Medical University of Vienna and Christian Doppler Laboratory on

Molecular Cancer Chemoprevention

Establishment of an In Vitro Co-culture Model to Study the Molecular Pathways Involved in Ulcerative Colitis Associated

Colorectal Cancer

Doctoral Thesis at the Medical University of Vienna

for obtaining the "Doctor of Philosophy" Degree resp. "PhD" Degree

Author:

Mag. Christoph Campregher

Supervisor: Prof. Christoph Gasche Division of Gastroenterology and Hepatology, Department of Medicine III

Medical University of Vienna, Austria

& Christian Doppler Laboratory on

Molecular Cancer Chemoprevention, Vienna, Austria

Wien, 2.Mai 2008

2

Referent 1: Prof. Brigitte Marian

Universitätsklinik für Innere Medizin I & Institut für Krebsforschung

1090 Wien, Borschkegasse 8a

Referent 2: Prof. Erich Heidenreich

Universitätsklinik für Innere Medizin I & Institut für Krebsforschung

1090 Wien, Borschkegasse 8a

Parts of this work are published: Campregher C, Luciani MG, Gasche C. Activated Neutrophils Induce an hMSH2-dependent G2/M Checkpoint Arrest and

Replication Errors at a (CA)13-repeat in Colon Epithelial Cells.J Biol Gut. 2008 Feb

13; [Epub ahead of print] PMID: 18272544 [PubMed - as supplied by publisher]

Campregher C, Luciani MG, Gasche "5-ASA and replication fidelity"

FALK SYMPOSIA Nr.158 (Springer) 2007

Contributions: Luciani MG, Campregher C, Gasche C. Aspirin blocks proliferation in colon cells by inducing a G1 arrest and apoptosis

through activation of the checkpoint kinase ATM.

Carcinogenesis. 2007 Oct;28(10):2207-17. Epub 2007 May 17.

PMID: 17510082 [PubMed - indexed for MEDLINE]

My contribution was the measurement of superoxide scavenging properties of aspirin

(Supplementary Figure S4)

Luciani MG, Campregher C, Fortune JM, Kunkel TA, Gasche C. 5-ASA affects cell cycle progression in colorectal cells by reversibly activating a

replication checkpoint.

Gastroenterology. 2007 Jan;132(1):221-35. Epub 2006 Oct 12.

PMID: 17241873 [PubMed - indexed for MEDLINE]

My contribution was the western blot analysis of mismatch repair proteins (Figure 3A)

In Preparation: Luciani MG, Campregher C, Biesenbach P, Gasche C. Differential Effects of 5-ASA Derivatives 4-ASA, 3-ASA and N-Acetyl-5-ASA on Cell

Cycle and Replication Fidelity in Colon Epithelial Cells

Campregher C, Honeder C, Gasche C Sequence, Composition and Length Affect the Stability of Nucleotide Repeats in

Colon Epithelial Cells

3

4

To Lena, Heidemarie, Adolfo and Julia A Life Vision finally fulfilled

INDEX

INDEX 1. SUMMARY………………………….……………………………………………..7 1. ZUSAMMENFASSUG……………………………………………………………8 2. INTRODUCTION………………………………………..……………………….10 2.1. THE HISTORY: LINKING INFLAMMATION TO CANCER……...………..10 2.2. CHRONIC INFLAMMATION AND COLORECTAL CANCER………….…10 2.2.1. The Inflammatory Process……………………………………..................10 2.2.2. Inflammation and Cancer Development…………………………………..12

2.2.3. Microsatellite Instability in Colorectal Cancer…………………………….14

2.3. DNA REPLICATION MACHINERY…………...……………………………..16

2.4. MISMATCH REPAIR SYSTEM………………….…………………….…….17 2.5. DNA DAMAGE AND CELL CYCLE REGULATION………………..……...19 2.6. CHEMOPREVENTION……..……………………………………………...…20 2.6.1. 5-ASA suppresses spontaneous mutations………………………………20 2.6.2. 5-ASA Counteracts Induced Mutations………………………….………..22 2.7. AIMS OF THIS STUDY……..………………………………………………...24 3. ORIGINAL ARTICLE.…………………..……………………………………….31 3.1 ABSTRACT……….……………………………………………...……………..32

3.2 INTRODUCTION…………..…………………………………...………..…….33 3.3 MATERIAL AND METHODS…………………………………...………..……35

3.3.1. Cell lines……………………………………..…………...………..………..35 3.3.2. Isolation and activation of PMNs………………………..………..………..35

3.3.3. Co-culture and cell cycle analysis……………..………..……..………….35 3.3.4. Western blot analysis………………………..………..…………………….36 3.3.5. Analysis of replication errors………………..………..…………………….36 3.3.6. Statistical analysis……………………..………..…………………..………36 3.4. RESULTS..……………………………………………………...……………..37 3.4.1. Establishment of an in-vitro co-culture system……………………..……37 3.4.2. Activated-PMNs cause an hMHS2-dependent G2/M arrest in colon

epithelial cells……………………………………………………………………….38 3.4.3. Activated-PMNs do not change the expression of MMR proteins…..…39

5

INDEX

6

3.4.4. Activation of the ATM/ATR-targets Chk1 and p53 is associated with the

PMN-induced G2/M arrest…………………………………………………………39 3.4.5. The PMN-induced G2/M arrest depends on the expression of p53 and

p21……………………………………………………………………………...…….41

3.4.6. Superoxide dismutase and catalase do not inhibit phosphorylation at

p53 Ser15 and increased levels of p21waf1/cip1……….….……………………….42 3.4.7. Activated-PMNs cause replication errors in colon epithelial cells……..43 3.5. DISCUSSION…………..……………………….…………………………..…45 CURRICULUM VITAE…………………………….……………………………….54 ERKLÄRUNG …..…………………………………….……………………………58 ACKNOWLEDGEMENTS ..………………………………………………………59

SUMMARY

1. SUMMARY Ulcerative Colitis is an inflammatory bowel disease with an elevated risk for colorectal

cancer. Chronic inflammation creates a hypermutable environment that fulfils the

criteria of a mutator phenotype. Inflammation causes not only direct DNA mutations

through oxidative stress but also an impairment of various repair mechanisms

through a mechanism that needs to be discovered. We intended to simulate such an

environment in cell culture and measure its ability to induce mutations at a defined

microsatellite. Thereby we transferred the clinical situation into the laboratory. We

focused our research on human polymorphonuclear neutrophils (PMN) because

these are the major contributing cell types in ulcerative colitis (UC). PMNs form crypt

abscesses which are characteristic for UC and are located in the mucosa of the large

intestine. Microsatellite instability is a hallmark of an imbalanced mismatch repair

system and is frequently observed in UC related colorectal cancer (CRC).

Our working hypothesis was that PMNs release reactive oxygen species

(ROS) as well as many other factors which can damage DNA. By using a co-culture

system with colon epithelial cells as targets and PMNs as effectors we were able to

analyze biological changes in the colon epithelial cells upon co-culture with PMNs.

PMNs induced a p53, p21 and hMSH2-dependent G2/M cell cycle arrest. This arrest

was paralleled by activation of typical DNA damage response pathways including

phosphorylation of p53 at site Ser15 and Chk1 at site Ser317. Interestingly,

components of the mismatch repair system were not affected and the scavengers

superoxide dismutase and catalase did not overcome the PMN induced G2/M arrest

suggesting other neutrophil-released ROS than superoxide anion or hydrogen

peroxide are responsible for the observed cell cycle arrest. Using an EGFP based

reporter system for mutations (Gasche et al., PNAS 2003) we observed the existence

of PMN induced transient frame-shift mutations at a (CA)13 polynucleotide repeat

(Campregher et al., Gut 2008).

Our model is useful for studying molecular mechanisms of colitis associated

cancer development and protective or chemopreventive effects of certain drugs (or

natural compounds). A proper understanding of the mechanisms by which UC-related

tumors develop might open new avenues for the primary prevention of cancer.

7

ZUSAMMENFASSUNG

1. ZUSAMMENFASSUNG Colitis Ulcerosa ist eine chronisch entzündliche Darmerkrankung mit einem erhöhten

Darmkrebsrisiko. Durch chronische Entzündung entsteht dabei ein hypermutables

Millieu welches alle Kriterien eines „Mutator Phänotyps“ erfüllt. Entzündung

verursacht nicht nur direkte DNA Schäden durch oxidativen Stress, sondern führt

auch zur Beeinträchtigung verschiedener Reparaturmechanismen durch bisher

ungeklärte Ereignisse. Wir haben dieses Milieu in Zellkultur imitiert und mit einem

eigens von uns entwickeltem Verfahren die Entstehung von DNA Mutationen

innerhalb eines definierten Mikrosatelliten gemessen. Dadurch konnten wir die

klinische Situation ins Labor tranferieren. Wir fokusierten uns dabei auf humane

polymorphkernige neutrophile Granuolzyten (PMN) da diese einen entscheidenden

Zelltyp während Colitis Ulcerosa ausmachen. PMNs bilden Krypt-Abszesse welche

charakteristisch für Colitis Ulcerosa sind. Diese Abszesse sind in der Mukosa des

Dickdarms lokalisiert. Mikrosatelliteninstabilität ist ein Kennzeichen eines

beeinträchtigten „mismatch“-Reparatur Systems und kommt häufig bei Colitis

Ulcerosa assoziiertem Kolorektalkarzinom vor.

Unsere Hypothese stützte sich auf die Gegebenheit, das PMNs Reaktive

Sauerstoff-Spezies (ROS) als auch viele andere potentiell DNA schädigende

Faktoren ausschütten. Wir verwendeten ein Kokultur-System mit Darmepithelzellen

als Zielzellen und PMNs als Effektorzellen und konnten so durch PMNs induzierte,

biologische Veränderungen der Epithelzellen messen. PMNs induzierten dabei einen

p53, p21 und hMSH2-abhängigen G2/M Zellzyklus Arrest. Dieser Arrest wurde von

der Aktivierung typischer DNA-Schädigungs Signalwege, wie die Phosphorylierung

von p53 an Serin 15 und Chk1 an Serin 317, begleitet. Interessanterweise waren die

Komponenten des „mismatch“-Reparatur Systems dabei nicht beeinträchtigt und die

Radikalfänger Superoxiddismutase und Katalase konnten einen G2/M Arrest nicht

verhindern. Dies ist ein Anzeichen dafür, das andere von PMNs sezernierte Radikale

oder Faktoren für den Zellzyklusarrest verantwortlich sind. Mit einem EGFP-basierten

Mutationsdetektionssystem (Gasche et al., PNAS 2003) konnten wir die Existenz von

PMN induzierten transienten Leserahmen-Mutationen in einer (CA)13

Tandemwiederholung nachweisen (Campregher et al., Gut 2008).

