Lecture 1

Ch1 Introduction:
the nature of
Cancer Definitions
Incidence = number of new cases in a particular time frame. So not the
total number or a percentage, it’s an absolute number of new cases in for example a year.
This doesn’t say much because countries differ in size, to compare different regions the incidence rate
is used which is normalized by the inhabitants, usually per 100 000.
Prevalence = total number of people who were diagnosed in a certain period of time. They may have
the disease at the moment, have had it and are now cured, etc. They have to still be alive. There isn’t
made an extinction between having cancer at the moment or not anymore. The period is usually 5
years.
Mortality = number of people who died as a result of cancer within a certain period, usually 1 year.
This is not the number of patients who had cancer and died, but who died as a result of cancer. You
don’t always now, for example when the patient is old it is not always clear if he/she would have died
anyways. Because of this the mortality is always corrected by age, so with the chance the person
would die at this age if it would not have had cancer.
Survival = the percentage of people that are still alive after a certain period of time after diagnosis.
This is always corrected for the chance someone would have died anyway.

There are large regional differences, in some country’s cancer is a high cause of death while in others
it is not. In the western world cancer is priority but that’s not globally, in some countries other
problems are bigger.

What is cancer?
It’s a group of diseases, there are many different types, usually defined by the type of tissue from
which the tumor develops bot there are many more differences so there also are subtypes per tissue.
They all are characterized by a uncontrolled cell growth and they are invasive and forming metastases.
If a patient has a tumor it does not always mean the patient has cancer. A tumor is simply a mass of
cells and they can also be benign, these are not invasive and not metastasizing.

Why is a malignant tumor life threatening?

o Invasion of organs disturbs organ function: It can take of functions of cells, produce growth
factors that influence cells in the tissue which causes malfunctioning of the organ
o Cancer cells compete with normal cells for nutrients and oxygen. This is not only for the organ
in which the tumor is growing but for the entire organism. Cancer patients are often very tired
because they are at a low energy level.
o Growing tumors can cause obstructions: for example pressing on a blood vessel which causes
that there is no blood supply for a certain tissue.

Difference between some cancers has to do with from which celtype/tissue the cancer developed;
o Carcinoma: developed from epithelia. It’s the majority of all cancers
o Adenocarcinoma: arises from glandular tissue
o Sarcoma: arises from mesodermal tissues like bone or muscle
o Lymphoma: arises from white blood cells or their progenitors.

The incidence of carcinoma’s is 85% so it is way higher than other cancers. Carcinomas are derived
from epithelial cells which line the body, all surfaces in- and outside your body that are exposed to
molecules from the outside. Therefore, these cells have a bigger chance of being exposed to
carcinogens. A carcinogen is any agent causing cancer, this can be a compound but also radiation etc.
they cause alterations in your DNA. All cancer cells contain alterations in their DNA, often a lot of
them. This accumulation of mutations causes the stepwise development of cancer, oncogenesis or
carcinogenesis.

Oncogenesis
In the normal epithelium you have a stepwise increase of changes upon accumulation of multiple
mutations. Only in the last phase where it starts to invade tissue it’s called cancer. Because there is
accumulation of mutations before we have full blown cancer, you see cancer is clonal and
heterogeneous at the same time. It all starts but one first mutation in a single cell. All tumor cells will
be derived from this single cell, so the cancer is clonal. However, after proliferation of the single cell,
one cell may have additional mutations. So, different cells all carry the original mutation that was in
the initial cell, but they have additional mutations because of which they differ from each other. So, in
the end you have a heterogenous mass of cells that have different mutations accumulated.

Is cancer inheritable?
The answer is no, it is not inheritable from itself so you cannot be born with cancer. You can inherit
mutations int the germline that increase the chance to develop cancer later in life.

Higher risk at older age

Because there is an accumulation of mutations, you have a higher chance to develop cancer later in
life because it needs time to build op all mutations. The incidence of cancer is increasing due to longer
life expectancy.
Also, the development of cancer is a matter of chance, it depends on the how much exposure to
carcinogens you have over life.

A tumor is more than just tumor cells. Apart from the heterogenous cancer cells a tumor also contains
non-malignant cells like fibroblasts, blood vessels and immune cells.
A cell can have different stages. You have an undifferentiated stem cell. This can differentiate in
mature cell types, mature cell types can de-differentiate to a more primitive cell type. But also, cells
can proliferate or die. In cancer, differentiation and apoptosis is stopped and proliferation is promoted.
Characteristics of cancer cells
Cancer cells;
o Have a different morphology
o Can grow at a low serum concentration. Because there is uncontrolled cell growth they don’t
need all the stimuli other cells need to proliferate. They can be grown in the lab with a low
concentration of serum so you don’t have to add many growth factors.
o Can grow in the absence of a substrate for attachment. Normal tissues bind to extracellular
matrix proteins to grow but cancer cells can also proliferate without.
o Show no/less contact inhibition

Factors playing a role in development of cancer

o Environment like sunlight and asbestos
o Diet and exercise
o Alcohol
o Smoking
o Reproduction, contraception, hormone replacement therapy
o Viruses
o Own metabolism: the majority of cancer are not caused by environmental factors like diet and
radiation but by your own metabolism. In your metabolism byproducts are produced that cause
error in DNA-replication. You can decrease the chance of developing cancer by leading a
healthy life but cannot prevent it because most of comes from within.

Treatment of cancer
o Surgery
o Radiotherapy
o Chemotherapy
o Prevention of cell division, the cytostatic effect
o Killing of cancer cells, the cytotoxic effect
The two possible effects of treatment, cytostatic and cytotoxic, are dose dependent. With an increase
in dose of treatment you get an increased effect on the cancer. But all these treatments also have an
effect on other cells so at a certain dose you also have killing of the healthy tissue.
Ideally, the difference in dose already treats the cancer and the dose that’s toxic to your healthy tissue
is large, so the further the curves are apart the better. For the majority of treatments the difference is
really small, which makes it hard to find a good dose with little side effects. The difference between
the doses can be defined in two ways, therapeutic index and the therapeutic window. The therapeutic
index is the distance between the two curves at a 50% toxic effect. The therapeutic window Is the
distance between the minimal dose you need to get an effect on the cancer and the maximum tolerated
dose so the dose in which get some toxicity which can be tolerated. Ideally the given dose is as close
to the maximum tolerated dose as possible.
The aim is to develop anti-cancer agents that are very effective against cancer and causing less
toxicity to normal tissue. Therefore, targeted drugs are developed. To develop targeted drugs you need
to know the differences between tumor and normal cells. This are the 10 hallmarks of cancer, what
distinguishes cancer cells from normal cells. So, every hallmark is a potential target for selective
therapy.

