Download The role of base excision repair and mismatch repair proteins in the

The University of Toledo
The University of Toledo Digital Repository
Theses and Dissertations
2014
The role of base excision repair and mismatch repair
proteins in the processing of cisplatin interstrand
cross-links
Akshada S. Sawant
University of Toledo
Follow this and additional works at: http://utdr.utoledo.edu/theses-dissertations
Recommended Citation
Sawant, Akshada S., "The role of base excision repair and mismatch repair proteins in the processing of cisplatin interstrand cross-links"
(2014). Theses and Dissertations. 1594.
http://utdr.utoledo.edu/theses-dissertations/1594
This Dissertation is brought to you for free and open access by The University of Toledo Digital Repository. It has been accepted for inclusion in Theses
and Dissertations by an authorized administrator of The University of Toledo Digital Repository. For more information, please see the repository's
About page.
A Dissertation
Entitled
The Role of Base Excision Repair and Mismatch Repair Proteins in the Processing of
Cisplatin Interstrand Cross-Links.
by
Akshada Sawant
Submitted to the Graduate Faculty as partial fulfillment of the requirements for the
Doctor of Philosophy Degree in Biomedical Science
Dr. Stephan M. Patrick, Committee Chair
Dr. Kandace Williams, Committee Member
Dr. William Maltese, Committee Member
Dr. Manohar Ratnam, Committee Member
Dr. David Giovannucci, Committee Member
Dr. Patricia R. Komuniecki, Dean
College of Graduate Studies
The University of Toledo
August 2014
Copyright 2014, Akshada Sawant
This document is copyrighted material. Under copyright law, no parts of this document
may be reproduced without the expressed permission of the author.
An Abstract of
The Role of Base Excision Repair and Mismatch Repair Proteins in the Processing of
Cisplatin Interstrand Cross-Links
By
Akshada Sawant
Submitted to the Graduate Faculty as partial fulfillment of the requirements for the
Doctor of Philosophy Degree in Biomedical Science
The University of Toledo
August 2014
Cisplatin is a well-known anticancer agent that forms a part of many combination
chemotherapeutic treatments used against a variety of human cancers. Despite successful
treatment, the development of resistance is the major limitation of the cisplatin based
therapy. Base excision repair modulates cisplatin cytotoxicity. Moreover, mismatch
repair deficiency gives rise to cisplatin resistance and leads to poor prognosis of the
disease. Various models have been proposed to explain this low level of resistance caused
due to loss of MMR proteins. In our previous studies, we have shown that BER
processing of the cisplatin ICLs is mutagenic. Our studies showed that these mismatches
lead to the activation and the recruitment of mismatch repair proteins. The role of MMR
proteins in the processing of cisplatin ICLs has not been studied before. Here, we show
that MMR proteins are required to maintain cisplatin ICLs on the DNA leading to
increased cellular sensitivity. We distinguished between the requirement of different
downstream MMR proteins for maintaining cisplatin sensitivity. We also show that,
MutSα heterocomplex is required to maintain cisplatin sensitivity whereas Mutsβ
iii
complex has no effect. These results can be correlated with the increased repair of
cisplatin ICLs and ICL induced double-strand breaks (DSBs) in the resistant cells.
Moreover, we show that MLH1 proficient cells displayed a cisplatin sensitive phenotype
when compared with the MLH1 deficient cells and the ATPase activity of MLH1 is
essential to mediate this effect. Based on our results, we propose that different
recognition heterocomplexes are required to mediate cisplatin cytotoxicity and an entire
functional MMR pathway is essential to maintain a cisplatin sensitive phenotype. In
addition, targeting XPF-ERCC1 and signaling kinases, including ATM and ATR lead to
sensitization of BER and MMR deficient cells to cisplatin highlighting the potential use
of their respective small molecule inhibitors in cisplatin combination therapy in the
treatment of BER and MMR deficient or mutated tumors.
iv
I lovingly dedicate my dissertation to my parents, Sudhakar Sawant and Shubhada
Sawant, and my sister Harshada Sawant for always supporting my decisions and for their
unconditional love and faith in my abilities.
v
Acknowledgements
I deeply acknowledge my advisor Dr. Stephan Patrick for his invaluable guidance and
advice over the past 4 years, which allowed me to improve my critical thinking skills and
grow as a research scientist. His direction, technical assistance and moral support through
multitude of challenges have added greatly to my graduate experience. I thank Dr.
Kandace Williams for her mentoring during my time away from U.T. and the enriching
scientific discussions we shared. Drs. William Maltese, Manohar Ratnam, Zijian Xie, and
David Giovannucci all contributed brilliant suggestions and excellent advice, which
helped me immensely in my research. I thank Dr. Randall Ruch for agreeing to be a
graduate school representative on my committee.
I am thankful for all past and present members of the Patrick lab. Drs. Anbarasi
Kothandapani, Mohan Dangeti and Sanjeevani Arora provided much-welcomed guidance
and encouragement. I thank Elaine Chalfin, Vivian Maltese, Shengnan Du and Katelyn
Marchal for their help and humor. I am grateful to the faculty, secretaries and students of
the Biochemistry and cancer biology department for their guidance and support.
I thank my parents Sudhakar Sawant and Shubhada Sawant, and my sister
Harshada Sawant for their love and encouragement. Special thanks to Dr. Mugdha Patki,
Shweta Aras, Gaurav Mehta and Archit Trivedi for being my 4 a.m. friends. Lastly, I
thank my cousins and friends back home in India for their moral support, and all my
friends from Toledo for years of friendship and fun.
vi
Table of Contents
Abstract…………………………………………………………………………………...iii
Acknowledgements………………………………………………………...…………......vi
Table of Contents……………...……………………………………………………........vii
List of Figures…………………………………………………………………………......x
List of Abbreviations……………………...………………….……………………….....xii
List of Symbols……………………………………………………………………..…...xiv
1 Literature Review…………………………………………………...……………….......1
1.1 Biological activity of platinum agents…………………………………….…….…..2
1.2 DNA is a primary biological target of platinum agents………………………...…...4
1.3 Non-DNA targets of cisplatin………………………………………………...…......4
1.4 Types of adducts formed by cisplatin………………………………………...……..5
1.5 Mechanisms of cisplatin resistance and role of DNA repair pathways………..…....6
1.6 Nucleotide excision repair and repair of cisplatin intrastrand adducts………..….…7
1.7 Repair of ICLs: a complicated process………………………………………….…10
1.8 NER dependent processing and translesion repair synthesis of the ICLs……….…11
1.9 Mechanism of the recombination dependent repair of ICLs……………………....13
1.10 FA pathway and ICL repair……………………………………………….……...14
vii
1.11 Base excision repair (BER): short and long patch repair pathways…………........15
1.12 Base excision repair and cisplatin cytotoxicity……………………………….…..18
1.13 Mismatch repair (MMR): a post replicative DNA repair pathway…………….…21
1.14 Mismatch repair in mammalian cells………………………………………….….23
1.15 Role of mismatch repair in tumor susceptibility……………………………….…27
1.16 Role of mismatch repair pathway in response to DNA damage……………….…28
1.17 Role of ATR and ATM signaling in cisplatin cytotoxicity………………...……..33
1.18 Novel role of BER and MMR in cisplatin sensitivity …………………………....35
2 Overlapping role of BER and MMR proteins in mediating cisplatin cytotoxicity…….37
2.1 Introduction…………………………………………………………………….…..38
2.2 Materials and methods………………………………………………………….….41
2.3 Results……………………………………………..………………………….……50
2.4 Discussion…………………………………………………………………….…....58
2.5 Figures………………………………………………………………………….......62
3 Role of downstream mismatch repair proteins in the processing of cisplatin interstrand
crosslinks………………………………………………………………………………....72
3.1 Introduction…………………………………………………………………….......73
3.2 Materials and methods…………………………………………………….……….75
3.3 Results………………………………………………………………………….…..80
3.4 Discussion……………………………………………………………………….....88
viii
3.5 Figures…………………………………………………………………….………96
4 Overall summary………………………………………………………………….…..117
References……………………………………………………………………………....120
ix
List of Figures
Figure 1: Mammalian nucleotide excision repair………………………………….…….9
Figure 2: Models for ICL repair…………………………………………………….……11
Figure 3: Basic mechanism of base excision repair pathway…………………………....19
Figure 4: Mismatch repair of a single base mismatch…………………………………...26
Figure 5: siRNA mediated downregulation of MSH2…………………………………...62
Figure 6: Cisplatin cytotoxicity in MEFs with MSH2 KD………………………………63
Figure 7A-C: Repair of cisplatin adducts in MEFs with MSH2 KD…………………….64
Figure 8: Colony survival assay in D256A cells with MSH2 KD…………………….....66
Figure 9: Biotin pull-down assay for cisplatin ICL DNA processed with BER proteins..67
Figure 10: BER and MMR play an epistatic role in mediating cisplatin cytotoxicity…..68
Figure 11: Targeting signaling kinases ATR and ATM in BER proficient and deficient
cells……………………………………………………………………………………....69
Figure 12: Downregulating XPF-ERCC1 in MMR proficient and deficient cells………71
Figure 13: shRNA mediated downregulation of MSH3 and MSH6……………………96
Figure 14: Cisplatin cytotoxicity in MDA-MB-231 cells (A) MSH6 KD (B) MSH3
KD………………………………………………………………………………………97
x
Figure 15: Cisplatin ICL repair in MDA-MB-231 cells (A) MSH6 KD (B) MSH3
KD……………………………………………………………………………………….98
Figure 16: DSB repair in MDA-MB-231cells (A) MSH6 KD (B) MSH3 KD………….99
Figure 17: knockdown efficiency in (A) wildtype and (B) Polβ -/- MEFs……….…….100
Figure 18: Colony survival assay in WT and Pol β -/- MEFs with MLH1 KD…….…..101
Figure 19: Cisplatin adduct repair in WT and Polβ -/- MEFs with MLH1 KD………...102
Figure 20: Graphical representation of the MLH1 ATPase mutants and expression
levels…………………………………………………………………………………....104
Figure 20 D: MTS assay in MLH1 ATPase mutant HCT116 cells……………..…...…105
Figure 21: Cisplatin adduct repair in HCT116 cells….………………...………………106
Figure 22: Downregulating XPF-ERCC1 in MLH1 proficient and deficient cells…….107
Supplementary Figure S1. The colony survival assay in MEFs with MSH3 KD……....108
Supplementary Figure S2. Oxaliplatin cytotoxicity (A) MSH3 KD (B) MSH6 KD
(C) HCT116 cells………..……………………………………………………...………109
Supplementary Figure S3A-C: MTS assay with Myricetin treatment in HCT116
cells…………………………………………………………………………...………...111
Supplementary Figure S4: MTS assay with MNNG treatment……..……………...…..113
Supplementary figure S5: Western blot analysis for (A) MGMT and (B) PMS2…..….114
Supplementary Figure S6: Expression of XPF and ERCC1in HCT116 cells……….…115
xi
List of Abbreviations
APE1……………………………………………... Apurinic/apyrimidinic endonuclease 1
CSA………………………………………………………………... cockayne syndrome A
CSB………………………………………………………………... cockayne syndrome B
DSB………………………………………………………………….. double-strand break
EXO I……………………………………………………………………… Exonuclease I
FA……………………………………………………………………….. Fanconi Anemia
FEN1…………………………………………………………………. flap endonuclease I
HNPCC……………………………………… hereditary non-polyposis colorectal cancer
HR…………………………………………………………… homologous recombination
ICLs………………………………………………………………… Interstrand crosslinks
IDLs………………………………………………………………. insertion deletion loops
KD…………………………………………………………………………….. knockdown
LP-BER…………………………………………………………………... long patch BER
MBD4………………………………………………. methyl CPG domain binding protein
MMC………………………………………………………………………… mitomycin C
MMR……………………………………………………………………...Mismatch repair
MLH1…………………………………………………………………… MutL homolog 1
MSH…………………………………………………………………… MutS homologues
MSI……………………………………………………………… microsatellite instability
Neil 1……………………………………………………….. Nei endonuclease VIII-like 1
NER…………………………………………………………… Nucleotide excision repair
PARP1……………………………………………………… Poly ADP ribose polymerase
PCNA…………………………………………………… proliferating cell nuclear antigen
Polβ………………………………………………………………………. Polymerase beta
RFC…………………………………………………………………… replication factor C
ROS……………………………………………………………... Reactive oxygen species
xii
RT………………………………………………………………………. room temperature
SMUG1…………….……single-strand selective monofunctional uracil DNA glycosylase
SSBs………………………………………………………………….. single-strand breaks
ssDNA………………………………………………………………. single-stranded DNA
TC-NER………………………………………………………. transcription coupled NER
TDG………………………………………………... thymine/uracil mismatch glycosylase
TFIIH……………………………………………………………. Transcription factor II H
TLS………………………………………………………………….Translesion synthesis
UNG……………………………………………………………... uracil DNA glycosylase
WT……………………………………………………………………………….wild type
xiii
List of Symbols
α .................................Alpha
β .................................Beta
γ ..................................Gamma
δ..................................Delta
ε .................................Epsilon
µ .................................Micro
η…………………… Eta
ζ……………………..Zeta
xiv
Chapter 1
Literature review
Cisplatin, as a chemotherapeutic agent, was discovered serendipitously. Since its
discovery, it has been widely used in the treatment of a number of human malignancies.
Platinum-based drugs like cisplatin are used against various solid tumors which include
genitourinary, colorectal and non-small cell lung cancers [1]. Cisplatin has been found to
be particularly effective in the treatment of testicular cancer with a cure rate of nearly
100% when the tumors are detected in an early stage and when used in combination with
bleomycin and etoposide [2]. It has become central to the treatment of testicular germ cell
tumors. However, cisplatin based chemotherapy has been associated with increased
nephrotoxicity, ototoxicity as well as nausea and vomiting [3]. In addition, cancer cells
exhibit either inherent or acquired resistance to cisplatin treatment leading to poor
prognosis [4, 5]. Attempts have been made to synthesize new platinum agents that would
compensate for the drawbacks of cisplatin. Currently, only four platinum agents have
been registered as marketed drugs which include cisplatin, carboplatin, oxaliplatin and
nedaplatin. Out of the new compounds, only oxaliplatin has been approved by the FDA
for the treatment of colorectal cancers. Recently formulated monofunctional platinum
1
agent, phenanthriplatin has showed a greater efficacy in an NCI 60 cell tumor panel
screen [6]. In addition, developments of nano particle based technologies that can
efficiently deliver cisplatin in small quantities and that are only activated near the target
are underway [7, 8]. In spite of the new advances, it is important to further study
platinum agents and understand how resistance to platinum is developed in different
types of cancers. Therefore, it becomes imperative to understand the molecular
mechanisms that are central to the acquired and intrinsic resistance to platinum agents
and use this information for the development of new targets that can potentiate cisplatin
chemotherapy. The main objective of this dissertation is to understand such
mechanisms and elucidate the pathways that can be targeted to improve platinum
therapy.
1.1 Biological activity of platinum agents
The transport of the platinum compounds across the cell membranes is the first
step for successful chemotherapy. Transport of cisplatin has been known to occur through
both active and passive mechanisms through the cell membrane. Once cisplatin enters the
aqueous solution inside the cell, it undergoes aqaution recations. In this reaction, the
chloride groups are replaced with water molecules. The high chloride concentration
(~100 mM) in the extracellular fluid prevents the formation of mono or diaquo cisplatin
species. However, inside the cell, the chloride concentration ranges from 2 to 30 mM. In
this scenario, the hydrolysis of cisplatin takes place in an efficient manner leading to the
formation of aqauted cisplatin in which one or both the chloride groups are replaced with
water molecules. This leads to the formation of [Pt(H2O)2(N H3)2]2+ cation. This aquated
cisplatin is very reactive towards the nucleophilic centers in DNA as H20 is a much better
2
leaving group compared to chloride ions. It is observed that ~42% of the total platinum is
maintained in the form of these aquated species [9]. These cationic molecules rarely
diffuse back out of the cell through the lipid bilayer membrane. Such trapped forms of
activated cisplatin, therefore, increase the chances of binding to its biological targets such
as DNA, which leads to increased cytotoxic potential of the drug.
Early studies have shown that cisplatin enters the cell through passive diffusion.
However, recent research has established a role of active transporters in the cell
membrane that are responsible for the uptake and transport of cisplatin and thus, affecting
the cytotoxicity of the drug. Import and export transport systems that are involved in the
maintenance of copper homeostasis have also been shown to be engaged in the transport
of platinum drugs [10]. The copper transporting p-type ATPase, ATP7B, is responsible
for the efflux of heavy metals outside the cell. It was observed that, overexpression of
ATP7B gave rise to resistance to cisplatin in ovarian cancer cell lines, possibly due to
enhanced efflux of the drug outside the cell [11]. On the other hand, studies in yeast have
shown that lack of ctr1, which is a protein involved in copper influx, also gave rise to
cisplatin resistance as it lead to decreased accumulation of the drug inside the cell which
reduced the amount of drug available to target the DNA [12]. However, studies also
showed that increased influx of cisplatin due to overexpression of ctr1 did not give rise to
any change in the level of cisplatin inflicted DNA damage [13]. These observations
suggest that the effects of the copper transporters on the cytotoxic potential of cisplatin
are still uncertain. In addition, the delivery of platinum through the cell membrane has
also been facilitated by organic cationic transporters (OCTs) [14]. These transporters
have been implicated in cisplatin nephrotoxicity. Studies have shown that renal secretion
3
of cisplatin has been abolished in the absence of transporters OCT1 and OCT2, thus
providing protection against the cisplatin induced renal damage [15]. Many of the studies
on these transporters have suggested OCT2 as the major transporter that protects against
renal toxicity [16]. Other transport systems that might play a role in the transport of
cisplatin are MRP2/cMOAT (multidrug associated protein 2)(cananicular multispecific
organic anion transporter), MRP1 and p-glycoproteins [17, 18]. These observations,
therefore, illustrate that platinum enters the cell through passive diffusion as well as
through active transport mechanisms.
1.2 DNA is a primary biological target of platinum agents
DNA is the primary target of platinum agents. As far as other biomolecules are
concerned, the levels of platinum atoms that bind to proteins and RNA are much less, and
they do not exhibit any inhibitory effects. In addition, the increased turnover of proteins
and RNA molecules might be responsible for the minimal effects of such binding on the
cellular processes [19]. Moreover, the levels of platinum DNA adducts correlate
significantly well with the sensitivity of the cells to the drug [20].
1.3 Non-DNA targets of cisplatin
Glutathione is an important antioxidant that protects the cell from the damage
caused by reactive oxygen species such as free radicals and peroxides. It is a tripeptide
that maintains the redox environment of the cells and provides protection from the
xenobiotic substances [21]. Glutathione binds to cisplatin in a reaction catalyzed by
glutathione-S-transferase (GST) resulting in the inactivation of the drug. Increased
expression of this enzyme can lead to increased inactivation of the drug giving rise to
4
cisplatin resistance. A significant correlation was found between an increased
concentration of glutathione or GST with increased resistance to cisplatin in cervical and
lung cancer cell lines [22, 23]. However, other studies pointed out that no correlation
could be demonstrated between the expression of GST and the response rate or survival
after platinum treatment in patients with cervical and non-small cell lung cancer [24, 25].
In addition, metallothionines are thiol rich proteins that can also bind to cisplatin
leading to drug inactivation as these proteins are involved in metal homeostasis and
detoxification processes in the cell [21]. A relationship between increased expression of
metallothionines and elevated resistance has been investigated with conflicting results.
Therefore, the metallothionines are considered to be a minor factor in the development of
platinum drug resistance.
1.4 Types of adducts formed by cisplatin
The reaction of cisplatin with DNA leads to the formation of different types of adducts.
Cisplatin reacts with the N7 atoms of guanine and adenine situated in the major groove of
the double helix. These sites are easily accessible and are reactive nucleophilic centers for
platinum DNA binding. The adduct formation takes place through the formation of
monofunctional adducts which later are converted to bifunctional adducts which include
intrastrand adducts and interstrand crosslinks. The intrastrand adducts are formed on the
same strand of DNA. These adducts include 1,2-d(GpG) intrastrand adducts (60-65%)
and 1,2-d(ApG) intrastrand adducts (20-25%). 1,3-d(GpXpG) intrastrand adducts form
about 5-10% of the total adducts. The interstrand crosslinks (ICLs), which form about 510 % of the total adducts, are formed between two adjoining strands of DNA at GpC
5
sites. The binding of cisplatin occurs with the N7 atoms of guanine residues in the
adjoining DNA strands. The resulting structure produced by an ICL is a bulky structure.
It distorts the DNA and the cytosine residues are flipped extrahelical. The crystal
structure also indicates that the DNA is unwound 70º and bent 47º causing significant
structural distortion [26]. This structure formed by cisplatin ICLs is distinct from the
ICLs formed by other DNA cross-linking agents including oxaliplatin and mitomycin C.
1.5 Mechanisms of cisplatin resistance and the role of DNA repair pathways in the
removal of cisplatin adducts
Impaired blood flow and decreased blood pressure can hinder effective delivery
of the drug to the distant tumor cells [27]. As noted earlier, mechanisms of decreased
drug influx and increased drug efflux can also result in decreased sensitivity of the drug.
In addition, the drug inactivating mechanisms also contribute to the development of
cisplatin resistance. Furthermore, decreased expression of pro-apoptotic proteins and
increased expression of apoptotic inhibitors can also result in tolerance to the drug. In
addition, platinum adducts can also be bypassed and can go unrecognized as the cells can
utilize translesion polymerases to tolerate the DNA damage. However, in vivo studies
using lung cancer mouse models have shown that enhanced DNA repair is a predominant
mechanism for the development of cisplatin resistance [28]. A number of studies have
shown the importance of enhanced DNA repair as an important mechanism for the
development of resistance [29, 30]. Studies in ovarian cancer cells have also reported
increased repair of platinum ICLs and as pointed out by the authors, this is likely a major
mechanism for clinical resistance to cisplatin [31].
6
1.6 Nucleotide excision repair and repair of cisplatin intrastrand adducts
Nucleotide excision repair (NER) is one of the repair pathways that removes
bulky damaged bases from the DNA (Figure 1)[32]. NER is a complex biochemical
process that requires the participation of several proteins [33]. NER detects the helical
distortion in the DNA produced by the bulky adducts caused by exogenous agents like
UV light and many DNA damaging agents used in chemotherapy [34]. In the global
genomic NER pathway, this damage is recognized by the XPC protein, which is in a
complex with another protein called HHR23B. This leads to the binding of several
downstream NER proteins including XPA, RPA, TFIIH and XPG. XPA is a zinc finger
protein that facilitates specific recognition of the base damage along with RPA, which is
a single-stranded DNA (ssDNA) binding protein. Bulky base damages promote some
degree of unwinding of the DNA structure, thereby promoting binding of RPA to the
single-stranded regions. The recognition of the damage is followed by the stepwise
assembly of the rest of the NER proteins. Among these are the Transcription factor II H
(TFIIH) which is a subcomplex of RNA polymerase II. TFIIH, therefore, is a part of NER
as well as the basal transcription machinery of the cell. It consists of six subunits and
contains two DNA helicases XPB and XPD that unwind the DNA duplex near the
damaged bases. This unwinding causes the formation of a bubble structure in the DNA
localized around the damaged site. These junctions facilitate the incision of the DNA at
both these ends by endonucleases XPF-ERCC1 and XPG. XPG endonucleolytic activity
cuts the DNA strand at the 3’ side of the base damage. On the other hand, XPF-ERCC1 is
a heterodimeric protein that cleaves the 5’ end of the damaged DNA strand. This leads to
the removal of a patch of about 27-32 nucleotides containing the damage. The excision of
7
this oligonucleotide fragment from the genome takes place as the repair synthesis is
initiated by DNA polymerases. The new DNA synthesis requires DNA polymerases δ
and ε along with the accessory factors PCNA, RPA and RFC. The ends of the newly
synthesized strands are then ligated by DNA ligase. In transcription coupled NER (TCNER), RNA polymerase acts as a sensor and initiates the repair process. In this type of
pathway, the adducts are repaired preferentially on the transcribed strand. The damage
recognition is independent of XPC and it involves cockayne syndrome A and B (CSA
and CSB) proteins. In conclusion, this entire process converts the damaged DNA to its
native chemistry and conformation. Cisplatin intrastrand adducts are a good substrate for
this repair process due to structural distortions produced by the adducts. Increased NER
of intra adducts has been shown to give rise to cisplatin resistance. On the other hand,
decreased expression of NER proteins, for example XPA and XPF-ERCC1, results in
increased cellular sensitivity to cisplatin [35-37]. As a consequence, cisplatin sensitivity
in testicular cancer cells has been attributed to the decreased expression of these proteins
and thus, reduced repair of cisplatin adducts from the DNA [30].
8
Figure 1
Figure 1: Mammalian nucleotide excision repair. a. Recognition, b. Unwinding, c.
Incision, d. Synthesis and ligation
Modified from: Proposed model for mammalian nucleotide excision repair (NER)
Yasuhiro Matsumura and Honnavara N. Ananthaswamy DOI:
10.1017/S1462399402005343
9
1.7 Repair of ICLs: a complicated process that involves the contribution from
multiple DNA repair pathways
A number of ICL inducing agents are used in the clinic for the treatment of
various types of malignancies in combination with other agents. These include cisplatin,
oxaliplatin, mitomycin C, psoralens and alkylating agents like nitrogen mustards. The
ICL formation can also occur naturally as a result of lipid peroxidation, byproducts of
acrolein, crotonaldehyde R and beta unsaturated aldehydes [38, 39]. The high cellular
cytotoxicity caused by these agents is attributed to the covalent linkages formed between
two complementary strands of the double helix. The ICLs produce a molecular constraint
in the helix and when unrepaired, they form a definite block to the cellular processes like
DNA replication and DNA transcription which renders them highly cytotoxic and
genotoxic. Due to this property, the repair of ICLs also becomes a complicated process
compared to the repair of monoaddcuts or intrastrand adducts which are formed only in
one strand of DNA. In most organisms, ICL repair takes place via two DNA repair
pathways, which can be either error free or error prone (Figure 2) [40, 40, 41]. The error
prone pathway is recombination independent where the crosslink is excised on one strand
by NER proteins and the resulting gap is filled by translesion polymerases. The crosslink
is then removed by a second round of NER [42, 43]. This pathway appears to be less
responsible for the repair of ICLs in cycling mammalian cells, however, this pathway
may be the only resort for the removal of ICLs in the G1 phase or in non-replicating cells.
A second repair pathway is initiated with a similar excision process but is dependent on
the homologous recombination (HR) pathway for the completion. As the HR pathway
10
requires a copy of homologous donor template, this pathway takes place in the S phase of
the cell cycle and can be initiated when the incoming replication fork detects an ICL.
Figure 2
a
b
Figure 2: Models for ICL repair. a. Recombination dependent ICL repair. b.