Unser Modell ist geeignet, spezifische Mechanismen der Krebsentstehung bei

Colitis ulcerosa zu studieren. Insbesonders aber werden wir Medikamente oder

natürliche Substanzen (z.B. Pflanzenextrakte oder Nahrungsbestandteile) auf deren

8

ZUSAMMENFASSUNG

9

Fähigkeit testen, die Genauigkeit der DNA-Replikation während der Zellteilung zu

verbessern. Die Ergebnisse dieser Tests können in der Klinik angewendet werden,

indem diese Substanzen in Zukunft zur Prävention von Dickdarmkrebs bei Colitis

ulcerosa und eventuell auch von anderen entzündungs-assoziierten Krebsarten im

Gastrointestinaltrakt (z.B. dem immer häufiger auftretenden Speiseröhrenkrebs)

eingesetzt werden.

INTRODUCTION

2. INTRODUCTION

2.1. THE HISTORY: LINKING INFLAMMATION TO CANCER In the 19th century Rudolf Virchow (Figure 2-1), a German doctor, noted

accumulations of white blood cells (leucocytes) in neoplastic tissue and hypothesized

that this “lymphoreticular infiltrate” reflects the basis of cancer at sites of chronic

inflammation1. With his most noteworthy publication

‘Cellularpathologie’ he took pathology to a cellular

rationale and it became the major basis for his research

in oncology. He also investigated aspects of

inflammation, regardless only a few links to tumor

pathology were known. The few links from viral or

bacterial infection and inflammation to tumor pathology

have almost been forgotten and have never been

evaluated and discussed sufficiently. Contagious

diseases such as syphilis and tuberculosis have

hallmarks of a ‘tumorigenic process’ and were therefore often difficult or unfeasible to

separate from a ‘genuine’ tumor process, which was primarily recognized by Virchow.

Figure 2-1: Rudolf Ludwig Karl Virchow

Over the past decades our knowledge about the connection of inflammation and

cancer development has supported Virchow´s hypothesis several times2, 3 and the

connection between this reaction of the immune system response and cancer starts

to have implications in developing strategies for treatment and prevention.

2.2. CHRONIC INFLAMMATION AND COLORECTAL CANCER

2.2.1. The Inflammatory Process The origin of inflammation can be of infectious and non-infectious character 4, 5 and is

initiated by injury or irritation. Cytokines or receptor molecules that sense microbes

lead to a recruitment of inflammatory cells (e.g. macrophages or leukocytes) to sites

of affected tissue. Among those inflammatory cells, neutrophils (PMNs) are the

primary cells which reach the site of lesion and are the most important immune cells

of the innate response during infection 6. These cells represent 50 to 60% of the total

circulating leukocytes and are known as the ''first line of defense'' against organisms

or substances that penetrate the body's physical barriers. The bone marrow of a

10

INTRODUCTION

healthy adult produces more than 1011 PMNs daily and more than 1012 per day in the

setting of acute inflammation. Upon release into circulation, PMNs are in a non-

activated state and have a half-life of only 4 to 10 h before marginating and entering

tissue pools, where they survive for further 1 to 2 days. Senescent PMNs are thought

to undergo apoptosis prior to removal by macrophages.

PMNs have an improved lifetime during infection and large numbers

accumulate at the inflammatory site to destroy pathogens which try to invade the

tissue. PMNs harbor granules (Figure 2-2) which are of major importance for their

function. The term “granules” is derived from morphological observations.

Figure 2-2: Segmented neutrophils. Two segmented neutrophils are shown in the middle of the image. White arrows show granules (violet) (GNU Free Documentation License)

PMNs can release different species of oxygen-dependent and oxygen-independent

molecular weapons to destroy microbes and to remove or neutralize infectious

agents7. Oxygen-dependent molecules are combined under the term reactive oxygen

species (ROS) such as hydrogen peroxide (H2O2), superoxide anion (O2−) or

hypochlorous acid (HOCl) which are known to be effective radicals against

microbes8. In contrast, PMN-released oxygen-independent molecules are important

for chemotaxis, degranulation, lysis and phagocytosis 7. Inflammatory cells, such as

macrophages, also release reactive nitrogen species (RNS) such as nitric oxide

(NO•) or peroxynitrite (ONOO−) 9 which are mainly biological messenger molecules

involved in physiological reactions of the human body (e.g. blood vessel dilatation).

However, the effect of RNS against bacterial invaders is not as effective compared to

11

INTRODUCTION

ROS. The majority of inflammatory cells which release high amounts of RNS are

macrophages whereas PMNs release only minor amounts.

The generation of ROS is achieved by an activation of a multiprotein enzyme

complex, the so called nicotinamide adenine dinucleotide phosphate (NADPH)

oxidase. The “respiratory” burst is a large increase in oxygen uptake by PMNs but

also macrophages through the activation of an NADPH-cytochrome-dependent

oxidase that reduces oxygen to O2−. Individuals with an inherited mutation in which

the oxidase that reduces oxygen to O2− is decreased or absent (chronic

granulomatous disease) often die as a result of regular microbial infections.

2.2.2. Inflammation and Cancer Development Patients with Barrett’s esophagus are at increased risk for development of

esophageal cancer, with chronic gastritis for gastric cancer, with chronic pancreatitis

for pancreatic cancer, with primary sclerosing cholangitis for cholangiocellular

carcinoma. The accumulation of cancers in the setting of chronic inflammation does

not seem to be organ specific but is generally associated with the inflammatory

process. Cancer is a disease which typically develops very slowly. For most human

tumors there is a 20 year delay between the initial mutation and the clinical detection

of cancer. Ulcerative colitis (UC) is a chronic disease of the colon that is noticeable

by inflammation and ulceration of the colon mucosa (Figure 2-3) and typically starts

in the second or third decade of life.

Figure 2-3: Ulcerative colitis of the colon (kindly provided by Prof. Christoph Gasche)

Tiny ulcers form on the surface of the lining, where they bleed and produce pus and

mucus. Because the inflammation leads to a frequent emptying of the colon,

symptoms are typically diarrhea (partly bloody) and cramps together with abdominal

12

INTRODUCTION

pain. Inflammation in the colon may be confined to the rectum or may continuously

extent to more proximal parts of the colon. If the total colon is involved the disease is

termed pancolitis. Rarely, the disease is associated with chronic inflammation of the

bile duct system, so called primary sclerosing cholangitis.

UC affects around 0.3% of Western population. The pathogenesis of this

disease is only partially understood regarding autoimmunity, genetic predisposition

and environmental based triggers (e.g. nutrition10). The disease is associated with an

enhanced colorectal cancer (CRC) (Figure 2-4) risk that increases with disease

extent (e.g. 19-fold in pancolitis) 11, young age at onset 12, family history of CRC 13,

presence of primary sclerosing cholangitis 14, 15 or backwash ileitis 16.

NO, oxidativestress

MSI, CIN,CIMP

MSI, CIN,CIMP

MSI, CIN,CIMP

Accumulationof mutations, clonal selection

Accumulationof mutations, clonal selection

Dysplasia Dysplasia

CancerCancer

Figure 2-4: Development of CRC in the context of persistent inflammation. Oxidative or NO induced stress leads to either DNA damage, inactivation of the mismatch repair system (MMR) or failures in checkpoints. The result can be microsatellite instability (MSI), chromosomal instability (CIN) or a CpG island methylator phenotype (CIMP). These can further lead to an accumulation of mutations and clonal selection which ultimately leads to dysplasia and cancer.

In UC, CRC development occurs at an elevated rate and speed. For many years it

was hypothesized that cancer cells exhibit a mutator phenotype17. The spontaneous

mutation rate of human cells is approximately 1.4 x 10-10 per base pair per cell

generation. The basic principle is that normal mutation rates are not enough to

account for the numerous mutations observed in cancer cells, and, therefore,

molecular changes that increase mutation rates are essential for tumor development.

13

INTRODUCTION

The mutator phenotype hypothesis proposes that the intrinsic genetic instability of

cancer cells drives tumorgenesis by producing a pool of mutations, some of which

confer a selective advantage, allowing cells to proliferate under adverse conditions.

The best example of a mutator phenotype in human cancer has been found in tumors

from patients with hereditary nonpolyposis colorectal cancer (HNPCC), which display

microsatellite instability (MSI) due to germline mutations in major mismatch repair

(MMR) genes.

2.2.3. Microsatellite Instability in Colorectal Cancer DNA microsatellites are tandem repeats composed of one to six nucleotide bases.

The most common types of repeats are (A)n and (CA)n, which are ubiquitously

spread all over the human genome about 105 times and occasionally occur in coding

regions of genes. In fact, most human colon cancer cell lines with MSI have

mutations of poly(A) repeats in codon 125 to 128 of the TGF-βRII gene 18. Mutations

in sequences of repeated nucleotides have been reported in numerous tumor

suppressor genes including ATR, IGFIIR, BAX, hMSH3, hMSH6, E2F4, TCF,

caspase-5, MBD4, and MLH3. Novel technologies such as inhibition of nonsense

mediated decay may lead to the discovery of further tumor suppressor genes that

harbor a microsatellite in its coding region 19. The lengths of the repeated sequences

are very polymorphic throughout the population. Microsatellite sequences of

individuals remain unchanged in every tissue; they are stably inherited and extremely

valuable for linkage analysis and also for genomic mapping. The use of

microsatellites in a genome-wide search for the HNPCC locus allowed Peltomaki et

al. 20 to map the first HNPCC gene (i.e. hMSH2) to chromosome 2p15-16 in two large

CRC-prone families and MSI was hereby recognized as a signature feature of

pathways of CRC development in which processes that determine replication fidelity

such as MMR are defect 21.

The current concept is that an impaired DNA MMR system leads to frame-shift

mutations (insertions or deletions) in tandem repeats of several tumor suppressor

genes and their inactivation, which progressively releases the cell from normal

growth restraints and eventually produces a clone of cells with a significant growth

advantage. Two forms of MSI have been recognized and classified according to the

number of mutated microsatellites: MSI-low when only one out of a panel of five

microsatellites (typically a dinucleotide repeat) is mutated, and MSI-high when two or

14

INTRODUCTION

more microsatellites (typically both mono- and dinucleotide repeats) are mutated 22.

In UC, MSI-low was not only found in dysplastic and cancerous tissue but also in

chronically inflamed non-dysplastic mucosa suggesting an impaired replication fidelity

which is a key mechanism early in the development of UC-associated CRC 23.

