Mutat Res Gen Tox En xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Mutat Res Gen Tox En

journal homepage: www.elsevier.com/locate/gentox

Minireview

Genetic variation of acquired structural chromosomal aberrations

⁎
Pavel Vodickaa,b,c, , Ludovit Musakd, Ludmila Vodickovaa,b,c, Sona Vodenkovaa,b,e,
Calogerina Catalanof, Michal Kroupaa,c, Alessio Naccaratia,g, Zdena Polivkovae,
Veronika Vymetalkovaa,b,c, Asta Förstif,h, Kari Hemminkif,h
a
Department of Molecular Biology of Cancer, Institute of Experimental Medicine, Czech Academy of Sciences, Prague, 14220, Czech Republic
b
Institute of Biology and Medical Genetics, First Faculty of Medicine, Charles University, Prague, 12800, Czech Republic
c
Faculty of Medicine and Biomedical Center in Pilsen, Charles University, Pilsen, 30605, Czech Republic
d
Biomedical Center Martin, Comenius University in Bratislava, Jessenius Faculty of Medicine, Martin, 03601, Slovakia
e
Department of Medical Genetics, Third Faculty of Medicine, Charles University, Prague, 10000, Czech Republic
f
Division of Molecular Genetic Epidemiology, German Cancer Research Center (DKFZ), Heidelberg, D69120, Germany
g
Italian Institute for Genomic Medicine (IIGM), Torino, 10126, Italy
h
Center for Primary Health Care Research, Lund University, Malmö, 214 28, Sweden

A R T I C LE I N FO A B S T R A C T

Keywords: Human malignancies are often hallmarked with genomic instability, which itself is also considered a causative
Chromosomal aberrations event in malignant transformation. Genomic instability may manifest itself as genetic changes in the nucleotide
Genetics sequence of DNA, or as structural or numerical changes of chromosomes. Unrepaired or insufficiently repaired
Xenobiotic metabolizing enzymes DNA double-strand breaks, as well as telomere shortening, are important contributors in the formation of
DNA repair
structural chromosomal aberrations (CAs). In the present review, we discuss potential mechanisms behind the
Mitotic checkpoints
Cyclin D1
formation of CAs and their relation to cancer. Based on our own studies, we also illustrate how inherited genetic
variation may modify the frequency and types of CAs occurring in humans. Recently, we published a series of
studies on variations in genes relevant to maintaining genomic integrity, such as those encoding xenobiotic-
metabolising enzymes, DNA repair, the tumour suppressor TP53, the spindle assembly checkpoint, and cyclin D1
(CCND1). While individually genetic variation in these genes exerted small modulating effects, in interactions
they were associated with CA frequencies in peripheral blood lymphocytes of healthy volunteers. Moreover, we
observed opposite associations between the CCND1 splice site polymorphism rs9344 G870A and the frequency of
CAs compared to their association with translocation t(11,14). We discuss the functional consequences of the
CCND1 gene in interplay with DNA damage response and DNA repair during malignant transformation.
Our review summarizes existing evidence that gene variations in relevant cellular pathways modulate the
frequency of CAs, predominantly in a complex interaction. More functional/mechanistic studies elucidating
these observations are required. Several questions emerge, such as the role of CAs in malignancies with respect
to a particular phenotype and heterogeneity, the formation of CAs during the process of malignant transfor-
mation, and the formation of CAs in individual types of lymphocytes in relation to the immune response.

1. Introduction malignant transformation or may directly trigger cancer onset, as in

case of chronic myeloid leukaemia.
Human cancers are often associated with chromosomal instability, Structural CAs may be specific, such as translocations and inver-
including both numerical and structural chromosomal aberrations sions, or nonspecific, such as chromatid breaks, fragmented or missing
(CAs) [1–3]; these are also considered causative events in malignant parts of chromosomes, and fusion resulting in dicentric and ring chro-
transformation [4]. Numerical chromosome variance refers to a mul- mosomes [5]. The former are often recurrent and they are analysed by
tiple of haploid sets or any non-euploid deviation in chromosome molecular cytogenetic methods (sequencing, fluorescent in situ hy-
number. Numerical aberrations are often incompatible with cell viabi- bridization). The later are scored by classical cytogenetic techniques,
lity. On the contrary, structural CAs are commonly presented in viable able to recognise chromosome type aberrations (CSAs) and chromatid
human cells. Structural CAs may be a secondary consequence of type aberrations (CTAs) according to morphological changes [6].

⁎
Corresponding author.
E-mail address: pvodicka@biomed.cas.cz (P. Vodicka).

https://doi.org/10.1016/j.mrgentox.2018.05.014
Received 22 January 2018; Received in revised form 24 April 2018; Accepted 10 May 2018
1383-5718/ © 2018 Elsevier B.V. All rights reserved.

Please cite this article as: Vodicka, P., Mutat Res Gen Tox En (2018), https://doi.org/10.1016/j.mrgentox.2018.05.014
P. Vodicka et al. Mutat Res Gen Tox En xxx (xxxx) xxx–xxx