The percentage of patients that benefit from treatment with targeted drugs is quite low. Only a quarter
of patients benefits. This is also seen in other diseases but in cancer it’s below the average. This is
because all tumors are different, the genomics of a tumor should be known to know which targeted
treatment to use, personalized medicine. So, the task is to define which patients will benefit from a
treatment and which will not. Therefore, diagnostics should be done like genetics, imaging and
immunohistochemistry

Ch2 DNA
structure and
stability: Every cell contains 2m of DNA in the nucleus so it has to be packed
densely but still able the be used. Therefore, it is packed around
mutations vs histones, octamers of 8 different histon proteins. Its wrapped around
repair histones twice. The histoncomplex with DNA wrapped around it is
called a nucleosome. The fiber of nucleosomes is further packed in
higher order structures. Eventually you get a chromosome. The shape of the chromosome in the
picture appears during metaphase. It is duplicated, there are two sister chromatids, kept together at the
centromere. At the centromere large protein complexes are being assembled, called kinetochores.
These complexes make the connection of the chromosome to the mitotic spindle which will pull the
sister chromatids apart at makes sure each will end up in a different daughter cell.

The backbone if DNA is made up of sugars and phosphates. The sugar contains an oxygen-atom and 5
carbon-atoms. The bottom side is called the 3’-end and the upper side the 5’-end. The opposing strand
is upside down. Each sugar is connected to a specific base, a pyrimidine or a purine. A pyrimidine is a
thymine or cytosine and purine is adenine or guanine. Cytosine binds to guanine by making 3
hydrogen bonds and thymine binds to adenine by making 2 hydrogen bonds, so the connection
between C and G is a little bit stronger.
A piece of DNA, the gene, is being transcribed into RNA. This RNA is spliced into mature messenger
RNA in which introns are removed and only exons are left. By doing alternative splicing you can
produce alternative gene products from one gene.
Upstream of the transcription side, at the 5’-end, there is the promotor sequence which contains
elements bound by transcription factors that regulate the expression of the gene. In addition, there are
regulatory sequences, these can be far away or in introns, that also help regulation the gene
expression.
DNA can fold and make loops which causes that pieces that are far away can come in proximity of the
promotor and transcription start side. Enhancer regions can affect the level of gene expression of this
gene.

Small DNA changes

In base pare substitution a base is changed so one particular codon is changed. So, one amino acid is
subituted in another amino acid. It can be that the change of one base can lead to introduction of a
stopcodon. Then the translation of the protein will be terminated and this can cause inactivation of the
protein function.
Insertion or deletion of a base causes the reading frame to be shift from that point onwards all amino
acids will be different.

Large DNA changes

o Aneuploidy: In aneuploidy a whole chromosome is gained or lost during cell division. The
expression of hundreds of genes will then be too much or too little, the gene dose is affected.
o Gene amplification: In gene amplifications pieces of DNA are copied and integrated in the
chromosome multiple times, intra-chromosomal, or are present outside in the chromosome,
extra-chromosomal. These extra-chromosomal pieces are circular and lack a centromere so
they are not equally distributed during the cell cycle, one daughter cell will receive more of
them than the other. The one with the most will gain a lot of amplification of genes on this
piece of DNA.
o Deletions: deletion of multiple genes at once
o Chromosome rearrangements (translocations): pieces of one chromosome end up in another
chromosome. This can cause amplification of genes.

Cancer results from alterations in cancer genes;

o Inactivation of repair/stability genes; DNA damage is not repaired so more mutations
accumulate
o Activation of oncogenes; overexpression of growth factors
o Inactivation of tumor suppressor genes

Carcinogenesis is a multi-step process

1. Initiation: DNA damage, more than 1 mutation
 2. Promotion: proliferation of tumor cells, influenced by interaction with the microenvironment
like growth factors and hormones
3. Progression: invasion, metastasis, angiogenesis

Causes of DNA aberrations

Exogenous;
o Smoking
o Alcohol
o Sunlight (UV)
o Radiation
Endogenous:
o DNA replication errors, the machinery is quite good but can still make mistakes
o Reactive oxygen species (ROS)
Specific sources cause specific types of damage.

Radiation
Radiation is energy in the form of waves. There are waves with long wavelengths and low frequencies
like radio, microwaves and infrared, and with shorter wavelengths and higher frequencies like
ultraviolet, X-ray and gamma ray. X-ray and gamma ray are ionizing radiation. When energy in the
form of ionizing radiation hits an atom an electron can be kicked out, you get a free negatively
charged electron and a positively charged atom which are both quite reactive. The electron can
damage the DNA either directly or indirectly by the production of reactive oxygen species (ROS). The
indirect damage can be caused by the radiolysis of water.

Radiolysis of water
A watermolecule consists of one oxygen-atom and two hydrogen-atoms. The oxygen-atom has 8
proteins and 8 electrons. The electrons in the outer circle form pairs with the electrons of the
hydrogen-atoms. When radiation hits the atom an electron is kicked out. This can lead to a multitude
of reactions. Example: the positively charged watermolecule interacts with another nearby
watermolecule. It will steal a hydrogen-atom from the nearby watermolecule, now it has 11 protons
and 10 electrons. The nearby watermolecule lacks an electron now, it looks for another electron to pair
with so it is really active. Therefore, it can be stabilized by multiple watermolecules together but it can
also with a watermolecule what leads to the removal of an hydrogen-atom.
So, the ionization of a watermolecule causes the emission of an electron and a cascade of subsequent
reactions. The ROS can interact with DNA.

Direct and indirect radiation-induced

DNA damage
Indirect: radiolysis of water will result in
ROS that can chemically react with the
DNA. Oxygen makes the DNA damage
permanent so there will be chemically
modified DNA, this can be in the backbone
Guanine can be modified by ROS,
then an oxygen is formed, 8-
oxoguanine. This is potentially
mutagenic because the DNA-
polymerase doesn’t recognize it as a
guanine but as a thymine. So, after
one round of replication an adenine
is put on the other side and after
another round it will be replicated
into thymine, so in the end there is a
mutation from G to T.

So, radiation is carcinogenic and how carcinogenic it is is dependent on the amount of energy. If there
is a lot of energy it can even cause direct breakage of the DNA. It is not only the energy, also the rate
at which it is released into the tissue. This Is described by the Linear Energy Transfer (LET), the rate
of energy loss among the track of an ionizing particle. Low LET are X-ray, gamma ray and protons.
High LET are a-particles, neutrons and carbon-ions. Low LET tracks travel through the tissue in a
diffuse way. This causes a little bit of damage while high LET tracks travel densely what causes a lot
of damage.
The amount of energy deposit is measured in Gray (Gy), in which 1Gy is 1 joule/kg tissue. This is
used in radiotherapy.
The amount of damage is not only determent by the rate and amount of energy but also by the type of
source. This is measured in Sievert (Sv), the absorbed dose in Gy multiplied by the radiation quality
factor. The radiation quality factor is 1 for protons and 20 for neutrons.

Linear Energy Transfer

The higher the LET of the radiation the higher the energy deposit over short distance. These are the
dense ionization tracks, so there is more cell kill per Gy because there is a lot of damage. But, there is
less penetration dept, it can already stopped by a sheet of paper.