However, scientists have been unable to find confirmation for inactivation of DNA

MMR genes in these tumors 24, 25. The prevailing hypothesis is that high loads of

ROS and RNS released by inflammatory cells overwhelm DNA repair pathways

leading to an accumulation of DNA lesions which can turn into mutations 26. Perhaps

the mechanism responsible for inflammation-associated mutagenesis is multifaceted

and involves a combined increase in the concentration of mutagens together with an

inactivation of the DNA repair apparatus by oxidative stress 27 or by promoter

hypermethylation 28. Furthermore, an adaptive imbalance in base excision-repair

(BER) enzymes was recently identified as a novel mechanism that may result in MSI-

low 29.

MSI was first described in CRC not selected on the basis of the diagnosis of

hereditary colon cancer. Depending upon the criteria used, 10-20% of colorectal

cancers have MSI. MSI is not an essential characteristic of CRC and is not limited to

tumors of this organ, nor limited to HNPCC. MSI can be found in gastric cancers,

endometrial cancers, ovarian tumors, urinary bladder tumors, non-small cell lung

cancers, small cell lung cancers, breast cancers, and other tumors 30. In some

tumors with MSI, inactivation or genetic deletion of a DNA MMR gene can be found,

but this is not always the case. In the majority of instances, there are no germ line

mutations at any of the known HNPCC loci when MSI is found. Even more

interestingly, MSI is frequently found in up to 50% of non-dysplastic chronic inflamed

mucosa 31. These tumors are mainly caused by CpG island hypermethylation of the

hMLH1 promoter 32.

There are certain genetic and phenotypic differences between colitis-associated

cancer and sporadic CRC, but these differences remain controversial 33-35. Most

importantly, polyp development does not usually occur in colitis-associated cancer,

which typically arises from flat mucosa, endoscopically recognized as DALMS

(dysplasia-associated lesion or mass). This is similar to tumors that develop within

the MSI pathway. Indeed, MSI can be found in dysplastic lesions or cancer

associated with UC 36. Currently, it is thought that oxidative stress may temporarily

inactivate the MMR system and thereby allow such mutations to accumulate 37, 38. In

15

INTRODUCTION

stable transfected DNA, no increase in mutation frequency to H2O2 induced oxidative

stress has been observed so far (C Gasche, unpublished data)39. It seems that H2O2

is insufficient to effectively inactivate the MMR system.

Another feature of UC-related CRC is early p53 mutations, which are also found

in non-dysplastic chronically inflamed mucosa 40. In contrast, mutations of the

gatekeeper gene, adenomatous polyposis of the colon (APC), are uncommon in UC-

related dysplasia or cancer 41. The mechanism behind inflammation-associated

tumor development is generally thought to be related to oxidative stress-induced

DNA (single or double) strand breaks, point- or frameshift mutations. However, the

exact mechanism has not yet been defined. Activation of AP-1 or NFκB may also

prevent cells from undergoing apoptosis 42, 43.

Mutations at microsatellites (i.e. MSI) are a function of polymerase errors and

post-replication MMR that is operative in every cell, and is responsible for correcting

certain types of mutations that occurs during the replication of DNA 44. MSI-low may

reflect a different pathway than MSI-high 45. The presence of MSI in the tissue is the

fingerprint of an ineffective MMR process.

2.3. DNA REPLICATION MACHINERY

RF-C

PCNA

Pol δ

Helicase(Mcm)

RNaseHFEN-1Pol δDNA Ligase

Pol δ

Polα-primase

RP-A RNA primer

3’5’

5’

5’3’

5’3’

Okazakifragment

Leading Strand

Lagging Strand

Figure 2-5: DNA replication fork. During replication the DNA double helix unwinds, with each single strand becoming a template for synthesis of a new, complementary strand. RP-A: single-stranded DNA binding protein; RF-C: “clamp loader” to assemble PCNA onto a template + primer; PCNA: ring-shaped factor (‘clamp’) for DNA polymerases; helicase: unwinds double-stranded DNA ahead of the fork - Mcm2-7

16

INTRODUCTION

High fidelity in DNA synthesis is important for preventing mutations that could initiate

and promote cancer development. The fidelity of DNA replication derives from

polymerase accuracy and its proofreading activity. Genome stability also requires the

ability to repair post-replicational DNA damage and the proficiency of the MMR

system46. In hereditary non-polyposis colorectal cancer (HNPCC) loss-of-function

mutations of DNA MMR proteins, such as hMLH1 or hMSH2, reduce the activity of

post-replicational DNA mismatch repair and strongly increase the spontaneous

mutation rate.

DNA replication is a semi-conservative process. Replication occurs only in the

S-phase during the cell cycle where one DNA strand serves as the template for the

second DNA strand. DNA exists in the nucleus as a compact, condensed structure.

To prepare the DNA for replication, a series of proteins are involved in the unwinding

and separation of the double-stranded DNA. DNA replication proteins such as the

single-strand DNA protein RPA, the clamp/clamp loader complex PCNA/RFC and

DNA polymerase α and δ are recruited to replication initiation sites47. The newly

synthesized DNA strand is generated as RNA-initiated discontinuous segments

called Okazaki fragments which later are joined by ligase activity (Figure 2-5).

2.4. MISMATCH REPAIR SYSTEM Prokaryotic and eukaryotic cells have several efficient repair systems to deal with

spontaneous or acquired DNA damage. In general, these systems detect and repair

errors in the DNA in order to prevent their correct propagation in daughter cells.

Genetic diversity would be much more dangerous for multicellular organisms, and the

DNA repair systems in eukaryotes normally remove mutations to limit genetic

variability. An imbalance between systems that regulate DNA fidelity might be

responsible for human diseases.

During DNA replication the MMR system serves as a control mechanism to

avoid mismatch errors which can lead to dysfunction or complete loss of proteins.

The MMR system requires several enzymes which can recognize and remove

mismatches that result in distortion of the DNA helix. Errors corrected by MMR

include not only single base pair mismatches (i.e. simple mispairings that are not

Watson-Crick matches), but also "loop-outs" of unpaired bases that may occur in the

newly synthesized DNA strand at repetitive DNA sequences called microsatellites,

which are extended repeats of one to six nucleotide sequences 48. The MMR system

17

INTRODUCTION

is just one of several DNA repair systems, some of which repair specific types of

damage.

The hMSH2 protein is the major mismatch recognition component of the

eukaryotic MMR. hMSH2 has a broad specificity in recognizing single base pair

mismatches and multiple base insertion-deletion loops (IDLs)49 at repetitive DNA

sequences called microsatellites50. hMSH2 and hMSH6 form heterodimer (called

hMutSβ), which bind to such mismatches or IDLs and initiate the repair on the newly

synthesized DNA strand. hMLH1 and hPMS2 form a heterodimer (called hMutLα),

which interacts with the hMSH2/hMSH6 complex and actually induce repair. Several

other enzymes, such as DNA endonuclease, helicase, polymerase, ligase, etc. are

required to complete the repair process, which proceeds on the strand of DNA that

contained the mismatch, and includes removal of misaligned nucleotides, de novo

DNA synthesis, and DNA ligation (Figure 2-6).

hMutSα binds tomismatches

hMSH6

hMSH2

ACCCGTAC

5’ and 3’ nickingby MutLα

hPMS2

hMLH1

C

TGGGCATG

ACCCGTACTGGGCATG

A

B

C

ATP

ADP

Correct re-synthesis of primer strand by DNA polymerase α holoenzyme

C

Figure 2-6: Recognition of mismatches or insertion-deletion loops (IDLs) by the mismatch repair (MMR) system. A) The MMR complex hMutSα (a heterodimer of hMSH2 and hMSH6) binds to a mismatch or IDL. B) hMutLα is recruited to the hMutSα complex and induces repair. C) The ExoI exonuclease activity removes the mispaired or looped basepairs and DNA polymerase α holoenzyme fills the gap on the primer strand.

18

INTRODUCTION

2.5. DNA DAMAGE AND CELL CYCLE REGULATION To maintain genome stability and monitor the structure of chromosomes, eukaryotic

cells have evolved surveillance mechanisms called cellular checkpoints51, 52.

Checkpoint pathways include damage sensors, signal transducers, and effectors51

and are activated by DNA damage and incomplete DNA replication53. ATM and ATR

are caffeine-sensitive PI 3-like kinases which act as transducers in the checkpoint

responses54, 55. The ATM pathway responds to the presence of DNA double-strand

breaks (DSBs)56, 57, whereas ATR (ATM and Rad3-Related) is mainly activated by

agents that interfere with replication forks, such as ultraviolet (UV) light and

hydroxyurea (HU)57, 58. If DNA synthesis is impaired (by the presence of stalled

replication forks or DNA damage that occurs during S-phase) the intra-S-phase

checkpoint stabilizes components of replication forks and prevents initiation of more

DNA origins. Activation of this checkpoint is mediated by ATR and leads to the down-

regulation of the S-phase kinases59, 60. Chk1 and Chk2 act downstream of ATM and

ATR61. These kinases work by phosphorylating and inactivating the CDC25

phosphatases. As a consequence, cells arrest in late G1, S and G2 phases62, 63

(Figure 2-7). Moreover, ATM and ATR directly phosphorylate and stabilize p53 in vivo

on Ser15 and Ser3764-66.

S-phase

M-phaseG1-phase

G2-phase

STOP

STOP

STOP

STOPSTOP

STOPSTOP

STOPSTOP

STOPSTOP

STOPSTOPSTOPSTOP

Figure 2-7: Cell Cycle Checkpoints sense DNA damage and ensure the integrity of the genome and fidelity of replication. The series of steps that a eukaryotic cell goes through to duplicate its genetic material and split into two daughter cells is called cell cycle. This is divided into 4 phases called: M-phase (Mitosis), G1 (Gap phase 1), S-phase (DNA Synthesis) and G2. Errors in regulation of the cell cycle can lead to uncontrolled growth and cancer. The cells presents protective mechanisms called checkpoints (animation) which become activated in case of cellular damage which interfere with cell cycle progression (animation) In order to ensure genomic stability.

19

INTRODUCTION

2.6. CHEMOPREVENTION Chemoprevention is the effort to use natural or synthetic compounds to interfere in

the early stages of cancer development, before an invasive disease starts. This

approach is involved in carcinogenesis -- the transformation of a normal cell into a

cancer cell. Currently, more than 1000 potential agents are under investigation

(Database of agents & diets ranked by efficacy, http://www.inra.fr/reseau-nacre/sci-

memb/corpet/indexan.html).

Chemopreventive agents can act in two different ways: they can prevent or

stop DNA alterations that lead to cancer, and they can prevent or stop processes that

lead to excessive replication of such transformed cells. Chemoprevention involves

administering nontoxic agents to otherwise healthy individuals who may be at

increased risk for cancer (e.g. ulcerative colitis).