In the present review, we first discuss the potential mechanisms involved 337 different genes. While gene fusions (i.e., the juxta posi-
behind the formation of structural CAs and their relation to cancer. tioning of two genes leading to the translation of a deregulated and/or
Later on the basis of available studies, we describe how inherited var- chimeric protein) were initially identified; and their clinical value was
iations in genes relevant to important cellular pathways (such as those recognized in malignant haematological disorders, they have become
encoding xenobiotic-metabolising enzymes, DNA repair, the tumour increasingly important also in solid tumors [26]. Mitelman et al. pre-
suppressor TP53, spindle assembly checkpoint and cyclin D1) modify dicted in 2007 that fusion genes account for 20% of human cancer
the frequency and types of CAs in humans. Genetic variation in relevant morbidity. In a more recent review from year 2015, they reported on a
genes has been shown to modulate inter-individual variability in their huge increase in identified fusion genes, amounting to 9000 during the
ultimate functions. Moreover, the knowledge of the full effects of gen- covered 3 years [27]. Currently the number comprises 11,124 identified
e–gene interactions on chromosomal aberrations remains scarce. fusion genes (Mitelman database of chromosomal aberrations and gene
fusions in cancer; https://cgap.nci.nih.gov/chromosomes/Mitelman).
The evidence linking nonspecific CAs with cancer is not as over-
1.1. Formation of structural CA and its relation to cancer risk
whelming as it is for specific, recurrent CAs. However, an increase in
the frequency of CAs has been found in incident cancer patients thus
Unrepaired or insufficiently repaired DNA double-strand breaks
closely linking CAs with cancer development [28,29]. Some of the CAs
(DSBs) as well as telomerase dysfunction are substantial players in the
observed are also generated during cancer development, nevertheless
formation of structural CAs. Morphologically distinct types of CSAs and
CA frequency in PBLs is considered to be an early marker of cancer
CTAs emerge depending on cell cycle stage and the type of DNA da-
susceptibility, based on the hypothesis that genetic damage in PBLs
mage: whereas CSAs arise during G0 or G1 phase on a basis of DSBs
reflects similar damage in other body cells undergoing carcinogenesis
induced by genotoxic damage, CTAs are formed due to insufficiently
[30].
repaired DSBs during late S or G2 phase of the cell cycle [5,7,8]. Taken
together in both mechanisms of CA formation, the repair of DNA da-
mage plays a critical role [7], and, as was recently postulated, telomere
1.2. Genetic predisposition to CAs
shortening also acts as an important contributor to CA formation
(Fig. 1). Telomeres are nucleoprotein structures composed of TTAGGG
There are only limited reports exploring the inherited genetic basis
tandem repeats that cap ends of eukaryotic chromosomes. The telomere
of acquired structural CAs. To achieve this goal a genome-wide asso-
complex regulates a “cellular mitotic clock” and protects chromosomes
ciation study (GWAS) may substantially contribute with an aim to
against exonucleolytic degradation, DNA damage and chromosomal
finding novel genetic variants associated with CAs, and potentially with
instability [9–11]. There is growing evidence of shorter telomeres being
cancer risk, to elucidate the possible functional effects of these variants
associated with the increased frequency of CAs in peripheral blood
by in silico predictions.
lymphocytes (PBLs), in particular with CSAs [12,13]. Telomere short-
However, through a Czech-Slovak-German collaboration, we re-
ening, long hypothesized as being associated with cancer development
cently published a series of studies related to DNA repair [47], meta-
[14,15], induces genomic and chromosomal instability [16,17]. Telo-
bolism [38] and mitotic checkpoint [72] (overviewed separately). We
mere becomes shorter at each round of replication, and critically
compared genotype frequencies between healthy volunteers whose
shortened telomeres may be poorly end-capped and recognized as DSB
total CA frequency was > 2% and those with lower frequency of CAs;
by repair machinery [18]. These processes may underlie end-to-end
for CSAs and CTAs the threshold was 1%. The numbers of volunteers
chromosome fusions, the initiation of breakage-fusion-breakage cycles,
genotyped ranged from 282 to 2200 depending on the SNP and DNA
and the development of numerical and structural CAs [19–22]. On the
availability (Table 2). All SNPs were individually genotyped and mul-
other hand, the reactivation of telomerase stabilizes telomeres and
tivariable logistic regression analyses adjusted for age, sex, occupa-
immortalizes premalignant cells, thus enabling cancer progression
tional exposure and smoking; conducted for total CAs, CTAs, and CSAs.
[23,24]. In this context, our recent report suggested that altered DSB
SNPs were considered individually, and in all possible pairwise com-
repair, measured by sensitivity towards mutagen bleomycin in PBLs,
binations within the three sets of genes. Those studies are reviewed
occurs particularly in colorectal carcinogenesis. Irrespective of cancer
below, but they appear to share some overall features: individual SNPs
type telomere shortening is associated with a decreased capacity to
had weak associations, but gene–gene interactions were common and
repair DSB [25].
often involved partners from different pathways.
Many human cancers and neoplastic cells exhibit chromosomal
abnormalities [4]. In their review from 2007, Mitelman et al. sum-
marized the role of translocations and related fusion genes in the in-
itiation of cancer. Their catalogue included 358 gene fusions, of which

Fig. 1. Single-stranded DNA damage (a consequence of oxidative

DNA damage in this particular case) if unrepaired gives rise to
DSBs and CTAs (as assayed for by chemiluminescent measurement
of γ-H2AX phosphorylation). Shortened telomeres may be poorly
end-capped and recognized as DSB by repair machinery, resulting
in the development of numerical and structural CAs.

2
P. Vodicka et al. Mutat Res Gen Tox En xxx (xxxx) xxx–xxx

Table 1
Background variables of age, smoking status, and occupational exposure ([34,45,70]) plus polymorphism rs1042522:G4C (Ex41119 G4C, Arg72Pro) in gene TP53 in
relation to chromosomal damage.
Variable Persons CAtot CSA CTA

95% 95% 95%

Odds Confidence Odds Confidence Odds Confidence
ratio interval P-value ratio interval P-value ratio interval P-value

CAtot (cases/control) 951/1245

CTA (cases/control) 1042/1154
CSA (cases/control) 983/1213
Age (min, max, mean) 18, 88, 43 1.06a 1.00–1.13 0.07 1.07 1.00–1.13 0.04 0.95 0.90–1.01 0.11
Occupational environment (exposed/unexposed) 1207/989 2.36 1.97–2.83 < 0.01 1.73 1.45–2.06 < 0.01 1.64 1.38–1.96 < 0.01
Sex (M/F) 1171/1025 1.03 0.86–1.23 0.77 1.05 0.88–1.26 0.59 0.83 0.69–0.99 0.04
Smoker (S/NS) 614/1557 1.19 0.97–1.45 0.09 0.95 0.78–1.16 0.63 1.13 0.93–1.38 0.23
Arg72Argb 278 – – – – – – – – –
Arg72Pro 269 0.91 0.64–1.29 0.60 1.14 0.81–1.62 0.45 1.11 0.79–1.58 0.52
Pro72Pro 46 1.03 0.54–1.96 0.93 0.82 0.92–1.60 0.55 1.42 0.76–2.66 0.27
Arg72Arg vs. Arg72Pro + Pro72Pro – 0.93 0.66–1.30 0.66 1.03 0.73–1.44 0.88 1.16 0.84–1.61 0.37

a
ORs for age were calculated for 10 year age difference.
b
rs1042522:G4C (Ex41119 G4C, Arg72Pro).

Table 2
Polymorphisms in relevant genes of important cellular pathways (encoding xenobiotic-metabolising enzymes, DNA repair, the tumour suppressor TP53, spindle
assembly checkpoint, and cyclin D1) evaluated in this study. The table additionally refers to the in silico predicted functional consequence of polymorphisms.
Genes SNP ID Amino Acid Alleles (major/ NCBI Chromosome Location Function* Subjects with
substitution minor) CAs