Recombination independent ICL repair (Modified from: Mutagenic repair of DNA
interstrand cross-links)[40]
1.8 NER dependent processing and translesion repair synthesis of the ICLs
Repair of ICLs appears to take place in all phases of the cell cycle. Increasing
evidence has shown that, ICL repair can occur independent of replication. In the absence
11
of a homologous donor sequence, lesion bypass polymerases seem to play a major role in
resolving the crosslink structure. Most ICLs are recognized by GG-NER [44], however,
cisplatin ICL removal has been recently shown to be dependent exclusively on TC-NER
and translesion polymerases. Reporter assays using plasmids containing site-specific
cisplatin ICLs have shown that recombination independent ICL repair is error free and is
dependent on the CSB, Pol ζ and Rev1 proteins [45]. In support of these studies,
downregulation of Rev3, an essential component of Pol ζ increases cellular sensitivity to
cisplatin and to other platinum agents in lung tumors and in HeLa cells [43, 46, 47]. In
addition, overexpression of Rev1 in ovarian cancer cell line 2008 has been shown to lead
to cisplatin resistance. This overexpression was also associated with increased
mutagenesis [48]. Plasmid reactivation assays also indicated increased repair of platinum
damaged DNA. However, even though this study did not differentiate between the repair
of intra and interstrand crosslinks, the overall involvement of Rev1 in the repair of
cisplatin adducts and its effect on cytotoxicity cannot be refuted [43, 47]. Studies by
Scharer and colleagues have shown that in addition to Rev1 and Pol ζ, a number of other
TLS polymerases can bypass ICLs, and this bypass is dependent on the degree of
distortion produced by the ICL inducing agent [49]. NER has been shown to be essential
for the recombination independent repair of ICLs produced by mitomycin C (MMC) [50].
MMC crosslinked luciferase plasmids showed inability of XPA, XPF, XPG and ERCC1
mutants to repair the ICL, indicating that the recognition and excinuclease activites of
NER are essential for the repair of ICLs in the G1 phase. Additionally, the helicase
acitivity of XPD was also shown to be important for the repair. A majority of the insights
into the mechanism of ICL repair have come from the studies involving wild type and
12
repair deficient cell lines and their response to crosslinking agents. Most NER deficient
cell lines with an exception of XPA and XPF-ERCC1 are less sensitive to cross-linking
agents [51, 52]. These studies indicated that an intact NER pathway and TLS
polymerases are essential for the replication/recombination independent repair of ICLs
produced by numerous crosslinking agents.
1.9 Mechanism of the recombination dependent repair of ICLs
ICL repair can be initiated in S phase as a consequence of the complete blockage
of the replication fork due to the ICL. This triggers a complex response which results in
the removal of the crosslink and reconstruction of the replication fork. Studies in
Xenopus egg extracts using site-specific ICL DNA substrates have provided strong
evidence for replication dependent ICL repair [53-55]. The ICL prevents unwinding of
the two DNA strands, thus, resulting in stalled replication forks. As the replication fork
approaches the ICL, the leading strand stalls approximately 20-40 nucleotides from the
lesion. As the replication fork encounters the ICL, it leads to the signaling and activation
of the Fanconi Anemia signaling (FA) pathway. Further processing by the FA proteins is
essential for the repair of ICLs. Recent studies have identified several HR proteins as a
part of the FA pathway. These include FANCD1/BRCA2 and FANCO/Rad51C among
others. According to one study, stalling of the replication fork leads to fork regression
and the formation of a DNA DSB. These DSBs are formed independent of XPF-ERCC1,
indicating that the initial DSBs are formed as a result of replication fork collapse [56].
Consequently, this formation of a DSB provides a substrate for XPF-ERCC1 which
cleaves at the 3’ end of the ICL. This leads to the unhooking of the ICL from one of the
strands followed by gap filling by one of the translesion polymerases. Similar to the G1
13
phase ICL repair, Rev1 and Pol ζ likely play key roles in this process. The leading strand
is thus extended, and it can be ligated to the first downstream Okazaki fragment giving
rise to an intact DNA that can serve as a donor template for the recombinational repair
and replication restart later [55]. The remaining ICL is removed by NER and in the
absence of damage the resulting gap can be filled by the replicative polymerases. The
repair of the DSB is mediated by HR pathway which involves 3’ end resection and strand
invasion followed by replication restart. RAD51 is shown to play a key role in this
process [54]. In addition to NER proteins, deficiencies in the RAD51 family of proteins,
BRCA1 and BRCA2 which take part in HR also showed hypersensitivity to the ICL
inducing agents indicating that these proteins are involved in the repair of the crosslinks
[51].
1.10 FA pathway and ICL repair
Fanconi Anemia (FA) is a rare, hereditary syndrome characterized by congenital
abnormalities and increased risk for hematological and squamous cell cancers. There are
14 complementation groups that have been associated with the disease. The products of
these FA genes act in a common cellular pathway that is essential for the replication
dependent repair of ICLs. Out of these FA genes, 7 gene products form an FA core
complex which includes FANCA, FANCB, FANC, FANCE, FANCG and FANCL. Cells
deficient in the FA pathway are hypersensitive to crosslinking agents, indicating that
these proteins play a crucial role in the repair of ICLs. The FA pathway is considered to
be involved in the coordination of DNA repair pathways such as HR, NER and TLS.
ICLs that are encountered in the S phase of the cell cycle are recognized by FANCM
which further activates the FA core complex and has ubiquitin ligase activity known as
14
FANCL. The core complex is responsible for the monoubiquitylation of FANCD2 and
FANCI. The entire FA core complex is essential for this process. The monoubiquitylation
of FANCD2 and FANCI stabilizes them on the chromatin. FAN1 is a 5’ endonuclease
associated with FANCD2, and it has been shown that lack of FAN1 leads to cisplatin
sensitivity. This study highlighted the role of FAN1 as one of the nucleases that are
required for the initial unhooking of the ICL [57]. Increased levels of XPF-ERCC1,
MUS81-EME1 and FA proteins have been observed in the resistant tumors associated
with chronic cisplatin exposure. These observations point towards the importance of
these nucleases in the increased repair of ICLs and consequently in increased resistance
to the platinum agents [58-60].
1.11 Base excision repair (BER): short and long patch repair pathways
BER is required for the repair of damaged bases that are generated due to
endogenous damage to the DNA. BER mainly deals with the repair of base lesions that
are generated in abundance in cells every day. Endogenous DNA damage is caused
predominantly by some of the products generated during normal metabolic processes.
These damaged lesions are caused by various events like deamination, alkylation and
oxidation of individual bases. Water and reactive oxygen species form a major group of
endogenous reactants that damage DNA in a variety of ways. In addition, lipid
peroxidation products also attack DNA and form a substrate for BER. Water generates
hydrolysis products which are responsible for the loss of purines and pyrimidines leading
to the generation of abasic sites [61]. Deamination of cytosine and 5-methylcytosine
gives rise to uracil and thymine, respectively. Similarly, deamination of adenine gives
rise to hypoxanthine and guanine deaminates to xanthine. Reactive oxygen species (ROS)
15
are formed as a by-product of oxidative metabolism. ROS cause oxidation of the bases
and lead to the formation of some of the highly mutagenic products including 8oxoguanine [61, 62]. All these products, if left unrepaired, can have serious cytotoxic and
mutagenic effects [63]. BER is essential for the removal of such endogenous damage as
well as base damage caused by exogenous environmental agents. The importance of BER
is highlighted by the fact that deficiencies in the major enzymes involved in this pathway
lead to embryonic lethality. BER proteins are involved in both direct and indirect proteinprotein interactions with other BER and non-BER proteins. BER forms an intricate
system that is highly regulated and the enzyme activities are linked to the cell cycle,
DNA replication and other cellular processes. However, the basic mechanism of BER
remains a highly conserved process (Figure 3). BER is initiated by DNA glycosylases
which recognize the damaged base. There are multiple DNA glycosylases that detect
different types of base damages. The damage recognition is followed by cleaving of the
glycosidic bond connecting the damaged base to the DNA phosphodiester backbone. Two
types of DNA glycosylases are found in nature. The monofunctional glycosylases include
uracil DNA glycosylase (UNG), thymine/uracil mismatch glycosylase (TDG), singlestrand selective monofunctional uracil DNA glycosylase (SMUG1) and methyl CPG
domain binding protein (MBD4). These all have uracil as a common substrate. Methyl
purine DNA glycosylase also known as AAG deals with the removal of a wide variety of
damaged purine bases. The other class of glycosylases contains lyase activity that cleaves
the DNA phosphate backbone 3’ of the abasic site created after the glycosylase reaction.
These are called bifunctional glycosylases and include hOGG1, which removes oxidized
purines like 8-oxoguanine, and hNTH1 which removes oxidized pyrimidines. The DNA
16
glycosylase hMYH1 is responsible for the recognition and repair of the 8-oxoG/ adenine
mismatches. The damage recognition and base removal by DNA glycosylases is followed
by the recruitment of Apurinic/apyrimidinic endonuclease (APE1), which cleaves the
DNA phosphodiester backbone 5’ of the abasic site and leaves a free 3’ OH group.
Further BER processing can take place via either short patch or long patch BER
pathways. In the short patch repair pathway, Polymerase beta (Polβ) carries out the DNA
synthesis. Polβ has polymerase and lyase activity. It uses its lyase activity to cleave the
dRP residue 3’ of the AP site. After DNA synthesis, DNA ligase III/XRCC1 seals the
gap, and the short patch BER mechanism is complete [64]. Short patch BER typically
involves single nucleotide synthesis by Polβ. However, in some cases, strand
displacement synthesis requires incorporation of multiple nucleotides. In these cases, the
long patch BER (LP-BER) pathway is employed, which involves DNA synthesis by
replicative polymerases like polymerases δ and/or ε with the help of proliferating cell
nuclear antigen (PCNA) and replication factor C (RFC). The strand synthesis on the
damaged strand involves displacement and creation of a 5’ flap junction. This flap
junction is removed by flap endonuclease I (FEN1) and the gap is again sealed by DNA
ligase I to complete the long patch repair [65]. The choice between short patch or long
patch repair pathway is dependent upon a number of factors like whether the AP site is
refractory to the lyase activity, the type of DNA glycosylase that removed the damaged
base and also the ATP concentration during the reaction. If the 5’ end is not processed,
then a polymerase switch can occur, bringing pol δ/ ε to continue the strand synthesis.
The long patch pathway can also be initiated by the presence of replication and repair
protein factors that interact with the proteins of LP-BER (e.g RFC, PCNA and RPA). The
17
employment of either of these sub-pathways is also influenced by the ratio of different
BER proteins available for the reaction. Poly ADP ribose polymerase (PARP1) acts like a
scaffolding protein and facilitates the recruitment of BER proteins like Polβ and XRCC1.
Therefore, excess PARP1 has been shown to block LP-BER [66]. On the other hand,
overexpression of APE1 promotes long patch repair [67]. To summarize, the basic
mechanism of BER and its sub-pathways has been depicted in Figure 3. BER involves
processes that are highly coordinated via protein-protein interactions where correct
proteins for each individual steps are recruited in a timely fashion [68].
1.12 Base excision repair and cisplatin cytotoxicity: Role of BER in the processing of
bulky lesions?
BER proteins have been implicated in the cytotoxic response to the ICL inducing agents.
Recent evidence suggests an emerging link between BER proteins and ICL processing.
AAG has been shown to recognize psoralen crosslinks and may be required for the
enhanced ICL DNA repair process [69]. Nei endonuclease VIII-like 1 (Neil 1) is a DNA
glycosylase/endonuclease VIII specific for oxidized and ring fragmented bases. Lack of
Neil 1 gives rise to hypersensitivity to the ICL agent psoralen in HeLa cells [70]. In
addition, Neil 1 can excise unhooked ICL fragments, suggesting that this glycosylase can
play an important role in the removal of psoralen monoadducts as well as in the later
stages of ICL repair [70]. Uracil DNA glycosyalse (UNG) deficiency gives rise to
cisplatin resistance in breast cancer cells as well as in mouse embryonic fibroblasts
(MEFs) [71]. This resistant phenotype correlated well with the increased repair of ICLs
suggesting a role for UNG in modulating cisplatin cytotoxicity [71].
18
Figure 3
1. DNA glycosylase removes the damaged base
2. AP endonuclease cleaves the abasic site
3. Polβ mediated DNA synthesis
4. Ligation by DNA ligase I
Figure 3: Basic mechanism of short-patch base excision repair pathway. Modified
from The Pathways of Double-Strand Break Repair. Emil Mladenov and George Iliakis
DOI: 10.5772/24572
19
In addition, APE1 was shown to cleave the resulting abasic sites from the ICL DNA
substrates. Methoxyamine, which binds to the abasic sites and blocks the action of APE1
gives rise to a cisplatin resistant phenotype [71]. In contrast, overexpression of APE1 has
been observed in cisplatin resistant head and neck cancer and non-small lung cancer cells
[72, 73]. On the other hand, APE/ref-1 expression was unchanged between platinumsensitive and platinum-refractory ovarian cancers [74]. Polβ has been implicated in
modulating cisplatin cytotoxicity. It has been shown that Polβ facilitates the error prone
translesion replication synthesis of the intrastrand adducts and thus, gives rise to DNA
damage tolerance, which ultimately results in drug resistance [75-77]. On the other hand,
recent reports have indicated that cisplatin treatment in Polβ knockout cells resulted in
increased cisplatin resistance [78]. In support of this, the data from our lab demonstrated
that Polβ knockout MEFs and Polβ deficient human breast cancer cells are resistant to
cisplatin [71]. We have described a novel role of Polβ in mediating cisplatin sensitivity.
We showed that Polβ has a low fidelity at the sites flanking cisplatin ICLs, and this BER
processing interferes with the repair of cisplatin ICLs. Our data suggests that these
responses are specific to cisplatin ICLs as the resistant phenotype in the absence of Polβ
was seen only with cisplatin and not with any other crosslinking agent including
mitomycin C or oxaliplatin.
Other BER proteins like XRCC1 and PARP are also implicated in the processing
of ICLs [79]. Lack of XRCC1 gave rise to a sensitive phenotype in hepatocellular
carcinoma cell line in response to cisplatin [80]. PARP is essential for the repair of
single-strand breaks (SSBs) and its downregulation has been shown to sensitize a number
20
of cancer cell lines to γ-irradiation and alkylating agents [81]. Photoaffinity crosslinking
experiments have revealed that proteins like XRCC1, DNA ligase III and PARP1 bind to
the cisplatin ICLs in vitro, highlighting a possible role of these proteins in the processing
of cisplatin ICLs [82]. These studies suggest that ICLs can be potential substrates of the
BER pathway. In addition, differential response of BER deficient cell lines to different
ICL inducing agents point toward the dependence
on the structure of the ICL to
influence which DNA repair pathway will be activated. Studies that show processing of
ICLs by Neil 1 and by UNG reveal both productive and non-productive repair of ICLs by
BER proteins.
1.13 Mismatch repair (MMR): a post replicative DNA repair pathway
Replicative polymerases possess 3’-5’ exonucleolytic activity to remove any
mismatched bases. However, when such mismatches escape the polymerase proofreading
activity, the MMR pathway is activated. The MMR pathway is responsible for
maintaining genomic stability by correcting the mismatches generated during replication
and by inhibiting recombination between non-identical sequences. MMR is highly
conserved through evolution and deficiency of MMR proteins leads to genome wide
instability and are therefore, responsible for giving rise to a mutator phenotype. Due to
this reason, defects in the MMR system have been associated with an increased risk of
cancer. Inheritance of MMR mutations is associated with hereditary non-polyposis
colorectal cancer (HNPCC), also called Lynch syndrome. This genetic condition also
leads to increased risk for the development of cancers of endometrium, stomach, ovaries,
small intestine and upper urinary tract. Sometimes somatic mutations in MMR genes give
rise to sporadic cancers. In addition, epigenetic modification of MLH1 has also been
21
observed in some cases of sporadic cancers [83, 84]. MMR also plays a key role in the
cytotoxic response to alkylating agents as well as to platinating agents, including cisplatin
and carboplatin. MMR deficient cells have been shown to be around 100 times resistant
to cell killing by alkylating agents [85], whereas the fold difference of this resistance to
platinating agents is around 2-3 fold [86, 87]. Recent evidence also shows the
involvement of the MMR pathway in the early steps of checkpoint and apoptotic
responses to DNA damaging agents [88].
Correction of the errors produced by DNA polymerases requires the directing of
the MMR system to the newly synthesized DNA strand. In E. coli, the new daughter
strand is transiently unmethylated at GATC sites, thus providing a means of strand
differentiation. These sites are normally methylated by the enzyme deoxyadenine
methylase. The activity of this enzyme lags behind the replication fork causing the newly
synthesized strand to be partially unmethylated. The recognition of the mismatch is
mediated by the MutS homodimer. MutS binding leads to the recruitment of MutL
homodimer. The formation of a MutS and MutL ternary complex leads to the activation
of MutH endonuclease. MutH binds to the hemimethylated GATC sites and cleaves the
unmethylated strand. The UvrD helicase then causes the unwinding of the mismatch
containing DNA strand. This unwinding provides access to a number of exonucleases
which digest the error containing strand. These nucleases can be 5’ to 3’ or 3’ to 5’
exonucleases and the selection depends upon whether the nearest GATC site lies 5’ or 3’
from the site of the mismatch. The mismatch removal stops the exonucleolytic cleavage,
and the resulting gap is filled by DNA PolIII followed by nick sealing by DNA ligase.
22
1.14 Mismatch repair in mammalian cells
Mismatch recognition in mammalian cells is mediated by MutS homologues
(MSH). There are five MSH proteins identified in human cells: MSH2, MSH3, MSH4,
MSH5 and MSH6. These proteins participate in the mismatch recognition in the form of
heterodimers. The two most common heterodimers required for mismatch repair and
present in all cells are MutSα and MutSβ. The most abundant form of heterodimer
present in cells is MutSα, which is composed of MSH2 and MSH6. MutSα is required for
the recognition of base-base mismatches and single insertion deletion loops (IDLs) [89].
Another heterodimer called MutSβ, which is composed of MSH2 and MSH3 is required
for the recognition of longer IDLs [90].
Single nucleotide IDLs are recognized by both MutSα and MutSβ heterodimers
giving rise to partial redundancy in their functions. This functional overlap has been
helpful in explaining the tumor phenotypes of MSH2, MSH6 and MSH3 null mice.
MSH2 forms a common partner of these two heterodimers and lack of MSH2 displays the
most severe tumor and mutator phenotype as it lacks all of the mismatch recognition
function. MSH6 deficiency displays a less severe mutator phenotype suggesting that
MSH3 can pair up with MSH2 to repair IDLs. In addition, MSH3 -/- mice show late stage
tumorigenesis indicating that MutSα can repair most of the mismatches generated during
DNA replication. In addition, MutSα has also been shown to recognize larger IDLs with
reduced affinity [91].
At the site of the mismatch, MutS heterodimers form a sliding clamp that initially
recognizes double-stranded DNA and later engages in ATP dependent sliding along the
23
DNA strand (Figure 4). MSH proteins have one ATP binding site per molecule and in the
presence of ADP, the MSH proteins form a tight complex with the mismatch. However,
in the presence of ATP, the MSH proteins form a sliding clamp. This movement is
essential at the site of the mismatch, as MSH proteins cannot recognize the newly
synthesized strand. The MutH protein found in prokaryotic cells is responsible for
making a nick in the error containing strand. There is no homologue for MutH in human
cells and the activity that helps to recognize the correct strand has not been discovered in
eukaryotic cells. Several models exist to explain how the mismatch containing strand
might be identified in human cells. Several in vitro studies have shown that the MMR
system is activated after it recognizes a nick in the mismatch containing strand, and this
nick serves as an entry point for the exonucleases which cleave the damaged DNA strand
[92]. These observations support the idea that the strand recognition can be dependent on
a single-strand nick. This nick can be provided by gaps between okazaki fragments in the
lagging strand during DNA replication. It has been observed that the yeast mismatch
repair system corrects mismatches more efficiently on the lagging strand during DNA
replication [93]. Furthermore, studies show that MutL has endonucleolytic activity in
MMR in yeast as well as in human cells [94, 95]. But, it is still not clear what drives
strand differentiation between newly synthesized and parental DNA strands.
There are three MutL activites present in mammalian cells. Like MutS, these
activites also occur in the form of heterocomplexes with MLH1 acting as a common
subunit. MutLα is composed of MLH1 and PMS2 and is a primary activity present in the
mammalian cells. MutLα comprises ~ 90% of all MLH1 in human cells [96]. MutLα can
continue the mismatch repair initiated by both MutSα and MutSβ [97]. MutLβ is
24
composed of MLH1 and PMS1 heterodimer. This heterodimer does not participate in
MMR in vitro. However, PMS1 knockout mice displayed microsatellite instability
highlighting some role of MutLβ in MMR in vivo [98]. MutLγ consists of MLH1 and
MLH3 subunits, and it has been shown to support MutSα initiated base-base and single
IDL repair to a low level in vitro [99]. MutL endonucleolytic activity has been attributed
to a divalent metal binding site within PMS2 which is called a DQHA(X2)E(X4)E motif.
This motif is evolutionarily conserved but is absent in bacterial systems which use GATC
hemimethylation to direct mismatch repair. Mutations in this motif lead to the inhibition
of MutL endonuclease activity as well as it abolishes mismatch repair activity in nuclear
extracts [94].
Downstream exonucleolytic degradation in MMR is dependent upon the nick
generated either 5’ or 3’ to the mismatch. Exonuclease I (EXO1) has been reported to
possess 5’ to 3’ exonuclease activity, and this activity has been implicated in MMR in
vivo and in vitro. EXO1 -/- mice are deficient in MMR of mononucleotide repeats. Some
of the recent studies have shown that the 3’ to 5’ exonuclease activity of Mre11 is
involved in MMR in vivo [100]. In addition, Mre11 deficiency leads to the reduction in
the efficiency of 3’ directed mismatch repair by ~ 40%, and this effect was rescued by the
addition of purified Mre11 protein [101]. EXO1 mediated excision of the damaged DNA
strand at a nick can span over 1000 nucleotides from the mismatch, and it stops about 150
nucleotides past the mismatch [102].
25
In addition to MutL, MutS and EXO I MMR requires other factors for
completion. These factors, which inlude PCNA, RFC, RPA and Polδ, are also involved in
DNA replication machinery and have been implicated in mismatch repair in human cell
extracts. PCNA is loaded onto the DNA by RFC and is essential for Polδ mediated DNA
repair synthesis [103, 104]. In addition, MSH proteins MSH6 and MSH3 have been
reported to be associated with PCNA, suggesting that MMR proteins may form a part of
replication machinery [105]. RPA binds to ss- DNA and it stabilizes the parent strand
while the error containing daughter strand is being removed. After DNA synthesis, DNA
ligase I seals the nick, thereby, completing the repair.
1. Mismatch recognition by MutS heterodimers
2. Recruitment of MutL complex
3. Exonuclease activity
4. DNA repair synthesis and ligation
Figure 4: Mismatch repair of a single base mismatch. Figure modified from [106]
26
1.15 Role of mismatch repair in tumor susceptibility
Knock-out mouse models have been developed for each of the MMR genes. Most
of these mouse models display a mutator phenotype and a predisposition to cancer. A
mutator phenotype is defined by the high microsatellite instability (MSI) caused by a lack
of MMR function in cells. Microsatellites are repeated sequences in DNA. The nucleotide
sequences within the repetitive components contain mononucleotide repeats of adenine
(A)n or cytosine-adenine (CA)n dinucleotide repeats. During DNA replication, due to
polymerase slippage, these repetitive sequences may decrease or increase in length. In the
absence of a functional MMR pathway, these mutations are fixed in the genome leading
to high MSI phenotype. The most widely used markers for the detection of MSI
phenotype are 2 mononucleotide repeat markers: BAT25 and BAT26; and 3 dinucleotide
repeat markers: D2S123, D5S346 and D17S250 [107].
MMR deficient mice typically develop lymphomas, sarcomas and gastrointestinal
tumors. MSH2 deficient mice have shorter life span, and they show high MSI and a
cancer phenotype. Lack of MSH3 does not give rise to cancer susceptibility, consistent
with the fact that no MSH3 mutations associated with human HNPCC syndrome have
been identified. MSH6 deficient mice displaying tumor susceptibility similar to MSH2
deficient mice, however, they do not share the typical MSI phenotype displayed by
MSH2 null mice. MLH1 deficient mice exhibit similar characteristics as MSH2 deficient
mice. In addition, MLH1 deficient mice are also infertile, consistent with the role of
MutLγ complex in meiosis. Exo1 -/- null mice display shorter life span, an increased
susceptibility to develop lymphomas and similar to MLH1 null mice, EXO1 -/- mice are
sterile [108-110]
27
In addition, epigenetic modifications also give rise to inactivation of MMR genes.
For example, studies have shown that hypermethylation of promoter sequences of MLH1
is responsible for lack of MLH1 function in sporadic colon cancers [111]. In these
studies, tumor cell lines with MLH1 promoter hypermethylation were examined. These
cell lines displayed loss of MLH1 function. When these cell lines were treated with the
demethylating agent 5-aza-deoxycytidine, the MLH1 function was restored indicating
that epigenetic modifications of MMR genes can be responsible for a mutator phenotype
in addition to genetic defects [112].
1.16 Role of mismatch repair pathway in response to DNA damage
In addition to a predisposition to cancer phenotype, mismatch repair deficiency
gives rise to resistance to alkylating agents and platinating agents like cisplatin and
carboplatin. MMR deficient human cells are ~ 100 fold resistant to the treatment with
alkylating agents as compared to MMR proficient cells [113, 114]. The commonly used
alkyaltina agents such as N-methyl-N-nitrosoguanidine (MNNG), N-methyl-Nnitrosourea (MNU) and temozolomide produce N’ subsitituted bases and methylate O6
position of guanine (O6MeG). O6MeG can also be removed by a suicide enzyme called
O6MeG methyl transferase (MGMT) [115]. In the absence of MGMT and during DNA
synthesis, mutations are generated as there is a misincorporation of thymine instead of
cytosine. The MMR system recognizes these mismatches, which leads to the activation of
cell cycle arrest pathways and cell death pathways [116]. Cells lacking MMR proteins are
resistant to treatment with alkylating agents [117-119]. Colorectal tumor cells deficient in
MLH1 showed a resistance to MNNG as compared to the chromosome 3 complemented
MLH1 proficient sub-line [120, 121].
28
Cisplatin forms DNA adducts that have been widely shown to be recognized by
the MMR system. These adducts, however, are not processed by MMR proteins. The
purified hMSH2 protein has been shown to bind to cisplatin adducts, especially to a
single 1,2 GG adduct [122]. This binding was specific to adducts formed by clinically
relevant anticancer agents like cisplatin, MNNG but not to clinically ineffective
transplatin [123]. MLH1 proficient and deficient cells were tested for their response to a
number of chemotherapeutic agents like ionizing radiation (IR), topoisomerase inhibitors,
MNNG and crosslinking agents like mitomycin C, psoralen and cisplatin. The results
showed that MMR deficiency gives rise to cellular resistance only to alkylating agents
and cisplatin [87]. Cell lines lacking a functional MSH2 or MLH1 are 2 to 4 fold resistant
to cisplatin [86, 87, 124, 125]. This low level of resistance leads to an enrichment of
MMR deficient cells in the heterogenous tumor population. In a xenograft model
generated using MSH2+/+ and 2 fold resistant MSH2 -/- cells, it was observed that
MSH2 deficient cells were less responsive to cisplatin treatment. These data suggested
that low level resistance caused by lack of MMR is clinically relevant. In a case where
MMR deficiency occurs early in the tumor development, the cells remain resistant to the
treatment but no further enrichment would occur after cisplatin treatment. On the other
hand, occurrence of a mutation in MMR in the later stages of tumor development, would
result in the enrichment of cells that are deficient in MMR following cisplatin treatment.
Re-expression of MLH1 in MLH1 deficient cells also leads to an increase in sensitivity to
cisplatin [87]. These results suggest that the MMR status of the cells affects the clinical
outcome of cisplatin therapy.