There is accumulating evidence for the chemopreventive activity of

mesalazine in UC-associated CRC 67. In one study this chemopreventive effect of

mesalazine was estimated to be as high as 91% risk reduction of CRC development 68. Anti-inflammatory, oxygen scavenging, or pro-apoptotic properties of mesalazine

have been contributed to this observation 69-71.

2.6.1. 5-ASA suppresses spontaneous mutations We recently developed a flow cytometry-based assay to study the fidelity on the

replication of microsatellite sequences72. Frameshift mutations were quantified at a

(CA)13 repeat that shifted an enhanced green fluorescence protein (EGFP) into a +2

position, thereby leading to expression of a non-fluorescent peptide72, 73. With this

assay, we detected three cell populations according to their fluorescence intensity:

non-fluorescent, non-mutant M0 cells; dimly fluorescent, intermediate mutant M1

cells; and strongly fluorescent, definitive mutant M2 cells72. We showed that

intermediate mutant M1 cells carry (CA)13-(GT)12 DNA heteroduplexes that are only

present immediately after the DNA replication, before repair takes place. A failure of

the MMR complex to recognize these heteroduplexes results in the generation of

mutant M2 cells carrying (CA)12-(GT)12 homoduplexes.

Treatment with 5-ASA leads to inhibition of cell growth and proliferation in

colon epithelial cells74-76. However, at concentrations ranging from 1.25 mM to 5 mM,

5-ASA caused a dose dependent reduction of cell proliferation and frameshift

mutations at a (CA)13 repeat (Figure 2-8). Treatment with 5-ASA, caused a

20

INTRODUCTION

significant dose dependent drop in the mutant cell fraction, suggesting a role of 5-

ASA in the reduction of spontaneous frameshift mutation rate in colon cells. As this

effect was equally observed in cells bearing the hMLH1 protein (data not shown) and

in cells in which hMLH1 was not expressed, we concluded that the effect of 5-ASA on

the increase of replication fidelity is independent on hMLH1. Recently, we have also

shown that 5-ASA does not directly interfere with DNA polymerase in an in-vitro

polymerase assay74. However, it is still possible that 5-ASA may interact with other

factors of the replication machinery such as PCNA or replication factor C, which may

alter the structure of the replication fork.

0.E+002.E+044.E+046.E+048.E+041.E+051.E+051.E+05

0 1.25 2.5 5

mesalazine (mM)

tota

l cel

ls

0.00000.00100.00200.00300.00400.00500.00600.00700.0080

0 1.25 2.5 5

mesalazine (mM)

mut

ant f

ract

ion

A

B

Figure 2-8: Effect 5-ASA on the mutation rate at a (CA)13 microsatellite. Non-fluorescent HCT116 cell were sorted into 24-well plates (1x103 cells per well) and cultured for 8 days with addition of various nontoxic concentrations of 5-ASA (0 to 5mM). Cells were harvested and fluorescent (mutant) cells were quantitated by flow cytometry. The mutant fraction was calculated as the number of fluorescent cells (M1 and M2) per total cells (M0+M1+M2). Treatment with 5-ASA caused a dose dependent drop in the mutant cell fraction (p

INTRODUCTION

2.6.2. 5-ASA Counteracts Induced Mutations 5-ASA counteracts the mutagenic effect of the intercalating agent 9-aminoacridine

It has been proposed that 5-ASA’s anti-mutagenic activity is explained by its excellent

oxygen scavenging properties. 9-Aminoacridine (9-AA) is an intercalating mutagen

and causes predominantly frameshift mutations. In our laboratory, we studied the

ability of 9-AA to increase frameshift mutations in colon epithelial cells and the effect

of 5-ASA during the exposure to such a strong frameshift mutagen.

HCT116+ch3 (an hMLH1-corrected colorectal cell line77) harboring a (CA)13

repeat, were treated with 9-AA alone or 9-AA in combination with 5-ASA. The total

cell number and the EGFP-positive fraction [mutant fraction (MF)] were analyzed by

flow cytometry as described earlier72. 9-AA caused a decrease of total cell number

and an increase of MF (up to 15-fold) in HCT116+chr3 cells. In our model, 5-ASA

reduces the 9-AA-induced mutation rate about 50% (Figure 2-9). These data support

the concept that 5-ASA counteracts the 9-AA induced mutation rate at a poly(CA)

tract independently of its anti-inflammatory properties (DDW 2007).

0.E+00

1.E+05

2.E+05

3.E+05

4.E+05

5.E+05

6.E+05

7.E+05

c o n tro l 5 µ M 9 -A A 5 µ M 9 -A A & 5 mM 5 A S A

tota

l cel

ls

0.0000

0.0040

0.0080

0.0120

0.0160

0.0200

c o n tro l 5 µ M 9 -A A 5 µ M 9 -A A & 5 m M 5 A S A

mut

ant f

ract

ion

A

B

control

5µM 9-AA

5µM 9-AA & 5mM 5ASA

Figure 2-9: Effects of 9-aminoacridine (9-AA) and 5-ASA on the mutation rate at a (CA)13 microsatellite. HCT116+chr3 cells (harboring a (CA)13 repeat) were plated into 6-well plates (5x103 cells per well) and cultured for 8 days with addition of 9-AA or 9-AA + 5-ASA. Cells were harvested and fluorescent (mutant) cells were quantitated by flow cytometry as described72. The mutant fraction was calculated as the number of fluorescent cells (M1 and M2) per total cells (M0+M1+M2). Treatment with 5-ASA caused a drop in the 9-AA induced mutant cell fraction (p

INTRODUCTION

Our recent results suggest that 5-ASA has anti-mutagenic properties and in this

respect, it is useful for prevention of colorectal cancer independently of its anti-

inflammatory properties. Other molecules, similar to 5-ASA, may have similar anti-

carcinogenic impact on the inflamed colon mucosa. 5-ASA is a key molecule for

chemical engineering to create new derivatives with similar or even stronger

efficiencies in cancer prevention. The understanding of the molecular mechanisms of

chemoprevention may enable the design of more active and less toxic drugs and the

hunt for the identification of novel targets78 will bring new insights into cancer

chemoprevention.

23

INTRODUCTION

2.7. AIMS OF THIS STUDY We tried here to contribute answering following essential questions in UC associated

CRC by using an in vitro co-culture system (Figure 2-10) which mimics the conditions

of this disease: Are human peripheral PMNs capable of inducing frameshift mutations

in colon epithelial cells and which molecular pathways are activated upon PMN

induced oxidative stress? A proper understanding of the mechanisms by which UC-

related tumors develop might open new avenues for the primary prevention of

cancer.

Analysis ofTarget Cells

Co-culture

Target : effector ratios0:1 – 100:1

HCT116 (hMLH1-/-)HCT116+chr3 (hMLH1+/-)

Semipermeable membrane

PMA activatedneutrophils (PMNs)

Addition of Effector Cells

Plating Target Cells

Figure 2-10: Co-culture system to simulate inflammatory conditions in ulcerative colitis. Colon epithelial cells are plated into a 6well plate. A semipermeable membrane is placed into the 6well. Into the so created upper chamber neutrophils are added in ratios of up to 100 effector cells per 1 epithelial cell. The effector cells are activated using Phorbol-myristate-acetate. After an appropriate time of co-culture the upper chamber is removed and epithelial cells can be analyzed for frameshift mutations, cell cycle changes or protein expression or phosphorylation levels.

24

INTRODUCTION

Reference List

1. Balkwill F, Mantovani A. Inflammation and cancer: back to Virchow? Lancet

2001;357:539-545.

2. Jackson AL, Chen R, Loeb LA. Induction of microsatellite instability by oxidative DNA damage. Proc Natl Acad Sci U S A 1998;95:12468-12473.

3. Campregher C, Luciani MG, Gasche C. Activated Neutrophils Induce an hMSH2-dependent G2/M Checkpoint Arrest and Replication Errors at a (CA)13-repeat in Colon Epithelial Cells. Gut 2008.

4. Coussens LM, Werb Z. Inflammation and cancer. Nature 2002;420:860-867.

5. Nathan C. Points of control in inflammation. Nature 2002;420:846-852.

6. Ricevuti G. Host tissue damage by phagocytes. Ann N Y Acad Sci 1997;832:426-448.

7. Witko-Sarsat V, Rieu P, scamps-Latscha B, Lesavre P, Halbwachs-Mecarelli L. Neutrophils: molecules, functions and pathophysiological aspects. Lab Invest 2000;80:617-653.

8. Roos D, van BR, Meischl C. Oxidative killing of microbes by neutrophils. Microbes Infect 2003;5:1307-1315.

9. Perwez HS, Harris CC. Inflammation and cancer: An ancient link with novel potentials. Int J Cancer 2007;121:2373-2380.

10. Bartel G, Weiss I, Turetschek K, Schima W, Puspok A, Waldhoer T, Gasche C. Ingested matter affects intestinal lesions in Crohn's disease. Inflamm Bowel Dis 2008;14:374-382.

11. Gyde SN, Prior P, Allan RN, Stevens A, Jewell DP, Truelove SC, Lofberg R, Brostrom O, Hellers G. Colorectal cancer in ulcerative colitis: a cohort study of primary referrals from three centres. Gut 1988;29:206-217.

12. Ekbom A, Helmick C, Zack M, Adami HO. Ulcerative colitis and colorectal cancer. A population-based study. N Engl J Med 1990;323:1228-1233.

13. Nuako KW, Ahlquist DA, Mahoney DW, Schaid DJ, Siems DM, Lindor NM. Familial predisposition for colorectal cancer in chronic ulcerative colitis: a case-control study. Gastroenterology 1998;115:1079-1083.

14. Broome U, Lofberg R, Veress B, Eriksson LS. Primary sclerosing cholangitis and ulcerative colitis: evidence for increased neoplastic potential [see comments]. Hepatology 1995;22:1404-1408.

15. Loftus EV, Jr., Aguilar HI, Sandborn WJ, Tremaine WJ, Krom RA, Zinsmeister AR, Graziadei IW, Wiesner RH. Risk of colorectal neoplasia in patients with primary sclerosing cholangitis and ulcerative colitis following orthotopic liver transplantation. Hepatology 1998;27:685-690.

25

INTRODUCTION

16. Heuschen UA, Hinz U, Allemeyer EH, Stern J, Lucas M, Autschbach F, Herfarth C, Heuschen G. Backwash ileitis is strongly associated with colorectal carcinoma in ulcerative colitis. Gastroenterology 2001;120:841-847.

