et al., 2004), but SP showed purified TFIIH to NESP was enough to induce a complete recov-
at least some NER inhibitory activity even at submicromolar ery of DNA repair. Importantly for the use of SP as a chemosen-
concentration. sitizer, its effect on XPB was shown to be reversible because
At the molecular level, SP was proposed in an in silico screen- normal cellular concentrations of TFIIH and NER activity were
ing approach to inhibit the interaction between ERCC1 and XPA quickly restored after SP-removal. SP also impaired the transac-
(Barakat et al., 2009). However, in our hands, SP targets an even tivation of t-RA responsive genes upon ligand addition ex vivo
upstream NER factor by inducing the potent degradation of XPB, as well as the basal transcription in vitro. Strikingly, addition of
a subunit of the basal transcription/repair factor TFIIH. Such purified TFIIH to SP-treated NE was sufficient to restore tran-
degradation seems to depend both on ubiquitin-activating en- scription. Together, these results pinpoint the high specific
zymes and on the 26S proteasome pathway. In the human ubiq- effect of SP on XPB at least in NER and in basal transcription
uitylation pathway, ubiquitin is adenylated either by UBA1 or processes. Thus, SP could serve as an exquisite tool to further
UBA6, the two only ubiquitin-activating enzymes (Jin et al., investigate the regulation and function of TFIIH. Indeed, drug-
2007). Because the degradation of XPB depends both on ubiqui- induced protein degradation offers the possibility of a fine
tin-activating enzymes and on the 26S proteasome, we conclude temporal control over protein levels, compared to RNAi and
that XPB undergoes rapid ubiquitylation after SP-treatment that biochemical methods that only allow for the constitutive removal
leads to its quick degradation by the proteasome pathway. of the protein.
Because TFIIH is also a transcription factor, treatment with SP Interestingly, the molecular disruption of TFIIH by SP sensi-
induces an inhibition of basal or activated transcription both tized tumor cells to platinum-based chemotherapy. The pres-
in vivo and in vitro. We noted, however, that the basal transcrip- ence of SP was associated with increased cytotoxicity of
tion was less hindered compared to DNA repair in vitro in the cisplatin and, to a lesser degree, oxaliplatin toward ovarian and
presence of SP. This was in agreement with the fact that SP- colon carcinoma cells, respectively. This finding may have
treated cells showed very little toxicity and as thus did not important clinical implications, because the combination of SP
seem to be much affected in their capacity to transcribe genes. and platinum could reduce the risk of relapse by increasing
Overall, it appears that a reduction in the amount of TFIIH primary tumor cell mortality. Alternatively, this might allow a shift to
affects DNA repair before transcription. This observation was lower dose of platinum-based compounds that would reduce
already made in trichotyodystrophy patients who display muta- the toxic side effects. Such strategy would be particularly attrac-
tions in subunits of TFIIH leading to a strong reduction of the tive for cancers such as ovarian carcinoma where SP could be
amount of TFIIH with a defect in NER but a rather stable administrated together with cisplatin as an intraperitoneal infu-
transcription activity in vitro (Vermeulen et al., 2000). sion, where very high drug concentrations can be reached and
One can then question the specificity of SP on TFIIH. Although where the transcriptional impact of SP on normal cells could
this is a difficult question to tackle, several arguments can be be limited. Ovarian cancer is the ninth most common cancer
made on this matter. First, using local UV irradiation, we have among women and ranks fifth in cancer deaths among women,
mapped the NER defect induced by SP and identified a defi- accounting for more deaths than any other cancer of the female
ciency in the recruitment of TFIIH and downstream NER factors reproductive system (Coleman et al., 2013). Thus, SP-mediated
(i.e., XPA or XPF) to the local damaged sites, while the TFIIH- inhibition of NER emerges as a possible strategy for enhancing
upstream NER factor XPC was normally recruited. Second, the efficacy of platinum derivatives although additional work
using an in vitro dual incision assay, we show that addition of will be necessary to determine the consequences of the

404 Chemistry & Biology 21, 398–407, March 20, 2014 ª2014 Elsevier Ltd All rights reserved
Chemistry & Biology
DNA Repair Inhibitor Targets XPB Helicase