29
Various models have been proposed to describe the mechanism by which MMR
deficiency confers cisplatin resistance. One of the models is the “futile cycle” of DNA
repair. Studies by Vaismann et al have shown that MLH1 and MSH6 deficient cells
displayed 2.5 to 6 fold increased replicative bypass of cisplatin adducts, and these cells
showed 2 to 5 fold increased resistance to cisplatin [124]. According to this model,
replicative bypass of the adducts leads to the generation of mismatches. This
modification is recognized by MutS and MutL complexes. Repair of these mismatches
fails because it is directed at the daughter strand. This leads to the removal of the
mismatch, but the adduct in the parent strand remains intact. The persistent action of
these “futile” cycles of replication and repair leads to the generation of DNA DSBs which
signal apoptosis. Oxaliplatin adducts were not recognized by mismatch repair proteins
and there was no significant difference between the replicative bypass of these adducts in
MMR proficient and deficient cells. The futile cycle model of mismatch repair has been
suggested in case of treatment with alkylating agents like MNNG. Studies have shown
that methylation damage leads to single-strand breaks in the newly synthesized DNA and
is dependent on the mismatch repair status of the cells [126]. MMR attempts to
correct MeG/T mispairs that arise during DNA replication. Due to the persistence of MeG
in the template strand, repair synthesis cannot take place and this creates single-stranded
gaps. During a second round of replication, these gaps cause replication fork collapse and
lead to the activation of recombination pathways and cell cycle arrest to allow repair.
These data implicate a futile repair model where MMR makes repeated, unsuccessful
attempts to correct the mismatches and in this process, strand breaks are produced which
cause cell cycle arrest and eventually apoptosis [127, 128].
30
Another model to describe drug resistance is the direct signaling model.
According to this model, the MMR system acts as DNA damage sensor and can activate
signal transduction pathways [129, 130]. MMR proteins are considered to be associated
with the replication machinery, but MMR proteins can also be recruited to DNA damage
independent of replication [131]. MMR proteins take part in a post replicative repair
pathway, but their expression remains constant throughout the cell cycle [132]. Studies
have shown that MutS proteins bind to damaged DNA as well as to damaged DNA
containing a mismatch [133]. MMR proteins are reported to be recruited at the sites of
DNA damage in human cells [134]. In this report, MMR proteins were shown to be
recruited immediately to the site of single-strand breaks, DSBs and UV induced DNA
damage in human cells indicating that MMR plays a role in a cellular response to a
variety of DNA damage. In addition, genetic evidence of the direct signaling model has
been provided by the generation of “separation of function” mutants of MMR proteins
[88, 135-137]. It was shown that DNA repair and DNA damage induced apoptosis
functions are uncoupled in the case of a missense mutation G674A in the ATPase domain
of MSH2. In this case, cells with this mutation remain responsive to chemotherapy even
in the absence of mismatch repair activity while still maintaining the function of
signaling apoptosis [88, 138]. Consistent with this idea, MLH1 G67R and MSH6
T1217D ATPase missense mutations also represent separation of function mutants where
they exhibit mismatch repair deficiency while still retaining sensitivity to
chemotherapeutic agents. These data implied that DNA damage response function and
mismatch repair mediated by MMR proteins are independent of each other [135, 136].
Moreover, PMS2 has been reported to interact with p73 which is a p53 related protein.
31
This interaction is essential for stabilization of p73 and cisplatin mediated stimulation of
p73 apoptotic function. These data indicate the role of PMS2 in DNA damage induced
apoptosis in addition to mismatch repair [139]. In a similar vein, studies have shown that
PMS2 R20Q variant is defective in inducing p73 mediated apoptotic response to cisplatin
further strengthening the role of MMR proteins in DNA damage response [137].
Cisplatin has been shown to induce ATR mediated DNA damage response leading to
apoptosis. MSH2 was recognized to be one of the ATR binding proteins. Furthermore,
ATR activation and its recruitment to the nuclear foci was inhibited in MSH2 deficient
cells. This report concluded that ATR activation and the downstream Chk2 signaling in
response to cisplatin treatment is dependent on MSH2 [130]. In another study, cisplatin
induced apoptotic signaling was shown to be dependent on the MMR status of the cells.
The localization of cytochrome C to cytoplasm as well as cleavage of caspase-9 and
caspase-3 was reported to be reliant on the presence of MMR proteins. Additionally, this
response was found to be independent of p53 [140]. It remains to be understood whether
MMR mediated apoptosis is dependent on p53. Studies have shown that this response can
be both dependent and independent of p53. Studies using a number of cell lines
containing MMR mutations showed that increased sensitivity in MMR proficient cells
was independent of p53 status [28, 140, 141].
These studies establish a definite role of MMR proteins in DNA damage response
after cisplatin treatment. MMR proteins have been shown to recognize both intrastrand
adducts and ICLs formed by cisplatin [82, 123, 142]. However, how MMR proteins
process different adducts formed by cisplatin and how they mediate DNA damage
signaling in response to different types of adducts is currently unknown. MMR seems to
32
play a role in both postreplicative mismatch repair as well as in DNA damage response
leading to apoptosis. The repair function allows faithful completion of replication, and
the apoptosis signaling function is required for the induction of cell death in heavily
damaged cells. Therefore, both these functions are required to maintain genomic stability.
MMR dependent responses to chemotherapy agents are clinically relevant, and thus,
further studies on MMR dependent signaling complex are required for the assessment of
the exact role of MMR in DNA damage response.
1.17 Role of ATR and ATM signaling in cisplatin cytotoxicity
DNA damage by platinum or other chemotherapeutic agents initiates a response
that activates DNA repair mechanisms that are specific for the type of lesions produced.
Along with these repair processes, DNA damage response also involves additional
processes. The generation of highly cytotoxic lesions involves activation of cell cycle
checkpoints and cell cycle arrest. Checkpoints form a cellular surveillance mechanism
and consist of pathways that coordinate DNA repair processes and cell cycle transitions.
Phosphoinositide-3- kinase related (PI3K) kinases such as ATM, ATR and DNA-PK
form central components of the checkpoint machinery. ATR is activated in response to
ssDNA and stalled replication forks. On the other hand, ATM and DNA-PK respond to
DNA DSBs. Recruitment of ATR and ATM to the damaged site leads to the activation of
a number of downstream substrates including checkpoint kinase 1 (Chk1) and checkpoint
kinase 2 (Chk2) respectively. Cisplatin treatment has been reported to activate checkpoint
signaling components including the ATR-Chk1 pathway. Studies have shown that ATR
plays a critical role in tumor cell survival however Chk1 inhibition did not sensitize cells
to cisplatin [143]. Moreover, one study showed that ATR activation in response to
33
cisplatin can be initiated by RPA/9-1-1 complex. The hMSH2 mediated ATR activation
leads to Chk2 phosphorylation (instead of canonical ATR-Chk1 pathway) culminating in
cell cycle arrest and apoptosis [130]. However, the role of Chk1 cannot be completely
ruled out as Chk1 has been reported to play an important role in the repair of ICLs by
phosphorylating Rad51 to stimulate HR and by phosphorylating FancE to repair the
crosslinks through the FA pathway [144, 145]. SiRNA mediated inhibition of ATR
sensitizes tumor cells to cisplatin and carboplatin, both in vitro and in vivo [146]. In
addition, ATR inhibition using a highly selective ATR inhibitor, VE-821, leads to cell
death, and this compound synergizes well with cisplatin and other chemotherapy agents.
This study showed a synthetic lethal interaction between ATR and ATM/p53 signaling
pathway, and thus, making them selective for tumor cells. ATM plays a key role in
mediating sensitivity to DNA damaging agents especially IR [147]. A recently developed
compound, KU60019, inhibited ATM kinase and resulted into radiosensitization of
human glioma cells. KU60019 treatment also resulted in inhibition of phosphorylation of
downstream DNA damage effectors like p53 and H2AX. However, beyond these studies,
little progress has been made in the development of new ATM kinase inhibitors. The
effect of ATM kinase inhibitors in combination with platinating agents is not well
established. As the repair of cisplatin adducts involves the generation of DSBs that are
detected by ATM, these inhibitors might prove beneficial to improve cisplatin
chemotherapy in combination treatment.
34
1.18 Novel role of BER and MMR in mediating cisplatin sensitivity
Cisplatin has been used in the clinic for almost 4 decades since its discovery and
FDA approval. It has been used successfully in the treatment of testicular cancer in
combination with other anticancer agents. However, cisplatin-based chemotherapy is
vulnerable to the development of resistance, recurrence and hence poor prognosis.
Cisplatin has been studied extensively but the mechanisms of resistance are still not
completely understood. DNA repair plays an important role in modulating cisplatin
cytotoxicity. Alteration in DNA repair pathways is one of the major mechanisms of
resistance. Therefore, it becomes important to understand how cisplatin adducts are
recognized and how different DNA repair pathways are activated.
The cisplatin ICL has a unique structure with cytosine residues flipped away from
the double helix [26]. These cytosine residues are more prone to undergo oxidative
deamination in the cellular environment. Using cisplatin ICL substrates in vitro, studies
from our lab have shown that these cytosine residues undergo preferential deamination to
uracil. These residues are removed by uracil DNA glycosylase to form an abasic site. We
have observed that cisplatin treatment leads to an increased generation of abasic sites as
compared to the other crosslinking agents including oxaliplatin, transplatin and
mitomycin C. This abasic site is recognized by APE1 which cleaves at the 5’ side of the
abasic site to generate a free 3’ OH group. This free end is used by Polβ to carry out
DNA synthesis. The lyase activity of Polβ is utilized for the removal of 5’ dRP residue
that is remaining in the DNA backbone. The BER proteins bind to the regions flanking
cisplatin ICL. We believe that this results in non-productive BER processing as BER
proteins do not repair the ICLs but interfere with the productive repair of ICLs by NER
35
and HR proteins. The repair of ICLs is hampered and in the presence of BER proteins the
cells show increased cellular sensitivity to cisplatin. Furthermore, DNA synthesis by Polβ
is error prone. We experimentally showed that Polβ can incorporate incorrect nucleotides
even in the presence of correct bases. Distortion created by ICL is unique to cisplatin as
the DNA helix is not shown to be distorted to this degree in case of other cross-linking
agents like oxaliplatin and mitomycin C. Our studies showed that Polβ processing is
more mutagenic near cisplatin ICL sites as compared to undamaged DNA duplex. The
misincorporation by Polβ generates mismatched bases, and we hypothesize that this will
result in MMR protein binding and further MMR processing. We believe that the binding
of MMR proteins would inhibit the productive repair of cisplatin ICLs by NER and HR
proteins resulting in decreased repair of ICLs and therefore increased cellular sensitivity.
The aim of this study is to understand whether BER and MMR can work together
in the same molecular mechanism. Our hypothesis is that BER and MMR
processing near the ICL sites would interfere with the “actual repair” of cisplatin
ICLs and mediate cisplatin sensitivity.
36
Chapter 2
Overlapping role of BER and MMR proteins in mediating cisplatin
cytotoxicity*
Akshada Sawant1, Anbarasi Kothandapani1, Robert W. Sobol2, 3, 4 and Steve M.
Patrick5
1. Department of Biochemistry and Cancer Biology, University of Toledo – Health Science
Campus, Toledo, OH 43614
2. Department of Pharmacology & Chemical Biology, University of Pittsburgh School of
Medicine, Pittsburgh, PA 15213
3. University of Pittsburgh Cancer Institute, Hillman Cancer Center, Pittsburgh, PA 15213,
USA
4. Department of Human Genetics, University of Pittsburgh Graduate School of Public Health,
Pittsburgh, PA 15213
5. Department of Oncology, Karmanos cancer institute, Wayne State University,
Detroit, MI 48201
*This work was published as a part of following manuscript
Epistatic role of base excision repair and mismatch repair pathways in mediating
cisplatin cytotoxicity.
Kothandapani A, Sawant A, Dangeti VS, Sobol RW, Patrick SM.
Nucleic Acids Res. 2013 Aug;41(15):7332-43. doi: 10.1093/nar/gkt479. Epub 2013 Jun
12.
37
2.1 Introduction
Cisplatin remains one of the most widely used chemotherapeutic agents for the treatment
of a variety of cancers. Cisplatin binds to DNA and forms intrastrand and interstrand
crosslinks (ICLs). These adducts hamper normal cellular functions like DNA replication
and RNA transcription which leads to cell cycle arrest and/or apoptosis. Cisplatin has
been used successfully in the treatment of testicular cancer as a part of combination
therapy along with other anticancer agents. However, there are various limitations to the
clinical use of cisplatin. Cisplatin based chemotherapy is susceptible to the development
of resistance, recurrence and hence poor prognosis. The mechanisms of development of
resistance are still not completely understood. One of the major mechanisms of resistance
is the alteration in the DNA repair capacity of the cells. Therefore, it becomes important
to understand how various DNA repair proteins recognize cisplatin adducts and how it
leads to the activation of a variety of DNA repair pathways. Nucleotide excision repair
(NER) is the major pathway responsible for the repair of cisplatin intrastrand adducts.
But the exact mechanisms of cisplatin ICL repair is poorly understood. NER, HR and FA
pathway play an essential role in the repair of cisplatin ICLs. Other DNA repair pathways
like base excision repair (BER) and mismatch repair (MMR) have also been implicated in
mediating cisplatin cytotoxicity. Recent studies from our lab and others have shown that
reduced expression of polymerase β (Polβ) gives rise to cisplatin resistance [71, 78]. In
addition, MMR deficiency also results in resistance to cisplatin and carboplatin [86].
However, the exact mechanism of how these proteins influence cisplatin cytotoxicity is
not well understood.
38
The MMR pathway maintains genomic stability via post replicative mismatch
repair. The mismatches can be recognized by one of the two heterocomplexes. MutSα is
composed of MSH2 and MSH6 and recognizes base-base mismatches and single IDLs.
On the other hand, MutSβ consists of MSH2 and MSH3 and is required for the repair of
IDLs larger than a single base pair. Mismatch recognition by MutS proteins is followed
by recruitment of MutL complexes which consist of essential downstream MMR
proteins. In addition to its role in repair of mismatches, the MMR pathway also responds
to DNA damage produced by a number of DNA damaging agents. MMR machinery is
responsible for the recognition of mismatches generated during the treatment with
alkylating agents. MMR deficient cells are resistant to treatment with methylating agents
[128]. Lack of MMR proteins gives rise to a low level of resistance to cisplatin and
carboplatin [86, 124]. Various models have been proposed to explain the mechanism of
MMR mediated cisplatin resistance, including futile cycle repair of cisplatin intrastrand
adducts, MMR dependent apoptotic signaling and a model involving DNA damage
shielding [148, 149]. The postulated hypotheses for these models involve binding of
MMR proteins to cisplatin intrastrand adducts. MMR recognition protein MutSα
recognizes cisplatin GG intrastrand adduct [122, 123, 142]. The binding of MMR
proteins to cisplatin ICLs is not very well studied. Photo-crosslinking experiments and
electromobility shift assays have also shown that hMutSβ is one of the proteins that binds
to cisplatin ICLs. The pathways involved in the recruitment of MMR proteins to cisplatin
ICLs in vivo remain to be explored.
Our previous studies have shown that the BER pathway modulates cisplatin
cytotoxicity [71]. Studies have shown that Polβ is capable of bypassing cisplatin
39
intrastrand adducts which may lead to DNA damage tolerance and increased
mutagenicity [150, 151]. Using cisplatin ICL DNA substrates, we have shown that BER
proteins process the regions near the ICL site and Polβ has a low fidelity near the DNA
flanking the ICL sites. This leads to misincorporation of nucleotides at the sites flanking
the cisplatin ICL and these mismatched bases can act as an initiating point for the MMR
pathway. Therefore, we targeted BER and MMR pathways separately and together to
assess whether these pathways work together in an epistatic manner to mediate cisplatin
sensitivity. Using mouse embryonic fibroblasts (MEFs), we show that BER and MMR
pathways are epistatic in mediating cisplatin cytotoxicity. Our studies show that, when
BER and MMR are downregulated separately there is cisplatin resistance, and when
downregulated together, no additional increase in the resistance was seen. This strongly
suggests that BER and MMR have overlapping roles in mediating cisplatin sensitivity.
We also show that MMR binding occurs downstream of BER processing, and activation
of MMR is dependent upon the mutagenic role of Polβ near cisplatin ICL sites. Our data
also demonstrate that loss of BER and MMR leads to faster repair of ICLs which appears
to be the cause of cisplatin resistance in these cells. Based on our results, we hypothesize
that BER and MMR pathways maintain cisplatin sensitivity by competing with the
“actual repair” of cisplatin ICLs. BER and MMR deficiency, therefore, leads to faster
repair of ICLs and hence cisplatin resistance.
40
2.2 Materials and Methods
Chemicals and Antibodies
Cisplatin was purchased from Sigma-Aldrich. VE-821 was purchased from Axon
Medchem. KU60019 and NU7441 were purchased from Tocris Bioscience. All other
chemicals and reagents were from standard suppliers. The antibody for MSH2 was from
Calbiochem and α-tubulin was from Sigma-Aldrich. For stock preparation, cisplatin was
diluted in 1X PBS and vortexed vigorously to dissolve the drug completely. The stock
concentration was 1mM and fresh stocks of cisplatin were prepared before each
experiment.
Cell lines
The human breast adenocarcinoma MDA-MB-231 cells were grown in RPMI 1640
containing 10% FBS and antibiotics. MDA-MB-231 Polβ knockdown cells (Polβ
lentiviral shRNA) were grown in the presence of 0.5μg/mL puromycin. The MDA-MB231 Polβ knockdown cells re-expressing WT Polβ and a variant deficient in polymerase
activity (D256A) were grown under similar conditions with the addition of 700μg/ml
geneticin. The development and characterization of the MDA-MB-231/Polβ-KD cells
were described previously [152]. Wild-type (92TAg) and Polβ null (88TAg) primary
MEFs were cultured in high glucose DMEM supplemented with 10% FBS plus
antibiotics at 37ºC in a humidified atmosphere under 10% CO2. The MSH2 deficient
human endometrial carcinoma cell line HEC59 and its chromosome 2 complemented
sub-line HEC59+2 were obtained from Dr. Kandace Williams, University of Toledo. The
41
chromosome complemented sub-lines were maintained in DMEM F-12 media
supplemented with 400 μg/ml of geneticin.
siRNA transfection
ON-TARGET plus SMART pool siRNAs specific for human MSH2, XPF and ERCC1
were purchased from Dharmacon RNAi technologies, Thermo Scientific. The nontargeting control siRNA was used as a control for non-specific effects. siRNA
transfection was carried out per the manufacturer’s protocol. Briefly, the cells were plated
in 6 well plates in the antibiotic free media. At the time of transfection, the cell density
was maintained at 60-70% and two transfections were done with an interval of 24 hrs.
Dharmafect transfection reagent 1 and 4 were used for MEFs and MDA-MB-231 cells,
respectively. The cells were harvested at 48 and 72 hr timepoints after transfection for the
detection of protein and transcript knockdown.
Western blot analysis
Cells were harvested at the indicated timepoints after the infection, washed with PBS and
lysed in lysis buffer (10 mM Tris pH 8.0, 120 mM NaCl, 0.5% NP-40, 1mM EDTA)
containing protease inhibitors (stock concentrations 1000X: 0.5 M phenyl methyl
sulphonyl fluoride PMSF, 1 mg/ml Leupeptin and 1 mg/ml pepstatin A). The protein
concentration was measured using Bradford assay with BSA as a standard.
Approximately 50 µg of total protein extract was used for loading. The proteins were
separated on 8% SDS-polyacrylamide gels and transferred onto Immobilon-P transfer
membranes (Millipore). After blocking in 2% non-fat dry milk, the membranes were
probed with primary antibodies recognizing human MSH2, with α-tubulin as loading
42
control. The primary antibody incubations were carried out overnight at 4ºC. The
membranes were incubated with appropriate secondary antibodies and the signal was
detected by using Enhanced chemiluminescence detection system.
Real time PCR for the measurement of transcript levels
At indicated post-transfection time points, cells were harvested and pelleted. RNA was
isolated using TRIzol reagent (Invitrogen) by standard procedures. The total RNA was
reverse transcribed using MMLV reverse transcriptase enzyme (Invitrogen) as per the
manufacturer’s protocol. The primers specific for human and mouse MSH genes were
designed and were analyzed using oligoanalyzer 3.0 (IDT technologies). The transcript
levels were quantified using iQ SYBR green supermix (Bio-Rad) in iCycler iQ System,
with GAPDH as an endogenous control for human cells and B actin for mouse cells. The
percent transcript knockdown was determined from 2^-∆∆CT values.
Colony survival assay
The cells were plated at a density of ~ 500 cells per dish. Cells were treated with
increasing concentrations of cisplatin for 2 hr in serum free media. After treatment, fresh
media was added and the cells were allowed to grow for 7-14 days. Colonies were fixed
with 95% methanol and stained with 0.2 % crystal violet. Colonies with ≥50 cells were
counted using a light microscope, and colony survival was expressed as the ratio of the
average number of colonies in drug treated cells versus control cells x 100. The
experiment was done in triplicate for each drug concentration. Statistical analyses were
performed using student’s t test comparing wild type and knock down cells.
43
Drug treatment
The cells were treated with VE-821, KU60019 or NU7441 alone to assess their effect on
cell viability (Data not shown). From these curves, a concentration (IC20) was chosen, so
as to have minimum influence on cell viability with the inhibitors alone. The cells were
pretreated with this concentration for 2 hrs followed by treatment with increasing
concentration of cisplatin for 2 additional hrs. The cells were kept in the inhibitor
containing media for total 24 hrs. The colony survival assays were performed as
described.
ELISA
Cisplatin intrastrand adduct measurement was carried out using an antibody that detects
the major cisplatin GG adduct. This antibody was kindly provided by Michael J. Tilby,
University of Newcastle, UK. Cells were treated with cisplatin for 2 hr in serum free
media. After completetion of 2 hr treatment, the drug containing media was replaced with
complete media. The cells were then harvested at timepoints 0, 12, 24, 48 hr post
treatment. The repair kinetics of cisplatin intrastrand adducts was assessed by ELISA.
The cell pellets were stored at -80 ºC until needed. The genomic DNA was harvested
using DNeasy blood and tissue DNA isolation kit from Qiagen. The DNA concentrations
were measured using a nanodrop instrument. The genomic DNA was coated onto 96 well
ELISA plates in a coating buffer containing 1M NaCl, 50 mM sodium Phosphate buffer,
pH 7.0 and 0.02% Sodium Azide. The coating reaction was done overnight at 4 ºC. The
next day the plate was washed using a wash buffer containing 0.1% Tween-20 in PBS.
After 3 washes of 10 mins each the blocking buffer (1% BSA) was added to the plates
44
and incubated for 1 hr at room temperature (RT). The plate was washed and was
incubated with primary antibody ICR4 which is specific for cisplatin GG adducts. After
incubation with HRP conjugated goat anti-rat antibody (Calbiochem), TMB (1 step ultra
TMB-ELISA, Thermo Scientific) was added. The reaction was stopped with 2M sulfuric
acid and absorbance was measured at 450 nm (Spectramax M5 plate reader, Molecular
Devices). The % intrastrand adducts were calculated using OD 450 nm where the 0 hr
time point was used as 100% intrastrand adducts in each cell line.
Alkaline comet assay
Alkaline comet assay was used to analyze the repair of cisplatin interstrand cross-links
(ICLs) as described [153, 154]. This assay is also called single-cell gel electrophoresis as
the analysis can be done at the single cell level. In this method, the cells were treated with
cisplatin for 2 hrs in serum free media and the assay was performed at 0, 24, 48 and 72 hr
time points post-treatment. At the start of each time point, the cells were treated with 50
μM hydrogen peroxide (Fisher) for 15 mins in serum free media. Cells were trypsinized,
harvested and were resuspended in cold PBS. Cell suspensions (~10000 cells) were
embedded on a microscopic slide in 0.5% ow melting point (LMP) agarose. A total three
layers of normal melting agarose (without cells) and low melting agarose (with cells)
were used to embed the cells. Each layer of agarose was allowed to solidify at RT for 10
mins each. The cell lysis was carried out in a lysis buffer containing 2.5 M NaCl, 100mM
EDTA, 10mM Tris Base, NaOH to pH 10 and 1% triton X100 for 1 hr at 4 C. The lysis
step removes all the cellular proteins. After the lysis, the slides were incubated in ice-cold
alkaline solution for 20 min to allow the DNA to unwind. The denatured DNA is then
subjected to electrophoresis in a buffer containing 0.3 M NaOH and 1 mM EDTA.
45
Electrophoresis was carried out for 30 min at 28 V, 300 mA. Slides were neutralized and
stained with SYBR green (Invitrogen).
During electrophoresis, broken DNA fragments migrate further than the
supercoiled undamaged DNA. The resulting structure resembles a comet with a bright
fluorescent head nucleus and the length and intensity of the tail part is determined by the
DNA strand breakage produced in the cell. Cisplatin treatment generates covalent ICLs
between two adjoining strands of DNA. Hydrogen peroxide treatment leads to the
formation of DNA DSBs. The presence of a covalent linkage in the DNA retards its
migration during electrophoresis. Hence, the DNA comets from these cells have a
reduced tail moment. On the other hand, lack of ICLs in the DNA due to increased repair
leads to their uninhibited migration during the electrophoresis giving rise to a larger tail
moment. The comets were scored using a Nikon epifluorescence microscope. At least
fifty cells were analyzed per slide using Komet Assay Software 5.5F (Kinetic Imaging,
Liverpool, UK). The data was expressed as the percentage of ICLs that remained at a
particular time point normalized to 100% at 0 hr.
Immunofluorescence
Cisplatin ICL processing leads to the formation of DSBs during the later stages of repair.
As the DSBs are formed, ATM protein undergoes autophosphorylation and activation
which in turn phosphorylates H2AX in the chromatin near the damage. The
phosphorylated H2AX is called γ-H2AX. These events can be detected as foci at the
damage site in the nucleus. This phosphorylation event can be monitored using an
antibody against γ-H2AX. The dephosphorylation of γ-H2AX is correlated with repair of
46
the ICL induced DSBs. DSB repair was assessed by monitoring the nuclear γ-H2AX foci
by immunofluorescence. The cells were plated in 60mm plates containing coverslips in
complete media. The next day cisplatin treatment was carried out for 2 hr in serum free
media. The cells were assayed at 0, 24, 48 and 72 hr post-treatment. For each timepoint,
cells were fixed using 4% paraformaldehyde in PBS, permeabilized using 0.3% Triton
X100 and blocked using 10% goat serum in PBS. The primary monoclonal anti γ-H2AX
antibody was used at a dilution of 1:500 (Millipore). Washes were done using 0.1%
Triton X 100 in PBS. The coverslips were then incubated with secondary Alexa Fluor®
488 Goat Anti-Mouse IgG (H+L) Antibody (Invitrogen). The nuclei were stained with
DAPI. The coverslips were mounted on the glass slides using DAKO anti-fade reagent.
The edges of the coverslips were sealed with nail polish. The images were then visualized
using a Nikon Eclipse T2000-U microscope. Foci were counted in at least 200 cells at
each time point per condition in each cell line and results are expressed as % γ-H2AX
foci positive nuclei.