17. Loeb LA, Springgate CF, Battula N. Errors in DNA replication as a basis of malignant changes. Cancer Res 1974;34:2311-2321.

18. Markowitz S, Wang J, Myeroff L, Parsons R, Sun L, Lutterbaugh J, Fan RS, Zborowska E, Kinzler KW, Vogelstein B. Inactivation of the type II TGF-beta receptor in colon cancer cells with microsatellite instability. Science 1995;268:1336-1338.

19. Ionov Y, Nowak N, Perucho M, Markowitz S, Cowell JK. Manipulation of nonsense mediated decay identifies gene mutations in colon cancer Cells with microsatellite instability. Oncogene 2004;23:639-645.

20. Peltomaki P, Aaltonen LA, Sistonen P, Pylkkanen L, Mecklin JP, Jarvinen H, Green JS, Jass JR, Weber JL, Leach FS. Genetic mapping of a locus predisposing to human colorectal cancer [see comments]. Science 1993;260:810-812.

21. Ionov Y, Peinado MA, Malkhosyan S, Shibata D, Perucho M. Ubiquitous somatic mutations in simple repeated sequences reveal a new mechanism for colonic carcinogenesis. Nature 1993;363:558-561.

22. Boland CR, Thibodeau SN, Hamilton SR, Sidransky D, Eshleman JR, Burt RW, Meltzer SJ, Rodriguez-Bigas MA, Fodde R, Ranzani GN, Srivastava S. 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 Res 1998;58:5248-5257.

23. Brentnall TA, Crispin DA, Bronner MP, Cherian SP, Hueffed M, Rabinovitch PS, Rubin CE, Haggitt RC, Boland CR. Microsatellite instability in nonneoplastic mucosa from patients with chronic ulcerative colitis. Cancer Res 1996;56:1237-1240.

24. Noffsinger AE, Belli JM, Fogt F, Fischer J, Goldman H, Fenoglio-Preiser CM. A germline hMSH2 alteration is unrelated to colonic microsatellite instability in patients with ulcerative colitis. Hum Pathol 1999;30:8-12.

25. Cawkwell L, Sutherland F, Murgatroyd H, Jarvis P, Gray S, Cross D, Shepherd N, Day D, Quirke P. Defective hMSH2/hMLH1 protein expression is seen infrequently in ulcerative colitis associated colorectal cancers. Gut 2000;46:367-369.

26. Loeb KR, Loeb LA. Genetic instability and the mutator phenotype. Studies in ulcerative colitis. Am J Pathol 1999;154:1621-1626.

27. Chang CL, Marra G, Chauhan DP, Ha HT, Chang DK, Ricciardiello L, Randolph A, Carethers JM, Boland CR. Oxidative stress inactivates the human DNA mismatch repair system. Am J Physiol Cell Physiol 2002;283:C148-C154.

26

INTRODUCTION

28. Fleisher AS, Esteller M, Harpaz N, Leytin A, Rashid A, Xu Y, Liang J, Stine OC, Yin J, Zou TT, Abraham JM, Kong D, Wilson KT, James SP, Herman JG, Meltzer SJ. Microsatellite instability in inflammatory bowel disease-associated neoplastic lesions is associated with hypermethylation and diminished expression of the DNA mismatch repair gene, hMLH1. Cancer Res 2000;60:4864-4868.

29. Hofseth LJ, Khan MA, Ambrose M, Nikolayeva O, Xu-Welliver M, Kartalou M, Hussain SP, Roth RB, Zhou X, Mechanic LE, Zurer I, Rotter V, Samson LD, Harris CC. The adaptive imbalance in base excision-repair enzymes generates microsatellite instability in chronic inflammation. J Clin Invest 2003;112:1887-1894.

30. Boland CR. Roles of the DNA mismatch repair genes in colorectal tumorigenesis. Int J Cancer 1996;69:47-49.

31. Brentnall TA, Crispin DA, Bronner MP, Cherian SP, Hueffed M, Rabinovitch PS, Rubin CE, Haggitt RC, Boland CR. Microsatellite instability in nonneoplastic mucosa from patients with chronic ulcerative colitis. Cancer Res 1996;56:1237-1240.

32. Nakagawa H, Nuovo GJ, Zervos EE, Martin EW, Jr., Salovaara R, Aaltonen LA, de la CA. Age-related hypermethylation of the 5' region of MLH1 in normal colonic mucosa is associated with microsatellite-unstable colorectal cancer development. Cancer Res 2001;61:6991-6995.

33. Rhodes JM, Campbell BJ. Inflammation and colorectal cancer: IBD-associated and sporadic cancer compared. Trends Mol Med 2002;8:10-16.

34. Xie J, Itzkowitz SH. Cancer in inflammatory bowel disease. World J Gastroenterol 2008;14:378-389.

35. Konishi K, Shen L, Wang S, Meltzer SJ, Harpaz N, Issa JP. Rare CpG island methylator phenotype in ulcerative colitis-associated neoplasias. Gastroenterology 2007;132:1254-1260.

36. Suzuki H, Harpaz N, Tarmin L, Yin J, Jiang HY, Bell JD, Hontanosas M, Groisman GM, Abraham JM, Meltzer SJ. Microsatellite instability in ulcerative colitis-associated colorectal dysplasias and cancers. Cancer Res 1994;54:4841-4844.

37. Chang CL, Marra G, Chauhan DP, Ha HT, Chang DK, Ricciardiello L, Randolph A, Carethers JM, Boland CR. Oxidative stress inactivates the human DNA mismatch repair system. Am J Physiol Cell Physiol 2002;283:C148-C154.

38. Gasche C, Chang CL, Rhees J, Goel A, Boland CR. Oxidative stress increases frameshift mutations in human colorectal cancer cells. Cancer Res 2001;61:7444-7448.

39. Yamada NA, Parker JM, Farber RA. Mutation frequency analysis of mononucleotide and dinucleotide repeats after oxidative stress. Environ Mol Mutagen 2003;42:75-84.

27

INTRODUCTION

40. Brentnall TA, Crispin DA, Rabinovitch PS, Haggitt RC, Rubin CE, Stevens AC, Burmer GC. Mutations in the p53 gene: an early marker of neoplastic progression in ulcerative colitis. Gastroenterology 1994;107:369-378.

41. Tarmin L, Yin J, Harpaz N, Kozam M, Noordzij J, Antonio LB, Jiang HY, Chan O, Cymes K, Meltzer SJ. Adenomatous polyposis coli gene mutations in ulcerative colitis- associated dysplasias and cancers versus sporadic colon neoplasms. Cancer Res 1995;55:2035-2038.

42. Shaulian E, Karin M. AP-1 as a regulator of cell life and death. Nat Cell Biol 2002;4:E131-E136.

43. Karin M, Lin A. NF-kappaB at the crossroads of life and death. Nat Immunol 2002;3:221-227.

44. Kunkel TA, Bebenek K. DNA replication fidelity. Annu Rev Biochem 2000;69:497-529.

45. Boland CR, Thibodeau SN, Hamilton SR, Sidransky D, Eshleman JR, Burt RW, Meltzer SJ, Rodriguez-Bigas MA, Fodde R, Ranzani GN, Srivastava S. 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 Res 1998;58:5248-5257.

46. Kunkel TA. Considering the cancer consequences of altered DNA polymerase function. Cancer Cell 2003;3:105-110.

47. Diffley JF, Labib K. The chromosome replication cycle. J Cell Sci 2002;115:869-72.

48. Kolodner RD, Marsischky GT. Eukaryotic DNA mismatch repair. Curr Opin Genet Dev 1999;9:89-96.

49. Fishel R, Wilson T. MutS homologs in mammalian cells. Curr Opin Genet Dev 1997;7:105-113.

50. Kolodner RD, Marsischky GT. Eukaryotic DNA mismatch repair. Curr Opin Genet Dev 1999;9:89-96.

51. Hartwell LH, Weinert TA. Checkpoints: controls that ensure the order of cell cycle events. Science 1989;246:629-634.

52. Hartwell LH, Kastan MB. Cell cycle control and cancer. Science 1994;266:1821-1828.

53. Zhou BB, Elledge SJ. The DNA damage response: putting checkpoints in perspective. Nature 2000;408:433-439.

54. Sarkaria JN. Identifying inhibitors of ATM and ATR kinase activities. Methods Mol Med 2003;85:49-56.

28

INTRODUCTION

55. Yang J, Xu ZP, Huang Y, Hamrick HE, Duerksen-Hughes PJ, Yu YN. ATM and ATR: sensing DNA damage. World J Gastroenterol 2004;10:155-160.

56. Khanna KK, Lavin MF, Jackson SP, Mulhern TD. ATM, a central controller of cellular responses to DNA damage. Cell Death Differ 2001;8:1052-1065.

57. Shiloh Y. ATM (ataxia telangiectasia mutated): expanding roles in the DNA damage response and cellular homeostasis. Biochem Soc Trans 2001;29:661-666.

58. Abraham RT. Cell cycle checkpoint signaling through the ATM and ATR kinases. Genes Dev 2001;15:2177-2196.

59. Jares P, Donaldson A, Blow JJ. The Cdc7/Dbf4 protein kinase: target of the S phase checkpoint? EMBO Rep 2000;1:319-322.

60. Weinreich M, Stillman B. Cdc7p-Dbf4p kinase binds to chromatin during S phase and is regulated by both the APC and the RAD53 checkpoint pathway. EMBO J 1999;18:5334-5346.

61. Bartek J, Lukas C, Lukas J. Checking on DNA damage in S phase. Nat Rev Mol Cell Biol 2004;5:792-804.

62. Hutchins JR, Dikovskaya D, Clarke PR. Dephosphorylation of the inhibitory phosphorylation site S287 in Xenopus Cdc25C by protein phosphatase-2A is inhibited by 14-3-3 binding. FEBS Lett 2002;528:267-271.

63. Raleigh JM, O'Connell MJ. The G(2) DNA damage checkpoint targets both Wee1 and Cdc25. J Cell Sci 2000;113 ( Pt 10):1727-1736.

64. Siliciano JD, Canman CE, Taya Y, Sakaguchi K, Appella E, Kastan MB. DNA damage induces phosphorylation of the amino terminus of p53. Genes Dev 1997;11:3471-3481.

65. Salles-Passador I, Fotedar A, Fotedar R. Cellular response to DNA damage. Link between p53 and DNA-PK. C R Acad Sci III 1999;322:113-120.

66. Waterman MJ, Stavridi ES, Waterman JL, Halazonetis TD. ATM-dependent activation of p53 involves dephosphorylation and association with 14-3-3 proteins. Nat Genet 1998;19:175-178.