BER
XRCC1 rs1799782 Arg194Trp C >T T > C 19q13.2 missense deleterious 639
XRCC1 rs25489 Arg280His G >A G > A 19q13.2 missense possibly deleterious 355
XRCC1 rs25487 Arg399Gln G >A A > G 19q13.2 missense benign 1760
OGG1 rs1052133 Ser326Cys C >G A > G 3p26.2 missense deleterious 1766
APEX1 (APE1) rs1130409 Asp148Glu T >G T > G/A 14q11.2 missense ambiguous 1299
NER
XPA rs1800975 G > A 9q22.3 5′UTR 1443
XPD rs13181 Lys751Gln G > A C>A 19q13.3 missense deleterious 1777
XPG rs17655 Asp1104His C > G C>G 13q33 missense deleterious 1756
XPC rs2228001 Lys939Gln A > C C>A 3p25 missense benign 1772
DSB
XRCC2 rs3218536 Arg188His G > A G > A 7q36.1 missense benign 813
XRCC3 rs861539 Thr241Met C > T C > T 14q32.3 missense deleterious 1715
NBN (NBS1) rs1805794 Glu185Gln C > G C > G 8q21 missense benign 966
RAD54L rs1048771 Ala730Ala C > T C > T/A 1p32 synonymous splicing regulation 282
DDR
TP53 rs1042522 Arg72Pro G>C C>G 17p13.1 missense deleterious 593
CCND1 rs9344 Pro241Pro G>A G>A 11q13.3 synonymous splicing regulation 730
XME
CYP1B1/432 rs1056836 Leu432Val C>G 2p22.2 missense drug response 549
CYP1B1/453 rs1800440 Asn453Ser A >G 2p22.2 missense benign/likely benign 550
EPHX1 rs1051740 Tyr113His T>C 1q42.12 missense drug-response 1120
EPHX1 rs2234922 His139Arg A >G 1q42.12 missense drug-response 1120
NQO1 rs1800566 Pro187Ser C>T 16q22.1 missense pathogenic 1018
GSTP1 rs1695 Ile105Val A >G 11q13.2 missense drug-response 1493
GSTM1 allele deletion 1p13.3 deleterious 1499
GSTT1 allele deletion 22q11.23 deleterious 1489
SAC
BUB1B rs1801376 Gln349Arg A >G 15q15.1 missense benign 330
BUB3 rs3808960 G >T 10q26.13 promoter region upstream variant 2KB 330
MAD2L1 rs903147 A >C 4q27 promoter region upstream variant 2KB 330
CENPF rs438034 Arg2943Gly G >A 1q41 missense missense 618
ESPL1 rs6580941 C >T 12q13.13 promoter region intron variant, upstream variant 2KB 656
NEK2 rs701928 T >A 1q32.3 promoter region intron variant, upstream variant 2KB 663
PTTG1 rs1862392 T >A 5q33.3 promoter region upstream variant 2KB 729
ZWILCH rs3087660 A >G 15q22.31 5′UTR nc transcript variant, upstream variant 674
2KB, utr variant 5 prime
ZWINT rs2241666 Arg187Gly G>A 10q21.1 missense intron variant, missense, nc transcript 662
variant

*SIFT, PolyPhen-2-algorithms; SNP single nucleotide polymorphism; BER base excision repair pathway; NER nucleotide excision repair pathway; DSB double-strand
break repair pathway; DDR DNA damage response pathway; SAC spindle assembly checkpoint genes XME xenobiotic-metabolising enzymes; Polymorphisms were
studied on subjects with CAs; their numbers are shown in the right column.

3
P. Vodicka et al. Mutat Res Gen Tox En xxx (xxxx) xxx–xxx

Fig. 2. Influence of metabolic gene polymorphisms on CAs in healthy volunteers [36]. The colour code indicates total CA (dark blue), CSA (blue), CTA (light blue),
and CSA & CTA (blue gradient). The arrows indicate significant gene–gene interactions.

1.3. CAs and variants in genes encoding xenobiotic-metabolising enzymes 1.4. CAs and variants in DNA repair
(XMEs)
Since human DNA is constantly exposed to physical (ultraviolet,
Only a limited number of reports have analysed effects of genetic ionizing radiation) and chemical (reactive oxygen species, alkylating,
predispositions on inter-individual variability in DNA and chromosomal and aralkylating) damaging agents, the arising damage (particularly
damage [31–35]. In a recent study, we tested the associations of var- DSBs) may be critical for establishing persistent chromosomal altera-
iants in genes encoding XMEs with CAs among the same cohort of vo- tions. CAs in PBLs reflect therefore inter-individual sensitivity to many
lunteers described above [36]. We selected 5 genes for analysis: cyto- genotoxic substances and serve as biomarkers of an early effect of
chrome P450 1B1 (CYP1B1, 2 separate polymorphisms), epoxide genotoxic carcinogens and carcinogenic risk [29,41–45]. Efficient DNA
hydrolase 1 (EPHX1), NAD(P)H:quinone oxidoreductase 1 (NQO1), each repair machinery, comprising several distinct pathways, effectively
coding for phase 1 enzymes, and glutathione S-transferase P1 (GSTP1), maintains genomic integrity. Alterations in the DNA repair increase the
glutathione S-transferases M1 (GSTM1), and T1 (GSTT1). All the 7 vulnerability of the cells, resulting in an accumulation of mutations in
polymorphisms tested in our study have previously been shown to be the genome, and may ultimately lead to tumorigenesis [46].
functionally relevant [32–34,48]. Cytochromes P450 are the most im- Individual DNA repair capacity in response to DNA damage, effec-
portant enzymes involved in the phase I biotransformation of various tively preventing an accumulation of CAs, is often modulated by the
compounds. They catalyse a number of reactions modifying dietary, gene variants in different DNA repair pathways [47]. Indeed, we
smoking, and environment-derived pre-carcinogens, and participate in documented a correlation between gene variants involved in base ex-
the metabolism of endogenous compounds including hormones and bile cision repair (BER) and the corresponding BER repair capacity a decade
acids. CYP1B1 is involved in activating polycyclic aromatic hydro- ago [48]. However, single-nucleotide polymorphisms (SNPs) in non-
carbons (PAHs) or heterocyclic amines to reactive metabolites that coding regions and even changes in wobble bases that do not affect
cause DNA damage [37]. Epoxide hydrolase (EPHX1) converts reactive amino acid sequence, may be important as well [49].
intermediates (epoxides) to less reactive, excretable diols. NAD(P) The DNA repair genes that we studied encode proteins involved in
H:quinone oxidoreductase (NQO1) is a two-electron reductase involved BER (represented by XRCC1, hOGG1 and APE1), NER (XPA, XPC, XPD
in the biotransformation of quinones [39]. Glutathione S-transferases and XPG) and in double-strand break repair DSB (XRCC2, XRCC3, NBS1
(GST), GSTM1, GSTP1, and GSTT1, belong to the most frequently stu- and RAD54L) [47]. Excision repair (both BER and NER) represents a
died XMEs in molecular epidemiology studies, playing a protective role crucial mechanism for a cells to preserve genomic integrity, since DSBs
against electrophilic chemicals. However, their outcomes in relation to or permanent mutations can arise as a consequence of unrepaired or
CAs are often controversial: while Musak et al. observed significant erroneously repaired single-stranded damage. BER removes small mo-
association of CA frequency with GSTT1-null genotype, particularly in lecular lesions and oxidative DNA damage, in which XRCC1 protein acts
smoking tire plant workers [33], Rossi et al. did not record such an as a coordinator of single-strand break repair proteins, hOGG1 is a
association [35]. In our analysis, we did not observe any modulating glycosylase removing oxidative DNA adducts, and APE1 acts as en-
effect of GST polymorphisms on CA frequency, as assayed on 488 donuclease toward single-strand breaks. NER is a mechanism used by
healthy individuals [40]. Our study describing the associations between mammals for the elimination of bulky DNA lesions. Its importance is
functional polymorphisms in 6 genes encoding XME and CAs revealed well documented on xeroderma pigmentosum or Cockayne syndrome
that only EPHX1 was individually associated with total CAs: high ac- disorders [50], associated with extreme UV light sensitivity and en-
tivity genotypes decreased CA frequency [33]. Tentative association of ormous skin cancer risk [51]. XPC (xeroderma pigmentosum com-
CAs with EPHX1 polymorphisms has been shown by us earlier plementation group C) protein participates in the recognition of lesions
[33,37,38]. We also tested pairwise interactions between all 7 variants. during global genomic-NER [52] – XPC enables the recruitment of two
A total of 12 nominally significant interactions were observed, 6 of helicases – while XPB and XPD initiate the opening of the DNA around
them at P < 0.01 [36]. In most of the interactions, a GST variant was the lesion [53,54]. XPA and XPG proteins subsequently form a stable
one partner of the pair (Fig. 2). The data provided evidence that var- open structure and recruitment of ERCC1-XPF by XPA initiates the dual
iants in genes coding for metabolic enzymes (which individually have incision of the DNA 5′ and 3′ to the lesion by ERCC1-XPF and XPG,
small effects) interact and are associated with CA frequencies in PBLs of respectively [55,56]. The replication machinery subsequently fills the
healthy volunteers. gap and seals the nick.