Figure 6. The SP-Induced Degradation of

A C XPB Is Reversible
(A) HeLa cells were pretreated with DMSO (lane 1)
or SP (lanes 2–5) (10 mM) for 2 hr. SP was removed
and fresh medium was added. Cells were allowed
to recover for the indicated period of recovery
times (RecTime). Extracts were resolved by SDS-
PAGE and analyzed by immunoblotting for either
XPB or TBP. The signals for XPB and TBP were
quantified using Genetool (Syngene) and plotted
on the graph relative to the signal in DMSO-treated
cells that is set to 1. RA, relative amount.
(B) HeLa cells were pretreated with DMSO (lanes
B 1–4) or SP (10 mM) (lanes 2, 3, 5, and 6) for 2 hr. SP
was then removed and fresh medium was added
with (lanes 4–6) or without (lanes 1–3) cyclohexi-
mide (CHX, 100 mM). Cells were allowed to recover
for 6 hr before lysing. Extracts were resolved by
SDS-PAGE and analyzed by immunoblotting for
either XPB or TBP.
(C) HeLa cells were treated with DMSO or SP
(10 mM) for 2 hr. SP was removed and fresh
medium was added. Cells were allowed to recover for different recovery times (RT) and subsequently UV-irradiated (60 J/m2), incubated for 3 hr (post-UV), and
fixed. Subsequently (6-4)PP lesions were labeled using anti-(6-4)PP antibodies and detected by HTM. The vertical axis represents the percent of (6-4)PP lesions
relative to the level of (6-4)PP lesions measured just after the irradiation that was set to 100 in all experiments; p value is indicated (**p < 0.01; ***p < 0.001).

administration of SP on XPB stability in vivo and on the transcrip- by XPC and is required for the subsequent recruitment of
tion activity of normal cells. XPA, XPG, and ERCC1-XPF. In the presence of spironolac-
Although administration of SP is not harmful in humans, prob- tone, XPC is still recruited to the damage but XPA, XPG,
ably due to its metabolization in the liver (Los et al., 1994), our and ERCC1-XPF are not. We also observed that the removal
study also addressed the consequences of long-term oral of spironolactone leads to a rapid recovery of XPB protein
administration of SP. This molecule has been used as a diuretic and NER activity showing the reversibility of spironolactone
and as an antihypertensive drug for more than 30 years. How- action. We also demonstrate that cotreatment of various
ever, an early study indicated a potential tumorigenic effect tumor cells with spironolactone and platinum derivatives
when administrated to rats in the diet (Lumb et al., 1978). Defect leads to reduced IC50 showing the potentialization of plat-
of NER in human causes genetic disorders that predispose inum treatment in the presence of our NER inhibitor.
patients to cancers (Lehmann, 2011). Because patients may
receive daily doses of spironolactone for several years, it would EXPERIMENTAL PROCEDURES
be important to reexamine the chronic effects of SP on patient
tissues following long-term SP treatment. Cell Lines
Parental A2780 and cisplatin-resistant A2780/CP70 ovarian carcinoma cells
were kindly provided by Robert Brown (Imperial College London, UK) while
SIGNIFICANCE the HCT-116 colon carcinoma cells were a gift from Bert Vogelstein (John
Hopkins Kimmel Cancer Center, Baltimore, MD). Oxaliplatin-resistant HCT-
116 (HCT-116/oxa) cells were developed in the laboratory of Annette K. Larsen
In this work, we have used a drug repositioning approach
as described (Petitprez et al., 2013). The nucleotide excision repair capacity
and drug screening to isolate inhibitors of the nucleotide of the four cell lines as determined by unscheduled DNA synthesis following
excision repair (NER) pathway that deals with bulky DNA le- UVC radiation has been described previously (Soares et al., 2011).
sions. NER is a fundamental repair pathway that particularly
repairs platinum-derived lesions used in anticancer chemo- Cell Viability Assays
therapy to kill tumor cells. The rationale behind our idea was The HeLa cell line was plated (25,000 cells per well in 6-well Petri dishes),
that any potent NER inhibitor could be used as an adjuvant in cultured overnight, and treated with 10 mM spironolactone or 0.1% DMSO
(control cells) for 4 hr before irradiation with various doses of UVC (Philips
platinum-based therapy to increase its efficiency by hinder-
TUV lamp, predominantly 254 nm) or exposition to various doses of cisplatin
ing cellular mechanisms that naturally reduce its efficacy. for 1 hr. Four days later, cells were stained with 0.2% Crystal Violet (Sigma).
The advantage of drug repositioning is that drugs tested After washing with tap water and drying, the stain was solubilized with 1%
are usually not cytotoxic, and their application for any given SDS and the optical density was measured by 96-well plate reader (595 nm).
new therapy may be quicker because these molecules are All values are averages of at least three independent experiments each done
already used in the human. Using this approach, we isolated in duplicate.
spironolactone as a potent NER inhibitor. The mechanism of
MTT Cytotoxicity Assay
spironolactone-induced NER inhibition was also investi-
Cancer cells were exposed to cisplatin or oxaliplatin in the presence of various
gated, and we found that one NER factor, the XPB helicase modulators (10 mM) for 120 hr and the cellular viability was determined by the
of TFIIH, was specifically degraded a few minutes after MTT tetrazolium dye [3-(4,5-dimethyl- thiazol-2-yl)-2,5-diphenyltetrazolium
addition of the drug. XPB is recruited to the damaged area bromide] assay (Sigma) as previously described (Poindessous et al., 2003).