Preparation of ICL DNA substrates and biotin DNA pull downs
Biotinylated ICL DNA substrates were used for the pull down assays. These substrates
were designed and synthesized to include undamaged DNA, ICL substrate, ICL substrate
containing uracil. The individual DNA oligos were purchased from IDT and were gel
purified to remove any degradation products. The substrates for ICLs were designed to
contain a single G in the top strand for the formation of a single ICL. The platination
reaction was carried out using platination buffer containing 10 mM sodium phosphate pH
7.5 and 30 mM NaCl at a drug to nucleotide ratio of 15:1 for ICL substrates. The reaction
was carried out at 37 ºC for around ~15 hrs to ensure adduct formation. After the reaction
47
was complete, the unreacted cisplatin was removed by filtration through tip20 columns
from Qiagen. The substrates were then ethanol precipitated in the presence of 0.3 M
sodium acetate pH 5.2 and annealed to the complementary strand to generate a doublestranded DNA substrate. The annealing reaction was carried out in annealing buffer
containing 10 mM Tris base pH 7.5 and 100 mM NaCl. The annealed substrate was then
subjected to overnight dialysis at 37 ºC in 100 mM sodium perchlorate and 10 mM TrisHCL (pH 7.5) to enhance the crosslinking reaction. The dialyzed substrate was then
ethanol precipitated and was sequencing gel purified. In a sequencing gel containing urea,
the two DNA strands are separated. However, the presence of an ICL in the DNA retards
the migration of the DNA leading to separation of the ICL product from uncrosslinked
DNA. The crosslinked substrate runs as a smear near the top of the gel, and this smeared
area can be excised from the gel. The DNA is eluted from the gel and ethanol
precipitated. Undamaged substrates were annealed without platination reaction.
BER processing of ICL substrates was carried out by treatment with purified UDG, APE1
and DNA Polβ in the presence of dNTPs. The reaction was carried out at 37 ºC for 1 hr.
The substrates were then recovered using ethanol precipitation. This BER processing of
the ICL U substrates leads to either correct or incorrect base insertion next to the cisplatin
ICL.
The biotinylated substrates were then bound to streptavidin beads (Dynal biotech) in the
presence of DNA binding buffer containing 20 mM HEPES pH 7.8, 2 mM DTT, 0.001 %
NP40, 100 mM NaCl and 200 mM MgCl2 at 4 ºC for 30 mins. The unbound DNA was
washed off using 3 washes with binding buffer. The tubes were kept in a magnetic
separation stand (Promega) during the washing steps. The beads were then incubated
48
with hMSH2-hMSH6 cell extract prepared from baculovirus infected SF9 insect cells. In
the same step, poly dI-dC competitive inhibitor (Sigma Aldrich) was added at 30 fold
excess concentration to reduce non-specific binding. The tubes were incubated for 1 hr at
4 ºC. To increase the stringency of binding, the next step of washes were done using
binding buffer containing 200 mM NaCl. The bound proteins were eluted in 1M NaCl.
The supernatants were collected from magnetically separated beads. The protein
precipitation was carried out using 10% trichloroacetic acid (TCA). The pellets were
washed using cold acetone, resuspended in 1X SDS buffer and were loaded onto 8% SDS
gels, transferred onto PVDF membrane (Milipore) and probed for MSH2 using western
blot analysis.
DNA substrates used for in vitro assays
Cisplatin ICL is formed by annealing VS IC 2.1 & VS IC 2.2
VS IC 2.1 is platinated before the annealing.
VS IC 2.1- 5’- CTCTTCCCCATCTCCTTTCCGCCTCCTTCCTTCCCCTTCCCT-3’
VS IC 2.2 AGGGAAGGGGAAGGAAGGAGGCGGAAAGGAGATGGGGAAGAG/BIOTIN
Undamaged duplex DNA is formed by annealing VS IC 2.1 & VS IC 2.2
Cisplatin ICL U substrate:
VS 2.1U – 5’CTCTTCCCCATCTCCTTTG/ideoxyU/CCCTCCTTCCTTCCCCTTCCCT
VS 2.2AGGGAAGGGGAAGGAAGGAGGGGCAAAGGAGATGGGGAAGAG/BIOTIN
49
2.3 Results
siRNA transfection for MSH2 and MLH1 knockdown:
WT and Polβ -/- MEFs were transfected with siRNA directed against MSH2. The
cells were transfected twice with an interval of 24 hours and were harvested at 48, 72, 96
and 120 hrs after first transfection. The protein was extracted and western blot analysis
was done to check for protein expression. Protein levels were found to be reduced to 9095% in both wild type (WT) and Polβ MEFs at all time-points (Figure 5) (Supplementary
Figure S1 in [155]). In addition, we quantified transcript levels of MSH2 and found that
90-95% of the transcript knockdown was achieved at 48 and 72 hr time point post
transfection.
Cisplatin cytotoxicity in BER and MMR deficient cells
To assess the effect of BER and MMR knockdown on cell viability in response to
cisplatin treatment, we performed colony survival assays (Figure 6) (Figure 1 A in [155]).
Polβ null cells showed resistance to cisplatin as compared to WT cells. When MSH2 was
knocked down in WT MEFs, cisplatin treatment gave rise to low level of resistance as
compared to the WT cells. These results are consistent with the literature [23, 24].
However, when MSH2 was knocked down in BER deficient Polβ null cells, no
significant increase in the resistance was observed. We performed similar studies in WT
and Polβ KD MDA-MB-231 cells, and we did not see any additive increase in the level
of resistance in response to cisplatin (Figure 1B in [155]). From these results, we
conclude that BER and MMR might have an overlapping role in mediating cisplatin
sensitivity.
50
Rate of cisplatin intrastrand adduct repair does not affect cisplatin cytotoxicity in
BER and MMR deficient cells
Cisplatin resistance is developed when there is an increased repair of DNA
adducts. Increased repair of cisplatin adducts has been correlated with decreased cisplatin
efficacy. As described earlier, NER is the major pathway responsible for the repair of
cisplatin intrastrand adducts. In our recent publication, we have shown that BER proteins
are not involved in the intrastrand adduct repair [71]. MMR proteins have been shown to
bind to cisplatin DNA adducts [122]. Our goal was to study whether BER and MMR
status of the cells affects intrastrand adduct repair. We studied the repair kinetics of
platinum intrastrand adducts after downregulation of BER and MMR pathways separately
and together. The cells were transfected for the knockdown of MSH2 and were treated
with cisplatin for 2 hrs. The genomic DNA was harvested at the specified time points
using Qiagen DNeasy DNA isolation kit. The DNA was then coated onto the 96 well
plates using a DNA binding buffer. To assess the repair of cisplatin intrastrand adducts,
we performed an ELISA using a monoclonal antibody specific for 1,2-GG intrastrand
adduct which is the major adduct formed by cisplatin (ICR4 antibody, Dr. Michael Tilby,
University of Newcastle, UK). After using the appropriate HRP conjugated secondary
antibody, the HRP signal was converted to a colored product. The percentage of adducts
at 0 hr time point were normalized to 100%. The number of adducts increased
significantly at 12 hr time point. However, there was an efficient repair of these adducts
from 12 hr to 48 hr time point. Similarly, we also knocked down MSH2 in WT and Polβ
deficient MDA-MB-231 cells and studied intrastrand adduct repair. The results show that
loss of BER and/or MMR displayed no significant difference in the rate of adduct
51
removal when compared to the WT cells. These data indicate that BER and MMR are not
involved in the repair of cisplatin intrastrand adducts. Moreover, Intrastrand adduct repair
does not contribute to the increased resistance in BER and/or MMR deficient cells
(Figure 7A) (Supplementary Figure S3 in [155]).
Cisplatin interstrand crosslink (ICL) repair
As intrastrand adduct repair kinetics are unchanged in BER and MMR deficient cells, we
studied cisplatin ICL repair in these cells. We have utilized alkaline comet assay to
examine the repair kinetics of ICLs. These results show that cisplatin ICLs are repaired
efficiently from 0 to 72 hr time points. At 24 hr time point, there was no significant
difference between proficient and deficient cells. However, from 48 to 72 hr BER and/or
MMR deficient cells showed increased repair of ICLs. ICL repair in mammalian cells
requires the participation of both NER and HR pathways. But downregulation of BER
and/or MMR resulted in increased repair of the ICLs (Figure 7B) (Figure 2A in [155]).
Studies have shown that increased ICL repair is associated with increased resistance to
cisplatin [31]. These results demonstrate that the increased resistance that we see in the
BER and MMR deficient cells is likely due to faster repair of ICLs as compared to the
WT cells. Several studies have suggested MMR protein binding to ICLs which points to a
possible role of MMR proteins in cisplatin ICL processing. Deficiency of both BER and
MMR shows similar repair kinetics as compared to the cells that are deficient in BER and
MMR alone and no increased repair capacity was observed in the absence of both BER
and MMR. These results imply that BER and MMR play an overlapping role, and they
work together in the same mechanism to maintain cisplatin sensitivity.
52
Cisplatin ICL induced double strand break (DSB) repair
Cisplatin ICL processing is known to give rise to DSBs [156]. The repair of DNA
DSBs thus can be correlated with cisplatin ICL repair [157]. Histone H2AX is
phosphorylated in response to DNA DSBs and is a part of foci formed at sites of these
breaks due to colocalization of DNA damage sensor proteins [157]. The
dephosphorylation or disappearance of these foci indicates the repair of cisplatin ICL
induced DSBs. After transfection, the cells were grown on coverslips and were treated
with cisplatin. The immunofluorescence assay was performed at specified time points to
monitor the repair of DSBs. The cells were probed with anti γH2AX monoclonal
antibody for 1 hour. Fluorescent secondary antibody alexafluor-488 was used and the
coverslips were then treated with DAPI. At least 200 cells were counted and were plotted
as percent foci positive cells at each post treatment incubation time point. Our results
show that the number of foci positive cells peaks at 24 hr and they are efficiently repaired
from 48 to 72 hr time point. When BER and MMR were downregulated separately and
together, faster repair of cisplatin induced DSBs was observed as compared to the BER
and MMR proficient WT cells. No increased DSB repair capacity was seen when BER
and MMR were downregulated together. These results suggest that the increased
resistance observed in the absence of BER and MMR is due to the increased repair of
cisplatin ICLs and ICL induced DSBs (Figure 7C) (Supplementary Figure S4 in [155]).
MMR requires polymerase activity of Polβ to mediate cisplatin cytotoxicity
DNA Polβ is a 39 KDa polypeptide that contains a 31 KDa polymerase domain. Asp 256
is one of the three active site aspartates in the polymerase domain, which is critical for
53
nucleotidyltransferase mechanism. D256A mutation has been shown to completely
abolish polymerase activity of Polβ [158]. This mutant lacks the polymerase activity
while still retaining the dRP lyase activity. To understand the importance of polymerase
domain of Polβ, we re-expressed WT Polβ and a mutant deficient in polymerase activity
(D256A) in Polβ knockdown cells. We wanted to understand the involvement of MMR in
response to cisplatin in context of these cells. For this reason, we knocked down MSH2
using siRNA and we performed colony survival assays using increasing concentration of
cisplatin. The Polβ KD and MSH2 KD cells showed increased resistance to cisplatin. In
addition, downregulation of both Polβ and MSH2 did not give rise to any additional
increase in cisplatin resistance indicating that these proteins work in the same pathway to
mediate cisplatin cytotoxicity. The D256A cells showed cisplatin sensitive phenotype
indicating that the polymerase dead mutant can mediate cisplatin sensitivity but loss of
Polβ can give rise to cisplatin resistance. Interestingly, when we knocked down MSH2 in
D256A mutant cells we did not see any resistance phenotype, indicating that the
involvement of MMR is dependent on polymerase activity and is likely dependent on
misincorporation at cisplatin ICL sites (Figure 8) ( Figure 3C in [155]).
In vitro Binding of MMR proteins to cisplatin ICL DNA with a mismatch
Our data suggested that BER and MMR play an overlapping role in mediating
cisplatin sensitivity. In this mechanism, our data support the hypothesis that BER and
MMR proteins are required to maintain cisplatin sensitivity as they block the repair of
cisplatin ICLs. Our previous studies have shown that BER processing occurs at the site
near cisplatin ICLs. In addition, Polβ incorporates incorrect nucleotides in the flanking
region near cisplatin ICLs even in the presence of correct nucleotides [71]. We
54
hypothesized that mutagenic role played by Polβ is responsible for the activation of
MMR. Error prone DNA synthesis mediated by Polβ in the region flanking the ICL is
essential for generation of a mismatch and for subsequent MMR binding. We
hypothesized that this binding by BER and MMR blocks the productive repair of ICLs.
To test this hypothesis, we checked for binding of MMR proteins after BER processing
of ICL substrates. For this experiment, we generated undamaged uracil and cisplatin ICL
DNA substrates containing a uracil adjacent to the ICL. The substrates were processed in
a BER reaction containing UDG, APE1 and Polβ proteins in the presence of dNTPs. The
BER processing involves the removal of uracil by UDG followed by cleavage by APE1
and DNA synthesis by Polβ. We used these substrates in biotin-streptavidin pull-down
experiments using purified hMutSα protein (Figure 9). We observed increased
recruitment of MSH2 on the undamaged uracil substrate compared undamaged processed
substrate (Lane 3, G:U unprocessed). This is consistent with previous studies that showed
increased hMSH2-hMSH6 binding to a G:U mismatch [159]. BER processing at this site
can result in the incorporation of a correct nucleotide by Polβ. This results in a reduced
binding of hMSH2-hMSH6 at this site (Lane 4, G:U processed). In the case of ICL DNA
substrates, however, BER processing leads to increased retention of hMSH2-hMSH6 on
the ICL substrates (Lane 6, ICL G:U processed) as compared to unprocessed substrates
(lane 5, ICL G:U unprocessed). These results suggest that, at the sites near cisplatin ICL,
MMR binding occurs as a result of misincorporation of nucleotides by Polβ.
Our data suggest that MMR binding near cisplatin ICLs occurs downstream of BER
processing. The role of MMR in mediating cisplatin sensitivity is dependent on the
mutagenic processing by Polβ. BER processing near the cisplatin ICLs and subsequent
55
MMR binding blocks the repair of ICLs and ICL induced DNA double strand breaks, and
thus, BER and MMR play epistatic roles in mediating cisplatin sensitivity (Figure 10)
(Figure 5C in [155]).
Targeting XPF-ERCC1, ATM and ATR sensitizes BER and MMR deficient cells to
cisplatin
Our data shows that BER and MMR deficiency gives rise to cisplatin resistance.
Signaling kinases like ATR and ATM play an essential role in signaling DNA damage
followed by cisplatin treatment. ATR inhibition results in cisplatin sensitivity and this
inhibition was shown to be synthetic lethal with ATM/p53 signaling defect [160]. We
utilized an ATR inhibitor VE-821, to study its cytotoxic effect in BER proficient and
deficient human breast cancer cells and MEFs. We found that ATR inhibition sensitized
Polβ deficient cells to cisplatin, and they were as sensitive to cisplatin as the WT cells.
ATR inhibition also sensitized MEFs to cisplatin under both BER proficient and deficient
background (Figure 11A).
In addition, in response to DSBs, ATM detects DNA damage and causes
phosphorylation of downstream target effectors such as Chk2, p53, H2AX and 53BP1.
These effectors cause cell cycle arrest and facilitate the repair of DNA damage. ICL
resolution leads to the formation of DSBs and persistence of DSBs makes cells more
sensitive to platinum treatment. ATM plays a key role in the repair of these breaks, so we
asked whether inhibiting ATM kinase function will result in sensitization to cisplatin
chemotherapy.
For this purpose, we used ATM kinase inhibitor KU60019 at IC20
concentration. We pretreated cells with this inhibitor before cisplatin treatment and kept
56
the cells in inhibitor containing media for a total of 24 hrs. Our colony survival assays
show that, ATM inhibition sensitizes human breast cancer cells to cisplatin under both
BER proficient and deficient background (Figure 11B). These results show that ATM and
ATR inhibitors can be utilized for the sensitization of BER and possibly MMR deficient
cells to cisplatin. However, when we pretreated the cells with DNA-PK inhibitor
NU7441, we did not see any sensitive phenotype in response to cisplatin (Figure 11C).
XPF-ERCC1 is a structure specific endonuclease which plays a key role in the
repair of cisplatin adducts. It is required as a part of the NER pathway for the removal of
cisplatin intrastrand adducts. Studies show that XPF-ERCC1 also plays an essential role
in providing substrates for homologous recombination and cisplatin ICL repair. We have
previously shown that XPF-ERCC1 knockdown results in an increased sensitive
phenotype in BER deficient human cancer cells [71]. For this reason we wanted to study
whether downregulation of XPF-ERCC1 would also sensitize MMR deficient cells to
cisplatin. Inhibition of XPF-ERCC1 in HEC59 cancer cells lines sensitized the cells to
cisplatin, indicating that XPF-ERCC1 inhibitors could potentially be used in cisplatin
combination therapy for BER and MMR deficient or mutated cancers (Figure 12).
57
2.4 Discussion
Cisplatin and related platinum compounds are widely used anticancer agents for
the treatment of solid tumors. The therapeutic effect is mediated by intrastrand adducts
and ICL DNA leading to stalled replication forks which result in cell cycle arrest and
apoptosis. Despite initial favorable response, platinum therapy often results in
development of resistance. This results in poor prognosis and recurrence of the disease.
Therefore, it becomes crucial to understand these mechanisms of cisplatin resistance.
Several studies have indicated that the rate of DNA repair is one of the major
determinants of cisplatin resistance and increased repair of cisplatin adducts leads to
enhanced resistance to the drug. Deficiency or loss of MMR proteins gives rise to
resistance to cisplatin. Cisplatin treatment results in the selection of resistant cells that are
MMR deficient. We have also shown that BER deficiency also contributes to a cisplatin
resistant phenotype. The mutagenic role played by Polβ in bypassing cisplatin adducts
suggests a role for the subsequent activation of the MMR pathway. For this reason, we
studied the effect of downregulation of BER and MMR pathways on cisplatin
cytotoxicity and their potential involvement in a single mechanistic pathway to mediate
cisplatin cytotoxicity.
Colony survival assays showed that loss of BER and MMR resulted in ~2 fold
resistance to cisplatin which is consistent with the literature. Interestingly, when both
BER and MMR pathways were downregulated together, no additional resistance to
cisplatin was seen. Therefore we hypothesized that BER and MMR pathways play an
overlapping role and function together in the same mechanism. This is the first study that
58
shows a possible epistatic role of two DNA repair pathways to mediate cisplatin
cytotoxicity in mammalian cells.
Numerous studies have shown that MMR proteins bind to cisplatin intrastrand
adducts. However, when we checked for the repair of cisplatin GG intrastrand adducts in
BER and MMR deficient cells, we did not see any significant difference in the rate of
repair in these cells. These data revealed that BER and MMR pathways are not involved
in the repair of cisplatin intrastrand adducts. Moreover, these studies also demonstrated
that the increased resistance seen in the BER and MMR deficient cells is not due to any
change in the rate of repair of intrastrand adducts.
In addition, BER and MMR deficiency results in the increased repair of cisplatin
induced ICLs as compared to the WT cells. We show that increased repair of cisplatin
ICLs contributes to increased resistance to cisplatin. MSH2 has been previously shown to
be involved in the recognition and processing of psoralen ICLs and loss of MSH2 results
in hypersensitivity to psoralen. However, in our study, MSH2 deficiency resulted in
increased repair of ICLs indicating that ICLs produced by different crosslinking agents
are processed differently. In the case of cisplatin, MMR deficiency gives rise to
resistance due to increased repair of ICLs and ICL induced DSBs.
We have previously shown that the BER pathway plays a novel role in processing
a uracil adjacent to the cisplatin ICL. This uracil is likely produced as result of the
deamination of extrahelical cytosines induced by the distorted structure of cisplatin ICLs.
Uracil acts as a substrate for the BER pathway and if BER processing leads to nucleotide
misincorporation and generation of a mismatch, this can act as a nucleation point for the
59
MMR pathway. Using ICL DNA substrates containing uracil, we studied the binding of
MutSα after BER processing. We observed increased retention of MSH2 on the BER
processed ICL substrates indicating that binding of MMR proteins occurs downstream of
BER processing. Furthermore, the mutagenic role of Polβ is essential for activation and
recruitment of MMR proteins. This is consistent with our data using polymerase dead
D256A mutant cell lines in which MMR knockdown results in no cisplatin resistance
[155]. We believe that BER and MMR binding near the cisplatin ICL sites interferes with
the actual processing of the cisplatin ICLs and therefore, lack of these pathways gives
rise to increased repair of ICLs and increased cellular resistance to the drug. This is a first
report that shows that MMR processing in the regions near cisplatin ICLs, affects the
repair of ICLs and influences cisplatin cytotoxicity. This study highlights a novel
mechanism where MMR binding occurs downstream of BER processing and these two
pathways play an epistatic role to mediate cisplatin cytotoxicity (Model Figure 10)
(Figure 6 [155]).
Our results show that BER and MMR deficiency in tumor cells involves increased
repair of cisplatin ICLs. Therefore, we asked whether inhibiting the proteins involved in
these DNA repair pathways would sensitize BER and MMR deficient cells to cisplatin.
We targeted ATR and ATM signaling kinases in BER proficient and deficient cells
human cancer cells. We found that ATR and ATM inhibition lead to sensitization of
these cells to cisplatin making them an attractive target for cisplatin combination therapy
in cisplatin resistant cells that are deficient in BER or MMR. DNA-PK also plays a key
role in non-homologous end joining pathways of DNA DSB repair. The role of DNA-PK
in processing of cisplatin mediated DNA damage is not clearly understood. Cisplatin
60
adducts themselves have been postulated to inhibit DNA-PK activity [161]. In one study,
shRNA mediated downregulation of DNA-PK activity lead to the sensitization of cancer
cells to cisplatin [162]. On the other hand, it has also been reported that cells deficient in
DNA-PK activity are markedly resistant as compared to their WT counterpart [163].
Moreover, reports have shown that DNA-PK inhibition sensitized breast cancer cells to
ionizing radiation and doxorubicin [164]. However, when we inhibited DNA-PK using
NU7441, we observed increased resistance to cisplatin as compared to the untreated cells.
However, the mechanism of cisplatin resistance in DNA-PK deficient cells remains to be
clearly established.
61
Figures
A
B
Transcript knock-down in MEFs
C
% Transcript knock-down
120
100
WT
Pol b -/-
48 hr
72 hr
80
60
40
20
0
Figure 5, Supplementary Figure S1 in [155]: siRNA mediated downregulation of
MSH2 in wildtype (A) and Polβ null (B) MEFs. Cells were transiently transfected with
siRNA directed against MSH2 as described in Materials and Methods. Proteins were
extracted at indicated time points and probed with MSH2 antibody with α-tubulin as a
loading control. Control cells were treated with nontargeting siRNA (Ctl si).
(C) Percent transcript knockdown in wildtype and Polβ-/- MEFs. Cells were transiently
transfected with siRNA directed against MSH2. At 48 and 72 hr post transfection, total
RNA was extracted from cells and the transcript levels were quantified as described in
Materials and Methods with β-actin as an endogenous control. The percent transcript
knockdown was determined from 2^-ΔΔCT values with non-targeting siRNA as controls.
62
Figure 6
% Colony Survival
A
Colony Survival assay
120
Wt + siCtrl
Wt + siMSH2
Polb -/- + siCtrl
Polb -/- + siMSH2
100
80
60
40
20
0
0
2
4
6
8
10
Cisplatin (µM)
Figure 6, Figure 1A in [155]: Cisplatin cytotoxicity - Colony survival assays in wild
type and Polβ null MEFs. Control and MSH2 siRNA transfected cells were treated with
increasing doses of cisplatin and cytotoxicity was determined by clonogenic assays.
Results are represented as mean ± SD from 3 independent experiments.
63
Figure 7
Intrastrand adduct repair
A
% of intrastrand adducts
160
140
NS NS
Wt + siCtrl
Wt + siMSH2
Polb -/- + siCtrl
Polb -/- + siMSH2
NS
120
100
NS
80
NS NS
60
40
NS NS NS
20
0
0 hr
12 hr
24 hr
48 hr
Post-incubation time
ICL Repair
B
Wt + siCtrl
Wt + siMSH2
Polb -/- + siCtrl
Polb -/- + siMSH2
% of ICLs
100
80
NSNS NS
60
40
20
*
* *
* * *
0
0 hr
24 hr
48 hr
72 hr
Post-incubation time
DSB repair
C
% g H2AX foci positive cells
80
NS
NS
Wt + siCtrl
Wt + siMSH2
Polb -/- + siCtrl
Polb -/- + siMSH2
NS
60
40
* * *
20
* *
*
0
0 hr
24 hr
48 hr
72 hr
Post-incubation time
64
Figure 7A, Supplementary Figure S3: Repair of cisplatin intrastrand adducts in
MEFs. Cells were treated with cisplatin and ELISAs were performed as described in
Materials and Methods at different time intervals (0, 12, 24, 48 hr). The percentage of
intrastrand adducts present at 0 hr were normalized to 100%. Results are represented as
mean ± SD of three independent experiments. Statistical analysis was performed by
student’s t test and comparisons are made between wildtype and proficient cells vs
deficient cells. NS – non-significant; * - P< 0.05.
Figure 7B, Figure 2A in [155]: Repair of cisplatin ICLs in MEFs. Cells were treated
with cisplatin and comet assays were performed as described at different time intervals
(0, 24, 48 and 72 hr). The percentage of interstrand crosslinks present at each time point
was calculated using olive tail moments. Results are represented as mean ± SD of three
independent experiments. Statistical analysis was performed by student’s t test and
comparisons are made between wildtype and proficient cells vs deficient cells. NS – nonsignificant; * - P< 0.05.
Figure 7C, Supplementary Figure S4 in [155]: Repair of DSBs in MEFs. Cells were
treated with cisplatin and immunofluorescence assays were performed at different time
intervals (0, 24, 48 and 72 hr). The percentage of foci positive cells were counted for
each timepoint. Results are represented as mean ± SD of three independent experiments.
Statistical analysis was performed by student’s t test and comparisons are made between
wildtype and proficient cells vs deficient cells. NS – non-significant; * - P< 0.05.
65
Figure 8
Colony Survival assay
120
Pol b KD + siCtrl
KD/Wt Polb + siCtrl
KD/D256A + siCtrl
Polb KD + siMSH2
KD/Wt Polb + siMSH2
KD/D256A + siMSH2
% Colony Survival
100
80
60
40
20
0
0
2
4
6
8
10
Cisplatin (µM)
Figure contributed by Dr. Anbarasi Kothandapani
Figure 8, Figure 3C in [155] Cisplatin cytotoxicity, Cells were transfected with siRNA
for MSH2, treated with cisplatin and colony survival assays were performed. Results are
represented as mean SD from three independent experiments.
66
Figure 9
ICL G
/U pr
o
ces se
d
proce
s sed
ICL G
/U Un
o ces s
ed
G/ U p
r
G/U
U npro
ces sed
aged
Un da
m
B ead
s al on
e
BER Processed Substrates
MSH2
1
2
3
4
5
6
Figure 9, Figure 6C in [155]: Interaction of MSH2 with cisplatin ICL DNA
processed with BER proteins. Undamaged and single cisplatin ICL biotinylated DNA
duplexes incubated with BER proteins UDG, Ape1, and Polβ were bound to streptavidin
magnetic beads and were incubated with equal amounts of overexpressed hMSH2hMSH6 insect cell extract. The proteins bound to the DNA beads were eluted and
immunoblotted with the antibody against MSH2.
67
Figure 10
Pol b misincorporation (1-2 extra C’s)
AB
Mutation Frequency?