Radiation induced cancer is dependent on many factors;

o The volume that is being exposed
o The total dose
o Which organ is being hit
o Genetic predisposition
o Biological factor
 o Gender
o Age

To estimate the carcinogenic risk of a certain radiation source you need;

o A large population that has been exposed,
o A large control group
o A long follow-up time
o The individual dosimetry data

Leukemia is the most frequent ionizing radiation-induced cancer. Children are the most sensitive and
women are more sensitive than men, particularly at a young age.

Non-ionizing radiation
Sunlight (UV) is carcinogenic whereas microwaves, radio waves and electro magnetic radiation from
a phone are non-carcinogenic. Ultra Violet B (UVB) is the most carcinogenic type of UV-radiation.
To most typical mutations of UVB are pyrimidine dimers. Two pyrimidines following after each other
can be linked by UV-light, which disturbed the helical shape of the DNA. Then the polymerases will
insert an adenine in the opposite strand. When this is not correct it will lead to a mutation. This
particularly leads to skin cancer.

Other genotoxic carcinogens include;

o Polycyclic Aromatic Hydrocarbons (PAHs), found in sigarette smoke
o Aromatic animes
o Nitrosamines
o Alkylating drugs
Genotoxic refers to capacity of the carcinogens to change the DNA.
The working mechanism is that there is a direct or indirect induction of damage to the DNA. The
drugs work by DNA adduct formation so a chemical group is added to the DNA and this either
disturbed the DNA structure or makes sure the bases are not properly replicated by the DNA
polymerizes. When there is no sufficient repair the mutation can become permanent.
Guanine can become methylated and is then called O6 adduct of guanine. This is then being
recognized as an adenine so there will be a T opposite of the G and after replication one of the
daughter cells will have A-T base pair instead of G-C.

Other carcinogens include;

o Fibrous minerals: ROS generation, inflammatory response
o Oncoviruses: block tumor suppressor proteins
o Hormones: high concentrations of estrogen a a risk in post-menopasual women for developing
breast cancer by causing enhanced cell proliferation and the genotoxic estrogen metabolites.
o Endogenous sources: ROS and cytosine deaminases
 Lecture 2
Ch2 Radiation
and chemical
DNA repair mechanisms
carcinogenesis A different type of lesion acquires a specific repair mechanism;
o Damaged base: base excision repair
o Bulky adducts: nucleotide excision repair
o Replication error: mismatch repair
o Double strand break: homologous recombination or non-homologous end joining
o Cross-link: complex repair coordinated by the FA pathway

Base excision repair

The cell suffers from exogenous and endogenous sources of DNA damage. The endogenous sources
damage the bases. The primary way to repair this is by base excision repair. For example, ROS lead to
the oxidation of guanine and the 8-oxoguanises that are developed need to be repaired by base
excision repair. Or for example deamination of cytosine to uracil. There, enzymes naturally occurring
in the body that normally act as a defense mechanism in viral infections and these can accidently also
deaminate cytosines leading to uracil.

Cells contain a number of DNA glycosylases that continuously scan the DNA for damaged bases.
When it recognizes a damaged base, the DNA flips the lesion outside of the helix and cuts the base
leaving an abasic (AP) site behind. The abasic site is recognized by other enzymes like AP
endonuclease which cuts next to the abasic side, leaving a 3’hydroxyl group that can be used by a
DNA-polymerize. So, because of the activity of AP endonuclease the 3’end becomes exposed. This
abasic nucleotide is removed and there is a new nucleotide inserted by polymerase. Then the backbone
is sealed again by ligase.
There are two types of base excision repair; short and long patch. Short-patch generates a gap of one
nucleotide and long patch involves the generation of bit larger capes, between 2-10 nucleotides. It is
the same mechanism, so a cut is made, a number of nucleotides are removed, new nucleotides are
inserted by a polymerase and then a ligase is needed.

A genetic disease is related to base excision repair deficiency. This is due to a mutation in a DNA
glycosylase called MUTYH. Patients with this mutation suffer from the disease called adenomatous
polyposis (FAP) characterized by many polyps in the colon which is a big risk in developing
colorectal cancer.

Nucleotide excision repair

Nucleotide excision repair is required for mutation that disturb
the helical shape of the DNA due to oxidative damages or UV-
induced pyrimidine dimers and large chemical adducts. Then the
entire nucleotide has to be removed. The lesion is detected by
specialized proteins including XPA and XPC. Then helicases
are included, specialized enzymes that can break the hydrogen
bonds between bases of opposing DNA strand, so they open up
the DNA. By endonucleases nucleotides are removed. After this
there is recognition, helicase activity, nuclease activity,
polymerase activity and then ligase to seal.

There are two types of nucleotide

excision repair; global genome NER and transcription-coupled NER. In global genome NER the entire
genome is continuously scanned for these types of lesions. Transcription-coupled NER is more
specialized, this includes damage that is occurred on the DNA and then the transcription machinery
collides in the damage and cannot longer transcribe the DNA. Then this type of repair is activated
which has the same mechanism as global genome NER but uses other proteins.
In the disease Xeroderma pigmentosum (XP) patients have maturation in nucleotide excision repair
genes. This leads to hypersensitivity to sunlight. The sunlight causes pyrimidine dimers that cannot be
repaired sufficiently because NER does not work properly. Patients with this disease have an
increased risk for developing skin cancer.

Mismatch repair
Mismatch repair is used to repair mismatch lesions that occur of the DNA replication machinery
makes a mistake, especially in repeat sequences. This includes specialized proteins, mismatch
recognition is done by MutSa and MuTSb, both are dimers of two different proteins but MutSa
consists of MSH2 and MSH6 whereas MutSb consists of MSH2 and MSH3. One of these is involved
in the recognition of the DNA mismatch. Once it has been recognized the MutS dimer recruits the
MuTl dimer. There are three types MutLa, MutLb and MutLy. Together these complexes make sure
there is made a nick next to the lesion after which nucleotides are removed by nuclease. By
polymerase nucleotides are included and a ligase seals the nick.
This is post replication repair, so after replication in the S-phase of the cell cycle. Germline mutations
in mismatch repair genes MSH2 and MLH1 are found in patients what causes Lynch syndrome. These
patients are at risk for developing colorectal cancer.

Double strand breaks

The most lethal type is damage is a double strand break. If not repaired, it is in most cases lethal. It
can cause large deletions, translocations etc so it’s the cause of many chromosomal abnormalities so it
can have a large carcinogenic effect. Therefore, the cell has evolved multiple mechanisms to repair
double strand breaks.