Chemistry & Biology 21, 398–407, March 20, 2014 ª2014 Elsevier Ltd All rights reserved 405
 Chemistry & Biology
DNA Repair Inhibitor Targets XPB Helicase

A B C

Figure 7. SP Sensitizes Human Carcinoma Cells to Platinum Derivatives

(A and B) The MTT assay was used to measure cell viability of either A2780 (A) or HCT-116 (B) human carcinoma cells after treatment with cisplatin or oxaliplatin
(0.1 to 50 mM), respectively, in the presence of AN (10 mM), SP (10 mM), or DMSO. Values represent averages (±SD) from three independent experiments. The
p values (*p < 0.05; **p < 0.01; ***p < 0.001) and IC50 are indicated. One hundred percent corresponds to the amount of cells without any drugs.
(C) A2780 and HCT-116 cells were incubated with DMSO or SP (10 mM) for 4 hr before lysis. Extracts were resolved by SDSPAGE and subsequently analyzed by
immunoblotting for XPB or TBP.

7-AAD Cytotoxicity Assay ascites. Anti-XPF (Ab-5 clone 51) antibody was obtained from Thermo
HeLa cells were treated with DMSO, 10 mM SP, or 2.5 mM cis-diammineplati- Fisher Scientific, anti-phospho-Histone H2A.X (Ser139) (clone JBW301) from
num (II) dichloride (Sigma) for 24 hr, stained with 7-aminoactinomycin D Millipore, anti-XPC (A301-122A) from Bethyl; XPB degradation was also
(7-AAD Viability Dye, Beckman Coulter) according to manufacturer’s instruc- verified using anti-XPB antibody (S-19) from Santa-Cruz Biotechnology.
tions, and analyzed by flow cytometry. Anti-6-4PP were from Cosmobio (64M-2).

Dual Incision SUPPLEMENTAL INFORMATION

Dual incision was carried out as described in Riedl et al. (2003). Briefly, dual
incision assay was carried out in 25 ml of incision buffer (45 mM HEPES- Supplemental Information includes Supplemental Experimental Procedures,
KOH [pH7.8], 5 mM MgCl2, 1 mM DTT, 0.3 mM EDTA, 10% glycerol, 2.5 mg two figures, and one movie and can be found with this article online at
BSA, and 50 mM KCl) supplemented with 2 mM ATP. Following preincubation http://dx.doi.org/10.1016/j.chembiol.2013.12.014.
for 10 min at 30 C, 30 ng of monodamaged platinum DNA was added and
reaction was continued for 30 min at 30 C. The excised fragment was detected AUTHOR CONTRIBUTIONS
on 14% urea-acrylamide after annealing with 9 ng of the complementary oligo-
nucleotide and addition of four radiolabeled dCMPa-P32 (3,000 mCi/mmol) res- S.A. and M.A. performed the experiments. L.B. provided the original idea of
idues by Sequenase V2.1 (USB). the screen and performed it. J.M.E. provided reagents. S.A., A.L., and F.C.
designed the experiments and analyzed the data. F.C. wrote the article.
Transactivation Assay
RAR-dependent gene activation was performed as described (Keriel et al., ACKNOWLEDGMENTS
2002). Briefly, 1 3 106 cells were incubated with medium in a 10 cm cell culture
dish for 24 hr. Twelve hours before ligand treatment, cells were incubated with We are grateful to Alain Sarasin and Alexandre Escargueil for discussions,
phenol red-free medium containing 10% charcoal-treated fetal calf serum Virginie Poindessous for expert technical assistance, and Zita Nagy for critical
(FCS) and 40 mg/ml gentamycin. Cells were treated with 10 mM all-trans reti- reading. We are thankful to Marc Koch for his help with the microscopy
noic acid (tRA). experiments, Claudine Ebel for the help with flow cytometry experiments,
and Amelie Weiss for the help with the drug screening. This study was
Run-Off Transcription supported by the Foundation ARC for the Cancer Research grant to S.A.
Reaction mixtures of 12 ml containing 50 ng of linear AdMLP DNA template and (PDF20120605005). The laboratory of J.M.E. and F.C. is supported by the
the indicated amount of HeLa nuclear extract was preincubated for 5 min at European Research Council (ERC; ERC-2008-TRANSREACT) and the Ligue
30 C in transcription buffer (20 mM HEPES [pH 7.9], 7 mM MgCl2, 55 mM Contre le Cancer.
KCl) and transcription was initiated by the addition of 2 ml nucleotide solution
to final concentrations of 600 mM UTP, ATP, GTP, and 0.6 mM [a-32P] CTP. Received: October 3, 2013
Reactions were carried out for 30 min and then stopped by the addition of Revised: December 9, 2013
0.5 ml of 0.5 M EDTA (pH 8). The resulting RNA transcripts were analyzed on Accepted: December 23, 2013
an 8% denaturing polyacrylamide gel. Published: February 6, 2014