4
P. Vodicka et al. Mutat Res Gen Tox En xxx (xxxx) xxx–xxx

Fig. 3. Influence of DNA repair gene polymorphisms on CAs in healthy volunteers [47]. The colour code indicates total CA (dark blue), CSA (blue), CTA (light blue),
and CSA & CTA (blue gradient). The arrows indicate significant gene–gene interactions.

DSB repair is a crucial mechanism for a cell to deal with the most frequencies suggesting that accumulation of CAs requires complex in-
serious DNA damage. Since DSBs have the potential to arrest cell cycle terplay between different DNA repair pathways.
or even kill the cell, they have to be efficiently repaired to avoid such
events. Different DNA repair mechanisms work in different stages of the 1.5. CAs and variation in TP53 gene (DNA damage response pathway)
cell cycle, namely homologous recombination (HR), non-homologous-
end joining (NHEJ), and highly inaccurate microhomology-mediated The tumour suppressor TP53 (MIM# 191170), located on chromo-
end joining (MMEJ). When DSB occurs, a plethora of DNA repair pro- some 17p13.1, is known as ‘the guardian of the genome’ or ‘the cellular
teins cooperate with each other to preserve genome integrity and gatekeeper of growth and division’ owing to its diverse functions. The
functionality. NBS1 is one of the first proteins to accumulate at DSB protein p53, a 393 amino acid long phosphoprotein, acts as a key
sites. It creates a complex with MRE11 and RAD50 proteins, which regulator of cellular growth control and plays a central role in the in-
recognizes the DNA damage site and promotes resection of its broken duction of genes that are important in cell cycle arrest and apoptosis
ends [57]. Interestingly, NBS1 is commonly overexpressed in different following DNA damage. Cell cycle arrest provides a time window for
types of cancer, probably due to its participation in inaccurate MMEJ the cell to repair genomic damage before entering the critical stages of
pathway [58,59]. Other key proteins such as XRCC2 and XRCC3 (the X- DNA synthesis and mitosis (previously reviewed [64]). The most fre-
ray repair complementing defective repair in Chinese hamster cell 2 quently investigated TP53 polymorphism is rs1042522, a G to C
and 3, respectively) play an irreplaceable role during HR when an transversion in codon 72 of exon 4, which results in an amino acidic
undamaged sister chromatid is invaded by single-stranded overhangs of change from arginine to proline (TP53 Arg72Pro). This rs1042522
the damaged strand to promote break repair. XRCC2 and XRCC3 polymorphism is located in a proline-rich region of the protein with
(RAD51B, RAD51C and RAD51D) are paralogues of RAD51 re- importance for the growth suppression and apoptotic functions.
combinase. XRCC2 and XRCC3 create complexes with other RAD51 Rs1042522 results in a structural change in the protein, giving rise to
paralogues, which in turn facilitate recruitment and stabilisation of variants of distinct electrophoretic mobility; and the two isoforms of
RAD51 at DSB sites [60,61]. The whole RAD51-mediated strand inva- p53 due to polymorphism at codon 72 differ in biochemical and bio-
sion is coordinated by RAD54L protein, which translocates along DSB logical properties. Apparently, the TP53 Arg72 form induces apoptosis
sites and has diverse functions during HR, e.g. remodelling of chro- more efficiently than the Pro72 form. The Arg72 variant, when in cis-
matin structure, Rad51 filament assembly, and D-loop dissolution form with certain tumour-derived mutations, might enhance tumour
[62,63]. suppressive function owing to increased ability to activate p53 [64,65].
By analysing SNPs individually, we observed significantly lower However, we did not observe any association between either chromo-
CTA frequencies in association with the XPD Lys751Gln homozygous somal damage and rs1042522 TP53 Arg72Pro (Tables 1 and 2) in 593
variant genotype (OR 0.64, P = 0.004). This association replicated with healthy individuals. Rs1042522 did not modulate the frequencies of
a similar significance level in a separate healthy population earlier CAs, CTAs, or CSAs, in all instances the dominant model of inheritance
[29]. An association of the heterozygous variant genotype in RAD54L was tested.
with increased CSA frequency (OR 1.96, P = 0.03) was also found. By
addressing gene–gene interactions, we discovered 14 interactions 1.6. CAs and variation in the major mitotic checkpoint genes
modulating CA frequencies, 9 modulating CTA and 12 modulating CSA
frequencies with nominal significance. The interactions with strong The fidelity of chromosome segregation during mitosis is ensured by
statistical support are highlighted in Fig. 3. For total CAs, highly sig- the spindle assembly checkpoint (SAC). In case of unattached kine-
nificant gene–gene interactions were observed mainly for genes parti- tochores to microtubules, the inhibition of anaphase promoting com-
cipating in BER (APE1, hOGG1, XRCC1), NER (XPC, XPD,XPG), and DSB plex (APC or cyclosome) together with its adaptor binding partner
repair (NBS1, XRCC2, XRCC3). In general, gene–gene interactions al- Cdc20 prevents an occurrence of aneuploidy. The mitotic checkpoint
ways included pairs from two different pathways. Although individual complex, composed of Mad2, Bub1, Bub3, NEK2, ZWILCH and ZWINT
variants in genes encoding DNA repair proteins modulated CAs only proteins [66,67], is generated as a result of SAC activation. Mad2 forms
modestly, several gene–gene interactions were associated with CA a complex with the Cdc20 protein and therefore enables binding of