Repair of double strand breaks

The repair of double strand breaks can be executed by different repair mechanisms. The preferred
pathway is homologous recombination because this is error free. The drawback is that is relies on the
sister chromatid so it can only take place during S- or G2 phase of the cell cycle in which there Is a
sister chromatid available.
After a break has been recognized few nucleotides are being removed by exonucleases. This leaves an
overhang of single stranded DNA. This overhang is coated with RAD51 proteins what searches for
homology in the sister chromatid. It uses the other chromatid as a template to copy the exact sequence.
The sequence will be exactly the same so it is error free.
When there is no sister chromatid available or the homologous recombination machinery does not
work properly the cell must use other mechanism. In non-homologous end joining the two ends of the
broken DNA are re-ligated. This can lead the loss of a few nucleotides and leave a scar in the DNA.
Alternative end joining requires a specialized polymerase called polQ or teta. This relies on
microhomology that usually occurs at both ends of a DNA break but that is also in most cases leading
to the loss/gain of some nucleotides so leaving a scar in the DNA.
Single strand annealing uses repeat sequences to find homology to relegate to two broken ends. This
leads to larger deletions.
Homologous recombination
The blue strand is broken, the red strand is the
still intact sister chromatid. ATM is a kinase
that is recruited to the break to phosphorylate
other DNA repair proteins and proteins that
stop cell cycle progression which is impotent to
give the cell time to repair the break.
Proteins that are recruited include components
of the MRN complex. These have exonuclease
activity so this leads to single strand overhang
of the blue strand. These overhangs are coated
with RAD 51 proteins what searches for
homology in the sister chromatid. Then
polymerases are required to extend the strand.
Then ligases and resolvases are required to
restore the two sister chromatids

Mutations in genes that are involved in the repair

mechanisms predispose in cancer development because then mutations accumulate. For example;
mutations in
o ATM cause ataxia telangiectasia which causes patients to be at risk for developing leukemia or
lymphoma.
o NBS1 cause Nijmegen breakage syndrome and cause lymphoma.
o BLM cause Bloom’s syndrome and patients can develop all kinds of cancer.
o BRCA/FA pathway cause Fanconi anemia. FA is a complex repair pathway that is needed for
the repair of interstrand crosslinks. This can be due to certain agents like cisplatin or is the
result of endogenous metabolites. What makes a gene a Fanconi anemia gene is not only being
involved in this repair pathway but also because its found retaited in Fanconi anemia patients.
These patients have anemia but also a high risk of developing cancer, particularly head and
neck cancer and leukemia. This is due to the abundant ability of their cells to repair these kinds
of lesions. Most mutations in patients are bi-alelic, so both allels must be mutated in order to
see the phenotype. BRCA1 and -2 are Fanconi anemia genes when biallelically mutated,
monoallelic mutations predispose to breast cancer. So, patients with a monoallelic mutation in
BRCA1/2 ado not have Fanconi anemia but are at risk for developing cancer.
So, different mutations have their own tumor spectrum.
Treatment of patients with homologous recombination defects depends on knowledge on the DNA
repair, personalized medicine is needed.
For example, when Fanconi anemia patients have cancer they cannot be treated by cisplatin, because
this causes the interstrand crosslinks that cannot be repaired so it will be toxic. This is also the case for
other predisposition syndromes.

Radiotherapy: cancer treatment using ionizing radiation

Radiation can lead to cancer because it can cause damage of the DNA. Radiation is also used as
treatment for cancer patients. When radiating a cell, there is damage in the cell. This can either be
handled because the cell has the proper repair mechanism and the DNA is repaired so the cell will
survive. This is the case in normal cells. Sometimes the damage is excessive, beyond repair or there
are genetic defects in the cell, then the cell is unable to repair the damaged caused by the radiation.
This can lead to mutations, chromosomal aberrations and the cancer induction, but it can also lead
directly to cell death. Sometimes there is a therapeutic window in patients, in which the tumor cells
are much more sensitive than normal cells and therefore radiation can be used as a treatment.

There is a controlled use of radiation in diagnostics and therapy. The benefits and risk should be
balanced. Benefits include therapy, cure, palliation and diagnostic information. Risks include the side
effects. Tissue/organ injury, carcinogenic effect, patients can develop secondary tumors as a result of
the treatment.

Chemotherapy
Also compounds that are carcinogenic on itself are also used as treatment. These ..
o Alkylating agents: Alkylating agents add DNA adducts via an alkyl group, a chemical
compound that includes carbohydrogen. This group can be large or small, for example
chlorambucil, a reactive compound because it has two chloride residues which can be removed
whereafter the carbon can be connected to the DNA. So, the compound can be bound
covalently to the DNA, this is a risk and has to be repaired.
These agents require metabolic activity.
o Platinum-based drugs: Cisplatin can cause crosslinks, it can be linked to two adjacent bases on
the same or opposite strand. There is no metabolization.
o Antimetabolites: include for example 5-FU what mimics a precursor of thymine. Because of
this it blocks the synthesis of thymine. If there are cells with 5-FU there will be a shortage of
T-nucleotides. This blocks the synthesis of new DNA.
o Organic drugs: including doxorubicin, a topoisomerase ll inhibitor. If the helix is opened the
twist of the DNA is increased what causes a risk for the DNA to break. Topoisomerase is
needed to release stress of DNA. The DNA is broken, twists are removed and sealed again.
When topoisomerase is not working there is an increased risk for breakage of the DNA.
 Some drugs interfere with microtubule dynamics. During mitotic cell division the separation of
sister chromatids is impaired. This can lead to lots of chromosomal aberrations.

Hormone therapy
Hormone therapy targets cancers that need hormones to grow. Some cancers are dependent on
continuous hormone signaling. Hormone therapy works via;
o Blocking the hormone receptor: there are drugs that prevent the signaling, for example
Tamoxifen. Tamoxifen interacts with the estrogen receptor to prevent biding with estrogen.
Normally estrogen binds to the receptor which after a cascade of reactions leads to the
expression of target genes that drive the proliferation of the genes. So, using tamoxifen the
expression of target genes is inhibited and the cell will not proliferate anymore.
o Interfere with the synthesis of hormones, these are called aromatase inhibitors.

Drug or radiation resistance

In every cancer treatment there is a risk of resistance. There are intrinsic and acquired reasons that
cause this.
Intrinsic resistance;
o Location of cells in a tumor: a cell that is really deep in a tumor is difficult to reach. This also
depends on the availability of blood vessels. If there are not so many blood vessels the drug
cannot reach the middle of the tumor.
Also, hypoxic cells are radio-resistant. This is due to the fact that the damage done by radiation
often uses oxygen to make it permanent. Radiation causes generation of ROS which damage
the DNA and in order to make it permanent oxygen needs to be present so in hypoxic
conditions this is not happening, the cells are more radio-resistant.
o Tumor cell heterogeneity: because of continuously accumulating mutations some cells will
become more resistant. Cancer stem cells are generally more resistant to radiation and
chemotherapy.
Acquired resistance;
Some cells have developed mechanism to pump out a drug in a higher rate, so the efflux is increased
or they have mechanisms to prevent the influx. Sometimes the gene that is being targeted is being
amplified, if more of a certain target protein is being produced it can outnumber the available
compounds of the drug that target these proteins.
A lot of drugs require intracellular metabolization, if the metabolization of the drug to an active form
the cell also becomes resistance. Some cells will restore certain DNA repair mechanism so they can
repair the damage more efficiently.