Statistical Analysis REFERENCES

The statistical analysis of experimental data was performed using a Student’s
paired t test, and results are presented as mean ± SD; p values are shown Barakat, K.H., Torin Huzil, J., Luchko, T., Jordheim, L., Dumontet, C., and
(*p < 0.05; **p < 0.01; ***p < 0.001). Tuszynski, J. (2009). Characterization of an inhibitory dynamic pharmacophore
for the ERCC1-XPA interaction using a combined molecular dynamics and
Antibodies virtual screening approach. J. Mol. Graph. Model. 28, 113–130.
Anti-XPB (1B3), anti-p62 (3C9), anti-p44 (1H5), anti-p52 (1D11), anti-XPA Barakat, K., Gajewski, M., and Tuszynski, J.A. (2012). DNA repair inhibitors:
(1E11), anti-RPA p32 (IE9), anti-Sug1 (1H9 and 1B8), anti-TBP (3G3), and the next major step to improve cancer therapy. Curr. Top. Med. Chem. 12,
anti-XPG (1B5) are in-house mouse monoclonal antibodies produced from 1376–1390.

406 Chemistry & Biology 21, 398–407, March 20, 2014 ª2014 Elsevier Ltd All rights reserved
Chemistry & Biology
DNA Repair Inhibitor Targets XPB Helicase

