La lecture en ligne est gratuite
Le téléchargement nécessite un accès à la bibliothèque YouScribe
Tout savoir sur nos offres
Télécharger Lire

Characterization of Cellular DNA Repair Capacity by Molecular and Cytogenetic Approaches: Suitability as Intermediate Phenotype of Cancer [Elektronische Ressource] / Harald Martin Surowy

153 pages
INSTITUTE OF H UMAN GENETICS OF THE U NIVERSITYH OSPITAL OF U LM MEDICAL DIRECTOR: PROF DR. C. KUBISCH CHARACTERIZATION OF CELLULAR DNA REPAIR CAPACITY BY MOLECULAR AND CYTOGENETICA PPROACHES : SUITABILITY ASI NTERMEDIATEP HENOTYPE OFC ANCER DISSERTATION FOR THE DOCTORAL DEGREE OF H UMAN B IOLOGY (DR. B IOL. H UM .) FACULTY OF M EDICINE , U LM U NIVERSITY SUBMITTED B Y H ARALD M ARTIN SUROWY B ORN IN M EMMINGEN , GERMANY 2011 Acting Dean: Prof. Dr. Thomas Wirth 1. Correspondent: Prof. Dr. Walther Vogel 2. Correspondent: PD Dr. Marcus Cronauer stDay of Promotion: October 21, 2011 TABLE OF C ONTENTS PAGE i Table of Contents LIST OFA BBREVIATIONS II I1 INTRODUCTION 1 1.1 Genetic Predisposition to Cancer 1 1.1.1 Cancer Heritability and Susceptibility 1 1.1.2 DNA Repair Disorders and Cancer Risk 3 1.2 Cellular DNA Repair Capacity and Cancer 6 1.3 Assays Used for the Measure of Cellular DNA RepaCiarp acity from Peripheral Blood Samples 7 1.3.1 Micronucleus Assay 7 1.3.2 Sister Chromatid Exchange Assay 8 1.3.3 Mitotic Delay Assay 10 1.3.4 Further Measures of DNA Repair Capacity 10 1.4 Aims of the Study 11 2 PROBANDS , MATERIALS AND METHODS 13 2.1 Probands 13 2.1.1 Cohorts with DNA Repair and Genotype Data 13 2.1.2 Cohorts with Genotype Data Only 14 2.2 Materials 16 2.2.1 Sources of Peripheral Blood Lymphocytes andN AD 16 2.2.
Voir plus Voir moins


INSTITUTE OF H UMAN GENETICS OF THE U NIVERSITYH OSPITAL OF U LM

MEDICAL DIRECTOR: PROF DR. C. KUBISCH




CHARACTERIZATION OF CELLULAR DNA REPAIR CAPACITY
BY MOLECULAR AND CYTOGENETICA PPROACHES :
SUITABILITY ASI NTERMEDIATEP HENOTYPE OFC ANCER