5
P. Vodicka et al. Mutat Res Gen Tox En xxx (xxxx) xxx–xxx

Bub1 and Bub3 to Cdc20 [68]. The human checkpoint kinase Bub1, and lymphomas. The majority of haematological neoplasms are caused by
its adaptor protein Bub3 phosphorylate and subsequently inhibit the clonal expansion of a single cell that has acquired a somatic mutation in
APC/Cdc20 complex during SAC signalling cascade [67,69]. When all one allele of a proto-oncogene. This mutated gene stimulates in-
chromosomes are properly attached in the mitotic spindle, MAD2 is appropriate cellular proliferation, leading to the development of cancer.
released from the mitotic checkpoint complex, resulting in the activa- Another mechanism of inappropriate stimulation of cellular prolifera-
tion of APC/C-CDC20, which in turn targets PTTG1 (securin) for de- tion is the misplacement of a proto-oncogene through CA under a strong
gradation. ESPL1 (separase) gets activated and can cleave the cohesin promoter, and which may lead to the uncontrolled expression of this
complexes that keep the sister chromatids together [70]. Aberrant ex- gene [73]. An example is translocation between chromosomes 11 and
pression of genes involved in the mitotic checkpoint complex (NEK2) is 14, abbreviated t(11;14). In t(11;14) the cyclin D1 (CCND1) gene from
related to carcinogenesis of various tumors [71]. chromosome 11 is juxtaposed next to the promoter of the heavy chain
In our study, we focused on genes constituting the mitotic check- immunoglobulin (IgH) gene, driving overexpression of cyclin D1 and
point machinery in view of its role in maintaining chromosomal in- stimulating cell division. It is found in several haematological malig-
tegrity with possible links to CAs [72]. We selected 9 variants in main nancies, including mantle cell lymphomas in which every patient has
checkpoint genes (BUB1B, BUB3, MAD2L1, CENPF, ESPL1/separase, this translocation. It is the most common translocation in multiple
NEK2, PTTG1/securin, ZWILCH and ZWINT) for a genotyping study in myeloma, found in 15% of all patients [74].
the above cohort of healthy individuals. The selected variants either Our study on translocations in myeloma patients for the first time
caused a potentially deleterious amino acid change or were located at described a germline genetic basis of a translocation in humans: the t
the core regulatory regions of the genes (Table 2). Genetic variation in (11;14) was regulated by a splice site polymorphism rs9344 (archive
individual genes exhibited a minor importance, with only a few nom- dbSNP ID rs603965) in cyclin D1 [74]. The odds ratio (OR) between the
inally significant associations. Such data are consistent with the high risk allele G and the other allele A was about 2.0. The c.870G allele
conservation and selection pressure of the checkpoint system. However, creates an optimal splice donor site at the intron 4/exon 5 boundary,
many gene pairs were significantly associated with CAs and two pairs – resulting in the cyclin D1a transcript (Fig. 5). By contrast c.870A hin-
PTTG1-ZWILCH and PTTG1-ZWINT – were associated with all types of ders splicing, allowing for read-through into intron 4 and the produc-
CAs. MAD2L1 and PTTG1 were the most common partners in any of the tion of the variant cyclin D1b transcript. Although c.870A is pre-
two-way interactions (Fig. 4). These results suggest that interactions at ferentially associated with D1b production, the A allele is not fully
the cohesin level (PTTG1) and kinetochore function (ZWINT, ZWILCH penetrant [75,76]. Cyclin D1b has lost its capacity to associate with
and MAD2L1) contribute to the frequency of CAs. The data imply that CDK4 and it does not enhance phosphorylation of Rb protein [77].
gene variants at different checkpoint functions appear to be required for While cyclin D1b is a potent oncogene operating through aberrant ki-
the formation of CA. nase activity, cyclin D1a recruits RAD51 to local chromatin in response
to DNA damage [78]. In our study, we showed that the levels of D1a
and D1b transcripts correlated with the GG, GA, and AA genotypes in
1.7. Cyclin D1 and CAs CD138-selected plasma cell [74]. D1a/D1b transcript ratios have been
shown to shift in some cancers, suggesting that differential alternative
Random CAs may occur in normal cells, but they are a characteristic splicing can influence tumourigenesis. It is intriguing that the G allele
finding in advanced malignancies. They may be tumour-specific and encoding the full length D1a would be associated with myeloma, be-
contribute directly to malignant transformation through a rearrange- cause it has a low transforming capacity compared to D1b [79]. As the
ment of specific underlying genes [4]. Many types of specific recurrent G allele is the risk factor for t(11;14) in myeloma, the D1a protein may
CAs, including deletions, amplifications, and translocations have been promote the translocation, which in turn would drive cyclin D1 ex-
described for a large number of haematological malignancies, including pression by the IgH promoter at the translocation locus. D1a may also
lymphoid and myeloid leukaemias, multiple myeloma, and malignant

Fig. 4. Influence of mitotic checkpoint gene polymorphisms on CAs in healthy volunteers [72]. The colour code indicates total CA (dark blue), CSA (blue), CTA (light
blue), and CSA & CTA (blue gradient). The arrows indicate significant gene–gene interactions.

6
P. Vodicka et al. Mutat Res Gen Tox En xxx (xxxx) xxx–xxx

Fig. 5. Cyclin D1 participation in DSB repair by binding to RAD51 (a main recombinase involved in HR). The induction of the DNA damage response is mediated by
cyclin D1a, associated with polymorphism rs9344 in CCND1 (c.870G allele), whereas cyclin D1b (c.870A) lacks this activity.

act through separate pathways, promoting chromosomal aberrations To prove the above assumption we tested the association of CAs,
because it is associated with double-strand break repair complexes of CTAs, and CSAs with CCND1 c.G870A polymorphism in an independent
RAD51, BRCA1 and BRCA2 [78]. group of 255 sporadic cancer patients (mainly breast and colorectal)
In myeloma the immunoglobulin gene translocations are generated and controls. Moderately increased frequencies of chromosomal da-
by abnormal class switch recombination events occurring during the mage seemed to be associated with c.870A allele (OR 1.96 95% CI
germinal centre stage of B cell development. Such abnormal events are 0.94–4.06, P = 0.073 for CAs; OR 1.62 95%CI 0.90–2.94, P = 0.231 for
usually present in all clonal myeloma cells, and are also detectable in CTAs; and OR 1.58 95% CI 0.77–3.21 for CSAs, P = 0.210, respec-
monoclonal gammopathy of unknown significance (MGUS), a pre- tively). Although the data showed the same tendency as described
malignant precursor of myeloma, and AL amyloidosis [80,81]. Normal above, the modulating effect was not statistically significant mainly due
class switch recombination involves a recognition step and double to a low number of observations.
strand DNA cleavage steps, driven by activation induced deaminase.
These strand breaks are then repaired using translocation prone path-
ways not involving classical non-homologous end joining (NHEJ) that 2. Conclusions
can result in the formation of reciprocal translocation to other strand
breaks located elsewhere in the genome [18]. CCND1 polymorphism is Cellular DNA is continually attacked by numerous electrophilic
also associated with t(11;14) in MGUS and AL amyloidosis, the G allele compounds, generated or detoxified via XMEs. There is ample evidence
being the risk allele as in myeloma [74,82,83]. We also tested genetic that DNA repair pathways are crucial mechanisms for preserving of the
associations with myeloma translocations other than t(11;14), but no genomic and chromosomal integrity. One source of variability in the
positive results were found. This suggests that the impact of the geno- function of XME and/or DNA repair is due to gene variants modulating
type on t(11;14) is not entirely through the facilitation of errors in class activities of the particular proteins. Our review provides evidence that
switch recombination, nor can it be explained through errors in a variants in the above genes may modulate the frequency of CAs.
general DNA repair mechanisms. However, even in the conditions of an optimal DNA repair mechanism,
As t(11;14) is found in all mantle cell lymphomas, we also tested CA may arise in a cell as a consequence of the improper mitotic divi-
these cells for the CCND1 polymorphism, and found no effect [74]. sion. Variants in major mitotic checkpoint genes, TP53 and cyclin D1
However, in mantle cell lymphomas the translocation is mediated via representing the DNA damage response pathway, may also contribute
abnormal VDJ recombination at the IgH locus. The breakpoints in to the frequency of CAs in PBLs. Our results show that variants in genes
mantle cell lymphomas are clustered, whereas the breakpoints in t encoding XMEs and DNA repair proteins as well as genes encoding key
(11;14) MM are heterogeneous spanning several hundred kb in the 5′ mitotic checkpoint proteins individually exerted small effects, while in
region of CCND1 [84]. The absence of an association is consistent with interactions with each other they were associated with CA frequencies
a different etiological basis for t(11;14) myeloma and mantle cell in PBLs of healthy volunteers. Obviously, the accumulation of CAs re-
lymphoma [18,85]. quires a complex interplay between genes involved in the maintenance
We also tested the possible impact of the CCND1 genotype on non- of genomic integrity. However, more functional/mechanistic studies
specific CAs in a large cohort of persons for whom CAs were measured supporting these observations are required.
[86]. The results for total CAs showed a significant association with the By assaying for the links between gene variants in the DNA damage
AA genotype (OR 1.85). The OR was almost as high as it was for t response pathway and CAs in PBLs, we did not find any association with
(11;14) in myeloma, but to our surprise it was to the opposite direction. the common functional TP53 Arg72Pro polymorphism. These data re-
A was the risk allele for non-specific CAs, while G was the risk allele for quire rather cautious interpretation, since other TP53 polymorphisms
translocation. How can these results be explained through known and haplotypes were not investigated. Opposite associations were ob-
biology of the splice site variants? We can speculate that there are two tained regarding the modulating effect of CCND1 splice site poly-
key mechanisms. One is that allele G, encoding cyclin D1a isoform, may morphism rs9344 G870A on CAs in healthy individuals and cancer
facilitate translocation through its DNA damage response functions. The patients compared to its association with t(11,14) translocation.
other is that specificity for t(11;14), rather than other translocations, Further study aimed at the disclosure of the functional consequences of
may depend on the juxtaposed CCND1 gene while in other myeloma the CCND1 gene and its variants in interplay with the DNA damage
translocations other proto-oncogenes are juxtaposed at this locus. The response pathway and DNA repair on chromosomal instability during
transforming potential of isoform D1b may cause CAs in dormant malignant transformation is warranted.
lymphocytes. An increased frequency of CAs has also been associated with a de-
creased relative telomere length in healthy subjects. Notably, this