Burnett, J.C., and Rossi, J.J. (2012). RNA-based therapeutics: current prog- Lumb, G., Newberne, P., Rust, J.H., and Wagner, B. (1978). Effects in animals
ress and future prospects. Chem. Biol. 19, 60–71. of chronic administration of spironolactone—a review. J. Environ. Pathol.
Coin, F., Marinoni, J.C., Rodolfo, C., Fribourg, S., Pedrini, A.M., and Egly, J.M. Toxicol. 1, 641–660.
(1998). Mutations in the XPD helicase gene result in XP and TTD phenotypes, Mori, T., Nakane, M., Hattori, T., Matsunaga, T., Ihara, M., and Nikaido, O.
preventing interaction between XPD and the p44 subunit of TFIIH. Nat. Genet. (1991). Simultaneous establishment of monoclonal antibodies specific for
20, 184–188. either cyclobutane pyrimidine dimer or (6-4)photoproduct from the same
Coin, F., Proietti De Santis, L., Nardo, T., Zlobinskaya, O., Stefanini, M., and mouse immunized with ultraviolet-irradiated DNA. Photochem. Photobiol.
Egly, J.M. (2006). p8/TTD-A as a repair-specific TFIIH subunit. Mol. Cell 21, 54, 225–232.
215–226. O’Donovan, A., Davies, A.A., Moggs, J.G., West, S.C., and Wood, R.D. (1994).
Coleman, R.L., Monk, B.J., Sood, A.K., and Herzog, T.J. (2013). Latest XPG endonuclease makes the 30 incision in human DNA nucleotide excision
research and treatment of advanced-stage epithelial ovarian cancer. Nat. repair. Nature 371, 432–435.
Rev. Clin. Oncol. 10, 211–224. Ogi, T., Limsirichaikul, S., Overmeer, R.M., Volker, M., Takenaka, K., Cloney,
Compe, E., and Egly, J.M. (2012). TFIIH: when transcription met DNA repair. R., Nakazawa, Y., Niimi, A., Miki, Y., Jaspers, N.G., et al. (2010). Three DNA
Nat. Rev. Mol. Cell Biol. 13, 343–354. polymerases, recruited by different mechanisms, carry out NER repair synthe-
de Boer, J., and Hoeijmakers, J.H. (2000). Nucleotide excision repair and hu- sis in human cells. Mol. Cell 37, 714–727.
man syndromes. Carcinogenesis 21, 453–460. Oksenych, V., Bernardes de Jesus, B., Zhovmer, A., Egly, J.M., and Coin, F.
de Gasparo, M., Joss, U., Ramjoué, H.P., Whitebread, S.E., Haenni, H., (2009). Molecular insights into the recruitment of TFIIH to sites of DNA damage.
Schenkel, L., Kraehenbuehl, C., Biollaz, M., Grob, J., Schmidlin, J., et al. EMBO J. 28, 2971–2980.
(1987). Three new epoxy-spirolactone derivatives: characterization in vivo Petitprez, A., Poindessous, V., Ouaret, D., Regairaz, M., Bastian, G., Guérin,
and in vitro. J. Pharmacol. Exp. Ther. 240, 650–656. E., Escargueil, A.E., and Larsen, A.K. (2013). Acquired irinotecan resistance
de Laat, W.L., Jaspers, N.G., and Hoeijmakers, J.H. (1999). Molecular mech- is accompanied by stable modifications of cell cycle dynamics independent
anism of nucleotide excision repair. Genes Dev. 13, 768–785. of MSI status. Int. J. Oncol. 42, 1644–1653.
Fousteri, M., and Mullenders, L.H. (2008). Transcription-coupled nucleotide Poindessous, V., Koeppel, F., Raymond, E., Cvitkovic, E., Waters, S.J., and
excision repair in mammalian cells: molecular mechanisms and biological Larsen, A.K. (2003). Enhanced antitumor activity of irofulven in combination
effects. Cell Res. 18, 73–84. with 5-fluorouracil and cisplatin in human colon and ovarian carcinoma cells.
Furuta, T., Ueda, T., Aune, G., Sarasin, A., Kraemer, K.H., and Pommier, Y. Int. J. Oncol. 23, 1347–1355.
(2002). Transcription-coupled nucleotide excision repair as a determinant of Riedl, T., Hanaoka, F., and Egly, J.M. (2003). The comings and goings of nucle-
cisplatin sensitivity of human cells. Cancer Res. 62, 4899–4902. otide excision repair factors on damaged DNA. EMBO J. 22, 5293–5303.
Galanski, M. (2006). Recent developments in the field of anticancer platinum Sijbers, A.M., de Laat, W.L., Ariza, R.R., Biggerstaff, M., Wei, Y.F., Moggs,
complexes. Recent Patents Anticancer. Drug Discov. 1, 285–295. J.G., Carter, K.C., Shell, B.K., Evans, E., de Jong, M.C., et al. (1996).
Gerard, M., Fischer, L., Moncollin, V., Chipoulet, J.M., Chambon, P., and Egly, Xeroderma pigmentosum group F caused by a defect in a structure-specific
J.M. (1991). Purification and interaction properties of the human RNA polymer- DNA repair endonuclease. Cell 86, 811–822.