5’
G CC
3’
CG
3’
A
5’
BER Processing & removal of AB site
MMR Deficiency
Role for MMR Mutagenic at ICL sites?
5’
G CC
3’
CG
3’
A
5’
MMR protein binding
G CC
5’
3’
C
GA
3’
Msh2
5’
Msh6
MutSa?
MutSb?
MMR Processing
MMR Processing Mutagenic?
Correct Repair
5’
GCT
3’
GA
C
3’
5’
5’
GCC
3’
3’
GG
5’
C
Persistent ICLs
Cisplatin Sensitivity
Figure 10: BER and MMR play an epistatic role in mediating cisplatin cytotoxicity.
BER processes the DNA flanking the ICL and Polβ produces a mutagenic spectrum at the
incision site. Incorporation of incorrect nucleotides would generate mismatched bases
and act as a nucleation point for MMR protein binding and further processing. Both BER
and MMR processing would result in non-productive repair of cisplatin ICLs. As a result
of blocking productive ICL repair pathways, there are persistent cisplatin ICLs which
enhance cisplatin cytotoxicity.
68
A
120
WT
WT + ATRi
Polb KD
Polb KD + ATRi
% Colony survival
100
80
60
40
20
0
0
1
2
3
4
Cisplatin (µM)
B
120
WT
WT + ATMi
Polb KD
Polb KD + ATMi
% Colony survival
100
80
60
40
20
0
0
1
2
3
4
Cisplatin (µM)
C
120
WT
WT + DNA-PKCi
Pol b KD
Pol b KD + DNA-PKCi
% Colony survival
100
80
60
40
20
0
0
2
4
6
8
10
Cisplatin (µM)
69
Figure 11: Targeting signaling kinases in BER proficient and deficient cells. The cells
were pretreated with IC20 concentration of VE821 (A), KU60019 (B) and NU7441 (C),
followed by cisplatin treatment for 2 hrs. The inhibitors were kept in the media for a total
of 24 hrs. The colony survival assays were performed with increasing concentrations of
cisplatin. Results are represented as mean ± SD from 3 independent experiments.
70
Figure 12
Colony Survival assay
120
Hec59+2 - siControl
Hec59+2 - siERCC1+siXPF
Hec59 - siControl
Hec59 - siERCC1+siXPF
% Colony Survival
100
80
60
40
20
0
0
5
10
15
20
25
Cisplatin (µM)
Figure contributed by Dr. Anbarasi Kothandapani
Figure 12: Downregulating XPF-ERCC1 in MMR proficient and deficient cells
siRNA mediated downregulation of XPF-ERCC1 was carried out in HEC59 and HEC59
+2 (A) cells. The colony survival assays were performed using increasing doses of
cisplatin. Results are represented as a mean SD of 3 independent experiments.
71
Chapter 3
ROLE OF DOWNSTREAM MISMATCH REPAIR PROTEINS IN THE
PROCESSING OF CISPLATIN INTERSTRAND CROSS-LINKS
Akshada Sawant1, Anbarasi Kothandapani1, Robert W. Sobol2, 3, 4, Anatoly
Zhitkovich5 and Steve M. Patrick6
1. Department of Biochemistry and Cancer Biology, University of Toledo – Health Science
Campus, Toledo, OH 43614
2. Department of Pharmacology & Chemical Biology, University of Pittsburgh School of
Medicine, Pittsburgh, PA 15213
3. University of Pittsburgh Cancer Institute, Hillman Cancer Center, Pittsburgh, PA 15213,
USA
4. Department of Human Genetics, University of Pittsburgh Graduate School of Public Health,
Pittsburgh, PA 15213
5. Department of Pathology and Laboratory Medicine, Brown Alpert Medical School, Brown
University, Providence, RI 02912
6. Department of Oncology, Karmanos cancer institute, Wayne State University,
Detroit, MI 48201
Running title: Differential role of MMR proteins in modulating cisplatin cytotoxicity
Keywords: BER, cisplatin, MMR, resistance, ATPase
72
3.1 Introduction
The DNA mismatch repair system, which is involved in the post replicative repair of
mismatches plays a crucial role in the maintenance of genomic stability [165]. In addition
to the recognition of mismatches, MMR proteins have also been involved in the
recognition and processing of DNA damage inflicted by a number of chemotherapeutic
agents, including cisplatin, carboplatin, alkylating agents and 5-Fluorouracil [86, 125,
166-168]. The MMR pathway is composed of recognition proteins with MSH2 as a
common partner in two heterocomplexes, namely MutSα (MSH2-MSH6) and MutSβ
(MSH2-MSH3) [169]. MutSα is required for the repair of base-base mismatches and one
base pair insertion deletion loops (IDLs). On the other hand, MutSβ carries out the repair
of IDLs with single or multiple base pairs. The mismatch recognition step is followed by
the recruitment of downstream MMR proteins MutLα (MLH1-PMS2), Exonuclease I
(EXO1), DNA Polδ and DNA ligase.
MMR has also been shown to participate in the DNA damage response after
treatment with certain chemotherapeutic agents. Loss of MMR proteins has been
associated with resistance to a number of anti-cancer agents (e.g. alkylating agents and
cisplatin) [86, 168]. Various models have been proposed for the possible role of MMR in
maintaining drug sensitivity. The adducts formed by alkylating agents can result in the
generation of mismatches. It has been suggested that MMR proteins take part in futile
cycles of repair of these mismatches in the daughter strand. It is believed that, the
resulting strand breaks signal apoptosis and loss of this function gives rise to drug
resistance. In addition, MMR proteins have been shown to bind to certain types of DNA
damage and activate the DNA damage response, eventually resulting in cell death.
73
However, these studies did not differentiate between different types of adducts that are
formed by cisplatin, namely intrastrand adducts, which are formed within the same DNA
strand, versus ICLs, which are formed between adjoining strands of DNA. MutSα
proteins have been shown to recognize cisplatin intrastrand adducts [122]. In addition,
MutSβ was found to be one of the proteins that interact with cisplatin ICLs [82].
However, the exact role of these proteins in the processing of cisplatin adducts has not
been clearly evaluated.
Recent studies have shown that MSH3 is required for the repair of DNA DSBs
induced by cisplatin and oxaliplatin treatment [170, 171]. Thus, the MMR pathway has
been shown to be required for the sensitization of colorectal cancer cells to cisplatin and
oxaliplatin, and this effect is believed to be independent of canonical MMR processing.
However, other studies have shown that MSH3 proficient cells, which were more
resistant to chemotherapy, expressed higher levels of NER proteins which could explain
the reason for increased resistance [172]. Thus, the exact role of MSH3 in modulating
platinum cytotoxicity remains to be clearly determined.
In our previous studies, we have shown that loss of base excision repair (BER)
and MMR proteins gives rise to resistance to cisplatin and these two pathways take part
in the same mechanism to mediate cisplatin sensitivity (Chapter 2) [71, 155]. In the
absence of these proteins, increased repair of cisplatin ICLs was observed which leads to
decreased cellular cytotoxicity and cisplatin resistance. We also showed that this
mechanism is dependent upon the low fidelity of Polβ, which leads to misincorporation
of bases and generation of base mismatches. These mismatches, in turn, activate the
mismatch repair pathway. In this report, we distinguish between the requirements of
74
different downstream MMR proteins to mediate this effect. We show that, in contrast to
previous studies, there is a clear distinction between MMR recognition heterocomplexes
where MutSα is required to maintain cisplatin sensitivity, and MutSβ seems to play no
role at least in breast cancer cell lines and mouse embryonic fibroblasts. Moreover, we
show that the ATPase activity of MLH1 is required for maintaining a cisplatin sensitive
phenotype, highlighting the importance of the mismatch repair activity in the processing
of cisplatin ICLs.
3.2 Materials and Methods
Chemicals and Antibodies
Cisplatin, oxaliplatin and myricetin were purchased from Sigma-Aldrich. All other
chemicals and reagents were from standard suppliers. Antibodies directed against MSH3,
MSH6 and MLH1 were from BD Pharmigen and α-tubulin was from Sigma-Aldrich. For
the stock preparation, cisplatin and oxaliplatin were diluted in 1X PBS and vortexed
vigorously until the drug dissolved completely. The stock concentration was 1 mM.
Cisplatin was prepared fresh before each experiment. Stock for oxaliplatin was stored at 80 ºC up to 6 months and thawed at room temperature (RT) when needed.
Cell lines
The human breast adenocarcinoma MDA-MB-231 cells were grown in RPMI 1640
containing 10% FBS and antibiotics. MDA-MB-231 Polβ knockdown cells (Polβ
lentiviral shRNA) were grown in the presence of 0.5μg/mL puromycin. The development
and characterization of the MDAMB-231/Polβ-KD cells were described previously [152].
75
Wild-type (92TAg) and Polβ null (88TAg) primary mouse embryonic fibroblasts were
cultured in high glucose DMEM supplemented with 10% FBS plus antibiotics at 37ºC in
a humidified atmosphere under 10% CO2. The HCT116 cells were grown in DMEM F-12
media with 10% FBS, antibiotics and 600 μg/ml of geneticin. The human colon cancer
HCT116 cell lines were characterized by Dr. Anatoly Zhitkovich, Brown University.
These HCT116 cells are deficient in MLH1. These cells were reconstituted with WT
MLH1, S44L ATPase mutant MLH1 and S44F ATPase mutant MLH1. The MLH1
deficient human colon cancer cells HCT116, chromosome 3 complemented sub-line
HCT116+3 were obtained from Dr. Kandace Williams, The University of Toledo. The
chromosome complemented sublines were maintained in 400μg/ml of geneticin.
shRNA transfection
Mission shRNA plasmid bacterial stocks directed against human MSH6 and MSH3 were
obtained from Sigma Aldrich.
The plasmid DNA was purified using a plasmid
purification maxi prep kit from Qiagen. Lentiviral particles were packaged using 293FT
cells with the help of 3rd generation packaging plasmids PMD2G, PMDLG/RRE and
PREV/RRE. Lipofectamine 2000 reagent (Invitrogen) was used for the transfection of the
plasmid DNA. The media was changed after 24 hrs of transfection. The viral particles
were harvested 48 hr and 72 hr after the transfection by centrifugation followed by
filtration through 0.2 micron filters. The viral stocks were stored as aliquots at -80ºC for
future use. At the time of the experiment, the viral stocks were used along with polybrene
(Sigma Aldrich) for the knock down of proteins of interest. Cells were harvested at the 72
hr timepoint post transduction to check for protein and transcript knockdown.
76
siRNA transfection
ON-TARGET plus SMART pool siRNAs specific for human MSH3 and MLH1 were
purchased from Dharmacon RNAi technologies, Thermo Scientific. The non-targeting
control siRNA was used as a control for non-specific effects. siRNA transfection was
carried out as per the manufacturer’s protocol. Briefly, the cells were plated in 6 well
plates in the antibiotic free media. At the time of transfection, the cell density was
maintained at 60-70% and two transfections were done with an interval of 24 hrs.
Dharmafect transfection reagent 1 and 4 were used for MEFs and MDA-MB-231 cells,
respectively. The cells were harvested at 48 and 72 hr timepoints after transfection for the
detection of protein and transcript knockdown.
Western blot analysis
Cells were harvested at 96, 120, 144 hrs after the infection, washed with PBS and lysed
in lysis buffer (10 mM Tris pH 8.0, 120 mM NaCl, 0.5% NP-40, 1 mM EDTA)
containing protease inhibitors (0.5 M phenyl methyl sulphonyl fluoride PMSF, 1 mg/ml
Leupeptin and 1 mg/ml pepstatin A). The proteins were separated on 8% SDSpolyacrylamide gels and transferred onto Immobilon-P transfer membranes (Millipore).
After blocking (2% non-fat dry milk), the membranes were probed with primary
antibodies recognizing human MLH1, with α-tubulin as loading control. The membranes
were incubated with appropriate secondary antibodies and the signal was detected by
using Enhanced chemiluminescence detection system.
77
Real time PCR for the measurement of transcript levels
At indicated post-transfection time points, cells were harvested and pelleted. RNA was
isolated using TRIzol reagent (Invitrogen) by standard procedures. The total RNA was
reverse transcribed using MMLV reverse transcriptase enzyme (Invitrogen) as per the
manufacturer’s protocol. The transcript levels were quantified using iQ SYBR green
supermix (Bio-Rad) in iCycler iQ System, with GAPDH as an endogenous control. The
percent transcript knockdown was determined from 2^-∆∆CT values.
Colony survival assay
Cells (~500) were treated with increasing concentrations of cisplatin for 2 hr. After
treatment, fresh medium was added, and the cells were allowed to grow for 7-14 days.
Colonies were fixed with 95% methanol and stained with 0.2 % crystal violet. Colonies
with ≥50 cells were counted, and colony survival was expressed as the ratio of the
average number of colonies in drug treated cells versus control cells x 100. The
experiment was done in triplicates for each drug concentration. Statistical analyses were
performed using student’s t test comparing wild type and knock down cells.
Alkaline comet assay
Alkaline comet assay was used to analyze the repair of cisplatin interstrand cross-links
(ICLs) as described [153, 154]. Cell suspensions (~10,000 cells) were embedded on a
microscopic slide, lysed and incubated in ice-cold alkaline solution for 20 min to allow
the DNA to unwind. Electrophoresis was carried out for 30 min at 28 V, 300 mA. Slides
were neutralized and stained with SYBR green (Invitrogen). The comets were scored
78
using a Nikon epifluorescence microscope. At least fifty cells were analyzed per slide
using Komet Assay Software 5.5F (Kinetic Imaging, Liverpool, UK). The data was
expressed as the percentage of crosslinks that remained at that particular time point
normalized to 100% at 0 hr.
Immunofluorescence
Double-strand break (DSB) repair was assessed by monitoring the nuclear γ-H2AX foci
by immunofluorescence. Cells were fixed using 4% paraformaldehyde, permeabilized
and probed with monoclonal anti γ-H2AX antibody (1:500, Millipore). The images were
visualized using a Nikon Eclipse T2000-U microscope. Foci were counted in at least 200
cells at each time point per condition in each cell line and results are expressed as % γH2AX foci positive nuclei.
MTS assays
The CellTiter 96 Aqueous Non-Radioactive Cell Proliferation Assay (Promega) was used
to evaluate cisplatin or oxaliplatin cytotoxicity. Cells were treated with increasing
concentration of cisplatin for 2 hr. After 72 hr, 20 μL of combined MTS/PMS solution
[(3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2Htetrazolium and phenazine methosulfate] was added to each well and incubated for 2-4 hr
at 37°C. The absorbance was measured at 490 nm in Spectramax M5 plate reader
(Molecular Devices). The percentage of cell survival was calculated from the average OD
of treated cells/average OD of control cells x 100. Statistical analyses were performed
using student’s t test comparing wild type and knock-down/ mutant cells.
79
Drug treatment
The cells were treated with myricetin alone to assess its effect on cell viability (Data not
shown). From these curves, a concentration of 10 µM (IC20) was chosen, so as to have
minimum influence on cell viability with Myricetin alone. The cells were pretreated with
this concentration for 2 hrs followed by treatment with increasing concentration of
cisplatin for 2 additional hrs. MTS assay was then performed as described.
3.3 Results
Requirement of different MMR recognition heterocomplexes to maintain cisplatin
sensitivity
To differentiate between the role of MutSα and MutSβ in mediating cisplatin
sensitivity, colony survival assays were performed using wild type and Pol β deficient
MDA-MB-231 cells. We knocked down MSH3 or MSH6 using shRNA lentivirus
directed against these proteins. The knockdown efficiency was analyzed at both the
protein and transcript level using western blot analysis and real time PCR, respectively
(Figure 13 A-F). The level of knockdown was found to be 80 – 90 % compared to the
controls. Using clonogenic assays, we found that MSH6 deficient cells were resistant to
cisplatin as compared to the wild type cells (Figure 14 A). Loss of Polβ also gave rise to
cisplatin resistance which is consistent with our previous results (Chapter 2) [71, 155].
However, down regulation of MSH6 in Pol β deficient cells did not give rise to any
additive increase in the degree of resistance reinforcing the epistatic role of MMR and
BER proteins in mediating cisplatin cytotoxicity. In addition, we also performed
clonogenic assays using MSH3 knockdown cells, and we found that down regulation of
80
MSH3 did not have any effect on cisplatin cytotoxicity. Similar to the wild type cells,
MSH3 deficient cells maintained a cisplatin sensitive phenotype (Figure 14 B).
Comparable results were observed in wild type and Polβ null mouse embryonic
fibroblasts (Supplementary Figure S1). These data indicate that MSH6 is required for
mediating cisplatin sensitivity, whereas, MSH3 does not influence cisplatin cytotoxicity
in the cell lines used in this study.
Previously, we have shown that BER and MMR processing of the ICLs is
dependent upon the unique structure produced by the cisplatin ICL. This distorted
extrahelical structure is not produced to the same extent by any other platinating agents.
For this reason, we studied cell viability in response to oxaliplatin in MSH3 and MSH6
knockdown cells using MTS assays (Supplementary Figure 2A, 2B). Cell survival studies
showed no statistical difference in the oxaliplatin cytotoxicity highlighting the fact that
the role of MMR proteins in modulating chemosensitivity is specific to cisplatin. In
support of these studies, Zdraveski et al have shown that the oxaliplatin adducts are not
recognized by MMR proteins [173].
Differential role of MMR recognition proteins in cisplatin ICL repair
Enhanced DNA repair has been shown to be a major mechanism in the
development of cisplatin resistance [174]. In previous studies, MMR proteins have been
shown to bind to the cisplatin GG adducts [122, 142]. In our previous studies, we have
shown that downregulation of MSH2, which is a common partner of MutSα and MutSβ
heterocomplexes, does not influence the rate of repair of cisplatin intrastrand adducts
(Chapter 2) [155]. Based on our previous observations, we did not expect MSH3 and
81
MSH6 to be involved in the repair of cisplatin intrastrand DNA adducts. As the repair of
intrastrand adducts was expected to be unaffected, we hypothesized that the rate of repair
of cisplatin ICLs would influence cisplatin cytotoxicity. For this reason, we studied
cisplatin ICL DNA repair using a modified alkaline comet assay in MSH3 or MSH6
knock down cells. Using this assay, we observed increased repair of cisplatin ICLs over
the time course of 0- 72 hr in all cell types with no difference in the repair rate up to the
24 hr time point. At the 48 hr and 72 hr time point, we observed increased repair of
cisplatin ICLs in Polβ deficient cells consistent with our previous studies [71, 155].
However, MSH6 knockdown in both WT and Polβ deficient cells showed increased
repair of cisplatin ICLs as compared to WT cells suggesting the involvement of MSH6 in
cisplatin ICL processing (Figure 15 A). Moreover, we did not see any additive increase in
the repair capacity in the absence of both MSH6 and Polβ compared to the repair capacity
in the cells lacking either MSH6 or Polβ alone. In addition, we also performed similar
assays using MSH3 knockdown cells. In contrast to MSH6 knockdown cells, we did not
see any change in the rate of repair of cisplatin ICLs in these cells (Figure 15 B). It
should be noted that the modified alkaline comet assay used in this report is sensitive to
the initial unhooking of the ICL from the double-stranded DNA and therefore, this assay
can represent the involvement of MMR proteins in the initial stages of the ICL repair.
Thus, these data suggested that MSH6 interferes with or inhibits the repair of cisplatin
ICLs whereas MSH3 does not participate in the cisplatin ICL processing which leads to a
similar rate of repair of ICLs as observed in WT cells. Moreover, increased repair rates
have been shown to be one of the mechanisms of cisplatin resistance and poor prognosis,
82
and these data correlate well with the increased cisplatin resistance observed after down
regulation of MSH6 (Figure 14A).
Cisplatin ICL processing leads to the generation of DNA DSBs. The ATM kinase
recognizes the DSBs and phosphorylates a histone variant H2AX at serine 139. This
causes recruitment of downstream DSB repair proteins. Phosphorylated γH2AX proteins
form distinct nuclear foci and can be studied using an immunofluorescence assay. In this
assay, we observed induction of γH2AX foci after cisplatin treatment followed by a
decrease in γH2AX positive cells from 24 hr to 72 hr time points indicating DNA repair.
The Polβ deficient cells showed a decreased percentage of γH2AX foci positive cells at
48 hr and 72 hr as compared to WT cells consistent with our previous report [71]. MSH6
knockdown cells also showed decreased percentage of γH2AX foci positive cells and
therefore, increased repair of cisplatin ICL induced DSBs as compared to the WT cells
(Figure 16 A). These data validate the involvement of MSH6 in the processing of
cisplatin ICLs. However, in contrast to some of the other studies done in colorectal
cancer cells lines [170, 171], MSH3 knockdown cells did not show any change in the rate
of repair DNA DSBs as compared to WT breast adenocarcinoma cell lines used in this
study (Figure 16 B). These results suggest that loss of MSH3 does not affect the repair of
cisplatin ICLs and ICL induced DSBs.
MutL homolog 1 (MLH1) plays a key role in maintaining cisplatin sensitivity
MLH1 is the human homologue of the E. coli MMR gene MutL. The MMR
pathway involves recognition of a base mismatch or insertion/deletion loop by a MutS
homolog followed by recruitment of a MutLα heterodimeric complex consisting of
83
MLH1 and PMS2 [169]. To understand the importance of MLH1 in mediating cisplatin
cytotoxicity, we knocked down MLH1 using siRNA in MDA-MB-231 breast cancer cells
and MEF cell lines. The knock down efficiency at both the transcript and protein level
were found to be 80-90% (Figure 17 A, B, C). Colony survival assay was performed to
address the effect of MLH1 downregulation on cell viability in response to cisplatin
(Figure 18). MLH1 knock down in WT cells showed ~2 fold resistance to cisplatin as
compared to the control cells. However, MLH1 knock down in Polβ deficient cells did
not give rise to any additional increase in cisplatin resistance indicating an overlapping
role of these two distinct DNA repair proteins in the same mechanism to mediate
cisplatin sensitivity. To understand the mechanism of resistance, we checked for the
effect of MLH1 knockdown on the rate of cisplatin adduct repair. As mentioned earlier,
increased rate of repair of intrastrand adducts and/or ICLs is concomitant with increased
drug resistance. We performed enzyme linked immunosorbent assays using a monoclonal
antibody specific for the cisplatin GG adduct, which is a major intrastrand adduct formed
by cisplatin. Knockdown of BER and MMR showed no difference in the repair of
intrastrand adducts indicating that these pathways do not influence the repair rate of
cisplatin GG adducts (Figure 19A). As cisplatin intrastrand adduct repair was unaffected,
we checked for the repair of cisplatin interstrand crosslinks. Modified alkaline comet
assays were used to evaluate the rate of repair of ICLs over a period of 0-72 hr time
points after cisplatin treatment. Downregulation of Polβ showed a decreased percentage
of ICLs at 48 hr and 72 hr time points. MLH1 has shown to be required for signaling
DNA damage and activating the DNA damage response pathway in reaction to psoralen
crosslinks [175]. In our studies in response to cisplatin, depletion of MLH1 in the WT as
84
well as Polβ deficient cells resulted in increased repair of ICLs (Figure 19 B). In addition,
similar results were observed in an immunofluorescence assay where MLH1 KD cells
showed increased repair of ICL induced DSBs as compared to the WT cells (Figure 19
C). These results indicate that increased repair of ICLs accounts for the cisplatin resistant
phenotype seen in these cells. Furthermore, we observed similar levels of cisplatin ICL
repair in the absence of both BER and MMR pathways as compared to the knockdown of
MMR and BER alone. These data indicate an epistatic relationship of these two pathways
in mediating cisplatin sensitivity.
ATPase activity of MLH1 is essential for cellular sensitivity to cisplatin.
We have shown that BER and MMR play an epistatic role in mediating cisplatin
sensitivity (Chapter 2) [155]. This overlapping role is dependent upon the error prone
nature of Pol β which generates a mismatch while processing cisplatin ICL flanking
DNA, thereby leading to the activation of the MMR pathway. Our results indicate that the
complete knockdown of MLH1 gives rise to a cisplatin resistant phenotype, suggesting
that a functional downstream MMR pathway is essential to maintain cellular sensitivity to
cisplatin. Next, to confirm the involvement of MMR and the actual processing of the
mismatch, we utilized human colon cancer cell lines that are deficient in the MLH1
ATPase activity (Figure 20 A). The HCT116 cells are deficient in MLH1. These cells
were reconstituted with plasmid vectors expressing WT MLH1, S44L and S44F ATPase
mutant MLH1. Owing to a point mutation that affects the serine residue at position 44 in
the ATPase domain, these cells are deficient in mismatch repair activity. We utilized
these cell lines to differentiate between the effects of possible binding of MLH1 to the
mismatch as opposed to the actual MMR processing of the mismatch by MLH1.
85
We studied the expression of MLH1 in these mutants and found both protein and
transcript levels to be comparable to the expression in cells reconstituted with WT MLH1
(Figure 20 B, 20 C). After that, we checked the effect of these mutations on cisplatin
cytotoxicity using an MTS assay. The MLH1 deficient empty vector cells showed a
cisplatin resistant phenotype as compared to the MLH1 proficient WT cells, consistent
with our previous results. Moreover, the MLH1 ATPase mutant cells also showed a
cisplatin resistant phenotype indicating the importance of the ATPase activity of MLH1
in mediating cisplatin sensitivity (Figure 20 D). In contrast, previous studies using mouse
embryonic fibroblasts have shown that loss of MLH1 ATPase activity still maintains
sensitivity to cisplatin, suggesting the direct DNA damage signaling by MMR proteins as
the mediator of cisplatin sensitivity [88, 135]. Our studies in human cells, however,
showed that this activity is essential for cellular sensitivity, indicating that different
ATPase mutations can have a differential effect on cisplatin cytotoxicity. Furthermore,
we did not see any significant difference in the survival profiles of these cells in response
to oxaliplatin (Supplementary Figure 2C). These results suggest that these effects are
specific to cisplatin and can be attributed to the unique structure produced by cisplatin
ICLs.
To understand the mechanism of resistance and to appreciate the events taking
place near the ICLs, we studied how these ATPase mutations affect the repair of cisplatin
ICLs. We used modified alkaline comet assays to study the rate of ICL repair up to the 72
hr time point after cisplatin treatment. MLH1 deficient empty vector cells showed
increased repair of cisplatin ICLs at 48 hr and 72 hr time point. Moreover, the MLH1
ATPase mutant cells also showed increased repair of cisplatin ICLs at 48 hr and 72 hr
86
time points as compared to the MLH1 proficient WT cells (Figure 21 A). In addition, in
an immunofluorescence assay MLH1 deficient empty vector cells and MLH1 ATPase
mutants also showed increased repair of ICL induced DSBs as compared to the WT
MLH1 proficient cells (Figure 21 B). These results indicated that the ATPase activity is
essential to maintain cisplatin ICLs on the DNA, and therefore, this increased repair can
be correlated with the resistant phenotype observed in the cell survival studies.
This role of MLH1 is dependent on the generation of a mismatch due to BER
processing. To understand whether in these cells MMR and BER pathways play an
overlapping role, we inhibited BER in HCT116 cells using myricetin. Myricetin is a
small molecule inhibitor of the repair function of Apurinic apyridiminic endonuclease 1
(APE1). Pretreatment with myricetin gave rise to a cisplatin resistant phenotype in MLH1
proficient WT cells, which is consistent with our previous findings. However, when we
used MLH1 deficient empty vector cells and MLH1 ATPase mutant cells, which are
already resistant to cisplatin, myricetin pretreatment did not give rise to any additive
increase in the degree of resistance (Supplementary Figure 3A-3C). These results suggest
that MMR processing occurs downstream of BER and these two pathways play an
epistatic role in mediating cisplatin cytotoxicity.