67. Eaden J. Review article: the data supporting a role for aminosalicylates in the chemoprevention of colorectal cancer in patients with inflammatory bowel disease. Aliment Pharmacol Ther 2003;18 Suppl 2:15-21.

68. Eaden J, Abrams K, Ekbom A, Jackson E, Mayberry J. Colorectal cancer prevention in ulcerative colitis: a case-control study. Aliment Pharmacol Ther 2000;14:145-153.

69. Bus PJ, Nagtegaal ID, Verspaget HW, Lamers CB, Geldof H, Van Krieken JH, Griffioen G. Mesalazine-induced apoptosis of colorectal cancer: on the verge of a new chemopreventive era? Aliment Pharmacol Ther 1999;13:1397-1402.

29

INTRODUCTION

30

70. Reinacher-Schick A, Schoeneck A, Graeven U, Schwarte-Waldhoff I, Schmiegel W. Mesalazine causes a mitotic arrest and induces caspase-dependent apoptosis in colon carcinoma cells. Carcinogenesis 2003;24:443-451.

71. Reinacher-Schick A, Seidensticker F, Petrasch S, Reiser M, Philippou S, Theegarten D, Freitag G, Schmiegel W. Mesalazine changes apoptosis and proliferation in normal mucosa of patients with sporadic polyps of the large bowel. Endoscopy 2000;32:245-254.

72. Gasche C, Chang CL, Natarajan L, Goel A, Rhees J, Young DJ, Arnold CN, Boland CR. Identification of frame-shift intermediate mutant cells. Proc Natl Acad Sci U S A 2003;100:1914-1919.

73. Gasche C, Chang CL, Rhees J, Goel A, Boland CR. Oxidative stress increases frameshift mutations in human colorectal cancer cells. Cancer Res 2001;61:7444-7448.

74. Luciani MG, Campregher C, Fortune JM, Kunkel TA, Gasche C. 5-ASA affects cell cycle progression in colorectal cells by reversibly activating a replication checkpoint. Gastroenterology 2007;132:221-235.

75. Luciani MG CCGC. Aspirin blocks Proliferation in Colon Cells by inducing a G1-Arrest and Apoptosis through Activation of the Checkpoint Kinase ATM. 2007.

76. Gasche C, Goel A, Natarajan L, Boland CR. Mesalazine improves replication fidelity in cultured colorectal cells. Cancer Res 2005;65:3993-3997.

77. Koi M, Umar A, Chauhan DP, Cherian SP, Carethers JM, Kunkel TA, Boland CR. 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 Res 1994;54:4308-4312.

78. Meisner NC, Hintersteiner M, Uhl V, Weidemann T, Schmied M, Gstach H, Auer M. The chemical hunt for the identification of drugable targets. Curr Opin Chem Biol 2004;8:424-431.

ORIGINAL ARTICLE

3. ORIGINAL ARTICLE Gut. 2008 Feb 13; [Epub ahead of print] Activated Neutrophils Induce an hMSH2-dependent G2/M Checkpoint Arrest and Replication Errors at a (CA)13-repeat in Colon Epithelial Cells Running title: Neutrophils, mismatch repair and cell cycle checkpoints

Christoph Campregher, Maria Gloria Luciani and Christoph Gasche

Medical University of Vienna, Department of Internal Medicine III, Division of

Gastroenterology and Hepatology and Christian Doppler Laboratory on Molecular

Cancer Chemoprevention, Vienna, Austria

LIST OF ABBREVIATIONS

ATM, Ataxia Telengectasia Mutated Kinase; ATR, ATM-and-Rad3-related Kinase;

Chk1, Checkpoint Kinase 1; CAT, Catalase; CRC, Colorectal Cancer; hMLH1, mutL

Homolgue 1; hMSH2, mutS Homologue 2; hMSH6, mutS Homologue 6; MSI,

Microsatellite Instability; MMR, Mismatch Repair; PCNA, proliferating cell nuclear

antigen; PMA, phorbol 12-myristate 13-actetate; PMN, polymorphonuclear

leukocytes; ROS, Reactive Oxygen Species; RPA, replication protein A; SOD,

superoxide dismutase

Correspondence: Prof. Christoph Gasche, AKH Wien, Division of Gastroenterology

and Hepatology, Währinger Gürtel 18, A-1090 Vienna, Austria; tel +43-1-404004764,

fax +43-1-404004735, christoph.gasche@meduniwien.ac.at

"The Corresponding Author has the right to grant on behalf of all authors and does

grant on behalf of all authors, an exclusive licence (or non exclusive for government

employees) on a worldwide basis to the BMJ Publishing Group Ltd and its Licensees

to permit this article (if accepted) to be published in Gut editions and any other

BMJPGL products to exploit all subsidiary rights, as set out in our licence

(http://gut.bmjjournals.com/ifora/licence.dtl)."

Word count: abstract 229, manuscript 3492 (without abstract, figure legends and

references); number of figures: 7

31

mailto:christoph.gasche@meduniwien.ac.at

ORIGINAL ARTICLE

3.1. ABSTRACT Objective: Chronic inflammation in ulcerative colitis is associated with increased risk

for colorectal cancer. Its molecular pathway of cancer development is poorly

understood. We investigated the role of neutrophil-derived cellular stress in an in-vitro

model of neutrophils as effectors, and colon epithelial cells as targets, and tested for

changes in cell cycle distribution and the appearance of replication errors. Design:

Colon epithelial cells with different mismatch repair phenotypes were co-cultured with

activated neutrophils. Target cells were analyzed for cell cycle distribution and

replication errors by flow cytometry. Changes in nuclear and DNA-bound levels of

mismatch repair- and checkpoint-related proteins were analyzed by western blot.

Results: Activated neutrophils cause an accumulation of target cells in G2/M,

consistent with an install of a DNA-damage checkpoint. Cells that do not express

hMSH2, p53 or p21waf1/cip1 failed to undergo the G2/M arrest. Phosphorylation of p53

at site Ser15 and Chk1 at Ser317 as well as accumulation of p21waf1/cip1 was

observed within 8-24 hours. Superoxide Dismutase and catalase were unable to

overcome this G2/M arrest, possibly indicating that neutrophil products other than

superoxide or H2O2 are involved in this cellular response. Finally, exposure to

activated neutrophils increased the number of replication errors. Conclusions: By

using an in vitro co-culture model that mimics intestinal inflammation in ulcerative

colitis, we provide molecular evidence for an hMSH2-dependent G2/M checkpoint

arrest and for the presence of replication errors.

32

ORIGINAL ARTICLE

3.2. INTRODUCTION Chronic inflammation leads to tumor development [1]. Ulcerative colitis (UC), is

associated with an increased risk of development of colorectal carcinoma (CRC).

One of the key features of UC is the presence of crypt abscesses, which are

accumulations of polymorphonuclear-cells (PMNs) within colonic crypts [2, 3].

Reactive oxygen species (ROS) released by PMNs are suggested to be one of the

main contributing factors to colon carcinogenesis [1]. Oxidative stress can alter

cellular components including proteins, mRNAs and DNA [4, 5, 6]. It is unclear,

however, whether oxidative stress on its own may cause mutations in cells [7, 8].

Activated PMNs not only produce ROS, but also excrete lactoferrin [9] and other

proteins including several cytokines [10, 11]. Thus, previous in vitro studies that

focused on H2O2-induced mutagenesis [8, 12] did only partially reflect the

pathophysiological condition of colon carcinogenesis.

The mismatch repair (MMR) system plays a central role in promoting genetic

stability by correcting DNA replication errors. Homologs of the bacterial MutS and

MutL MMR proteins in eukaryotes, form heterodimers with discrete roles in MMR-

related processes. The discovery of a link between human cancer and MMR defects

has led to an increased interest in eukaryotic MMR [13]. Frameshift mutations of

short-tandem repetitive sequences indicate instability of these sequences

(microsatellite instability - MSI) and represent a hallmark of MMR deficiency in human

cancers [14, 15]. Since MSI can be detected in colitis tissue without dysplasia,

inactivation of the MMR system must be an early event in colon carcinogenesis in

UC. However, the nature of inflammation-induced microsatellite mutations is still

obscure. The MMR-system can be activated after replication to repair DNA errors.

Evidence suggested that the PCNA is required for MMR recruitment prior to DNA

repair synthesis [16], leading to the hypothesis that replication and MMR may be

coupled and that the replication fork provide the strand discrimination signal for repair

[17].

Exposure of eukaryotic cells to agents that alter the DNA structure results in

transient arrest of the progression through the cell cycle. The Ataxia Telangiectasia

Mutated kinase (ATM) acts as a sensor of oxidative damage, coordinating stress

responses with cell cycle checkpoint control and repair of such damage [18]. Cell

cycle checkpoints give the cell the opportunity to, either mend the DNA damage or

undergo apoptosis. In particular, the G2/M checkpoint allows cells to overcome

33

ORIGINAL ARTICLE

replication errors before entering mitosis, thereby ensuring genomic integrity. Apart

from ATM, key components of the G2/M cell cycle checkpoint include the ATM-and-

Rad3-related kinase (ATR), the downstream checkpoint kinases Chk1 and Chk2 [19,

20] and the tumor suppressor protein p53 [21], which is stabilized by phosphorylation

at ATM and ATR sites [22, 23]. Phosphorylation of p53 correlates with enhanced

transcription of the cyclin-dependent kinase inhibitor p21waf1/cip1 [24, 25]. DNA-

alkylating agents induce phosphorylation and activation of p53, leading to an

increased expression of p21waf1/cip1. Cell lines with MMR-deficiency are resistant to

these alkylating agents and bypass the cell cycle arrest, indicating that the MMR has

a role in post-replication checkpoints [26, 27]. However, nitric oxide (NO) and H2O2

are capable of arresting hMLH1-mutant cells in G2/M [4, 28]. No information exists on

the role of hMSH2 in mediating such a cell cycle arrest.

In this work, we hypothesize that the chronic exposure of the intestinal mucosa

to activated-PMNs leads to DNA damage, which may activate checkpoint kinases

and initiate MMR, or if this is inefficient, may drive colon carcinogenesis. In order to

simulate the carcinogenic environment in UC, we established an in vitro co-culture

system with primary PMNs as effector cells and various colon cell lines as targets.

Our results show that exposure of colon cells to activated-PMNs install a G2/M cell

cycle checkpoint, indicative of DNA damage, through a mechanism that does not

require hMLH1, but rather p53/p21 and hMSH2. This G2/M arrest is associated with

an increase in replication errors.