Synthetic lethality
A tumor cell has a defect through which a pathway is not working properly. In some case this can
make it more vulnerable for inhibiting a second pathway. For example cells with defects in
homologous recombination, for example a mutation in BRCA1, are more sensitive for PARP
inhibitors. If PARP is inhibited, single strand breaks can no longer be repaired. Because of DNA
replication single strand breaks are then converted into double strand breaks, these require
homologous recombination or BRCA pathway to be repaired. However, when this is mutated the
combination will be lethal.
Mutational signatures
Cells that have as specific mutations can be linked to the cause of those mutations. Cells that suffer
from excessive DNA damage will show particular types of mutations

Ch3 Regulation of
gene expression
Cancer is a DNA disease. mutations change the gene expression
directly, for example by making a stop codon in a tumor suppressor gene, or indirectly.
Epigenetic events alter the gene expression. Changes in gene expression turn a normal cell into a
normal cell by activating oncogenes and inactivating tumor suppressor genes.

Transcription
The promotor is the region of the gene that lies directly upstream of the transcription start site, the
place where RNA polymerase starts the transcription. Transcription factors bind to response elements,
so to the promotor but also to other structures in the genome. These transcription factors are necessary
for the activation of RNA polymerase ll, so it’s actually a complex of a lot of proteins with RNA
polymerase as a business end. In some cases this enzyme binds to the TATA-box, a lot of genes don’t
have a TATA-box. Once RNA polymerase binds and gets positive signals through coactivators from
the transcription factors, the transcription can start.

Transcription factors
There are 3000 transcription factors encoded in the human genome, there are 20 000 genes in the
genome so a 1/6 of that are transcription factors. All of the TF are DNA binding proteins and most of
them have specific sequences they can recognize, response elements. When TF find these response
elements a signal needs to go to the RNA polymerase. This is done through co-activators and -
repressors, additional molecules that bridge the signal from the TF to the RNA polymerase.
Besides the TF binding sides in the proximal promotors there also may be other regulatory elements
that lie further in the gene. These are called enhancers and silencers. Interactions with these induce
recruitment of transcription factors.

TF regulation
o Dimerization, only function when they are in a dimer
o Ligand binding, a specific molecule is needed for its activity
o Expression restricted to particular cell types. Certain TF are specific for some cell types and
different cell types have different expression.
o Covalent modifications (phosphorylation), some TF require phosphorylation and some only
function when they are not phosphorylated.
o Cellular localization, for a TF to be functional it needs to be in the nucleus. Many are in the
cytoplasm and have to be activated to move into the nucleus, and some TF are moved out of
the nucleus to make sure they don’t function anymore.

TF are made up of a number of domains with specific functions. All TF’s have a DNA binding and
transcriptional activation domain. Some domains are very common but not present in all TF’s, this are
a dimerization and ligand binding domain.

DNA binding domains

There are 4 DNA binding domains;
o Zinc finger
o Helix-loop-helix
o Helix-turn-helix
o Leucine zipper
The aminoacid sequence gives specificity for the sequence it has to recognize. The response elements
have a certainse sequence, aminoacid tails in the domein have specificity for this sequence.

Ligand binding domains

An example is the steroid hormone receptor superfamily with 48 members. All of these have a ligand
binding domain to which the steroid can bind. The retinoic acid receptor is a nuclear receptor and a TF
as well. It can bind a ligand that moves all the way through the cell because a steroid is lipid soluble
so it can move through the membrane. so, retinoic acid is the ligand of this receptor, it can move to the
nucleus and bind the retinoic acid receptor which is bound to another TF, RXR. If they are not bound
by retinoic acid they are in a repressive conformation but when retinoic acid bounds it becomes active.
The glucocorticoid receptor is a little bit different, it is generally in the cytoplasm and needs
glucocorticoid to enter the nucleus.

Dimerization domains
For example, you have three factors. If these would only homodimerize, so only bind themselves, you
only have three variances possible to bind the DNA. But when they can also heterodimerize there are
6 different variances with all different functions.
An example is AP-1, made out of either Jun-Fos heterodimers or Jun-Jun homodimers. In total there
are 18 possible combinations of which some are similar and some differ a lot. Whereas usually the
Jun-Fos heterodimers are strong activators, when Jun homodimerizes to Jun B it’s an anti-proliferative
TF.

All domains act independently. This can be used in the lab by making dimeric receptors, for example
retinoic acid and thyroid hormone receptor. You can make a chimera of these two receptors by
switching the domains so by putting the ligand binding domain of the retinoic acid receptor on the
thyroid hormone receptor. This way you will get a receptor that recognizes the DNA binding sites for
the thyroid hormone receptor but is activated by retinoic acid, and the other way around.

How to examine TF binding, techniques to study TF and the sites they bind;
o Electrophoretic mobility shift essay (EMSA); DNA is negatively charged, so when put on the
gel with the negative pole at the top site the DNA will move to the bottom of the gel. How fast
it moves depends on the size of the DNA and whether its bound to proteins. A DNA probe is
used with a radioactive label. If you rub this through the gel it gives single band. When there is
a probe with a binding site for a TF and this TF is present in the nuclear extract, the TF can
bind and you see two bands, the unbound probe at the end and the complex a bit higher in the
gel. This demonstrates that this probe is being bound by a TF in the nuclear extract and now
you can start mutating the probe to see which sequences are important for the binding.
o DNAse footprinting: again, a probe is used and ran over gel but now the probe is a little bit
digested with DNA. They are cut only once so there are different sizes. If the probe is
incubated with nuclear extract you can protect the DNA from being degraded by the DNAs.
So, the DNAs cannot knick the sequence and this will give an empty spot on the gel where no
fragments are being found. You see fragments but when protein extracts are added these
fragments are gone so this piece of DNA is being protected by bound protein.
This is commonly done by ChIP-seq. The DNA is incubated with the protein but now its
crosslinked, so all the protein is crosslinked to the DNA and can now be fragmented. The
protein will stick to the DNA when it is crosslinked but it can be pulled out when you have an
antibody against the protein of interest. So, a TF is incubated with DNA, the DNA is
fragmented and all the DNA that’s is bound to the TF is pulled out which can then be released.
All of this DNA can be sequenced and mapped to the human genome. Now we can see exactly
where the protein binds in the genome.
 o (luciferase) reporter assays; Instead of looking for a protein you can look more specifically to
the promotor. The promotor is cloned in front of a reporter gene, which you can easily detect.
Usually luciferase is used as reporter gene because this gives a very strong signal in
combination with luciferin. You can make variances to the promotor like deletions etc.