DISSERTATION

FOR THE DOCTORAL DEGREE OF H UMAN B IOLOGY (DR. B IOL. H UM .)
FACULTY OF M EDICINE , U LM U NIVERSITY





SUBMITTED B Y

H ARALD M ARTIN SUROWY
B ORN IN M EMMINGEN , GERMANY
2011























Acting Dean: Prof. Dr. Thomas Wirth

1. Correspondent: Prof. Dr. Walther Vogel

2. Correspondent: PD Dr. Marcus Cronauer


stDay of Promotion: October 21, 2011

TABLE OF C ONTENTS PAGE i
Table of Contents
LIST OFA BBREVIATIONS II I
1 INTRODUCTION 1
1.1 Genetic Predisposition to Cancer 1
1.1.1 Cancer Heritability and Susceptibility 1
1.1.2 DNA Repair Disorders and Cancer Risk 3
1.2 Cellular DNA Repair Capacity and Cancer 6
1.3 Assays Used for the Measure of Cellular DNA RepaCiarp acity from
Peripheral Blood Samples 7
1.3.1 Micronucleus Assay 7
1.3.2 Sister Chromatid Exchange Assay 8
1.3.3 Mitotic Delay Assay 10
1.3.4 Further Measures of DNA Repair Capacity 10
1.4 Aims of the Study 11
2 PROBANDS , MATERIALS AND METHODS 13
2.1 Probands 13
2.1.1 Cohorts with DNA Repair and Genotype Data 13
2.1.2 Cohorts with Genotype Data Only 14
2.2 Materials 16
2.2.1 Sources of Peripheral Blood Lymphocytes andN AD 16
2.2.2 Cell Lines 16
2.2.3 Laboratory Material and Resources 16
2.3 Experimental Methods 21
2.3.1 Cell Culture Procedures 21
2.3.2 Micronucleus Assay 22
2.3.3 Sister Chromatid Exchange Assay 26
2.3.4 Mitotic Delay Assay 29
2.3.5 Isolation of DNA from Whole Blood Samples 32
2.3.6 Polymerase Chain Reaction (PCR) 33
2.3.7 Genotyping of DNA Polymorphisms 34
2.4 Statistical Procedures 39
2.4.1 Heritability Estimates 39
2.4.2 Linear Regression and Variance Analyses 41
2.4.3 Differences of DNA Repair Assay Results Betwene Groups 42
2.4.4 Association Studies 44
TABLE OF C ONTENTS PAGE ii
3 R ESULTS 48
3. 1 Differences of Cellular DNA Repair Capacity Betwee nBreast Cancer Cases
and Controls 49
3. 2 The Impact of Low-Frequency Coding Variants iRne pDaNiAr Genes on
Cellular DNA Repair Capacity 56
3.2.1 Gene and Variant Selection 56
3.2.2 Genotyping Results 58
3.2.3 Association of Breast Cancer With Variants in the Selected DNA Repair Genes 60
3.2.4 Association Between Genotypes and DNA Repair Capacity 69
3.2.5 Further Analyses of SNP 09 73
3. 3 Characterization of Assays on Cellular DNA Repaira pCacity 75
3.3.1 Heritability of Measures of Cellular DNA Repa ir Capacity 75
3.3.2 Correlation Between the DNA Repair Assays 85
3.3.3 Correlation Between Lymphoblastoid Cell Lines and Peripheral Blood Lymphocytes 86
3.3.4 Study of the Intrinsic Variability of ther oMniuccleus Assay 88
4 D ISCUSSION 97
4. 1 Association Studies 97
4.1.1 Associations Between Cellular DNA Repair Capa city and Sporadic Breast Cancer 98
4.1.2 Coding Variants in DNA Repair Genes, Breast C ancer Risk and
Cellular DNA Repair Capacity 101
4. 2 Characteristics of DNA Repair Capacity Measures 111
4.2.1 Heritability of Cellular DNA Repair Capacity 111
4.2.2 Correlations Between the Measured DNA Repair End Points 115
4.2.3 Lymphoblastoid Cell Lines as Surrogates fore Pripheral Blood Samples 116
4.2.4 Intrinsic Variability of the Micronucleus Aayss 118
4. 3 Conclusions and Prospects 121
5 SYNOPSIS 123
6 LITERATURE 125
ATTACHMENT 143
LIST OFA BBREVIATIONS PAGE iii
List of Abbreviations
a adenine (DNA base) G /M second gap and mitosis phases (cell 2
A alanine (amino acid) cycle)
aa amino acids G -S border of the second gap and DNA 2
ac autoclaved synthesis phases (cell cycle)
ACE model additive genetic, common G /S ratio of cells in the second gap and 2
environment, unique environment the DNA synthesis phases (cell cycle)