7
P. Vodicka et al. Mutat Res Gen Tox En xxx (xxxx) xxx–xxx

interesting association was not observed in patients with incident breast [22] M.J. Jones, P.V. Jallepalli, Chromothripsis: chromosomes in crisis, Dev. Cell 23 (5)
and colorectal cancers. The mechanisms underlying CA formation in (2012) 908–917.
[23] J.W. Shay, Molecular pathogenesis of aging and cancer: are telomeres and telo-
association with telomere length in various pathologies deserves further merase the connection? J. Clin. Pathol. 50 (10) (1997) 799–800.
attention. [24] A.F. Vallarelli, P.S. Rachakonda, J. Andre, et al., TERT promoter mutations in
In the context of the etiological role of suboptimal DNA processes in melanoma render TERT expression dependent on MAPK pathway activation,
Oncotarget 7 (33) (2016) 53127–53136.
carcinogenesis, we recently suggested that altered DSB repair measured [25] M. Kroupa, Z. Polivkova, S. Rachakonda, et al., Bleomycin-induced chromosomal
by a sensitivity towards mutagen in PBL occurs particularly in color- damage and shortening of telomeres in peripheral blood lymphocytes of incident
ectal tumours. Irrespective of cancer type, telomere shortening is as- cancer patients, Genes Chromosomes Cancer 57 (2) (2018) 61–69.
[26] F. Mertens, C.R. Antonescu, F. Mitelman, Gene fusions in soft tissue tumors: re-
sociated with a decreased capacity to repair DSB. There are several current and overlapping pathogenetic themes, Genes Chromosomes Cancer 55 (4)
emerging questions, such as the role of CAs in malignancies with re- (2016) 291–310.
spect to a particular phenotype and heterogeneity, and the formation of [27] F. Mertens, B. Johansson, T. Fioretos, F. Mitelman, The emerging complexity of
gene fusions in cancer, Nat. Rev. Cancer 15 (6) (2015) 371–381.
CAs during the process of malignant transformation. Addressing the
[28] P. Vodicka, Z. Polivkova, S. Sytarova, et al., Chromosomal damage in peripheral
above points is unconditionally required prior to any translations into blood lymphocytes of newly diagnosed cancer patients and healthy controls,
the disease prevention. Carcinogenesis 31 (7) (2010) 1238–1241.
[29] S. Vodenkova, Z. Polivkova, L. Musak, et al., Structural chromosomal aberrations as
potential risk markers in incident cancer patients, Mutagenesis 30 (4) (2015)
Acknowledgement 557–563.
[30] P. Rossner, P. Boffetta, M. Ceppi, et al., Chromosomal aberrations in lymphocytes of
Support from grants No. 15-14789S (Czech Science Foundation, healthy subjects and risk of cancer, Environ. Health Perspect. 113 (5) (2005)
517–520.
Czech Republic), 15-27580A (Ministry of Health, Czech Republic), and [31] P. Vodicka, R. Kumar, R. Stetina, et al., Genetic polymorphisms in DNA repair genes
grant Biomedical Center Martin, Slovakia, ITMS 26220220187 (EU and possible links with DNA repair rates, chromosomal aberrations and single-
Commission) are acknowledged. The study was also partially supported strand breaks in DNA, Carcinogenesis 25 (5) (2004) 757–763.
[32] A. Naccarati, P. Soucek, R. Stetina, et al., Genetic polymorphisms and possible
by the projects Research program of Charles University PROGRES Q28 gene–gene interactions in metabolic and DNA repair genes: effects on DNA damage,
(Oncology) and “Centrum of clinical and experimental liver surgery”, Mutat. Res. 593 (1-2) (2006) 22–31.
UNCE/MED/006, and by the National Sustainability Program I (NPU I) [33] L. Musak, P. Soucek, L. Vodickova, et al., Chromosomal aberrations in tire plant
workers and interaction with polymorphisms of biotransformation and DNA repair
Nr. LO1503 provided by the Ministry of Education Youth and Sports, genes, Mutat. Res. 641 (1-2) (2008) 36–42.
Czech Republic. [34] C.F. Skjelbred, M. Svendsen, V. Haugan, et al., Influence of GSTM1, GSTT1, GSTP1,
NAT1, NAT2, EPHX1, MTR and MTHFR polymorphism on chromosomal aberration
frequencies in human lymphocytes, Carcinogenesis 32 (3) (2011) 399–405.
References
[35] A.M. Rossi, I.L. Hansteen, C.F. Skjelbred, et al., Association between frequency of
chromosomal aberrations and cancer risk is not influenced by genetic polymorph-
[1] P.A. Futreal, L. Coin, M. Marshall, et al., A census of human cancer genes, Nat. Rev. isms in GSTM1 and GSTT1, Environ. Health Perspect. 117 (2) (2009) 203–208.
Cancer 4 (3) (2004) 177–183. [36] K. Hemminki, C. Frank, A. Forsti, et al., Metabolic gene variants associated with
[2] H. Rajagopalan, C. Lengauer, Aneuploidy and cancer, Nature 432 (7015) (2004) chromosomal aberrations in healthy humans, Genes Chromosomes Cancer 54 (4)
338–341. (2015) 260–266.
[3] R.A. Burrell, S.E. McClelland, D. Endesfelder, et al., Replication stress links struc- [37] P. Vodicka, P. Soucek, A.D. Tates, et al., Association between genetic polymorph-
tural and numerical cancer chromosomal instability, Nature 494 (7438) (2013) isms and biomarkers in styrene-exposed workers, Mutat. Res. 482 (1-2) (2001)
492–496. 89–103.
[4] F. Mitelman, B. Johansson, F. Mertens, The impact of translocations and gene fu- [38] P. Vodicka, R. Kumar, R. Stetina, et al., Markers of individual susceptibility and
sions on cancer causation, Nat. Rev. Cancer 7 (4) (2007) 233–245. DNA repair rate in workers exposed to xenobiotics in a tire plant, Environ. Mol.