DNA packaging
The human genome is over 2 million base pairs and 1-2 meters long. In eukaryotic cells the packaging
is done using histones. DNA and histones form a structure called chromatin with the nucleosome as
base unit, an octamer consisting of 2 copies of H2A/H2B/H3/H4, with 127 dp of DNA wrapped
around it 1.7 times. In between the nucleosomes is H1 which acts as a linker that protects the DNA
that’s between the nucleosomes.
DNA – nucleosome (beads on a string) – 30 nm fibers – radial loops – long range fiber-fiber
interactions

Epigenetic control
Epigenetic is control over genetic, so over mutations. Heritable epigenetic information does not
modify the DNA but it can still be passed on to a daughter cell. So, it is not heritable from one person
to the next but form one cell to the next. There are different covalent modifications that make
epigenetic marks;
 o Histon modifications: histons have tails that can be acetylated, phosphorylated, methylated and
ubiquitated.
o DNA methylations: the DNA itself can be covalently modified by methylation. Both are
reversible, this is different from mutations.

Histone modifications
It is necessary to make to promotor associable to transcription factors so the chromatin needs to be
remodeled, for which histone modifications are used. These modifications needed to be added and
removed specifically by chromatin modifying enzymes.
Enzymes that make sure that histones get acetylated are histone acetyltransferases (HATs). These are
activating mark. The acetylation can also be removed, this is done by histone deacetylases (HDAC), a
repressing mark.
Histon tails are rich in lysine, a positively charged amino acid. DNA is negatively charged so the
histones can very tightly pack the DNA. HATs acetylase the lysins which takes away the positive
charge and opens up the chromatin. HCADs remove the acetylation and close the chromatin.
These changes are frequently observed in cancer, altered HAT and HDAC activity is found in many
tumors, also because of mutations. This causes activation or silencing of oncogenes and tumor
suppressor genes. So, the recruitment of HDACs to the wrong gene promotors can cause the silencing
of tumor suppressor genes.

DNA methylation
This is directly to the DNA and is done by DNA methyltransferases (DNMTs). These marks are
heritable from one cell to the other. After replication you have a new strand of DNA that is then called
hemi-methylated DNA. Then DNA methyltransferase 1 is responsible for immediately putting a
methyl mark on it.
The methylation is very specific for Cs preceding a G, the C is 5’ upstream of the gene. CpGs are by
themselves underrepresented because methylated Cytosine is very easily deaminated what makes it
turn into a thymine so causing a mutation. So, CpGs are naturally selected out of the genome but the
ones that remain have an important function. Most of them can be found in CpG island, clusters of
CpGs of a few hundred bp long, found in 50% of gene promotors. So, methylation inhibits the
transcription. DNA methylation is kind of reversible but not really. The only ways to lose the
methylation mark is if there is no DNA methyltransferase 1 activity after replication or through TET
enzymes which remove the methylation.

How does DNA methylation inhibit transcription?

o Binding of proteins that recognize methylated CpG DNA: methylated CpG can recruit other
proteins like proteins that can prevent the binding of TFs and recruit HDACs that are close the
chromatin so TFs cannot bind.
o Directly interfere with the binding of TFs: methylated C has a different structure than
unmethylated C, so it is possible the sequences is not recognized by the TFs. However, there
are some TFs that specifically recognize methylated C.

In the upper strand there are no methylation marks on the DNA. The HATs have acetyl groups on the
DNA which opens it up and gives room for TFs to bind and makes sure RNA polymerase can initiate
transcription.
The bottom strand is silenced. Silencing a gene can be done by methylation, so putting methyl groups
on the DNA. Proteins with methyl binding domains are recruited which can in turn recruit HDACs
that take the acetyl away. This completely closes the DNA.
In cancer DNA, in front of the start site there are much more methylated Cs that are inhibiting
transcription. This is of course a tumor suppressor gene which is silenced and contributes to
carcinogenesis. So, hypermethylation of promotors is specifically seen in tumor suppressor genes.
Generally, we see global hypomethylation, they have lost most methylation marks but in tumor
suppressor you see promotor hypermethylation.

Methylated DNA reacts different to sodium bisulfite treatments, a chemical reaction that specifically
changes unmethylated Cytosines to Uracil, so deamination. Methylated cytosines are protected from
deamination. So, after treatment with sodium bisulfate you can distinguish the methylated form the

unmethylated DNA.
The difference can be picked up by sequencing pr by PCR. In sequencing you can’t see the
methylmark so you first have to do the bisulfate treatment. In the top you can see the methylated DNA
still has cytosines in it so these were protected by the methyl mark. In the bottom you can see all of
the cytosines are changed into uracil which are picked up as Ts, so these were all unmethylated.

In PCR you use methylation specific primers so you use primers that are complementary to part of the
DNA and then you do amplification. After the bisulfate treatment you can now use primers to target
methylated or unmethylated DNA.

microRNA
MicroRNAs are small non-coding RNA molecules, usually of 22 nucleotides long. They are not
translated in amino acids and do not code for any protein.
They are transcribed from genes by RNA polymerase,
the same as other RNAs. The difference is that they are
already processed in the nucleus by an effector
complex called microprocessor. This complex consists
of DGCR8 and DROSHA. DROSHA cleaves the
hairpin structures that these RNAs make. It is a hairpin
because there is complementarity to itself when falling
over. The cleaved hairpin gives the pre-miRNA which
A couple of microRNAs are oncogenes, so highly expressed in cancer, but some microRNAs function
as tumor suppressor genes.

Because microRNAs are transcribed like protein coding genes, they are also regulated in that way.
In cancer there is often a lower expression of DICER than in healthy tissue, which reduces the impact
of microRNA. So, microRNAs generally have a tumor suppressive function, but there are specific
microRNAs that do function as oncogenes.
MicroRNAs are quite versatile because they don’t need full complementarity to its target. The first
part is most important for it’s binding and this is only 7 nucleotides, these need to be perfectly
complementary but there can be a couple of mismatches in the rest of the sequence. Many mRNAs
have target sites for many microRNAs and there are many different microRNAs for a specific mRNA
so, it are many-to-many interactions.

Sometimes microRNAs have functional consequences but they can also function as biomarkers. Two
microRNAs where studied of which one had an higher expression in cervical cancer and one was
lower. To ratio between these gave a distinction between abnormal and normal cytology of cervical
tissue.

Long noncoding RNAs

Long noncoding RNAS (lncRNAs) are longer then 200 bases. They are not created by a certain
mechanism. They all have their own individual function, of many is their function unknown.
LncRNAs are processed the same as coding the DNA but do not produce any protein. Some function
in the nucleus or example one is involved in silencing the second X-chromosome in female tissue.
Xist is present on one of the chromosomes where it will code the chromosome and attract protein
complexes that put repressive marks on the DNA. Eventually this chromosome shrinks and turns into
heterochromatin. Another example is HOTAIR, this is expressed on a chromosome and functions on
the other chromosome. It puts a repressive mark on a gene on the different chromosome. This
important in the development and is overexpressed in breast cancer. TERRA is transcribed from
telomeres and has an important function in repressing telomerase activity.
So individual lncRNAs have their own function. A function that several lncRNAs share is function as
decoys for microRNAs. A lot of these might look like pseudogenes because they do not code for a
protein but still contain the 3’UTR with all the miRNA target sites because of which it can derepress
another protein. An example of this is the pseudogene ..PTEN which has the same target sites as
PTEN itself and can like this derepress PTEN.