model G /S-b baseline ratio of cells in the second 2
ANOVA analysis of variance gap and the DNA synthesis phases
aqua dd. double distilled water (cell cycle)
AUC area under the curve G /S-i induced ratio of cells in the second 2
BNC binucleated cells gap and the DNA synthesis phases
bp base pair (cell cycle)
BPDE Benzo[a]pyrenediepoxide GWAS genome-wide association study
BrdU 5-Bromo-2’-deoxyuridine Gy Gray (radiation dose)
c cytosine (DNA base) H histidine (amino acid)
2
C cysteine (amino acid) h heritability
CEU Utah Residents with Northern and HCR host-cell reactivation
Western European Ancestry HGVS Human Genome Variation Society
CI confidence interval HR homologous recombination
Coef. Var. coefficient of variation HUGO Human Genome Organization
CP control pairs HWE Hardy-Weinberg Equilibrium
Cyt-B Cytochalasin B I isoleucine (amino acid)
DAPI 4’,6-Diamidino-2-phenylindol- I-A, I-B Investigator A, Investigator B
dihydrochloride ICC intraclass correlation coefficient
dbSNP single nucleotide polymorphisms L leucine (amino acid)
database (NCBI) LCL lymphoblastoid cell lines
DMSO Dimethyl sulfoxide LD linkage disequilibrium
DNA deoxyribonucleic acid m male
dNTP deoxynucleoside triphosphate M molar
DSB double strand breaks M phase mitosis phase (cell cycle)
DZ dizygous twin pairs MAF minor allele frequency
E glutamic acid (amino acid) Max. maximum
EBV Epstein-Barr Virus MD mitotic delay
EDTA Ethylendiamide tetraacetate Min. minimum
f female MN micronucleus, micronuclei
F phenylalanine (amino acid) MN-b baseline micronucleus assay result
FA Fanconi Anemia (frequency)
FCS fetal calf serum MN-i induced micronucleus assay result
g guanine (DNA base) (frequency)
G glycine (amino acid) mRNA messenger ribonucleic acid
G phase first gap phase (cell cycle) MSA mean square error among twin pairs 1
G -S border of the first gap and DNA MSW mean square error within twin pairs 1
synthesis phases (cell cycle) MZ monozygous twin pairs
G phase second gap phase (cell cycle) n number of probands 2
LIST OFA BBREVIATIONS PAGE iv
N asparagine (amino acid) SCE sister chromatid exchange
NCBI National Center for Biotechnology SCE-b baseline sister chromatid exchange
Information, USA assay result (frequency)
NN wild-type genotype SCE-i induced sister chromatid exchange
Nr. Number assay result (frequency)
NV heterozygous genotype SD standard deviation
OD optical density SDS Sodium dodecylsulfate
OR odds ratio SNP single nucleotide polymorphism
p short arm (chromosome) SSC saline sodium citrate
P proline (amino acid) STC short-time cultures
P value probability value (significance) Std. Error standard error
P-trend Armitage’s test for trend, probability t thymine (DNA base)
value (significance) T threonine (amino acid)
P1, P2, P3 Preparator 1, 2, 3 TRIS Trishydroxy methylaminomethane
PAR population attributable risk V valine (amino acid)
PBL peripheral blood lymphocytes VV homozygous variant genotype
PBS phosphate buffered saline V within-pair variance of control pairs CP
PCR polymerase chain reaction V within-pair variance of dizygous DZ
pH potentia hydrogenii twin pairs
q long arm (chromosome) V within-pair variance of monozygous MZ
Q glutamine (amino acid) twin pairs
r correlation coefficient VAR variant allele
2
r regression coefficient vs. versus
R arginine (amino acid) W tryptophan (amino acid)
RNA ribonucleic acid
ROC receiver operating characteristic WT wild-type allele
S serine (amino acid) x -fold
S phase DNA synthesis phase (cell cycle)