ase B(II) general transcription factor BTF2. J. Biol. Chem. 266, 20940–20945. Soares, D.G., Machado, M.S., Rocca, C.J., Poindessous, V., Ouaret, D.,
Gillet, L.C., and Schärer, O.D. (2006). Molecular mechanisms of mammalian Sarasin, A., Galmarini, C.M., Henriques, J.A., Escargueil, A.E., and Larsen,
global genome nucleotide excision repair. Chem. Rev. 106, 253–276. A.K. (2011). Trabectedin and its C subunit modified analogue PM01183 atten-
Hickson, I., Zhao, Y., Richardson, C.J., Green, S.J., Martin, N.M., Orr, A.I., uate nucleotide excision repair and show activity toward platinum-resistant
Reaper, P.M., Jackson, S.P., Curtin, N.J., and Smith, G.C. (2004). cells. Mol. Cancer Ther. 10, 1481–1489.
Identification and characterization of a novel and specific inhibitor of the Sugasawa, K., Ng, J.M., Masutani, C., Iwai, S., van der Spek, P.J., Eker, A.P.,
ataxia-telangiectasia mutated kinase ATM. Cancer Res. 64, 9152–9159. Hanaoka, F., Bootsma, D., and Hoeijmakers, J.H. (1998). Xeroderma pigmen-
Jin, J., Li, X., Gygi, S.P., and Harper, J.W. (2007). Dual E1 activation systems tosum group C protein complex is the initiator of global genome nucleotide
for ubiquitin differentially regulate E2 enzyme charging. Nature 447, 1135– excision repair. Mol. Cell 2, 223–232.
1138. Vermeulen, W., Bergmann, E., Auriol, J., Rademakers, S., Frit, P., Appeldoorn,
Kelland, L. (2007). The resurgence of platinum-based cancer chemotherapy. E., Hoeijmakers, J.H., and Egly, J.M. (2000). Sublimiting concentration of TFIIH
Nat. Rev. Cancer 7, 573–584. transcription/DNA repair factor causes TTD-A trichothiodystrophy disorder.
Keriel, A., Stary, A., Sarasin, A., Rochette-Egly, C., and Egly, J.M. (2002). XPD Nat. Genet. 26, 307–313.
mutations prevent TFIIH-dependent transactivation by nuclear receptors and Volker, M., Moné, M.J., Karmakar, P., van Hoffen, A., Schul, W., Vermeulen,
phosphorylation of RARalpha. Cell 109, 125–135. W., Hoeijmakers, J.H., van Driel, R., van Zeeland, A.A., and Mullenders, L.H.
Köberle, B., Tomicic, M.T., Usanova, S., and Kaina, B. (2010). Cisplatin resis- (2001). Sequential assembly of the nucleotide excision repair factors in vivo.
tance: preclinical findings and clinical implications. Biochim. Biophys. Acta Mol. Cell 8, 213–224.
1806, 172–182. Weeda, G., Rossignol, M., Fraser, R.A., Winkler, G.S., Vermeulen, W., van ’t
Le May, N., Mota-Fernandes, D., Vélez-Cruz, R., Iltis, I., Biard, D., and Egly, Veer, L.J., Ma, L., Hoeijmakers, J.H., and Egly, J.M. (1997). The XPB subunit
J.M. (2010). NER factors are recruited to active promoters and facilitate chro- of repair/transcription factor TFIIH directly interacts with SUG1, a subunit of
matin modification for transcription in the absence of exogenous genotoxic the 26S proteasome and putative transcription factor. Nucleic Acids Res.
attack. Mol. Cell 38, 54–66. 25, 2274–2283.
Le May, N., Iltis, I., Amé, J.C., Zhovmer, A., Biard, D., Egly, J.M., Schreiber, V., Willmore, E., de Caux, S., Sunter, N.J., Tilby, M.J., Jackson, G.H., Austin, C.A.,
and Coin, F. (2012). Poly (ADP-ribose) glycohydrolase regulates retinoic acid and Durkacz, B.W. (2004). A novel DNA-dependent protein kinase inhibitor,
receptor-mediated gene expression. Mol. Cell 48, 785–798. NU7026, potentiates the cytotoxicity of topoisomerase II poisons used in the
Lehmann, A.R. (2011). DNA polymerases and repair synthesis in NER in human treatment of leukemia. Blood 103, 4659–4665.
cells. DNA Repair (Amst.) 10, 730–733. Yang, Y., Kitagaki, J., Dai, R.M., Tsai, Y.C., Lorick, K.L., Ludwig, R.L., Pierre,
Los, L.E., Pitzenberger, S.M., Ramjit, H.G., Coddington, A.B., and Colby, H.D. S.A., Jensen, J.P., Davydov, I.V., Oberoi, P., et al. (2007). Inhibitors of ubiqui-
(1994). Hepatic metabolism of spironolactone. Production of 3-hydroxy-thio- tin-activating enzyme (E1), a new class of potential cancer therapeutics.
methyl metabolites. Drug Metab. Dispos. 22, 903–908. Cancer Res. 67, 9472–9481.

Chemistry & Biology 21, 398–407, March 20, 2014 ª2014 Elsevier Ltd All rights reserved 407