As with miRNA, lncRNAs can be potential biomarkers in cancer. For example, PCA3 is quite specific
for prostate cancer. Or for example NORAD, a lncRNAs that is upregulated among chemotherapy and
when it is not present (because of a knock-out) cells start becoming aneuploid, they lose their
chromosomal integrity. So, NORAD has a critical function in response to DNA damage and is
necessary to prevent aberrant mitosis events. This is the case because PUMILIO binds to NORAD and
is sequestered away. Without NORAD there is too much PUMILIO and you get aberrant mitosis.

Telomeres
Telomeres are several thousand repeats of TTAGGG at the ends of chromosomes. Telomeres have a
different structure than regular they make a T-loop what protects the end of the chromosome. Without
telomeres the end of the chromosome would just be a loose piece of DNA which is sensitive for
degradation or it would be picked up as a double strand break what leads to fusion with other parts.
The shelterin complex maintains the stability of the T-loop.
To replicate the telomere there is a problem. DNA replication happens in a bubble and there is a RNA
primer used as a starting point to start the replication of DNA. DNA polymerase only works from the
5’ to the 3’. This is a problem for the lagging strand, here it needs to start all over with little pieces.
This process works fine until you get to the end of the chromosome because then you end up with an
RNA primer but you can’t add a new one because the chromosome ended. So, after every DNA
replication you lose a little bit of the end of the chromosome, a part of the telomere. Upon multiple
division telomeres get shorter, with every cell division. With every division 100-200 bases are lost of
the telomere. At some point you run out of telomere and you start cutting away the DNA. Before this
happens it is already detected as a double stranded DNA break and the cell will go into apoptosis. This
is unless you extend the telomeres using telomerase, this is important in the germline. Telomerase
does not use a complementary strand as template but uses a piece of RNA. So, it is actually a reverse
transcriptase using a RNA template to make DNA. This RNA template is called the telomerase RNA
(TR) and it has parts that are complementary to the telomerase repeat and an extra part that is
extended. In this way you get extra telomere repeat on the telomeres.
Telomerase consists of two components; the enzyme hTERT and the RNA template hTR. Normally it
is only expressed in stem cells and germline cells.

Telomere shortening is caused by the end replication problem and by oxidative stress. After 50-500
cell divisions you get critically short telomeres which leads to cells going into senescence.
In somatic tissue you lose telomeres and at some
point the cells will go into senescence. Thee only
way to overcome this is by is to start expressing
telomerase, by adding it in the lab. In cancer, the
cells will pass senescence by inactivating tumor
suppressor genes. P16, pRb and p53 are able to let
cells go into senescence and stop them from dividing
when telomeres get critically lost, but when these are
lost the cells will keep dividing and getting shorter
telomeres till the point of crisis. In crisis most of the
cells will die, this is a natural protection against
cancer. Some cells manage to escape this, usually by
re-expressing telomerase. In 90% of cancers
telomerase is expressed again.

One way telomerase can be expressed is

through telomerase promotor mutations. Some bases in the promotor of TERT, so just upstream the
transcription start site, that are hotspots for cancer mutations. All of these are deamination changes
which mutate Cs to Ts. When this happens the resulting sequence makes transcription factor binding
sites for Ets, an activating TF that leads to the expression of telomerase. This is found in 50% of
cancer patients.

Because in 90% of cancers telomerase is reactivated and the expression is limited in adult tissues it
seems to be the perfect cancer target, but there are some problems;
o Inhibitors struggled due to bioavailability and toxicity. So there were problems in getting the
right concentrations and patients that were getting the drugs had more side effects.
o Cancers may have already produced cells with relatively long telomeres. Because they have
reactivated telomerase cells may have already lengthened their telomeres for many divisions to
come.
o It is easy to get resistance, because there are also other mechanisms to lengthen the telomeres.

Therapeutic options
In cancer there is always deregulated transcriptions, related to epigenetic changes, because of
oncogenes that shouldn’t be expressed and tumor suppressors that are shut down. Besides mutation
this can be caused through promotor hypermethylation. There are therapeutic options that target these
epigenetic changes because these are reversible;
o DNA methylation inhibitors: prevent methylation, as tumor cells divide they can’t copy the
DNA methylation mark because they are inhibited
o HDAC inhibitors: the DNA opens up so suppressed tumor suppressor will now again be
expressed.
One of the most effective cancer treatments is re-expressing the tumor suppressor that were being
repressed.
Besides therapeutic options, these epigenetic alterations can also be used as disease markers. For
example hypermethylation is a good biomarker for cancer and can be used to detect cancer at an early
state.

Use of epigenetic drugs

DNA-methyl transferase inhibitors (DNMTi) like 5-Azacytidine induce the expression of viral genes
and transposable elements which yield dsRNA. dsRNA causes an interferon response which leads to
antitumor immunity.

Lecture 3
Ch4 Growth factor signaling and oncogenes
Two general statements describing the molecular basis of cancer;
o Aries by the accumulation of genetic and epigenetic changes in the DNA
o Arises by deregulation of signal transduction pathways

So, it’s a signaling disease. 1/3 of all genes are involved in signal transduction. This makes sense
because we are multicellular organisms so communication between cells is needed.
Signals lead to cellular responses. Cellular responses are changes in;
o Apoptosis
o Proliferation, cell division
o Differentiation
o Kill something, for example another cell
o Metabolism
o Protein production and secretion
o Depolarization and contraction
o Motility, detach and move
o Depolarization and contraction
All these changes are changed by protein activity. Protein activity can be changed at RNA or at
protein level.
RNA level;
o Gene transcription regulation
o Transcript degradation regulation by microRNAs
o Differential splicing
Protein level;
o Production
o Modification
o Degradation
o Translocation, protein is kept in the cytoplasm while it’s active in the nucleus so the
translocation determines its activity
The initiation of translation is completely changed in cancer cells.

In signaling, proteins are molecular switches, you can put them on and off. Protein changes are caused
by;
o Chemical modification
Phosphorylation
Acetylation
 Mono-ubiquitination
Methylation
Sumoylation
o Interaction
o Binding to small molecules

Protein phosphorylation
The phosphorylation is governed by protein kinases. These use a free hydroxyl-group and ATP is used
as a cofactor to change the hydroxyl-group into phosphorylated oxygen.
Amino acids with a free hydroxyl-group, so amnio acids that can be phosphorylated are serine,
tyrosine and threonine. There are 518 protein kinases in the genome. There is a functional and
homology classification. The functional classification is based on the amino acid;
o Tyrosine kinases (PTKs) phosphorylate tyrosine
o Serine/threonine kinases (PSKs) always go together, they phosphorylate serines and threonines
o Dual specificity; can do both

Phosphorylating can cause activation and deactivation, so be the on and off switch, but is also done by
dephosphorylation using phosphatases;
o Protein tyrosine phosphatases (PTP)
o Protein serine/threonine phosphatases (PSPs)
o Dual specificity phosphatases
Phosphatases are less specific than kinases.