In the manuscript, the official gene symbols proedv idby the Human Genome Organization Gene
Nomenclature Consortium are used in italics. The oficial full names of the genes are listedA titna chment A .
1. NITRODUCTION PAGE 1
1 Introduction
With an incidence of 12.7 million new cases and 7 .m6 illion deaths in 2008, cancer is the third
leading cause of death worldwide, after cardiovascu lar diseases and infections [248]. The most
common cancers are lung, breast, colorectal, stomach and prostate cancer (see Figure 01). In
Germany, approximately 480,000 newly diagnosed canc ers and 212,000 cancer-related deaths
were reported in 2008 [66].


Figure 01: Incidence of the most common cancers idweor (ldA)w and in Germany G(rBap)h.s were generated with GLOBOCAN
2008 provided by the International Agency for the eRsearch on Cancer, World Health Organization,
(http://globocan.iarc.fr, [66], accession date: Neomvber 05, 2010.)


1.1 Genetic Predisposition to Cancer
1.1. 1 Cancer Heritability and Susceptibility
Aside from clearly heritable cancer syndromes such as the Li-Fraumeni syndrome [131], familial
clustering of cancer cases has also been described for most common cancers [76,85]. For example
10 % of all breast and colorectal cancers are estimated to have a familial background. Thus, a
contribution of genetic factors to cancer format ioisn evident. Large studies on the co-occurrence
of cancer in families and twin pairs revealed a singificant degree of heritability, which is defineds a
the proportion of the variance of a trait that ist traibutable to genetic factors, opposite to
environmental factors. Prostate cancer (42 %), corleoctal cancer (35 %) and breast cancer (27 %)
were found to be the common cancers with the higth edsegrees of heritability [132], followed by
numerous other cancers with a significant contribuiotn of heritable factors to cancer risk [48].
Linkage analyses in families resulted in the identification of the highly penetrating susceptibility
genes for familial breast and ovarian cancer, BRCA1 and BRCA2 [154,247], which belong to the
1. NITRODUCTION PAGE 2
most prominent and most thoroughly characterized cancer susceptibility genes. Mutation
analyses revealed that germline mutations in BRCA1 are found in 43 % and ofBR CA2 in 10 % of
the families with a high risk of breast and ovarian cancer in the German population [111,152].
Recently, highly germline mutations of thReA D51C gene have been identified in further 1.3 % of
those families [153]. Other high risk susceptiybi ligtenes have been described for colorectal and
endometrial cancer [124,166] as well as melanoma [107]. In many other cancers, e.g. prostate
cancer, no high risk susceptibility genes have bee nfound at all.
A second class of cancer susceptibility genes whic hconfer a low to moderate risk increase have
been found by gene-centered association studies and mutation screening. The genesA TM ,
BARD1 , BRIP1 , CHEK2, NBN and PALB2 are well-characterized examples of such susceptiblity
genes for breast cancer [25,151,175,180,197,214B]AR.D 1 has additionally been associated to
neuroblastoma susceptibility [39], CHEK2 to prostate, kidney and thyroid cancer susceptibility [47]
and PALB2 has also been identified as a susceptibility gene for pancreatic cancer [106]. ThNeB N
gene is associated with an increased risk of prosttae cancer, too [46]. Biallelic mutations oATfM ,
NBN and PALB2 or BRIP1 result in the genetic disorders Ataxia Telangiecta sia, Nijmegen Breakage
Syndrome and Fanconi Anemia, respectively (see1 .1.2). Many other putative susceptibility genes
have been identified, however in most cases the rkis increase is marginal and studies often
produce contradictory results for different popuiloatns, indicating a very minor contribution of
these genes to cancer causation [144].
Recently, genome-wide association studies (GWAS) have yielded numerous further variants
associated with an increased risk for various types of cancer. In contrast to gene-centered studies,
GWAS do not rely on a pre-selection of genes afltienrk age results or a functional involvement in
cancer formation. The risk loci of breast and ovarni cancer, prostate cancer and colorectal cancer
comprise approximately half of all associations fonud by GWAS [89,99]. In several instances, two
or more unrelated GWAS could replicate significant associations in the same genomic regions, to a
part also independently in different populations. uFrthermore, independent associations of the
same region to multiple cancers have been found, rf oexample in the chromosomal region 8q24
which contains common risk variants of prostate, ebrast, colorectal and bladder cancer
[191,224,227,243]. However, the associated allelesp roduced by GWAS are almost exclusively
common variants and the increase in cancer risk is very low, with a simulated median odds ratio
of ~ 1.2 [99]. In addition, most of the genom ases-owcidateed variants are situated in intronic or
intergenic regions and tag haplotype blocks which are supposed to carry the true disease allele in
a coding sequence or genic region [54,88,89]. cThiricsu mstance makes the clarification of the
respective functional effects of the associated i loinc cancer development difficult, and implies the
1. NITRODUCTION PAGE 3
need for further analyses by genotyping or sequenncgi analyses in the vicinity of the associated
variant.
Taken together, the known cancer susceptibility genes and loci only explain a minority of up to 20
% of the underlying genetic variation in cancer rkis. The major proportion of the reported
heritability of common cancers is still missing, wichh is especially in the case of GWAS a currently
discussed issue [113,167]. Thus, the image is conlisdoating that cancer is a genetically
heterogeneous disease which involves many risk loc ithat interact in a complex manner. A further
degree of complexity is added by the fact that theg enetically determined variation of cancer
represents only the basis which modulates the altogether larger effects of the specific individual
influences of environmental factors [48,70,132].
As the genetic background of most familial cancer acses and of the vast majority of sporadic
cancer cases still remains unexplained, a more comprehensive characterization of individual
cancer risk is necessary to elucidate the influenc eof genetic factors to cancer formation.