[5] L.P. Bignold, Mechanisms of clastogen-induced chromosomal aberrations: a critical Mutagen. 44 (4) (2004) 283–292.
review and description of a model based on failures of tethering of DNA strand ends [39] I. Hlavata, D. Vrana, Z. Smerhovsky, et al., Association between exposure-relevant
to strand-breaking enzymes, Mutat. Res. 681 (2–3) (2009) 271–298. polymorphisms in CYP1B1, EPHX1, NQO1, GSTM1, GSTP1 and GSTT1 and risk of
[6] L. Hagmar, U. Stromberg, H. Tinnerberg, Z. Mikoczy, Epidemiological evaluation of colorectal cancer in a Czech population, Oncol. Rep. 24 (5) (2010) 1347–1353.
cytogenetic biomarkers as potential surrogate end-points for cancer, IARC Sci. Publ. [40] P. Vodicka, A. Naccarati, L. Vodickova, et al., Do GST polymorphisms modulate the
157 (2004) 207–215. frequency of chromosomal aberrations in healthy subjects? Environ. Health
[7] A.T. Natarajan, F. Palitti, DNA repair and chromosomal alterations, Mutat. Res. 657 Perspect. 117 (9) (2009) A384–A385 (author reply A385).
(1) (2008) 3–7. [41] R.A. Mateuca, I. Decordier, M. Kirsch-Volders, Cytogenetic methods in human
[8] M. Durante, J.S. Bedford, D.J. Chen, et al., From DNA damage to chromosome biomonitoring: principles and uses, Methods Mol. Biol. (Clifton, NJ) 817 (2012)
aberrations: joining the break, Mutat. Res. 756 (1–2) (2013) 5–13. 305–334.
[9] E.H. Blackburn, Switching and signaling at the telomere, Cell 106 (6) (2001) [42] M. Dusinska, M. Barancokova, A. Kazimirova, et al., Does occupational exposure to
661–673. mineral fibres cause DNA or chromosome damage? Mutat. Res. 553 (1-2) (2004)
[10] J. Maciejowski, T. de Lange, Telomeres in cancer: tumour suppression and genome 103–110.
instability, Nat. Rev. Mol. Cell Biol. 18 (3) (2017) 175–186. [43] M. Dusinska, A. Collins, A. Kazimirova, et al., Genotoxic effects of asbestos in hu-
[11] B. Heidenreich, R. Kumar, TERT promoter mutations in telomere biology, Mutat. mans, Mutat. Res. 553 (1–2) (2004) 91–102.
Res. 771 (2017) 15–31. [44] A. Kazimirova, M. Barancokova, Z. Dzupinkova, L. Wsolova, M. Dusinska,
[12] H. Li, H.T. Hilmarsen, M.B. Hossain, et al., Telomere length and LINE1 methylation Micronuclei and chromosomal aberrations, important markers of ageing: possible
is associated with chromosomal aberrations in peripheral blood, Genes association with XPC and XPD polymorphisms, Mutat. Res. 661 (1–2) (2009) 35–40.
Chromosomes Cancer 52 (1) (2013) 1–10. [45] L. Musak, Z. Smerhovsky, E. Halasova, et al., Chromosomal damage among medical
[13] K. Hemminki, S. Rachakonda, L. Musak, et al., Telomere length in circulating staff occupationally exposed to volatile anesthetics, antineoplastic drugs, and for-
lymphocytes: association with chromosomal aberrations, Genes Chromosomes maldehyde, Scand. J. Work Environ. Health 39 (6) (2013) 618–630.
Cancer 54 (3) (2015) 194–196. [46] R. Hakem, DNA-damage repair; the good, the bad, and the ugly, EMBO J. 27 (4)
[14] K. Broberg, J. Bjork, K. Paulsson, M. Hoglund, M. Albin, Constitutional short telo- (2008) 589–605.
meres are strong genetic susceptibility markers for bladder cancer, Carcinogenesis [47] P. Vodicka, L. Musak, C. Frank, et al., Interactions of DNA repair gene variants
26 (7) (2005) 1263–1271. modulate chromosomal aberrations in healthy subjects, Carcinogenesis 36 (11)
[15] J.S. Jang, Y.Y. Choi, W.K. Lee, et al., Telomere length and the risk of lung cancer, (2015) 1299–1306.
Cancer Sci. 99 (7) (2008) 1385–1389. [48] P. Vodicka, R. Stetina, V. Polakova, et al., Association of DNA repair polymorphisms
[16] R.A. DePinho, The age of cancer, Nature 408 (6809) (2000) 248–254. with DNA repair functional outcomes in healthy human subjects, Carcinogenesis 28
[17] J.W. Shay, Role of telomeres and telomerase in aging and cancer, Cancer Discov. 6 (3) (2007) 657–664.
(6) (2016) 584–593. [49] V. Simonelli, F. Mazzei, M. D'Errico, E. Dogliotti, Gene susceptibility to oxidative
[18] M. Gostissa, F.W. Alt, R. Chiarle, Mechanisms that promote and suppress chromo- damage: from single nucleotide polymorphisms to function, Mutat. Res. 731 (1-2)
somal translocations in lymphocytes, Annu. Rev. Immunol. 29 (2011) 319–350. (2012) 1–13.
[19] R.S. Maser, R.A. DePinho, Connecting chromosomes, crisis, and cancer, Science [50] J.E. Cleaver, Defective repair replication of DNA in xeroderma pigmentosum,
(New York, NY) 297 (5581) (2002) 565–569. Nature 218 (5142) (1968) 652–656.
[20] A.K. Meeker, J.L. Hicks, C.A. Iacobuzio-Donahue, et al., Telomere length abnorm- [51] J.J. DiGiovanna, K.H. Kraemer, Shining a light on xeroderma pigmentosum, J.
alities occur early in the initiation of epithelial carcinogenesis, Clin. Cancer Res. 10 Invest. Dermatol. 132 (3 Pt. (2)) (2012) 785–796.
(10) (2004) 3317–3326. [52] L. Nemzow, A. Lubin, L. Zhang, F.X.P.C. Gong, Going where no DNA damage sensor
[21] J. Maciejowski, Y. Li, N. Bosco, P.J. Campbell, T. de Lange, Chromothripsis and has gone before, DNA Repair (Amst). 36 (2015) 19–27.
kataegis induced by telomere crisis, Cell 163 (7) (2015) 1641–1654. [53] S. Benhamou, A. Sarasin, ERCC2/XPD gene polymorphisms and cancer risk,