34

ORIGINAL ARTICLE

3.3. MATERIALS AND METHODS 3.3.1. Cell lines The human colorectal carcinoma cell lines HCT116hMLH1-/- and their derivatives

HCT116+chr3hMLH1+/- [26], HCT116+chr3 A3.1, HCT116+chr3 A3.7 [29], HCT116-

mlh1-2hMLH1+/-, HCT116p53-/-, HCT116p21-/- [30] and LovohMSH2-/- and their derivatives

Lovo+chr2hMSH2+/- Lovo(DT40.2)-4-1hMSH2-/- [31], as well as the human endometrial

adenocarcinoma cell line HEC59hMSH2-/- and HEC59+chr2hMSH2+/- [32] were grown in

IMDM (Gibco/Invitrogen) containing 10% fetal bovine serum (FBS, Biochrom). The

medium for HCT116+chr3 contained 400µg/ml and for HEC59+chr2 and Lovo+chr2

cells 700µg/mL G418 (Gibco), respectively. The medium for HCT116-mlh1-2 cells

[33] contained 100µg/mL hygromycin B (Invitrogen). The clones HCT116+chr3 A3.1

and HCT116+chr3 A3.7 were grown with 150µg/mL hygromycin B and 400µg/mL

G418. The promyelocytic leukemia cells line HL60 (ATCC CCL-240) was cultured in

RPMI (Gibco/Invitrogen) supplemented with 10% FBS.

3.3.2. Isolation and activation of PMNs PMNs were freshly isolated from heparinized blood of healthy volunteers by dextran

T500 (Pharmacia) sedimentation followed by density gradient centrifugation through

Ficoll-Paque (Amersham) or HL60 cells were derived to granulocyte-like neutrophils

by differentiation as described previously [34]; erythrocytes were lysed in NaCl

(0.2%) followed by NaCl (1.6%) and cells were washed in Ca/Mg-free HBSS (Gibco).

CD66b mAb (BD 55572) was used to confirm the purity of isolated PMNs by flow

cytometry [35]. PMNs were activated in HBSS containing 50ng/mL phorbol 12-

myristate 13-actetate (PMA, Sigma) at 37°C and 5% CO2 for 30 min. PMA was

removed by washing the PMNs twice with HBSS. ·O2¯ production was determined by

lucigenin-enhanced chemiluminescence as described [36, 37]. 1000 U/mL

superoxide dismutase (SOD, Sigma) and 1000 U/mL catalase (CAT, Sigma) were

used as scavengers for ·O2¯ and H2O2, respectively.

3.3.3. Co-culture and cell cycle analysis 7x104 target cells were seeded onto 6-well plates. 24 hours later, PMNs were added

into the upper chamber of a Transwell 0.45µm microporous insert preventing cell-to-

cell contact. PMNs and target cells were co-cultured at effector:target ratios of 0:1 to

35

ORIGINAL ARTICLE

100:1 for up to 24 hours. Target cells were harvested using accutase (PAA

Laboratories GmbH), fixed and analyzed for cell cycle distribution as described [36].

3.3.4. Western blot analysis Cell lysates and DNA-bound fractions were obtained as described [36]. For western

blots analysis 50–150µg of lysates were used with the following antibodies: rabbit

polyclonal antibody (pAb) anti-phospho-p53 Ser15 (Cell Signaling Technology,

Danvers, MA); pAb anti-cleaved caspase-7 (Cell Signaling); mAb anti-p21waf1/cip1 (Cell

Signaling); mAb anti-hMSH2 (Becton-Dickinson), mAb anti-hMHS6 (Becton-

Dickinson), mAb anti-hPMS2 (Becton-Dickinson), mAb anti-hMre11 (Becton-

Dickinson) and mAb anti-hMLH1 (Becton-Dickinson); mAb anti-tubulin (Abcam); mAb

anti-actin (Sigma); pAb anti-phospho-Chk1 Ser317 (Cell Signaling).

3.3.5. Analysis of replication errors Five thousand non-fluorescent HCT116+ch3 cells, bearing the EGFP-based plasmid

pIREShyg2-EGFP/CA13 (clones A3.1, 1 plasmid copy, and A3.7, 2 plasmid copies)

[29] were sorted on a FACSVantage SE using CloneCyt Plus sorting technology

(Becton Dickinson Immunocytometry Systems), and PMNs were activated with PMA

as described above. Cells were cultured at 100:1 ratios for 24 hours and PMNs were

removed. Target cells were grown for additional 7 days. The EGFP-positive

population with low fluorescence intensity was considered as “transiently mutated

fraction” (M1), and that with high fluorescence intensity as “permanently mutated

fraction” (M2) [29].

3.3.6. Statistical analysis Experiments were carried out at least in triplicates and repeated twice. Data are

represented as mean ± S.D. and compared by using the students t test. p values

ORIGINAL ARTICLE

3.4. RESULTS 3.4.1. Establishment of an in-vitro co-culture system To simulate the carcinogenic environment in UC, we established an in-vitro co-culture

system in which aPMN were co-cultured with various colon epithelial cells separated

by a semi-permeable membrane. Activation of PMNs with PMA is followed by

oxidative burst [38]. A strong and sustained ·O2¯ release was observed for at least 2

hours after PMA removal (fig 1A). The expression or absence of MMR components,

p53 and p21 was tested (fig 1B) in the various cell lines used in this work.

Figure 1. Characterization and analysis of colon epithelial cell lines and ·O2¯ release by activated PMNs. (A) Freshly isolated PMNs were activated with 50ng/mL PMA for 30 min. Cells were washed twice and ·O2¯ release was measured by lucigenin-amplified chemiluminescence. A strong induction of ·O2¯ release was observed upon activated PMNs ( ) that lasted for more than 2 hours. Data are given as relative light units (RLU) per 1x106 cells. Non-activated cells ( ) were subjected to the same procedures (isolation and washing) and served as a control. A slightly increased level of ·O2¯ release by non-activated cells was observed. Each data point presents the mean (± SD) out of three independent measurements. (B) Western blot analysis of hMLH1, hMSH2, p21waf1/cip1 and p53 expression in target cell lines. The western blot confirmed the expected loss or reinstalled expression of protein expression of these cells (HCT116hMLH1-/-, HCT116+chr3hMLH1+/-, LovohMSH2-/-, Lovo+chr2hMSH2+/-, HEC59hMSH2-/-, HEC59+chr2hMSH2+/-, HCT116p21-/-, HCT116p53-/- and HCT116-mlh1-2)

37

ORIGINAL ARTICLE

3.4.2. Activated-PMNs cause an hMHS2-dependent G2/M arrest in colon epithelial cells Oxidative stress induces cellular checkpoints, leading to cell cycle arrest and

preventing mitosis of cells with defective DNA replication [39]. Activated-PMNs were

co-cultured with HCT116, HCT116+chr3, Lovo or Lovo+chr2 cells for 24 hours. All

cell lines, except Lovo, displayed an increase in the G2/M population within 24h at

20:1 ratios (aPMN:target cells) (fig 2A), consistent with the MMR component hMSH2,

but not hMLH1, being involved in the G2/M arrest. No such effect was observed

when using non-activated PMNs. A dose effect was observed when HCT116 were

cultured at different effector:target ratios (fig 3B). A similar experiment conducted with

HEC59 cells or HEC59+chr2, showed a G2/M arrest only in cells, in which an extra

chromosome 2 has been introduced, and therefore express hMSH2 (Figures 2A and

1B). The same results were observed with DMSO-differentiated neutrophils (d HL60)

derived from HL60 cells (fig 2B). Taken together, these results suggest that

activated-PMNs induce a G2/M arrest, independent of hMLH1- but dependent of

hMSH2-expression. Moreover, co-culture of Lovo(DT40+2)-4-1 cells (a Lovo+chr2-

derived cell line lacking hMSH2 expression [31]) with activated-PMNs revealed no

G2/M arrest (similar to Lovo and HEC59 cells; fig 2A), consistent with the assumption

that the observed G2/M arrest is hMSH2-dependent.

38

ORIGINAL ARTICLE

Figure 2. PMNs cause a G2/M arrest in hMSH2-expressing colon epithelial cells without affecting the expression levels or DNA binding activity of MMR proteins. (A) HCT116, HCT116+chr3, HCT116-mlh1-2, Lovo, Lovo+chr2, Lovo(DT40+2)-4-1, HEC59 and HEC59+chr2 cells were co-cultured at 20:1 ratios with non-activated or activated PMNs for 24 hours and cell cycle distribution was analyzed by flow cytometry. A significant increase of the G2/M population was observed in all cell lines upon co-culture with activated PMNs except for Lovo, Lovo(DT40+2)-4-1 and HEC59 (cells lacking wild-type hMSH2). Each column presents the mean (± SD) of at least three experiments. (B) HCT116 and HCT116+chr3 cells were co-cultured at 20:1 ratios with non-differentiated (nd) or DMSO-differentiated (d) HL60 cells, both treated with 100ng/mL PMA. A significant increase of the G2/M population was observed with DMSO-differentiated HL60 cells. Each column presents the mean (± SD) of at least three experiments. (C) HCT116+chr3 and Lovo+chr2 cells were co-cultured in the presence of activated PMNs for 24 hours and changes in the total protein levels of MMR proteins (hMre11, hMLH1, hPMS2, hMSH2 and hMSH6) were analyzed by western blot. None of the investigated proteins showed a change in expression levels under these conditions; α-tubulin served as a loading control.

3.4.3. Activated-PMNs do not change the expression of MMR proteins It was previously suggested that oxidative stress relaxes the MMR system and

reduces hMSH6 [40, 41]. As hMSH2 is a potential candidate for the install of the

G2/M arrest upon exposure to activated-PMNs, we tested for changes in the

expression levels of MMR proteins in target cells. However, no changes in MMR

protein levels were detectable under these conditions (fig 2C).

3.4.4. Activation of the ATM/ATR-targets Chk1 and p53 is associated with the PMN-induced G2/M arrest It has been previously established that ATM and ATR are required to activate a p53-

and Chk1-dependent G2-arrest upon DNA damage [42]. Upon 8 hours co-culture

with activated-PMNs, phosphorylation of Chk1 at Ser317 and of p53 at Ser15 was

detected in all cells but HCT116 and Lovo whereas an accumulation of p21 was only

seen in HCT116 and HCT116+chr3 cells (fig 3A). However, at 24 hours, a dose

dependent phosphorylation of p53 at Ser15 and expression of the p53-downstream

CDK-inhibitor p21waf1/cip1 was observed in HCT116 cells, which was paralleled by

an increase in G2/M arrest (Figures 3B and 3C). These results indicate the activation

of a DNA-damage checkpoint in colon cells independent of hMLH1. To control for a

possible role of additional genes transferred through chromosome 3 we tested also

HCT116 cells that had been transfected with an hMLH1 construct (HCT116-mlh1-2),

expressing wildtype hMLH1 and displaying MMR proficiency [33]. When co-cultured

in the presence of activated-PMNs, they exhibited a G2/M increase, similar to that

39

ORIGINAL ARTICLE

described for the parental cell line (fig 2A). Western blot analysis of lysates from

HCT116-mlh1-2 cells, co-cultured with activated-PMNs, showed also a similar

activation of Chk1 and p53 and accumulation of p21waf1/cip1 (fig 3A). Taken together,

these results suggest that the presence of hMLH1 accelerates the activation of a

checkpoint response but is not essential to achieve such.