1.1. 2 DNA Repair Disorders and Cancer Risk
Numerous genetic disorders with recessive inheritance, especially those which include
chromosomal instability as a cellular phenotype, al so mean an increased risk of cancer for the
affected individuals as one of their symptoms. Onf,t ethe susceptibility is also present in the
healthy heterozygous carriers of one mutated allele of the respective disease gene. Therefore,
many genes causing such disorders are found among the high or moderate risk cancer
susceptibility genes.
The common denominator of a large number of genesn violved in cancer susceptibility and
chromosomal instability disorders is their involvement in DNA repair processes. While by the year
2001 about 70 human DNA repair genes had been chara cterized [184], over 150 were known in
2005 [246]. Currently, the Entrez Gene database ofth e National Center for Biotechnology
th
Information (NCBI), National Institutes of HealthU,S A [142] (accessed November 10 , 2010), lists
468 gene products with a connection to human DNA rpeair. This high number of genes underlines
that the cellular DNA repair system constitutes a large network of proteins with a high degree of
complexity that comprises a significant portion otfhe estimated 20,000-25,000 human genes [98].
The processes induced upon the occurrence of DNA dma age can broadly be attributed to several
functional levels. Besides the various pathways ava ilable for the recognition and removal of the
specific DNA lesions, those include cell cycle alrtaetions, chromatin remodeling, the induction of
apoptosis and changes in transcription. Also pathwa ys and factors responsible for the general
1. NITRODUCTION PAGE 4
maintenance of chromosomal stability, such as chromatid cohesion and the conservation of
chromosome ends, are recognized as processes involved in DNA repair. As there are numerous
overlaps and interconnections between the single fcators and pathways, they form a cellular DNA
repair network (for reviews, see [192] and [10 0]).

The factors involved in the repair of DNA damage depend on the various types of DNA lesions and
on many other circumstances, for example the celyl ccle phase in which the damage occurs. The
direct repair of few of particular DNA lesions su chas thymine dimers can be facilitated by
specialized proteins. The mismatch repair system detects improper base pairing between the DNA
strands and thus acts in a proof-reading function.M utations in the mismatch repair genes cause
the Lynch syndrome, which is classified as a famali licancer syndrome [139].
Structural aberrations of the DNA, such as base add ucts, disrupted nucleotides and single strand
breaks are removed via the base excision repair and nucleotide excision repair systems. The
different complementation groups of Xeroderma Pigmnetosum are caused by a homozygous loss-
of-function of genes which are all involved inn tuhcele otide excision repair pathway (XP and ERCC
genes), however defects in some of these genes aclsaon be the cause for other syndromes,
including the Cockayne Syndrome, Trichothiodystrophy and the Cerebro-Oculo-Facio-Skeletal
Syndrome [119]. The Xeroderma Pigmentosum Variant hpenotype is caused by defects in the
POLH gene, a polymerase involved in a translesion synthesis mechanism [148] which allows the
replication fork to overcome such structural aberrtaions during the DNA replication.
Biallelic mutations in genes which are involved in the maintenance of telomeres can be the cause
of dyskeratosis congenita [41], while the Li-Fraunmi esyndrome, a familial cancer syndrome, is
attributable to mutations in the TP53 gene which is involved in the activation of apopstios upon
DNA damage [209].

Double strand breaks (DSB) constitute the most dram atic type of DNA lesion, and a large
spectrum of genetic disorders with an increased inicdence in cancer, also in heterozygous carriers,
is caused by mutations in genes involved in their erpair. DSB can lead to chromosomal aberrations
or loss of chromosomal material in the daughter celsl, if not or improperly repaired [161]. DSB can
arise by the direct actions of mutagens such as ioinzing radiation, but also the repair processes of
other DNA lesions can result in DSB.
DSB are also generated during the repair of inter-tsrand DNA cross-links via the Fanconi Anemia
(FA) pathway [116]. All of the 13 complementation grou posf FA are caused by biallelic mutations
in one of the corresponding 13 genes, tFhANe C genes, which concertedly function in the removal
of inter-strand DNA-cross-links [155]. Many FA peatnits develop various types of childhood cancer

Un pour Un
Permettre à tous d'accéder à la lecture
Pour chaque accès à la bibliothèque, YouScribe donne un accès à une personne dans le besoin