8
P. Vodicka et al. Mutat Res Gen Tox En xxx (xxxx) xxx–xxx

Mutagenesis 17 (6) (2002) 463–469. anaphase and cytokinesis in human cells, Science (New York, NY) 293 (5533)
[54] M. Spies, Two steps forward, one step back: determining XPD helicase mechanism (2001) 1320–1323.
by single-molecule fluorescence and high-resolution optical tweezers, DNA Repair [71] L. Fu, S. Liu, H. Wang, et al., Low expression of NEK2 is associated with hepato-
(Amst) 20 (2014) 58–70. cellular carcinoma progression and poor prognosis, Cancer Biomark. 20 (1) (2017)
[55] O.D. Scharer, XPG: its products and biological roles, Adv. Exp. Med. Biol. 637 101–106.
(2008) 83–92. [72] A. Forsti, C. Frank, B. Smolkova, et al., Genetic variation in the major mitotic
[56] N. Sugitani, R.M. Sivley, K.E. Perry, J.A. Capra, W.J. Chazin, XPA: a key scaffold for checkpoint genes associated with chromosomal aberrations in healthy humans,
human nucleotide excision repair, DNA Repair (Amst). 44 (2016) 123–135. Cancer Lett. 380 (2) (2016) 442–446.
[57] K. Komatsu, NBS1 and multiple regulations of DNA damage response, J. Radiat. [73] A.M. Rajan, S.V. Rajkumar, Interpretation of cytogenetic results in multiple mye-
Res. 57 (Suppl. 1) (2016) i11–i17. loma for clinical practice, Blood Cancer J. 5 (2015) e365.
[58] Y. Wang, M. Li, J. Long, et al., Clinical significance of increased expression of [74] N. Weinhold, D.C. Johnson, D. Chubb, et al., The CCND1 G870A polymorphism is a
Nijmegen breakage syndrome gene (NBS1) in human primary liver cancer, Hepatol. risk factor for t(11;14)(q13;q32) multiple myeloma, Nat. Genet. 45 (2013)
Int. 8 (2) (2014) 250–259. 522–525.
[59] D.S. Hsu, S.Y. Chang, C.J. Liu, et al., Identification of increased NBS1 expression as [75] K.E. Knudsen, J.A. Diehl, C.A. Haiman, E.S. Knudsen, Cyclin D1: polymorphism,
a prognostic marker of squamous cell carcinoma of the oral cavity, Cancer Sci. 101 aberrant splicing and cancer risk, Oncogene 25 (2006) 1620–1628.
(4) (2010) 1029–1037. [76] E.K. Millar, J.L. Dean, C.M. McNeil, et al., Cyclin D1b protein expression in breast
[60] N. Suwaki, K. Klare, M. Tarsounas, RAD51 paralogs: roles in DNA damage signal- cancer is independent of cyclin D1a and associated with poor disease outcome,
ling, recombinational repair and tumorigenesis, Semin. Cell Dev. Biol. 22 (8) (2011) Oncogene 28 (2009) 1812–1820.
898–905. [77] C.J. Kim, K. Nishi, T. Isono, et al., Cyclin D1b variant promotes cell invasiveness
[61] J.F. Carvalho, R. Kanaar, Targeting homologous recombination-mediated DNA re- independent of binding to CDK4 in human bladder cancer cells, Mol. Carcinog. 48
pair in cancer, Expert Opin. Ther. Targets 18 (4) (2014) 427–458. (10) (2009) 953–964.
[62] A.V. Mazin, O.M. Mazina, D.V. Bugreev, M.J. Rossi, Rad54, the motor of homo- [78] R.G. Pestell, New roles of cyclin D1, Am. J. Pathol. 183 (1) (2013) 3–9.
logous recombination, DNA Repair (Amst) 9 (3) (2010) 286–302. [79] N.A. Olshavsky, C.E. Comstock, M.J. Schiewer, et al., Identification of ASF/SF2 as a
[63] S.J. Ceballos, W.D. Heyer, Functions of the Snf2/Swi2 family Rad54 motor protein critical, allele-specific effector of the cyclin D1b oncogene, Cancer Res. 70 (2011)
in homologous recombination, Biochim. Biophys. Acta 1809 (9) (2011) 509–523. 3975–3984.
[64] A. Naccarati, V. Polakova, B. Pardini, et al., Mutations and polymorphisms in TP53 [80] W.J. Chng, O. Glebov, P.L. Bergsagel, W.M. Kuehl, Genetic events in the patho-
gene–an overview on the role in colorectal cancer, Mutagenesis 27 (2) (2012) genesis of multiple myeloma, Best Pract. Res. Clin. Haematol. 20 (2007) 571–596.
211–218. [81] J. Sawyer, The prognostic significance of cytogenetics and molecular profiling in
[65] X. Tian, S. Dai, J. Sun, S. Jiang, Y. Jiang, The association between the TP53 multiple myeloma, Cancer Genet. 204 (2011) 3–12.
Arg72Pro polymorphism and colorectal cancer: an updated meta-analysis based on [82] M.I. da Silva Filho, A. Försti, N. Weinhold, et al., Genome-wide association study of
32 studies, Oncotarget 8 (1) (2017) 1156–1165. immunoglobulin light chain amyloidosis in three patient cohorts: comparison to
[66] J.M. Kasuboski, J.R. Bader, P.S. Vaughan, et al., Zwint-1 is a novel Aurora B sub- myeloma, Leukemia 31 (8) (2017) 1735–1742.
strate required for the assembly of a dynein-binding platform on kinetochores, Mol. [83] I. Meziane, S. Huhn, M. Filho, et al., Genome-wide association study of clinical
Biol. Cell 22 (18) (2011) 3318–3330. parameters in immunoglobulin light chain amyloidosis in three patient cohorts,
[67] L. Jia, B. Li, H. Yu, The Bub1-Plk1 kinase complex promotes spindle checkpoint Haematologica 102 (10) (2017) e411–e414.
signalling through Cdc20 phosphorylation, Nat. Commun. 7 (2016) 10818. [84] R. Fonseca, B. Barlogie, R. Bataille, et al., Genetics and cytogenetics of multiple
[68] M. Simonetta, R. Manzoni, R. Mosca, et al., The influence of catalysis on mad2 myeloma: a workshop report, Cancer Res. 64 (2004) 1546–1558.
activation dynamics, PLoS Biol. 7 (1) (2009) e10. [85] P. Perez-Galan, M. Dreyling, A. Wiestner, Mantle cell lymphoma: biology, patho-
[69] K. Overlack, T. Bange, F. Weissmann, et al., BubR1 promotes bub3-dependent APC/ genesis, and the molecular basis of treatment in the genomic era, Blood 117 (2011)
C inhibition during spindle assembly checkpoint signaling, Curr. Biol. 27 (19) 26–38.
(2017) 2915–2927 (e2917). [86] K. Hemminki, L. Musak, V. Vymetalkova, et al., Cyclin D1 splice site variant triggers
[70] S. Hauf, I.C. Waizenegger, J.M. Peters, Cohesin cleavage by separase required for chromosomal aberrations in healthy humans, Leukemia 28 (2014) 721–722.