Targeting XPF-ERCC1 in MLH1 deficient cancer cells
Our studies showed that MLH1 deficiency gave rise to cisplatin resistance. Or
results suggest that in the absence of MLH1, cisplatin ICLs are repaired at a faster rate
compared to WT cells. XPF-ERCC1 is essential for various stages of ICL repair. Our
previous studies have shown that downregulation of XPF ERCC1 can be used to sensitize
87
BER deficient cancer cells [71]. When we knocked down XPF-ERCC1 in MLH1
deficient HCT116 cells, we observed increased sensitivity to cisplatin (Figure 22 A).
Similarly, a double knockdown of MLH1 and ERCC1 gave rise to a sensitive phenotype
in response to cisplatin (Figure 22 B). These results suggested that inhibition of XPFERCC1 can potentiate cisplatin cytotoxicity in MLH1 deficient cancer cells.
In conclusion, our results suggest that downstream MMR proteins play a
differential role in mediating cisplatin cytotoxicity. In addition, the ATPase activity of
MLH1 is essential for the processing of the BER induced mismatch near the cisplatin ICL
region and this MMR processing hinders the productive repair of cisplatin ICLs and thus,
gives rise to a cisplatin sensitive phenotype (Model Figure 10, Chapter 2). In the absence
of MMR processing, ICLs are repaired at an increased rate giving rise to cisplatin
resistance, and these cells can be sensitized by inhibition of XPF-ERCC1.
3.4 Discussion
The role of MutS and MutL homologues in the repair of ICLs has been previously
studied using psoralen as a DNA damaging agent [175, 176]. However, the role of
individual MMR proteins in the processing of cisplatin ICLs remains unexplored.
Cisplatin forms ICLs that have a unique structure as compared to most other platinum
agents as well as other cross linking agents. As a consequence, cisplatin treatment leads
to the recruitment of different DNA repair proteins and initiation of different DNA repair
pathways. Previously, we have shown for the first time the crosstalk between BER and
MMR pathways in response to cisplatin ICLs. We have shown that the cisplatin ICL
structure, which involves the flipping out of the flanking cytosine residues, makes them
88
prone to undergo oxidative deamination. These deamination events lead to the conversion
of cytosine residues to uracil. Uracil acts as a substrate for the initiation of the BER
pathway. BER processing near the cisplatin ICLs involves error prone DNA synthesis by
Polβ. The low fidelity nature of Pol β generates a mismatch, which in turn, leads to the
activation of the MMR pathway. We have shown an increased recruitment of the MMR
protein MSH2 on the BER processed substrates as compared to BER unprocessed
substrates [155]. The aim of this work was to understand the requirement of the
downstream MMR proteins in mediating cisplatin sensitivity. It was anticipated that this
information would help us better understand the mechanism of the crosstalk between
BER and MMR pathways near a cisplatin ICL site. Furthermore, it will help us to
elucidate the role of MMR proteins in the complex process of cisplatin ICL repair. Repair
of ICLs is a complicated process and the exact mechanism of the ICL repair and how
different repair proteins play a role is still unknown. From a clinical point of view,
increased repair of ICLs has been shown to be associated with resistance to the drug and
poor prognosis [31, 157]. Therefore, understanding the mechanisms of cisplatin ICL
repair and evaluating the role of different proteins in these pathways becomes clinically
relevant.
We are not the first to study the effect of an MSH3 defect on cisplatin
cytotoxicity. Recently, it was shown that MSH3 is involved in the repair of cisplatin ICL
induced DNA DSBs [170, 171]. In contrast, Vaismann et al have shown that MSH3
deficiency does not have any significant effect on cisplatin or oxaliplatin cytotoxicity
[124]. Consistent with these studies, when we knocked down MSH3 in the breast
adenocarcinoma MDA-MB-231 cell line, we did not see any effect on cisplatin or
89
oxaliplatin cytotoxicity indicating that MSH3 does not influence platinum cytotoxicity at
least in this cell line. These observations suggest that the MSH3 deficiency should still
retain sensitivity to cisplatin based chemotherapy. Depletion of MSH3 in Polβ deficient
cells did not give rise to a cisplatin sensitive phenotype indicating that MSH3 is not
essential and does not play any significant role in pathways that maintain cisplatin
sensitivity. In addition, MSH6 knockdown cells showed a cisplatin resistant phenotype
consistent with the literature [124, 136]. Furthermore, MSH6 downregulation in Pol β
knock down cells did not give rise to any additive increase in the cisplatin resistance,
indicating that these proteins play an overlapping role and are required to maintain
cisplatin sensitivity.
In accordance with the cell survival studies, we wanted to elucidate the
mechanism of cisplatin resistance in MSH6 deficient cells. Increased repair of adducts
from the DNA has been shown to be one of the major mechanisms of cisplatin resistance.
MSH2 forms a common partner in MutSα and MutSβ heterocomplexes. Previously, we
have shown that lack of MSH2 does not affect the rate of cisplatin intrastrand adduct
repair [155]. For this reason, we did not expect MSH3 and MSH6 to affect intrastrand
adduct repair either. When we studied ICL repair in MSH3 knockdown cells, we did not
see any change in the rate of cisplatin ICL repair as well as ICL induced DNA DSBs. Zhu
and Lippard have shown that MutSβ is one of the proteins that bind specifically to
cisplatin ICLs [82], indicating that MSH2-MSH3 complex might play some role in the
repair of cisplatin ICLs. Based on these observations, we thought that MSH3 might be
one of the proteins that is recruited to the cisplatin ICL site. However, it is not absolutely
essential, as lack of MSH3 did not affect cellular response to cisplatin within our
90
experimental conditions. MSH6 deficiency has been known to cause cisplatin resistance.
Various models have been proposed to explain the mechanism. There have been studies
that link futile cycles of mismatch repair of cisplatin intrastrand adducts with increased
sensitivity. Some other studies have shown that MSH6 is essential to signal apoptosis.
These studies have been attributed to all kinds of adducts formed by cisplatin overall.
However, we found that MSH6 knockdown resulted in the increased repair of cisplatin
ICLs and ICL induced DNA DSBs. These results suggested that the drug sensitizing
effects of MMR proteins are dependent on the processing of cisplatin ICLs with MutSα
and not MutSβ being required to maintain a cisplatin sensitive phenotype.
Inactivation of MLH1 has been shown to give rise to cisplatin resistance. In our
studies, knockdown of MLH1 in Pol β deficient cells did not give rise to any additional
increase in the degree of resistance, suggesting that these two proteins are involved in the
same pathway that mediates cisplatin cytotoxicity. The difference in the sensitivities
could be correlated with the difference in the rate of cisplatin ICL repair and ICL induced
DSBs. These results are consistent with an involvement of the entire downstream MMR
pathway in maintaining cisplatin sensitivity. We hypothesized that, the MMR pathway
tries to repair the mismatch that is generated due to error prone DNA synthesis by Polβ.
This mismatch repairing ability of MMR proteins is dependent upon the ATPase domain
of these proteins. This activity is essential for the initiation of the repair processes that
recruit downstream repair proteins and facilitate the removal of the mismatch. To
understand the importance of these processes in response to cisplatin, we utilized the
MLH1 ATPase mutant cell lines. MLH1 missense mutations form about 65% of all the
MMR mutations associated with HNPCC syndrome [177]. Many of these mutations
91
occur in the ATPase domain of this protein giving rise to mismatch repair deficiency and
a high level of microsatellite instability [178]. S44F mutations in the ATPase domain
have frequently been associated with the HNPCC syndrome [177]. In our cell lines, these
mutations did not have any significant effect on the level of mutant MLH1 protein. The
absence of ATPase activity led to a cisplatin resistant phenotype, indicating that this
mismatch repairing ability represents the essential activity that maintains a cisplatin
sensitive phenotype.
Previous studies have shown that an MLH1 ATPase G67R mutation still retains
sensitivity despite having a defect in the mismatch repair capacity [135]. These studies
point toward the role of MMR proteins in the activation of processes that directly signal
apoptosis. Different models have been proposed to explain the role of MMR proteins in
cisplatin mediated cytotoxicity. The futile cycle model involves the futile repair of base
mismatches generated in response to alkylating agents, which gives rise to strand breaks
ultimately leading to cell death. In our studies, the MLH1 deficient and MLH1 ATPase
mutant cells showed a resistant phenotype in response to alkylating agent MNNG
(Supplementary Figure 4). The expression of MGMT in the ATPase mutant cells was
comparable to that in the WT cells (Supplementary Figure 5A). On the other hand, a
second model involves the role of MMR proteins as DNA damage sensors which directly
signal apoptosis in response to alkylating and platinating agents [88, 135, 136, 140].
PMS2 binds to MLH1 and forms MutLα. This complex is involved in all MMR
processes. In response to cisplatin treatment, PMS2 has been shown to interact with p73
to signal apoptosis [137, 139]. ATP binding to MLH1 has been shown to enhance the
interaction between MLH1 and PMS2. In our cell lines, an ATPase mutation in MLH1
92
did not have any effect on the stability of PMS2 which would mean the signaling
function of PMS2, if any, remained unaffected (Supplementary Figure 5B). Our studies
suggest that, the MMR protein binding near the ICL site hinders the repair of ICLs which
in turn results in cisplatin sensitivity. The ATPase mutant MLH1 protein lacks the repair
capacity but retains its function as a DNA damage sensor, if such an activity is required
for cisplatin sensitivity. Cell survival studies in the ATPase mutant cell line show that
this activity is essential for mediating cisplatin cytotoxicity. This would predict that either
MLH1 does not have any DNA damage signaling role in response to cisplatin ICLs or
ATPase mutants do not stick around at the damage site long enough as they are unable to
repair the damage. These possibilities will be examined in future studies. Alternatively,
our study shows that MLH1 ATPase S44L and S44F mutations behave differently as
compared to the MLH1 ATPase G67R mutation studied before, suggesting that different
MLH1 ATPase mutations could respond differently to cisplatin therapy.
Initially, we set out to differentiate between the futile repair cycle model and the
direct DNA damage signaling model involving MMR proteins in response to cisplatin
ICLs. However, one caveat here is that the specific domains in the MMR proteins that are
required for sensing and signaling DNA damage are still unknown. These domains can be
attributed to any residues in the ATPase domain itself, because of this, the role of MMR
proteins as DNA damage sensors cannot be refuted. Therefore, we still cannot make any
definite conclusions about the role of direct signaling by MMR proteins in the
development of cisplatin resistance. Considering that the MMR deficient ATPase mutants
showed increased cisplatin resistance, we believe the mismatch repair ability and MMR
93
processing near the ICL sites plays a crucial role in maintaining cisplatin sensitivity
consistent with our ICL repair data.
When we looked at how these mutations affect the events taking place near the ICL site
by using alkaline comet assay, we found that these mutant cells as well as MLH1 null
cells showed increased repair of ICLs and ICL induced DNA DSBs. These results
suggest that the repair activities by MMR proteins or the MMR processing near the ICL
are essential to maintain ICLs on the DNA and therefore, to maintain a cisplatin sensitive
phenotype. This increased repair of cisplatin ICLs can be due to reduced binding of the
MLH1 mutant proteins at these sites due to an inability to process and repair the
mismatch. This, in turn, results in increased repair of ICLs through an enhanced
accessibility of NER and HR proteins to the ICL site. On the other hand, the increased
repair of ICLs can also be due to increased expression of NER and HR proteins which are
involved in the repair of ICLs. In contrast, there have been some other studies which
showed that the increased resistance in the MSH3 proficient cells was dependent on the
increased expression of XPF and ERCC1 proteins which are involved in ICL repair
[172]. Various studies have shown that increased expression of these proteins is
responsible for the development of cisplatin and oxaliplatin resistance [30, 36]. We
studied the expression of XPF and ERCC1 proteins for a period of 0 to 72 hr after
cisplatin treatment, and we did not see any change in the expression levels of these
proteins in cisplatin treated (Supplementary Figure 6) or in untreated cells (Data not
shown) which rules out enhanced HR or NER protein expression. These studies indicate
that the MMR proteins process the Polβ induced mismatch near the cisplatin ICL site,
block productive ICL repair and give rise to cisplatin sensitivity (Model figure 10).
94
Whereas, lack of MMR provides enhanced accessibility to ICL DNA repair proteins
which results in increased repair of cisplatin ICLs.
95
Figures
MSH3 Knockdown in Polb KD MDA-MB-231
MSH3 Knockdown in wt MDA-MB-231
A
shC96
shC120
96
120
144
B
168
shC96
shC120
96
120
144
168
MSH3
MSH3
α-tubulin
α-tubulin
MSH6 Knockdown in Polb KD MDA-MB-231
MSH6 Knockdown in wt MDA-MB-231
shC96
shC120
96
120
144
D
168
96
120
144
M SH6
α-tubulin
α -tubulin
MSH3 Transcript knockdown in MDA MB 231 cells
MSH6 Transcript knockdown in MDA MB 231 cells
F
120
100
% Transcript knockdown
shC 120
MSH6
E
WT
Polb KD
80
60
40
20
0
shC96
% Transcript knockdown
C
96 hr
80
60
40
20
0
120 hr
WT
Polb KD
100
96 hr
120 hr
Figure 13: shRNA mediated downregulation of MSH3 in wildtype (A) and Polβ KD (B) MDAMB-231 cells and MSH6 KD in wildtype (C) and Polβ KD (D) MDA-MB-231 cells. Cells were
transfected with shRNA directed against MSH3 or MSH6 as described in Materials and Methods.
Proteins were extracted at indicated time points and probed with MSH3 and MSH6 antibody with
α-tubulin as loading control.
Percent transcript knockdown in wildtype and Polβ KD MDA-MB-231 cells: MSH3 KD (E)
MSH6 KD (F). Cells were transiently transfected with shRNA directed against MSH3 and
MSH6. At 96 and 120 hr post transfection, total RNA was extracted from cells and the transcript
levels were quantified as described in Materials and Methods with GAPDH as an endogenous
control. The percent transcript knockdown was determined from 2^-ΔΔCT values with nontargeting shRNA as controls.
96
Figure 14
Colony survival assays
A
120
WT + shControl
WT + shMSH6
Polb KD + shControl
Polb KD + shMSH6
% Colony survival
100
80
60
40
20
0
0
1
2
3
4
5
6
Cisplatin (µM)
B
120
WT + shControl
WT + shMSH3
Polb KD + shControl
Polb KD + shMSH3
% Colony survival
100
80
60
40
20
0
0
1
2
3
4
5
6
Cisplatin (µM)
Figure 14: Cisplatin cytotoxicity. Colony survival assay in MDA-MB-231 cells (A).
Control (closed circles), MSH3 (open triangles), Polβ (closed squares) and Polβ MSH3
KD (open circles). shRNA transfected cells were treated with increasing doses of
cisplatin and cytotoxicity was determined by colony survival assay. (B) Colony survival
assay in MDA-MB-231 cells: Control (closed circles), MSH6 (open triangles), Polβ
(closed squares) and Polβ MSH6 KD (open circles). Results are represented as mean ±
SD from 3 independent experiments.
97
Figure 15
A
ICL repair
120
WT + shControl
WT + shMSH6
Polß KD + shControl
Polß KD + shMSH6
% ICLs remaining
100
80
60
40
* *
20
0
0 hr
24 hr
*
* **
48 hr
72 hr
Post incubation time
B
120
WT + shControl
WT + shMSH3
PolßKD + shControl
PolßKD + shMSH3
% ICLs remaining
100
80
60
40
**
20
0
0 hr
24 hr
48 hr
*
*
72 hr
Post incubation time
Figure 15. Repair of cisplatin ICLs in MDA-MB-231 (A) MSH3 KD in MDA-MB-231
(B) MSH6 KD in MDA-MB-231 Cells were treated with cisplatin for 2 hrs and comet
assays were performed as described at different time intervals (0, 24, 48 and 72 hr). The
data was collected using komet 5.5 software. The percentage of interstrand crosslinks
present at each time point was calculated using olive tail moments. Results are
represented as mean ± SD of three independent experiments. Statistical analysis was
performed by student’s t test and comparisons are made between wildtype and proficient
cells vs deficient cells. NS – non-significant; * - P< 0.05.
98
Figure 16
A
DSB repair
% g H2AX foci positive cells
70
WT + shControl
WT + shMSH6
Polb KD + shControl
Polb KD + shMSH6
60
50
40
*
30
*
*
20
** *
10
0
0 hr
24 hr
48 hr
72 hr
Post incubation time
B
% g H2AX foci positive cells
80
WT + shControl
WT + shMSH3
Polb KD + shControl
Polb KD + shMSH3
60
40
**
**
20
0
0 hr
24 hr
48 hr
72 hr
Post incubation time
Figure 16: Repair of γH2AX foci in MDA-MB-231 (A) MSH3 KD (B) MSH6 KD
Cells were treated with cisplatin for 2 hrs and immunofluorescence was performed as
described at different time intervals (0, 24, 48 and 72 hr). A minimum of 200 cells were
analyzed for each time point. The percentage of γH2AX foci positive cells at each time
point was calculated. Results are represented as mean ± SD of three independent
experiments. Statistical analysis was performed by student’s t test and comparisons are
made between wildtype and proficient cells vs deficient cells. NS – non-significant; * P< 0.05
99
Figure 17
A
MLH1 Knockdown in wt MEF
UT
48h
siCtrl
72h
96h
120h
MLH1
α-tubulin
B
MLH1 Knockdown in Polb -/- MEF
siCtrl
UT
48h
72h
96h
120h
MLH1
α-tubulin
Transcript knockdown
C
% MLH1 transcript knockdown
120
48 hr
72 hr
100
80
60
40
20
0
WT MEFs
Pol ß -/- MEFS
Figure 17: knockdown efficiency in wildtype (A) and Polβ -/- (B) MEFs, Cells were
transiently transfected with siRNA directed against MLH1. At 96 and 120 hr post
transfection, Proteins were extracted at indicated time points and probed with MLH1
antibody with α-tubulin as loading control.
(C) Percent transcript knockdown in wildtype and Polβ -/- MEF cells: Cells were
transiently transfected with siRNA directed against MLH1. At 96 and 120 hr post
transfection, total RNA was extracted from cells and the transcript levels were quantified
as described in materials and methods with B actin as an endogenous control. The percent
transcript knockdown was determined from 2^-ΔΔCT values with non-targeting siRNA
as controls.
100
Figure 18
Colony Survival assay
% Colony Survival
120
Wt + siCtrl
Wt + siMLH1
Polb -/- + siCtrl
Polb -/- + siMLH1
100
80
60
40
20
0
0
2
4
6
8
10
Cisplatin (µM)
Figure 18: Colony survival assay in WT and Pol β null MEFs with MLH1 KD. Cells
were treated with increasing doses of cisplatin for 2 hrs. Results are represented as mean
SD of 3 independent experiments.
101
Figure 19
A
Intrastrand adduct repair
% of intrastrand adducts
160
140
NS NS
Wt + siCtrl
Wt + siMLH1
Polb -/- + siCtrl
Polb -/- + siMLH1
NS
120
100
80
NS NS NS
60
NS
40
NS NS
20
0
0 hr
12 hr
24 hr
48 hr
Post-incubation time
B
ICL repair
Wt + siCtrl
Wt + siMLH1
Polb -/- + siCtrl
Polb -/- + siMLH1
100
% of ICLs
80
NS
60
NS NS
40
20
0
* * *
* * *
0 hr
24 hr
48 hr
Post-incubation time
102
72 hr
C
DSB repair
% g H2AX foci positive cells
80
NS NS
NS
60
Wt + siCtrl
Wt + siMLH1
Polb -/- + siCtrl
Polb -/- + siMLH1
40
* *
*
20
* * *
0
0 hr
24 hr
48 hr
72 hr
Post-incubation time
Figure 19: Cisplatin adduct repair in WT and Polβ null MEFs with MLH1 KD (A).
ELISA: MLH1 KD in MDA-MB-231 cells. Cells were treated with cisplatin for 2 hrs and
genomic DNA was extracted at the indicated time points. ELISA was performed using an
antibody specific for cisplatin GG intrastrand adduct and the percentage of GG adducts
remaining was calculated as described. Results are represented as mean ± SD of three
independent experiments. Statistical analysis was performed by student’s t test and
comparisons are made between wild type and proficient cells vs deficient cells. NS –
non-significant (B) Alkaline comet assay was performed in WT and Pol β null MEFs as
described previously. (C) Immunofluorescence assay results for MLH1 KD in WT and
Pol β null MEF cells.
103
Figure 20
A
ATPase
domain
N
PM S2 binding
domain
M utS binding
domain
C
S44L
S44F
B
M LH 1 protein Expression
WT 3.6
Empty
vector
S44L
S44F
MLH1
α-tubulin
% MLH1 Transcript knockdown
C
M LH 1 Transcript levels
140
120
100
80
60
40
20
0
Empty WT 3.6
vector
S44L
S44F
Figure 20: Graphical representation of the MLH1 ATPase mutants (A), (B) Western
blot analysis and (C) real-time PCR were performed to check the expression levels of
MLH1 transcript and protein.
104
MTS Assay
D
120
Empty vector
WT 3.6
S44L
S44F
% Survival
100
80
60
40
20
0
0
20
40
60
Cisplatin (µM)
80
100
Figure 20 D: MTS assay was performed by treating the HCT116 cells using increasing
concentration of cisplatin for 2 hrs. WT cells (closed diamonds), Empty vector (open
triangles), S44L cells (closed squares) and S44F cells (open circles). Results are
represented as mean ± SE from 3 independent experiments.
105
Figure 21
ICL repair
A
120
WT
Empty vector
S44L
S44F
% ICLs remaining
100
80
60
40
**
*
*
20
0
0 hr
24 hr
48 hr
**
72 hr
Post incubation time
DSB repair
B
% g H2AX foci positive cells
100
WT 3.6
Empty vector
S44L
S44F
80
60
40
*
20
0
0 hr
36 hr
**
48 hr
***
72 hr
Post incubation time
Figure 21: Repair of interstrand crosslinks in HCT116 cells (A) Alkaline comet assay
was performed as described previously. Repair of DNA DSBs in HCT116 cells was
measured using an immunofluorescence assay with an antibody specific for γH2AX as
described previously (B)
106
Figure 22
Colony Survival assay
A
120
Hct116+3 - siControl
Hct116+3 - siERCC1+siXPF
Hct116 - siControl
Hct116 - siERCC1+siXPF
% Colony Survival
100
80
60
40
20
0
0
10
20
30
40
50
Cisplatin (µM)
Colony survival assay
B
120
WT
WT + siMLH1
WT siERCC1
WT siMLH1 + siERCC1
% Colony survival
100
80
60
40
20
0
0
2
4
6
8
Cisplatin (µM)
Figure 22: Downregulating XPF-ERCC1 in MLH1 proficient and deficient cells
siRNA mediated downregulation of XPF-ERCC1 was carried out in HCT116 and
HCT116 +3 (A) cells. (Figure contributed by Dr. Anbarasi Kothandapani). (B) siRNA
mediated downregulation of MLH1 and ERCC1 separately and together. The colony
survival assays were performed using increasing doses of cisplatin. Results are
represented as a mean SD of 3 independent experiments.
107
Supplementary Figure S1
A
Colony survival assay
120
WT + siControl
WT + siMSH3
Polb -/- + siControl
Polb -/- +siMSH3
% Colony survival
100
80
60
40
20
0
0
2
4
6
8
10
Cisplatin (µM)
Transcript levels
% MSH3 transcript knockdown
B
120
WT
Polß -/-
100
80
60
40
20
0
48 hr
72 hr
Supplementary Figure S1, The colony survival assay was performed in WT and Polβ /- MEF using an MSH3 specific siRNA (A). The cells were treated with increasing
concentration of cisplatin to assess cytotoxicity. Results are represented as mean SD of 3
independent experiments. Percent transcript levels show the knockdown efficiency of
MSH3 (B)
108
Supplementary Figure S2
A
MTS assay
120
WT + shControl
WT + shMSH3
Polb KD + shControl
Polb KD + shMSH3
100
% Survival
80
60
40
20
0
0
20
40
B
60
80
100
Oxaliplatin (µ M)
MTS assay
120
WT + shControl
WT + shMSH6
Polb KD + shControl
Polb KD + shMSH6
100
% Survival
80
60
40
20
0
0
20
40
60
80
Oxaliplatin (µ M)
109
100
MTS assay
C
120
WT 3.6
Empty vector
S44L
S44F
% Survival
100
80
60
40
20
0
0
5
10
15
20
25
Oxaliplatin (µ M)
Supplementary Figure S2. Oxaliplatin cytotoxicity was measured by using MTS assay
for MSH3 KD in WT and Polβ deficient MDA-MB-231 cells (A), MSH6 KD in WT and
Polβ deficient MDA-MB-231 cells (B) and HCT116 cells (C). The cells were treated
with increasing concentrations of oxaliplatin for 2 hrs.
110
Supplementary Figure S3
A
MTS assay
120
WT 3.6 + Pt
WT 3.6 + Myricetin + Pt
% Survival
100
80
60
40
20
0
0
50
100
150
200
250
Cisplatin (µM)
B
MTS assay
120
Empty vector + Pt
Empty vector + Myricetin + Pt
% Survival
100
80
60
40
20
0
0
50
100
150
200
Cisplatin (µM)
111
250
C
MTS assay
120
S44L + Pt
S44L + Myricetin + Pt
% Survival
100
80
60
40
20
0
0
50
100
150
200
250
Cisplatin (µM)
Supplementary Figure S3: MTS assay was used to assess cisplatin cytotoxicity after
myricetin pretreatment. The cells were pretreated with myricetin for 2 hrs followed by
treatment with increasing concentration of cisplatin in WT 3.6 (A), Empty vector (B) and
S44L cells (C).
112
Supplementary Figure S4
MTS assay
120
WT 3.6
Empty vector
S44L
S44F
% Survival
100
80
60
40
20
0
0
2
4
6
8
10
MNNG (µM)
Supplementary Figure S4: MTS assay was used to assess MNNG cytotoxicity. The
cells were treated with MNNG for 45 mins. MTS assays were performed as described in
Materials and Methods. The results are represented as mean SD of 3 independent
experiments.
113
Supplementary Figure S5
A
B
Supplementary figure S5: Western blot analysis for MGMT (A) and PMS2 (B)
expression in HCT116 cells. Approximately 50 µg of protein extract was loaded and
western blot analysis was performed as described in materials and methods.
114
Supplementary Figure S6
XPF-ERCC1 expression in HCT116 cells after cisplatin treatment
WT 3.6
EV
S44L S44F WT 3.6 EV S44L S44F
WT 3.6 EV S44L
S44F WT 3.6 EV S44L S44F
XPF
ERCC1
α-tubulin
48 hr
24 hr
0 hr
72 hr
Post incubation time
Post incubation time
Supplementary Figure S6: Expression of XPF and ERCC1 was checked in HCT116
cells after cisplatin treatment. The cells were treated with cisplatin for 2 hrs and were
harvested over a period of 0 – 72 hr time points. The western blot analysis was performed
as described previously.
115
Chapter 4
Overall summary
Cisplatin is still one of the most widely used chemotherapeutic agents for the treatment of
a variety of human malignancies. It forms a basis of a number of treatment protocols and
combination therapies. Despite its successful use, cisplatin based chemotherapy often
develops resistance. This drug resistance can be acquired or it can be intrinsic in nature.