Figure 3. Activated PMNs induce a dose-dependent activation of checkpoint components in colon cells (A) HCT116, HCT116+chr3, HCT116-mlh1-2, Lovo and Lovo+chr2 cells were co-cultured at a 20:1 ratios for 8 hours; cells were harvested and cell

40

ORIGINAL ARTICLE

lysates were analyzed for total levels and phosphorylation status of checkpoint proteins. An increased phosphorylation of Chk1 at Ser317 and p53 Ser15 was detected in HCT116+chr3, HCT116-mlh1-2 and Lovo+chr2 cells and an accumulation of p21waf1/cip1 in HCT116 and HCT116+chr3 cells. 25µJ/m2 UV light was used as positive control. α-tubulin served as loading control. (B) HCT116 cells were co-cultured from 1:1 to 10:1 ratios with non-activated or activated PMNs for 24 hours and cell cycle distribution was measured by flow cytometry. A dose-dependent increase of G2/M arrest was observed upon co-culture with activated PMNs. (C) Total lysates of cells exposed to activated PMNs (1:1 to 10:1 ratios) for 24 hours were analyzed by western blot. A dose dependent increase of p53 phosphorylation at the site Ser15 and total levels of p21waf1/cip1 was detectable upon co-culture. Co-culture with non-activated PMNs is indicated as (-) and co-culture with activated PMNs is indicated as (+). 200µM H2O2 was used as positive control. Tubulin was used as a loading control.

3.4.5. The PMN-induced G2/M arrest depends on the expression of p53 and p21 Several reports suggest an essential role for p53 and p21waf1/cip1 in the installment

of the G2/M checkpoint [30, 43]. Indeed, cells in which the p53 or the p21 gene had

been disrupted failed to arrest in G2 following γ-ionizing radiations (IR) [30]. In fact

our experiments demonstrated p53 phosphorylation at Ser15 (an ATM- and ATR

target site) and p21 accumulation (Figures 3A and 3C). In order to investigate the

importance of p53 and p21 in our system, the isogenic cell lines HCT116p53-/- and

HCT116p21-/-, in which the p53 or p21waf1/cip1 genes had been disrupted [30], were

co-cultured as described above. Both cell lines failed to undergo a G2/M arrest (fig

4), suggesting that the G2/M arrest caused by activated-PMNs depends on the

expression of both p53 and p21waf1/cip1.

Figure 4. The PMN-induced G2/M arrest is dependent on the expression of p53 and p21waf1/cip1. HCT116, HCT116p21-/- and HCT116p53-/- cell lines were cultured with activated PMNs at 20:1 ratios for 24 hours and analyzed for cell cycle distribution by flow cytometry. HCT116p21-/- and HCT116p53-/- cells failed to undergo a G2/M arrest under these conditions. Each column represents the mean (± SD) of three independent experiments.

41

ORIGINAL ARTICLE

3.4.6. Superoxide dismutase and catalase do not inhibit phosphorylation at p53 Ser15 and increased levels of p21waf1/cip1 SOD catalyzes the reduction of ·O2¯ into oxygen and H2O2, whereas catalase (CAT)

catalyzes the reduction of H2O2 to water and oxygen. ·O2¯ release by activated-

PMNs was measured by lucigenin-amplified chemiluminescence in the presence of

SOD, CAT or both enzymes. SOD but not CAT showed a strong ·O2¯ scavenging

effect (fig 5A). To test the effect of these enzymes on the G2/M arrest, total cell

lysates of HCT116+chr3 cells were analyzed following co-culture in the presence of

both CAT and SOD. Although the addition of CAT and SOD during co-culture with

activated-PMNs reduced the phosphorylation of p53 at site Ser15 and accumulation

of p21waf1/cip1 (fig 5B), it had no effect on the G2/M arrest (fig 5C), suggesting that

PMN products in addition to ·O2¯ or H2O2 activate the p53/p21 pathway and are

sufficient for the installment of the cell cycle arrest.

Figure 5. SOD and CAT do not prevent G2/M arrest but reduce p53 phosphorylation and p21waf1/cip1 accumulation. (A) PMNs were activated with PMA (50ng/mL) for 30 min. After removal of PMA by washing the cells twize with HBSS, ·O2¯ -release was measured by lucigenin-amplified chemoluminescence in presence or absence of SOD (1000 U/mL) or CAT (1000 U/mL). A strong ·O2¯-scavenging effect was observed in the presence of SOD and, to a lower extent, of CAT. Very low ·O2¯ production was measurable in the absence of PMA. (B) HCT116+chr3 cells were co-cultured with activated PMNs in the presence of SOD and CAT as in (A) and the total levels of p21 and p53 phosphorylation at Ser15 were analyzed by

42

ORIGINAL ARTICLE

western blotting. The addition of SOD and CAT considerably reduced both p53 phosphorylation and accumulation of p21waf1/cip1 within 8 hours. ß-actin served as a loading control. (C) HCT116+chr3 cells were co-cultured as in (A) or exposed to 200µM H2O2, in the presence of SOD and CAT, and cell cycle distribution was measured by flow cytometry. Under these conditions, SOD and CAT enzymes blocked H2O2-dependent, but not PMN-induced, G2/M arrest upon 24 hours.

3.4.7. Activated-PMNs cause replication errors in colon epithelial cells Recruitment of the MMR complex following DNA replication errors, leads to cell cycle

arrest [44]. Our experiments so far show an hMSH2-dependent G2/M arrest, which

may be a consequence of an increase in replication errors upon exposure to

activated-PMNs (fig 2A). HCT116+chr3 (clones A3.1 and A3.7) bearing a GFP-

expressing plasmid in which the EGFP sequence is kept out of frame by a (CA)13

repeat [29], were exposed to activated-PMNs for 24 hours, and then expanded for 7

days. Analysis of the fluorescent fraction by flow cytometry revealed a significant rise

in the number of transiently mutated (M1) fraction whereas the increase in the highly

fluorescent population (M2) was not significant (fig 6). No changes in the mutant

fraction were observed with non-activated PMNs. This result could suggest that the

PMN-induced G2/M arrest is a consequence of increased replication errors. As PMN

did not increase the number of permanent mutations (M2 cells), it looks like that the

DNA repair system was functional.

43

ORIGINAL ARTICLE

Figure 6. Activated PMNs increase replication errors. (A) HCT116+chr3 cells (clones A3.1 and A3.7 harboring the pIREShyg2-EGFP/(CA)13 construct) were co-cultured with activated PMNs at 100:1 ratios for 24 hours. After removal of the PMNs, the target cells were expanded for another 7 days and analyzed by flow cytometry. Activated PMNs caused a significant increase in the M1 mutant fraction (i.e. transient mutations (replication errors) that may get repaired) but not in the M2 mutant fraction (i.e. permanent DNA mutations that are transmitted to daughter cells and may expand) in both clones.

44

ORIGINAL ARTICLE

3.5. DISCUSSION UC is a chronic inflammatory disease of the large intestine that is associated with

increased CRC risk [1, 45]. The mucosal injury of active UC is characterized by

enhanced transepithelial migration of activated-PMNs forming crypt abscesses [3]. In

this study, we have developed an in-vitro co-culture model in which primary PMNs

acted as effectors and colon cell lines as targets of inflammation-driven

carcinogenesis. In our study colon cells responded to activated-PMNs or granulocyte-

like HL60 cells by slowing proliferation and arresting in the G2/M phase of the cell

cycle (fig 2A and 2B). This observation is consistent with previous reports in which

exposure of colon epithelial cells to H2O2 and macrophages lead to G2/M arrest [4,

28].

Cellular damage induces responses that enable the organism either to

eliminate or cope with the damage. DNA damage response reactions include:

removal of damaged DNA and restoration of the continuity of the DNA structure;

activation of a DNA damage checkpoint, which arrests cell cycle progression in order

to allow repair and transmission of damaged or incompletely replicated

chromosomes; or apoptosis, which eliminates seriously damaged cells [39]. In most

of the cell lines analyzed in this study, exposure to activated-PMNs caused an arrest

of cell cycle in G2/M, consistent with the activation of a post-replication DNA-damage

checkpoint. Evidence for the install of such a checkpoint [46], apart from the G2/M

arrest (fig 2A), include: phosphorylation of Chk1 at the ATM and ATR target sites

Ser317 (fig 3A) and Ser345 (data not shown) [47]; phosphorylation of the p53 tumor

suppressor protein at site Ser15 (Figures 3A and 3C) [48]; increased expression of

the cyclin dependent kinase-inhibitor p21waf1/cip1 (Figures 3A and 3C); phosphorylation

of the histone isoform γ-H2AX at Ser 139 (data not shown) and cleavage of caspase-

7 (data not shown).

p53 is essential for the maintenance of a G2/M arrest following oxidative

stress. In fact, p53 contributes to the inhibition of cdc2, the mitotic cyclin-dependent

kinase through Gadd45, p21waf1/cip1, and 14-3-3σ. Cyclin B1 is required for cdc2

activity, and repression of the cyclin B1 gene by p53 also contributes to blocking

entry into mitosis [49]. After disruption of either the p53 or the p21waf1/cip1 gene,

gamma radiated cells progressed into mitosis in spite of extensive damage [30]. In

our system, p21waf1/cip1 expression and p53 phosphorylation increased upon exposure

to activated-PMNs and cells lacking p53 or p21waf1/cip1 failed to undergo a G2/M arrest

45

ORIGINAL ARTICLE

(fig 4), suggesting that the p53 pathway is required for the response to PMN-induced

damage. However, although SOD and CAT scavenge ·O2¯ and H2O2 (fig 5A), and

completely revert the H2O2 – induced G2/M arrest (fig 5C), they only partially reduced

the PMN-induced phoshporylation of p53 (fig 5B) and were unable to revert the cell

cycle arrest caused by PMNs (fig 5C) suggesting that activated-PMNs release also

additional molecules which can induce a G2/M arrest. Another potentially harmful

species released by PMNs are chlorinating agents (containing an active chloride in a

formal +1 oxidation state: e.g. HOCl) [50