Numerous studies have shown the correlation between increased repair of cisplatin
adducts and increased resistance to the drug. Earlier, cisplatin intrastrand adducts were
considered to be responsible for the clinical efficacy of the drug. It was believed that
these adducts are clinically most relevant as they are formed more frequently than
cisplatin ICLs, their cellular levels correlated well with clinical efficacy and clinically
ineffective transplatin did not form these adducts. Recent studies have considered
cisplatin ICLs to be more cytotoxic as compared to the other types of adducts formed by
cisplatin. These ICLs are formed at a lower rate and the synthesis of defined ICL
substrates has been difficult, thus limiting the studies related to the biological and clinical
importance of cisplatin ICLs. Nonetheless, cisplatin ICL repair rate has been correlated
with sensitivity of tumor cells. Furthermore, recent studies have reported that cisplatin
ICL repair is dependent on TC-NER, but not GG-NER, outside of replication. These
116
studies predicted that when cells enter S phase, this would lead to higher relative levels of
ICLs compared to intrastrand adducts, thus sensitizing tumor cells. These studies
highlight the clinical importance of ICLs. However, the exact mechanism and the repair
pathways involved in the repair of ICLs are still not very well understood. Enhanced
repair of cisplatin ICLs has been correlated with increased cisplatin resistance.
Understanding the mechanisms that affect the repair of ICLs, therefore, is clinically
relevant. DNA repair pathways including BER and MMR have shown to modulate
cisplatin cytotoxicity. Our studies have shown how these pathways influence cisplatin
ICL repair and affect cisplatin cytotoxicity.
Lack of BER and MMR pathways give rise to cisplatin resistance. Studies have shown
that cisplatin treatment leads to the enrichment of MMR deficient drug resistant cells.
These effects are detrimental to cisplatin based chemotherapy. Our studies showed that,
increased resistance in case of MMR and BER deficient cells can be attributed to
increased repair of cisplatin ICLs. The rate of intrastrand adduct repair was not affected
in BER and MMR deficient background. In addition, we did not see any resistant
phenotype in response to oxaliplatin. These results suggested that the influence of BER
and MMR status on cisplatin cytotoxicity was cisplatin ICL specific. In addition, our
studies highlighted a novel mechanism where two DNA repair pathways are involved in
modulating cisplatin ICL repair. Our data suggested that this possibility was created by
mutagenic role of Polβ. Polβ mediated misincorporation of nucleotides leads to the
generation of mismatches, and these mismatched bases activate and recruit MMR
proteins to the regions near the ICLs. We showed that BER processing in these regions is
essential for the binding and recruitment of MMR proteins. In support of these
117
observations, using Polβ D256A polymerase mutant cells, we showed that MMR
activation and involvement in this pathway is dependent on the polymerase activity of
Polβ. In these studies, Polβ D256A cells which are deficient in polymerase activity were
shown to be sensitive to cisplatin. Alterations in Polβ activity and Polβ mutations are a
common occurrence in cases of colon cancer. Currently, majority of the colon cancer
cases are treated with oxaliplatin. It will be interesting to study how different Polβ
mutations respond to cisplatin and whether these subsets of patients can be treated more
effectively with cisplatin.
Subsequently, we also studied the requirement of different MMR proteins in this
mechanism. We showed that MutSα is required to maintain a cisplatin sensitive
phenotype, whereas MutSβ had no role in mediating cisplatin cytotoxicity. In addition,
MLH1 is required to maintain a sensitive phenotype, suggesting that the entire MMR
pathway is essential for this process. Clinically relevant MLH1 ATPase mutants, which
are deficient in MMR also showed increased resistance owing to increased repair of ICLs
when compared with WT cells. From our studies, we predict that BER and MMR
proteins process the regions that flank the cisplatin ICLs, and this processing interferes
with the productive ICL repair by NER and HR proteins. As a consequence, ICLs are
maintained on the DNA for a longer time giving rise to increased cytotoxicity in response
to cisplatin. When we targeted XPF-ERCC1 in MMR deficient cells, we observed that
inhibition of adduct repair sensitizes the cells to cisplatin. Moreover, targeting signaling
kinases ATR and ATM also sensitized BER deficient cells to cisplatin, indicating the
potential use of these inhibitors in cisplatin based combination therapy. We believe that,
understanding the exact mechanisms of the pathways that affect ICL repair will lead to
118
better treatment protocols and generation of new protein targets that will increase the
sensitivity and efficacy of cisplatin based chemotherapy in BER and MMR deficient
cancer cells.
119
Reference List
[1] T.Boulikas, M.Vougiouka. Recent clinical trials using cisplatin, carboplatin and
their combination" chemotherapy drugs (review), Oncol. Rep. 11 (2004) 559-595.
[2] Y.Jung, S.J.Lippard. Direct cellular responses to platinum-induced DNA damage,
Chem. Rev. 107 (2007) 1387-1407.
[3] C.J.Williams, J.M.Whitehouse. Cis-platinum: a new anticancer agent, Br. Med. J.
1 (1979) 1689-1691.
[4] P.A.Andrews, S.B.Howell. Cellular Pharmacology of Cisplatin - Perspectives on
Mechanisms of Acquired-Resistance, Cancer Cells-A Monthly Review 2 (1990)
35-43.
[5] Z.H.Siddik. Cisplatin: mode of cytotoxic action and molecular basis of resistance,
Oncogene 22 (2003) 7265-7279.
[6] G.Y.Park, J.J.Wilson, Y.Song, S.J.Lippard. Phenanthriplatin, a monofunctional
DNA-binding platinum anticancer drug candidate" with unusual potency and
cellular activity profile, Proc. Natl. Acad. Sci. U. S. A 109 (2012) 11987-11992.
[7] S.Dhar, F.X.Gu, R.Langer, O.C.Farokhzad, S.J.Lippard. Targeted delivery of
cisplatin to prostate cancer cells by aptamer functionalized Pt(IV) prodrug-PLGAPEG nanoparticles, Proc. Natl. Acad. Sci. U. S. A 105 (2008) 17356-17361.
120
[8] S.Dhar, N.Kolishetti, S.J.Lippard, O.C.Farokhzad. Targeted delivery of a cisplatin
prodrug for safer and more effective prostate cancer therapy in vivo, Proc. Natl.
Acad. Sci. U. S. A 108 (2011) 1850-1855.
[9] M.C.Lim, R.Bruce Martin. The nature of< i> cis</i> amine Pd (II) and
antitumor< i> cis</i> amine Pt (II) complexes in aqueous solutions, Journal of
Inorganic and Nuclear Chemistry 38 (1976) 1911-1914.
[10] R.Safaei, S.B.Howell. Copper transporters regulate the cellular pharmacology and
sensitivity to Pt drugs, Critical reviews in oncology/hematology 53 (2005) 13-23.
[11] G.Samimi, N.M.Varki, S.Wilczynski, R.Safaei, D.S.Alberts, S.B.Howell. Increase
in expression of the copper transporter ATP7A during platinum drug-based
treatment is associated with poor survival in ovarian cancer patients, Clinical
Cancer Research 9 (2003) 5853-5859.
[12] S.Ishida, J.Lee, D.J.Thiele, I.Herskowitz. Uptake of the anticancer drug cisplatin
mediated by the copper transporter Ctr1 in yeast and mammals, Proceedings of
the National Academy of Sciences 99 (2002) 14298-14302.
[13] A.K.Holzer, G.Samimi, K.Katano, W.Naerdemann, X.Lin, R.Safaei, S.B.Howell.
The copper influx transporter human copper transport protein 1 regulates the
uptake of cisplatin in human ovarian carcinoma cells, Molecular Pharmacology
66 (2004) 817-823.
[14] G.Ciarimboli, T.Ludwig, D.F.Lang, H.Pavenstadt, H.Koepsell, H.J.Piechota,
J.Haier, U.Jaehde, J.Zisowsky, E.Schlatter. Cisplatin nephrotoxicity is critically
mediated via the human organic cation transporter 2, American Journal of
Pathology 167 (2005) 1477-1484.
[15] R.M.Franke, A.M.Kosloske, C.S.Lancaster, K.K.Filipski, C.Hu, O.Zolk,
R.H.Mathijssen, A.Sparreboom. Influence of Oct1/Oct2-deficiency on cisplatininduced changes in urinary N-acetyl-+¦-D-glucosaminidase, Clinical Cancer
Research 16 (2010) 4198-4206.
121
[16] C.S.Lancaster, C.Hu, R.M.Franke, K.K.Filipski, S.J.Orwick, Z.Chen, Z.Zuo,
W.J.Loos, A.Sparreboom. Cisplatin-induced downregulation of OCTN2 affects
carnitine wasting, Clinical Cancer Research 16 (2010) 4789-4799.
[17] K.Koike, T.Kawabe, T.Tanaka, S.Toh, T.Uchiumi, M.Wada, S.Akiyama, M.Ono,
M.Kuwano. A canalicular multispecific organic anion transporter (cMOAT)
antisense cDNA enhances drug sensitivity in human hepatic cancer cells, Cancer
Res. 57 (1997) 5475-5479.
[18] J.Wang, H.Wang, L.Zhao, S.Fan, Z.Yang, F.Gao, L.Chen, G.G.Xiao, J.Molnar,
Q.Wang. Down-regulation of P-glycoprotein is associated with resistance to
cisplatin and VP-16 in human lung cancer cell lines, Anticancer Res. 30 (2010)
3593-3598.
[19] M.Akaboshi, K.Kawai, H.Maki, K.Akuta, Y.Ujeno, T.Miyahara. The number of
platinum atoms binding to DNA, RNA and protein molecules of HeLa cells
treated with cisplatin at its mean lethal concentration, Cancer Science 83 (1992)
522-526.
[20] P.Boffetta, A.M.J.Fichtinger-Schepman, E.Weiderpass, H.C.M.van DijkKnijnenburg, G.Stoter, A.T.Van Oosterom, H.J.Keizer, S.D.Fossa, J.Kaldor,
P.Roy. Cisplatin-DNA adducts and protein-bound platinum in blood of testicular
cancer patients, Anti-cancer drugs 9 (1998) 125-130.
[21] B.Koeberle, M.T.Tomicic, S.Usanova, B.Kaina. Cisplatin resistance: Preclinical
findings and clinical implications, Biochimica et Biophysica Acta-Reviews on
Cancer 1806 (2010) 172-182.
[22] K.J.Mellish, L.R.Kelland, K.R.Harrap. In vitro platinum drug chemosensitivity of
human cervical squamous cell carcinoma cell lines with intrinsic and acquired
resistance to cisplatin, British Journal of Cancer 68 (1993) 240-250.
122
[23] C.Meijer, N.H.Mulder, H.Timmer-Bosscha, W.J.Sluiter, G.J.Meersma, E.G.de
Vries. Relationship of cellular glutathione to the cytotoxicity and resistance of
seven platinum compounds, Cancer Research 52 (1992) 6885-6889.
[24] K.Miyatake, K.Gemba, H.Ueoka, K.Nishii, K.Kiura, M.Tabata, T.Shibayama,
N.Takigawa, M.Kawaraya, M.Tanimoto. Prognostic significance of mutant p53
protein, P-glycoprotein and glutathione S-transferase-pi in patients with
unresectable non-small cell lung cancer, Anticancer research 23 (2002) 28292836.
[25] I.Konishi, K.Nanbu, M.Mandai, Y.Tsuruta, N.Kataoka, Y.Nagata, T.Mori. Tumor
response to neoadjuvant chemotherapy correlates with the expression of Pglycoprotein and PCNA but not GST--Ç in the tumor cells of cervical carcinoma,
Gynecologic Oncology 70 (1998) 365-371.
[26] F.Coste, J.M.Malinge, L.Serre, W.Shepard, M.Roth, M.Leng, C.Zelwer. Crystal
structure of a double-stranded DNA containing a cisplatin interstrand cross-link at
1.63 angstrom resolution: hydration at the platinated site, Nucleic Acids Research
27 (1999) 1837-1846.
[27] M.Suzuki, K.Hori, I.Abe, S.Saito, H.Sato. A new approach to cancer
chemotherapy: selective enhancement of tumor blood flow with angiotensin II, J.
Natl. Cancer Inst. 67 (1981) 663-669.
[28] T.G.Oliver, K.L.Mercer, L.C.Sayles, J.R.Burke, D.Mendus, K.S.Lovejoy,
M.H.Cheng, A.Subramanian, D.Mu, S.Powers, D.Crowley, R.T.Bronson,
C.A.Whittaker, A.Bhutkar, S.J.Lippard, T.Golub, J.Thomale, T.Jacks, E.A.SweetCordero. Chronic cisplatin treatment promotes enhanced damage repair and tumor
progression in a mouse model of lung cancer, Genes Dev. 24 (2010) 837-852.
[29] D.B.Zamble, S.J.Lippard. Cisplatin and Dna-Repair in Cancer-Chemotherapy,
Trends in Biochemical Sciences 20 (1995) 435-439.
123
[30] S.Usanova, A.Piee-Staffa, U.Sied, J.Thomale, A.Schneider, B.Kaina, B.Koeberle.
Cisplatin sensitivity of testis tumour cells is due to deficiency in interstrandcrosslink repair and low ERCC1-XPF expression, Molecular Cancer 9 (2010).
[31] P.Wynne, C.Newton, J.Ledermann, A.Olaitan, T.Mould, J.Hartley. Enhanced
repair of DNA interstrand crosslinking in ovarian cancer cells from patients
following treatment with platinum-based chemotherapy, British Journal of Cancer
97 (2007) 927-933.
[32] E.C.Friedberg. How nucleotide excision repair protects against cancer, Nature
Reviews Cancer 1 (2001) 22-33.
[33] R.D.Wood. Nucleotide excision repair in mammalian cells, J. Biol. Chem. 272
(1997) 23465-23468.
[34] T.Riedl, F.Hanaoka, J.M.Egly. The comings and goings of nucleotide excision
repair factors on damaged DNA, EMBO J. 22 (2003) 5293-5303.
[35] B.Koberle, J.R.W.Masters, J.A.Hartley, R.D.Wood. Defective repair of cisplatininduced DNA damage caused by reduced XPA protein in testicular germ cell
tumours, Current Biology 9 (1999) 273-276.
[36] S.Arora, A.Kothandapani, K.Tillison, V.Kalman-Maltese, S.M.Patrick.
Downregulation of XPF-ERCC1 enhances cisplatin efficacy in cancer cells, Dna
Repair 9 (2010) 745-753.
[37] C.Welsh, R.Day, C.McGurk, J.R.W.Masters, R.D.Wood, B.Koberle. Reduced
levels of XPA, ERCC1 and XPF DNA repair proteins in testis tumor cell lines,
International Journal of Cancer 110 (2004) 352-361.
[38] I.D.Kozekov, L.V.Nechev, M.S.Moseley, C.M.Harris, C.J.Rizzo, M.P.Stone,
T.M.Harris. DNA interchain cross-links formed by acrolein and crotonaldehyde,
Journal of the American Chemical Society 125 (2003) 50-61.
124
[39] I.D.Kozekov, L.Nechev, H.Wang, A.Sanchez, R.S.Lloyd, C.M.Harris,
T.M.Harris, C.J.Rizzo. Formation of DNA interchain crosslinks by the
deoxyguanosine adducts of 2,3-unsaturated aldehydes, Abstracts of Papers of the
American Chemical Society 224 (2002) U369.
[40] X.Shen, L.Li. Mutagenic Repair of DNA Interstrand Crosslinks, Environmental
and Molecular Mutagenesis 51 (2010) 493-499.
[41] E.M.Hlavin, M.B.Smeaton, P.S.Miller. Initiation of DNA Interstrand Cross-link
Repair in Mammalian Cells, Environmental and Molecular Mutagenesis 51
(2010) 604-624.
[42] M.Berardini, W.Mackay, E.L.Loechler. Evidence for a recombinationindependent pathway for the repair of DNA interstrand cross-links based on a
site-specific study with nitrogen mustard, Biochemistry 36 (1997) 3506-3513.
[43] X.Shen, S.Jun, L.E.O'Neal, E.Sonoda, M.Bemark, J.E.Sale, L.Li. REV3 and
REV1 play major roles in recombination-independent repair of DNA interstrand
cross-links mediated by monoubiquitinated proliferating cell nuclear antigen
(PCNA), Journal of Biological Chemistry 281 (2006) 13869-13872.
[44] P.A.Muniandy, D.Thapa, A.K.Thazhathveetil, S.T.Liu, M.M.Seidman. Repair of
laser-localized DNA interstrand cross-links in G1 phase mammalian cells, J. Biol.
Chem. 284 (2009) 27908-27917.
[45] M.Enoiu, J.Jiricny, O.D.Scharer. Repair of cisplatin-induced DNA interstrand
crosslinks by a replication-independent pathway involving transcription-coupled
repair and translesion synthesis, Nucleic Acids Research 40 (2012) 8953-8964.
[46] J.Doles, T.G.Oliver, E.R.Cameron, G.Hsu, T.Jacks, G.C.Walker, M.T.Hemann.
Suppression of Rev3, the catalytic subunit of Pol zeta, sensitizes drug-resistant
125
lung tumors to chemotherapy, Proceedings of the National Academy of Sciences
of the United States of America 107 (2010) 20786-20791.
[47] S.Sharma, N.A.Shah, A.M.Joiner, K.H.Roberts, C.E.Canman. DNA Polymerase
zeta Is a Major Determinant of Resistance to Platinum-Based Chemotherapeutic
Agents, Molecular Pharmacology 81 (2012) 778-787.
[48] X.J.Lin, T.Okuda, J.Trang, S.B.Howell. Human REV1 modulates the cytotoxicity
and mutagenicity of cisplatin in human ovarian carcinoma cells, Molecular
Pharmacology 69 (2006) 1748-1754.
[49] T.V.Ho, A.Guainazzi, S.B.Derkunt, M.Enoiu, O.D.Scharer. Structure-dependent
bypass of DNA interstrand crosslinks by translesion synthesis polymerases,
Nucleic Acids Research 39 (2011) 7455-7464.
[50] H.Y.Zheng, X.Wang, A.J.Warren, R.J.Legerski, R.S.Nairn, J.W.Hamilton, L.Li.
Nucleotide excision repair- and polymerase eta-mediated error-prone removal of
mitomycin C interstrand cross-links, Molecular and Cellular Biology 23 (2003)
754-761.
[51] I.U.De Silva, P.J.Mchugh, P.H.Clingen, J.A.Hartley. Defining the roles of
nucleotide excision repair and recombination in the repair of DNA interstrand
cross-links in mammalian cells, Molecular and Cellular Biology 20 (2000) 79807990.
[52] G.Damia, L.Imperatori, M.Stefanini, M.DIncalci. Sensitivity of CHO mutant cell
lines with specific defects in nucleotide excision repair to different anti-cancer
agents, International Journal of Cancer 66 (1996) 779-783.
[53] M.Raeschle, P.Knipsheer, M.Enoiu, T.Angelov, J.Sun, J.D.Griffith,
T.E.Ellenberger, O.D.Schaerer, J.C.Walter. Mechanism of replication-coupled
DNA interstrand crosslink repair, Cell 134 (2008) 969-980.
126
[54] D.T.Long, M.Raschle, V.Joukov, J.C.Walter. Mechanism of RAD51-Dependent
DNA Interstrand Cross-Link Repair, Science 333 (2011) 84-87.
[55] M.Raschle, P.Knipscheer, M.Enoiu, T.Angelov, J.Sun, J.D.Griffith,
T.E.Ellenberger, O.D.Scharer, J.C.Walter. Mechanism of replication-coupled
DNA interstrand crosslink repair, Cell 134 (2008) 969-980.
[56] L.J.Niedernhofer, H.Odijk, M.Budzowska, E.van Drunen, A.Maas, A.F.Theil,
J.de Wit, N.G.J.Jaspers, H.B.Beverloo, J.H.J.Hoeijmakers, R.Kanaar. The
structure-specific endonuclease Ercc1-Xpf is required to resolve DNA interstrand
cross-link-induced double-strand breaks, Molecular and Cellular Biology 24
(2004) 5776-5787.
[57] K.Kratz, B.Schopf, S.Kaden, A.Sendoel, R.Eberhard, C.Lademann, E.Cannavo,
A.A.Sartori, M.O.Hengartner, J.Jiricny. Deficiency of FANCD2-Associated
Nuclease KIAA1018/FAN1 Sensitizes Cells to Interstrand Crosslinking Agents,
Cell 142 (2010) 77-88.
[58] T.Taniguchi, M.Tischkowitz, N.Ameziane, S.V.Hodgson, C.G.Mathew, H.Joenje,
S.C.Mok, A.D.D'Andrea. Disruption of the Fanconi anemia-BRCA pathway in
cisplatin-sensitive ovarian tumors, Nature Medicine 9 (2003) 568-574.
[59] N.Bhagwat, A.L.Olsen, A.T.Wang, K.Hanada, P.Stuckert, R.Kanaar, A.D'Andrea,
L.J.Niedernhofer, P.J.Mchugh. XPF-ERCC1 Participates in the Fanconi Anemia
Pathway of Cross-Link Repair, Molecular and Cellular Biology 29 (2009) 64276437.
[60] M.Selvakumaran, D.A.Pisarcik, R.Bao, A.T.Yeung, T.C.Hamilton. Enhanced
cisplatin cytotoxicity by disturbing the nucleotide excision repair pathway in
ovarian cancer cell lines, Cancer Research 63 (2003) 1311-1316.
[61] T.Lindahl. Instability and Decay of the Primary Structure of Dna, Nature 362
(1993) 709-715.
127
[62] T.Lindahl, R.D.Wood. Quality control by DNA repair, Science 286 (1999) 18971905.
[63] L.J.Marnett. Oxyradicals and DNA damage, Carcinogenesis 21 (2000) 361-370.
[64] E.Cappelli, R.Taylor, M.Cevasco, A.Abbondandolo, K.Caldecott, G.Frosina.
Involvement of XRCC1 and DNA ligase III gene products in DNA base excision
repair, Journal of Biological Chemistry 272 (1997) 23970-23975.
[65] A.Klungland, T.Lindahl. Second pathway for completion of human DNA base
excision-repair: Reconstitution with purified proteins and requirement for DNase
IV (FEN1), Embo Journal 16 (1997) 3341-3348.
[66] T.A.Ranalli, S.Tom, R.A.Bambara. AP endonuclease 1 coordinates flap
endonuclease 1 and DNA ligase I activity in long patch base excision repair,
Journal of Biological Chemistry 277 (2002) 41715-41724.
[67] M.V.Sukhanova, S.N.Khodyreva, N.A.Lebedeva, R.Prasad, S.H.Wilson,
O.I.Lavrik. Human base excision repair enzymes apurinic/apyrimidinic
endonuclease1 (APE1), DNA polymerase beta and poly(ADP-ribose) polymerase
1: interplay between strand-displacement DNA synthesis and proofreading
exonuclease activity, Nucleic Acids Research 33 (2005) 1222-1229.
[68] S.H.Wilson, T.A.Kunkel. Passing the baton in base excision repair, Nat. Struct.
Biol. 7 (2000) 176-178.
[69] A.Maor-Shoshani, L.B.Meira, X.M.Yang, L.D.Samson. 3-methyladenine DNA
glycosylase ill important for cellular resistance to psoralen interstrand cross-links,
Dna Repair 7 (2008) 1399-1406.
128
[70] S.Couve-Privat, G.Mace, F.Rosselli, M.K.Saparbaev. Psoralen-induced DNA
adducts are substrates for the base excision repair pathway in human cells,
Nucleic Acids Research 35 (2007) 5672-5682.
[71] A.Kothandapani, V.S.M.N.Dangeti, A.R.Brown, L.A.Banze, X.H.Wang,
R.W.Sobol, S.M.Patrick. Novel Role of Base Excision Repair in Mediating
Cisplatin Cytotoxicity, Journal of Biological Chemistry 286 (2011) 14564-14574.
[72] D.Wang, D.B.Xiang, X.q.Yang, L.S.Chen, M.X.Li, Z.Y.Zhong, Y.S.Zhang.
APE1 overexpression is associated with cisplatin resistance in non-small cell lung
cancer and targeted inhibition of APE1 enhances the activity of cisplatin in A549
cells, Lung Cancer 66 (2009) 298-304.
[73] M.I.Koukourakis, A.Giatromanolaki, S.Kakolyris, E.Sivridis, V.Georgoulias,
G.Funtzilas, I.D.Hickson, K.C.Gatter, A.L.Harris. Nuclear expression of human
apurinic/apyrimidinic endonuclease (HAP1/REF-1) in head-and-neck cancer is
associated with resistance to chemoradiotherapy and poor outcome, International
Journal of Radiation Oncology Biology Physics 50 (2001) 27-36.
[74] S.Freitas, D.H.Moore, H.Michael, M.R.Kelley. Studies of apurinic/apyrimidinic
endonuclease/ref-1 expression in epithelial ovarian cancer: Correlations with
tumor progression and platinum resistance, Clinical Cancer Research 9 (2003)
4689-4694.
[75] J.S.Hoffmann, M.J.Pillaire, G.Maga, V.Podust, U.Hubscher, G.Villani. DnaPolymerase-Beta Bypasses In-Vitro A Single D(Gpg)-Cisplatin Adduct Placed on
Codon-13 of the Hras Gene, Proceedings of the National Academy of Sciences of
the United States of America 92 (1995) 5356-5360.
[76] F.Boudsocq, P.Benaim, Y.Canitrot, M.Knibiehler, F.Ausseil, J.P.Capp, A.Bieth,
C.Long, B.David, I.Shevelev, E.Frierich-Heinecken, U.Hubscher, F.Amalric,
G.Massiot, J.S.Hoffmann, C.Cazaux. Modulation of cellular response to cisplatin
by a novel inhibitor of DNA polymerase beta, Molecular Pharmacology 67 (2005)
1485-1492.
129
[77] V.Bergoglio, Y.Canitrot, L.Hogarth, L.Minto, S.B.Howell, C.Cazaux,
J.S.Hoffmann. Enhanced expression and activity of DNA polymerase beta in
human ovarian tumor cells: impact on sensitivity towards antitumor agents,
Oncogene 20 (2001) 6181-6187.
[78] G.P.Raaphorst, C.E.Ng, D.P.Yang. Comparison of response to radiation,
hyperthermia and cisplatin in parental and polymerase beta knockout cells,
International Journal of Hyperthermia 18 (2002) 33-39.
[79] S.Gurubhagavatula, G.Liu, S.Park, W.Zhou, L.Su, J.C.Wain, T.J.Lynch,
D.S.Neuberg, D.C.Christiani. XPD and XRCC1 genetic polymorphisms are
prognostic factors in advanced non-small-cell lung cancer patients treated with
platinum chemotherapy, Journal of Clinical Oncology 22 (2004) 2594-2601.
[80] R.Zhang, Y.J.Niu, Y.K.Zhou. Increase the cisplatin cytotoxicity and cisplatininduced DNA damage in HepG2 cells by XRCC1 abrogation related mechanisms,
Toxicology Letters 192 (2010) 108-114.
[81] J.M.de Murcia, C.Niedergang, C.Trucco, M.Ricoul, B.Dutrillaux, M.Mark,
F.J.Oliver, M.Masson, A.Dierich, M.LeMeur, C.Walztinger, P.Chambon, M.G.de.
Requirement of poly(ADP-ribose) polymerase in recovery from DNA damage in
mice and in cells, Proc. Natl. Acad. Sci. U. S. A 94 (1997) 7303-7307.
[82] G.Zhu, S.J.Lippard. Photoaffinity Labeling Reveals Nuclear Proteins That
Uniquely Recognize Cisplatin-DNA Interstrand Cross-Links, Biochemistry 48
(2009) 4916-4925.
[83] J.Jiricny, M.Nystrom-Lahti. Mismatch repair defects in cancer, Current Opinion
in Genetics & Development 10 (2000) 157-161.
[84] J.R.Eshleman, S.D.Markowitz. Mismatch repair defects in human carcinogenesis,
Human Molecular Genetics 5 (1996) 1489-1494.
130
[85] W.E.Glaab, J.I.Risinger, A.Umar, J.C.Barrett, T.A.Kunkel, K.R.Tindall.
Resistance to 6-thioguanine in mismatch repair-deficient human cancer cell lines
correlates with an increase in induced mutations at the HPRT locus,
Carcinogenesis 19 (1998) 1931-1937.
[86] D.Fink, S.Nebel, S.Aebi, H.Zheng, B.Cenni, A.Nehme, R.D.Christen,
S.B.Howell. The role of DNA mismatch repair in platinum drug resistance,
Cancer Research 56 (1996) 4881-4886.
[87] E.Papouli, P.Cejka, J.Jiricny. Dependence of the cytotoxicity of DNA-damaging
agents on the mismatch repair status of human cells, Cancer Research 64 (2004)
3391-3394.
[88] D.P.Lin, Y.X.Wang, S.J.Scherer, A.B.Clark, K.Yang, E.Avdievich, B.Jin,
U.Werling, T.Parris, N.Kurihara, A.Umar, R.Kucherlapati, M.Lipkin,
T.A.Kunkel, W.Edelmann. An Msh2 point mutation uncouples DNA mismatch
repair and apoptosis, Cancer Research 64 (2004) 517-522.
[89] J.T.Drummond, G.M.Li, M.J.Longley, P.Modrich. Isolation of An Hmsh2-P160
Heterodimer That Restores Dna Mismatch Repair to Tumor-Cells, Science 268
(1995) 1909-1912.
[90] F.Palombo, I.Iaccarino, E.Nakajima, M.Ikejima, T.Shimada, J.Jiricny. hMutSbeta,
a heterodimer of hMSH2 and hMSH3, binds to insertion/deletion loops in DNA,
Curr. Biol. 6 (1996) 1181-1184.
[91] J.Genschel, S.J.Littman, J.T.Drummond, P.Modrich. Isolation of MutSbeta from
human cells and comparison of the mismatch repair specificities of MutSbeta and
MutSalpha, J. Biol. Chem. 273 (1998) 19895-19901.
[92] D.C.Thomas, J.D.Roberts, T.A.Kunkel. Heteroduplex Repair in Extracts of
Human Hela-Cells, Journal of Biological Chemistry 266 (1991) 3744-3751.
131
[93] Y.I.Pavlov, I.M.Mian, T.A.Kunkel. Evidence for preferential mismatch repair of
lagging strand DNA replication errors in yeast, Current Biology 13 (2003) 744748.
[94] F.A.Kadyrov, L.Dzantiev, N.Constantin, P.Modrich. Endonucleolytic function of
MutL alpha in human mismatch repair, Cell 126 (2006) 297-308.
[95] F.A.Kadyrov, S.F.Holmes, M.E.Arana, O.A.Lukianova, M.O'Donnell,
T.A.Kunkel, P.Modrich. Saccharomyces cerevisiae MutL alpha is a mismatch
repair endonuclease, Journal of Biological Chemistry 282 (2007) 37181-37190.
[96] M.Raschle, G.Marra, M.Nystrom-Lahti, P.Schar, J.Jiricny. Identification of
hMutL beta, a heterodimer of hMLH1 and hPMS1, Journal of Biological
Chemistry 274 (1999) 32368-32375.
[97] G.M.Li, P.Modrich. Restoration of Mismatch Repair to Nuclear Extracts of H6
Colorectal Tumor-Cells by A Heterodimer of Human Mutl Homologs,
Proceedings of the National Academy of Sciences of the United States of America
92 (1995) 1950-1954.
[98] T.A.Prolla, A.Abuin, A.Bradley. DNA mismatch repair deficient mice in cancer
research, Seminars in Cancer Biology 7 (1996) 241-247.
[99] E.Cannavo, G.Marra, J.Sabates-Bellver, M.Menigatti, S.M.Lipkin, F.Fischer,
P.Cejka, J.Jiricny. Expression of the MutL homologue hMLH3 in human cells
and its role in DNA mismatch repair, Cancer Research 65 (2005) 10759-10766.
[100] A.T.Vo, F.X.Zhu, X.L.Wu, F.H.Yuan, Y.Gao, L.Y.Gu, G.M.Li, T.H.Lee,
C.T.Her. hMRE11 deficiency leads to microsatellite instability and defective
DNA mismatch repair, Embo Reports 6 (2005) 438-444.
[101] Y.Yamaguchi-Iwai, E.Sonoda, M.S.Sasaki, C.Morrison, T.Haraguchi, Y.Hiraoka,
Y.M.Yamashita, T.Yagi, M.Takata, C.Price, N.Kakazu, S.Takeda. Mre11 is
132
essential for the maintenance of chromosomal DNA in vertebrate cells, Embo
Journal 18 (1999) 6619-6629.
[102] J.Genschel, L.R.Bazemore, P.Modrich. Human exonuclease I is required for 5 '
and 3 ' mismatch repair, Journal of Biological Chemistry 277 (2002) 1330213311.
[103] J.Genschel, P.Modrich. Mechanism of 5 '-directed excision in human mismatch
repair, Molecular Cell 12 (2003) 1077-1086.
[104] L.Y.Gu, Y.Hong, S.McCulloch, H.Watanabe, G.M.Li. ATP-dependent interaction
of human mismatch repair proteins and dual role of PCNA in mismatch repair,
Nucleic Acids Research 26 (1998) 1173-1178.
[105] H.E.Kleczkowska, G.Marra, T.Lettieri, J.Jiricny. hMSH3 and hMSH6 interact
with PCNA and colocalize with it to replication foci, Genes & Development 15
(2001) 724-736.
[106] G.M.Li. Decoding the Histone Code: Role of H3K36me3 in Mismatch Repair and
Implications for Cancer Susceptibility and Therapy, Cancer Research 73 (2013)
6379-6383.
[107] C.R.Boland, S.N.Thibodeau, S.R.Hamilton, D.Sidransky, J.R.Eshleman,
R.W.Burt, S.J.Meltzer, M.A.Rodriguez-Bigas, R.Fodde, G.N.Ranzani,
S.Srivastava. A National Cancer Institute Workshop on Microsatellite Instability
for Cancer Detection and Familial Predisposition: Development of International
Criteria for the Determination of Microsatellite Instability in Colorectal Cancer,
Cancer Research 58 (1998) 5248-5257.
[108] K.C.Wei, A.B.Clark, E.Wong, M.F.Kane, D.J.Mazur, T.Parris, N.K.Kolas,
R.Russell, H.Hou, B.Kneitz, G.Yang, T.A.Kunkel, R.D.Kolodner, P.E.Cohen,
W.Edelmann. Inactivation of Exonuclease 1 in mice results in DNA mismatch
repair defects, increased cancer susceptibility, and male and female sterility,
Genes & Development 17 (2003) 603-614.
133
[109] H.T.Lynch, T.C.Smyrk, P.Watson, S.J.Lanspa, J.F.Lynch, P.M.Lynch,
R.J.Cavalieri, C.R.Boland. Genetics, Natural-History, Tumor Spectrum, and
Pathology of Hereditary Nonpolyposis Colorectal-Cancer - An Updated Review,
Gastroenterology 104 (1993) 1535-1549.
[110] J.C.Boyer, A.Umar, J.I.Risinger, J.R.Lipford, M.Kane, S.Yin, J.C.Barrett,
R.D.Kolodner, T.A.Kunkel. Microsatellite Instability, Mismatch Repair
Deficiency, and Genetic-Defects in Human Cancer Cell-Lines, Cancer Research
55 (1995) 6063-6070.
[111] M.F.Kane, M.Loda, G.M.Gaida, J.Lipman, R.Mishra, H.Goldman, J.M.Jessup,
R.Kolodner. Methylation of the hMLH1 promoter correlates with lack of
expression of hMLH1 in sporadic colon tumors and mismatch repair-defective
human tumor cell lines, Cancer Research 57 (1997) 808-811.
[112] J.G.Herman, A.Umar, K.Polyak, J.R.Graff, N.Ahuja, J.P.J.Issa, S.Markowitz,
J.K.V.Willson, S.R.Hamilton, K.W.Kinzler, M.F.Kane, R.D.Kolodner,
B.Vogelstein, T.A.Kunkel, S.B.Baylin. Incidence and functional consequences of
hMLH1 promoter hypermethylation in colorectal carcinoma, Proceedings of the
National Academy of Sciences of the United States of America 95 (1998) 68706875.
[113] L.Stojic, R.Brun, J.Jiricny. Mismatch repair and DNA damage signalling, Dna
Repair 3 (2004) 1091-1101.
[114] P.F.Swann, T.R.Waters, D.C.Moulton, Y.Z.Xu, Q.G.Zheng, M.Edwards, R.Mace.
Role of postreplicative DNA mismatch repair in the cytotoxic action of
thioguanine, Science 273 (1996) 1109-1111.
[115] P.Karran. Mechanisms of tolerance to DNA damaging therapeutic drugs,
Carcinogenesis 22 (2001) 1931-1937.
134
[116] M.T.Hawn, A.Umar, J.M.Carethers, G.Marra, T.A.Kunkel, C.R.Boland, M.Koi.
Evidence for A Connection Between the Mismatch Repair System and the G(2)
Cell-Cycle Checkpoint, Cancer Research 55 (1995) 3721-3725.
[117] N.Dewind, M.Dekker, A.Berns, M.Radman, H.T.Riele. Inactivation of the Mouse
Msh2 Gene Results in Mismatch Repair Deficiency, Methylation Tolerance,
Hyperrecombination, and Predisposition to Cancer, Cell 82 (1995) 321-330.
[118] S.D'Atri, L.Tentori, P.M.Lacal, G.Graziani, E.Pagani, E.Benincasa, G.Zambruno,
E.Bonmassar, J.Jiricny. Involvement of the mismatch repair system in
temozolomide-induced apoptosis, Molecular Pharmacology 54 (1998) 334-341.
[119] H.S.Friedman, S.P.Johnson, Q.Dong, S.C.Schold, B.K.A.Rasheed, S.H.Bigner,
F.AliOsman, E.Dolan, O.M.Colvin, P.Houghton, G.Germain, J.T.Drummond,
S.Keir, S.Marcelli, D.D.Bigner, P.Modrich. Methylator resistance mediated by
mismatch repair deficiency in a glioblastoma multiforme xenograft, Cancer
Research 57 (1997) 2933-2936.
[120] K.R.Tindall, W.E.Glaab, A.Umar, J.I.Risinger, M.Koi, J.C.Barrett, T.A.Kunkel.
Complementation of mismatch repair gene defects by chromosome transfer,
Mutation Research-Fundamental and Molecular Mechanisms of Mutagenesis 402
(1998) 15-22.
[121] M.Koi, A.Umar, D.P.Chauhan, S.P.Cherian, J.M.Carethers, T.A.Kunkel,
C.R.Boland. Human-Chromosome-3 Corrects Mismatch Repair Deficiency and
Microsatellite Instability and Reduces N-Methyl-N'-Nitro-N-Nitrosoguanidine
Tolerance in Colon-Tumor Cells with Homozygous Hmlh1 Mutation, Cancer
Research 54 (1994) 4308-4312.
[122] J.A.Mello, S.Acharya, R.Fishel, J.M.Essigmann. The mismatch-repair protein
hMSH2 binds selectively to DNA adducts of the anticancer drug cisplatin,
Chemistry & Biology 3 (1996) 579-589.
135
[123] D.R.Duckett, J.T.Drummond, A.I.H.Murchie, J.T.Reardon, A.Sancar, D.M.Lilley,
P.Modrich. Human MutS alpha recognizes damaged DNA base pairs containing
O-6-methylguanine, O-4-methylthymine, or the cisplatin-d(GpG) adduct,
Proceedings of the National Academy of Sciences of the United States of America
93 (1996) 6443-6447.
[124] A.Vaisman, M.Varchenko, A.Umar, T.A.Kunkel, J.I.Risinger, J.C.Barrett,
T.C.Hamilton, S.G.Chaney. The role of hMLH1, hMSH3, and hMSH6 defects in
cisplatin and oxaliplatin resistance: Correlation with replicative bypass of
platinum-DNA adducts, Cancer Research 58 (1998) 3579-3585.
[125] D.Fink, H.Zheng, S.Nebel, P.S.Norris, S.Aebi, T.P.Lin, A.Nehme, R.D.Christen,
M.Haas, C.L.MacLeod, S.B.Howell. In vitro and in vivo resistance to cisplatin in
cells that have lost DNA mismatch repair, Cancer Research 57 (1997) 1841-1845.
[126] N.Mojas, M.Lopes, J.Jiricny. Mismatch repair-dependent processing of
methylation damage gives rise to persistent single-stranded gaps in newly
replicated DNA, Genes & Development 21 (2007) 3342-3355.
[127] P.Karran, M.Bignami. Dna-Damage Tolerance, Mismatch Repair and Genome
Instability, Bioessays 16 (1994) 833-839.
[128] T.Yan, S.E.Berry, A.B.Desai, T.J.Kinsella. DNA mismatch repair (MMR)
mediates 6-thioguanine genotoxicity by introducing single-strand breaks to signal
a G2-M arrest in MMR-proficient RKO cells, Clin. Cancer Res. 9 (2003) 23272334.
[129] Y.Wang, J.Qin. MSH2 and ATR form a signaling module and regulate two
branches of the damage response to DNA methylation, Proceedings of the
National Academy of Sciences of the United States of America 100 (2003)
15387-15392.
136
[130] N.Pabla, Z.Ma, M.A.McIlhatton, R.Fishel, Z.Dong. hMSH2 Recruits ATR to
DNA Damage Sites for Activation during DNA Damage-induced Apoptosis,
Journal of Biological Chemistry 286 (2011) 10411-10418.
[131] A.G.Schroering, K.J.Williams. Rapid induction of chromatin-associated DNA
mismatch repair proteins after MNNG treatment, Dna Repair 7 (2008) 951-969.
[132] M.Meyers, M.Theodosiou, S.Acharya, E.Odegaard, T.Wilson, J.E.Lewis,
T.W.Davis, C.W.VanPatten, R.Fishel, D.A.Boothman. Cell cycle regulation of
the human DNA mismatch repair genes hMSH2, hMLH1, and hPMS2, Cancer
Research 57 (1997) 206-208.
[133] D.Mu, M.Tursun, D.R.Duckett, J.T.Drummond, P.Modrich, A.Sancar.
Recognition and repair of compound DNA lesions (base damage and mismatch)
by human mismatch repair and excision repair systems, Molecular and Cellular
Biology 17 (1997) 760-769.
[134] Z.Hong, J.Jiang, K.Hashiguchi, M.Hoshi, L.Lan, A.Yasui. Recruitment of
mismatch repair proteins to the site of DNA damage in human cells, Journal of
Cell Science 121 (2008) 3146-3154.
[135] E.Avdievich, C.Reiss, S.J.Scherer, Y.Zhang, S.M.Maier, B.Jin, H.Hou, Jr.,
A.Rosenwald, H.Riedmiller, R.Kucherlapati, P.E.Cohen, W.Edelmann, B.Kneitz.
Distinct effects of the recurrent Mlh1(G67R) mutation on MMR functions,
cancer, and meiosis, Proceedings of the National Academy of Sciences of the
United States of America 105 (2008) 4247-4252.
[136] G.Z.Yang, S.J.Scherer, S.S.Shell, K.Yang, M.Kim, M.Lipkin, R.Kucherlapati,
R.D.Kolodner, W.Edelmann. Dominant effects of an Msh6 missense mutation on
DNA repair and cancer susceptibility, Cancer Cell 6 (2004) 139-150.
[137] I.Marinovic-Terzic, A.Yoshioka-Yamashita, H.Shimodaira, E.Avdievich,
I.C.Hunton, R.D.Kolodner, W.Edelmann, J.Y.Wang. Apoptotic function of
human PMS2 compromised by the nonsynonymous single-nucleotide
137
polymorphic variant R20Q, Proceedings of the National Academy of Sciences of
the United States of America 105 (2008) 13993-13998.
[138] J.E.Clodfelter, M.B.Gentry, K.Drotschmann. MSH2 missense mutations alter
cisplatin cytotoxicity and promote cisplatin-induced genome instability, Nucleic
Acids Research 33 (2005) 3323-3330.
[139] H.Shimodaira, A.Yoshioka-Yamashita, R.D.Kolodner, J.Y.J.Wang. Interaction of
mismatch repair protein PMS2 and the p53-related transcription factor p73 in
apoptosis response to cisplatin, Proceedings of the National Academy of Sciences
of the United States of America 100 (2003) 2420-2425.
[140] R.P.Topping, J.C.Wilkinson, K.D.Scarpinato. Mismatch Repair Protein
Deficiency Compromises Cisplatin-induced Apoptotic Signaling, Journal of
Biological Chemistry 284 (2009) 14029-14039.
[141] G.Aquilina, S.Ceccotti, S.Martinelli, S.Soddu, M.Crescenzi, P.Branch, P.Karran,
M.Bignami. Mismatch repair and p53 independently affect sensitivity to N-(2chloroethyl)-N'-cyclohexyl-N-nitrosourea, Clin. Cancer Res. 6 (2000) 671-680.
[142] M.Yamada, E.ORegan, R.Brown, P.Karran. Selective recognition of a cisplatinDNA adduct by human mismatch repair proteins, Nucleic Acids Research 25
(1997) 491-495.
[143] J.M.Wagner, L.M.Karnitz. Cisplatin-Induced DNA Damage Activates Replication
Checkpoint Signaling Components that Differentially Affect Tumor Cell
Survival, Molecular Pharmacology 76 (2009) 208-214.
[144] C.S.Sorensen, L.T.Hansen, J.Dziegielewski, R.G.Syljuasen, C.Lundin, J.Bartek,
T.Helleday. The cell-cycle checkpoint kinase Chk1 is required for mammalian
homologous recombination repair, Nature Cell Biology 7 (2005) 195-U121.
138
[145] X.Z.Wang, R.D.Kennedy, K.Ray, P.Stuckert, T.Ellenberger, A.D.D'Andrea.
Chk1-mediated phosphorylation of FANCE is required for the Fanconi
anemia/BRCA pathway, Molecular and Cellular Biology 27 (2007) 3098-3108.
[146] K.A.Lewis, K.K.Lilly, E.A.Reynolds, W.P.Sullivan, S.H.Kaufmann, W.A.Cliby.
Ataxia telangiectasia and rad3-related kinase contributes to cell cycle arrest and
survival after cisplatin but not oxaliplatin, Mol. Cancer Ther. 8 (2009) 855-863.
[147] M.D.Rainey, M.E.Charlton, R.V.Stanton, M.B.Kastan. Transient inhibition of
ATM kinase is sufficient to enhance cellular sensitivity to ionizing radiation,
Cancer Research 68 (2008) 7466-7474.
[148] M.Kartalou, J.M.Essigmann. Recognition of cisplatin adducts by cellular proteins,
Mutat. Res. 478 (2001) 1-21.
[149] S.G.Chaney, S.L.Campbell, E.Bassett, Y.B.Wu. Recognition and processing of
cisplatin- and oxaliplatin-DNA adducts, Critical Reviews in Oncology
Hematology 53 (2005) 3-11.
[150] J.S.Hoffmann, M.J.Pillaire, D.GarciaEstefania, S.Lapalu, G.Villani. In vitro
bypass replication of the cisplatin-d(GpG) lesion by calf thymus DNA polymerase
beta and human immunodeficiency virus type I reverse transcriptase is highly
mutagenic, Journal of Biological Chemistry 271 (1996) 15386-15392.
[151] A.Vaisman, S.G.Chaney. The efficiency and fidelity of translesion synthesis past
cisplatin and oxaliplatin GpG adducts by human DNA polymerase beta, Journal
of Biological Chemistry 275 (2000) 13017-13025.
[152] R.N.Trivedi, X.H.Wang, E.Jelezcova, E.M.Goellner, J.B.Tang, R.W.Sobol.
Human methyl purine DNA glycosylase and DNA polymerase beta expression
collectively predict sensitivity to temozolomide, Molecular Pharmacology 74
(2008) 505-516.
139
[153] S.Arora, A.Kothandapani, K.Tillison, V.Kalman-Maltese, S.M.Patrick.
Downregulation of XPF-ERCC1 enhances cisplatin efficacy in cancer cells, Dna
Repair 9 (2010) 745-753.
[154] I.U.De Silva, P.J.Mchugh, P.H.Clingen, J.A.Hartley. Defects in interstrand crosslink uncoupling do not account for the extreme sensitivity of ERCC1 and XPF
cells to cisplatin, Nucleic Acids Research 30 (2002) 3848-3856.
[155] A.Kothandapani, A.Sawant, V.S.M.N.Dangeti, R.W.Sobol, S.M.Patrick. Epistatic
role of base excision repair and mismatch repair pathways in mediating cisplatin
cytotoxicity, Nucleic Acids Research 41 (2013) 7332-7343.
[156] P.L.Olive, J.P.Banath. Kinetics of H2AX Phosphorylation After Exposure to
Cisplatin, Cytometry Part B-Clinical Cytometry 76B (2009) 79-90.
[157] P.Clingen, J.Wu, J.Miller, N.Mistry, F.Chin, P.Wynne, K.Prise, J.Hartley. Histone
H2AX phosphorylation as a molecular pharmacological marker for DNA
interstrand crosslink cancer chemotherapy, Biochemical Pharmacology 76 (2008)
19-27.
[158] K.L.Menge, Z.Hostomsky, B.R.Nodes, G.O.Hudson, S.Rahmati, E.W.Moomaw,
R.J.Almassy, Z.Hostomska. Structure-function analysis of the mammalian DNA
polymerase beta active site: Role of aspartic acid 256, arginine 254, and arginine
258 in nucleotidyl transfer, Biochemistry 34 (1995) 15934-15942.
[159] L.Y.Gu, J.X.Wu, L.Qiu, C.D.Jennings, G.M.Li. Involvement of DNA mismatch
repair in folate deficiency-induced apoptosis, Journal of Nutritional Biochemistry
13 (2002) 355-363.
[160] P.M.Reaper,
M.R.Griffiths,
J.M.Long,
J.D.Charrier,
S.MacCormick,
P.A.Charlton, J.M.C.Golec, J.R.Pollard. Selective killing of ATM- or p53deficient cancer cells through inhibition of ATR, Nature Chemical Biology 7
(2011) 428-430.
140
[161] J.J.Turchi, K.M.Henkels, Y.Zhou. Cisplatin-DNA adducts inhibit translocation of
the Ku subunits of DNA-PK, Nucleic Acids Research 28 (2000) 4634-4641.
[162] J.Dejmek, J.D.Iglehart, J.B.Lazaro. DNA-Dependent Protein Kinase (DNA-PK)Dependent Cisplatin-Induced Loss of Nucleolar Facilitator of Chromatin
Transcription (FACT) and Regulation of Cisplatin Sensitivity by DNA-PK and
FACT, Molecular Cancer Research 7 (2009) 581-591.
[163] R.Jensen, P.M.Glazer. Cell-interdependent cisplatin killing by Ku/DNAdependent protein kinase signaling transduced through gap junctions, Proceedings
of the National Academy of Sciences of the United States of America 101 (2004)
6134-6139.
[164] W.M.Ciszewski, M.Tavecchio, J.Dastych, N.J.Curtin. DNA-PK inhibition by
NU7441 sensitizes breast cancer cells to ionizing radiation and doxorubicin,
Breast Cancer Research and Treatment 143 (2014) 47-55.
[165] R.R.Iyer, A.Pluciennik, V.Burdett, P.L.Modrich. DNA mismatch repair:
Functions and mechanisms, Chemical Reviews 106 (2006) 302-323.
[166] J.Y.J.Wang, W.Edelmann. Mismatch repair proteins as sensors of alkylation DNA
damage, Cancer Cell 9 (2006) 417-418.
[167] E.Pani, L.Stojic, M.El-Shemerly, J.Jiricny, S.Ferrari. Mismatch repair status and
the response of human cells to cisplatin, Cell Cycle 6 (2007) 1796-1802.
[168] D.Fink, S.Aebi, S.B.Howell. The role of DNA mismatch repair in drug resistance,
Clinical Cancer Research 4 (1998) 1-6.
[169] J.Jiricny. The multifaceted mismatch-repair system, Nature Reviews Molecular
Cell Biology 7 (2006) 335-346.
141
[170] M.Takahashi, M.Koi, F.Balaguer, C.Boland, A.Goel. MSH3 Mediates
Sensitization of Colorectal Cancer Cells to Cisplatin, Oxaliplatin, and a
Poly(ADP-ribose) Polymerase Inhibitor, Journal of Biological Chemistry 286
(2011).
[171] J.M.Park, S.Huang, D.Tougeron, F.A.Sinicrope. MSH3 Mismatch Repair Protein
Regulates Sensitivity to Cytotoxic Drugs and a Histone Deacetylase Inhibitor in
Human Colon Carcinoma Cells, Plos One 8 (2013) e65369.
[172] L.Tentori, A.Muzi, A.S.Dorio, S.Dolci, F.Campolo, P.Vernole, P.M.Lacal,
F.Praz, G.Graziani. MSH3 expression does not influence the sensitivity of colon
cancer HCT116 cell line to oxaliplatin and poly(ADP-ribose) polymerase (PARP)
inhibitor as monotherapy or in combination, Cancer Chemother. Pharmacol. 72
(2013) 117-125.
[173] Z.Z.Zdraveski, J.A.Mello, C.K.Farinelli, J.M.Essigmann, M.G.Marinus. MutS
preferentially recognizes cisplatin- over oxaliplatin-modified DNA, Journal of
Biological Chemistry 277 (2002) 1255-1260.
[174] A.Eastman, N.Schulte. Enhanced Dna-Repair As A Mechanism of Resistance to
Cis-Diamminedichloroplatinum(Ii), Biochemistry 27 (1988) 4730-4734.
[175] Q.Wu, K.M.Vasquez. Human MLH1 Protein Participates in Genomic Damage
Checkpoint Signaling in Response to DNA Interstrand Crosslinks, while MSH2
Functions in DNA Repair, Plos Genetics 4 (2008).
[176] Q.Wu, L.A.Christensen, R.J.Legerski, K.M.Vasquez. Mismatch repair
participates in error-free processing of DNA interstrand crosslinks in human cells,
Embo Reports 6 (2005) 551-556.
[177] C.E.Bronner, S.M.Baker, P.T.Morrison, G.Warren, L.G.Smith, M.K.Lescoe,
M.Kane, C.Earabino, J.Lipford, A.Lindblom, P.Tannergard, R.J.Bollag,
142
A.R.Godwin, D.C.Ward, M.Nordenskjold, R.Fishel, R.Kolodner, R.M.Liskay.
Mutation in the Dna Mismatch Repair Gene Homolog Hmlh1 Is Associated with
Hereditary Nonpolyposis Colon-Cancer, Nature 368 (1994) 258-261.
[178] A.R.Ellison, J.Lofing, G.A.Bitter. Human MutL homolog (MLH1) function in
DNA mismatch repair: a prospective screen for missense mutations in the ATPase
domain, Nucleic Acids Research 32 (2004) 5321-5338.
143