Kinases need ATP in the ATP-binding pocket. They are very interesting drug targets for cancer
treatment. Lots of Kinase inhibitors have been generated, these are ATP analogs and fit into the
binding pocket.

Phosphorylation of a protein causes;

o Change of conformation
o Creation of docking site, a phosphor-tyrosine can be recognized by another protein. It are
particularly the SH2 domain of proteins that are able to recognize.

Interaction
There are different kinds of interaction;
o Interaction with inhibiting proteins: Two proteins can be in a complex. The active protein is
inhibited by being in a complex. One of the proteins is phosphorylated, the complex opens and
the active protein is available.
o Interaction with activating protein: Two proteins are not a complex, when one of the proteins
gets phosphorylated which leads to formation of a complex and become active.
So, it can be activating and inactivating.

There are many protein binding domains, for example;

o SH2 domains bind to phosphotyrosines
o SH3 domains bind to proline rich areas in other proteins
o WW domains bind
o Pleckstrin Homology domain
o Leucine zipper domain
o Etc

Because interactions site are small domains they can be combined in one protein. There are many
proteins that contain a lot of these domains together. So, the protein can bind to all kind of things
based on what is present, modification state etc.
Binding to small molecules
Many small molecules can bind to protein and change their activity. Small molecules are released by
enzymes and then flow freely through the cells having impact on all proteins with domains binding
these second messengers. For example GTP is always present, this binds to heterotrimeric G-proteins
and Ras-like GTPases, very important signaling proteins particularly in cancer. Another example are
lipid derivatives.

RAS signaling
RAS protein is a small protein that sits into the membrane. There is a fatty acid linked to one of the
amino acids, with this fatty acid it sticks into the inner membrane. usually RAS is bound to GDP, then
it is inactive. By a protein GEF GDP is kicked out and GTP is put in. when bound to GTP, RAS is
active. This is the on-switch. The off-switch is the GTPase activity of RAS itself, it hydrolyzes GTP
into GDP. This is done using help from GTPases activating proteins (GAPS), that help RAS to
activate GTPase activity. Then the GTP changes into a GDP and the protein is inactive again.

The membrane is made out of phospholipids. A subgroup of these phospholipids is bound to inositol,
a sugar. This is phosphorylated on two places so its called PIP2. The enzyme phosphatidylinositol 3-
kinase (PI 3-kinase) is able to turn PIP2 into PIP3, with one extra phosphate groups as only difference.

Relay systems
There are three relay systems;
o Steroid hormones
o Protein phosphorylation cascades: EGFR-receptor
o Second messengers

Steroid hormones
There is a TF in the nucleus that is inactive till a co-factor is bound that binds the steroid hormone. So
the hormone comes from outside and when it enters the nucleus and it binds to nuclear receptors.
When the TF is bound to hormone it becomes active and steers gene transcriptions so gene that are
sensitive to these hormones will be transcribed.

Protein phosphorylation cascades: EGFR-receptor

The EGFR-receptor is a molecule in the membrane with a tyrosine-kinase part intracellular and a
ligand-binding domain extracellular, it binds EGF. When a growth factor is presented to the cell and it
binds to the extracellular part, the tyrosine-kinase domains will close and cross-phosphorylate each
other. After this you have a docking site for SH2 domain proteins. A protein with a SH2 domain can
bind and cause down signaling. The protein that binds to the phosphorylated EGFR-receptor is Grb2.
When this binds it attracts another protein called SOS, a guanine exchange factor (GEF). SOS kicks
out GDP from RAS and brings in GTP which activates RAS. Activated RAS binds Raf. Raf is a
kinase that phosphorylates and downstreams molecules called MEK. MEK is also a kinase and
phosphorylates ERK. ERK is also a kinase, it translocates to the nucleus and phosphorylates TFs to let
them bind to the target genes. Via all these steps the signal is transferred from the growth factor to TFs
in the nucleus that drive gene expression. So, EGF is an important stimulating factor for cell division.
Besides EGFR there are many other receptor tyrosine-kinases of which some work with adaptors etc.
This makes no difference for the concept.
There are four ErBB receptors, ErB1-4. ErB1 is EGFR, so there are more names for the same protein.
They can dimerize with the same molecule but also with others. It depends on the context of the cell,
whether it has for example only EGFR or also ErB2 or ErB3. And what happens in the cell depends
on this. It also depends on the ligands that are present.
All epithelial cells respond to EGF but cells might respond differently, not every cell has very
component.

RAS has not only the effect activating RAF-MAK-ERK-cell cycle, but it can also activate PI 3-kinase
what modifies PIP2 into PIP3 and more other effects, so, RAS has also multiple effects. All
components interact with each other so it’s a complex network.

Second messenger
The pleckstrin homology domain binds to PIP3. The protein that does this is AKT at when it’s
activated. This is a kinase so when activated it phosphorylates other proteins and gives a survival
signal in the cell. For cancer cells survival us the main interest.
The off-switch is a lipid phosphatase what can dephosphorylate proteins, and is called PTEN, an
important gene in cancer.
Here, the second messenger is PIP3.

Relay systems: why?

o To transfer signals
o To amplify the signals
o To integrate and regulate the signals

Receptors
o Growth factor receptors
o Photon receptors, in the eyes, signaling of vision. For example the rhodopsin receptor for light.
o Adhesion molecules can grab something and can feel and give signal about that. For example
the integrins that bind to the matrix.
o Ion channel receptors, regulate polarization in the nerves.

DNA changes and oncogenes

Cancer is a signaling disease. All the oncogenes and tumor suppressor genes act in signaling pathways
and they rewire the signaling systems in the cell to become a cancer cell. These changes in the
signaling pathways are caused;
o Point mutations: In oncogenes mostly activating mutations are found, missense mutations.
Missense mutations cause a codon to code for a different amino acid. There are also
inactivating mutations, like nonsense mutations which change the codon in a stopcodon TGA,
TAA and TAG, and like splice site mutations in nucleotides before and after an exon which
cause splicing to go wrong.
o Chromosomal translocations: For example the Philadelphia chromosome, a translocation
between chromosome 9 and 22.
o Chromosomal gains: gene amplification
o Viral integration: This means the viral genome inserts into the genome of the host cell. This
does play a role in mouse cancer, not in humans.

Example oncogenic activation BRAF

Raf is activated by RAS bound to ATP and it is an oncogene that is often mutated in melanoma. It has
the V6000E mutation so number 600 gets mutated from a valine to a glutamine. This is in the kinase
domain and the consequence of this mutation is that BRAF is constantly active. Amino acid number
one is always methionine.
Usually, missense mutations occur in oncogenes. However, it does occur a nonsense mutation is
found. Normally this causes that the protein stops but is does occur that receptor tyrosine-kinases are
mutated and only the active kinase domain is made and the ligand-binding part disappears.

Example translocation BCR-Abl

The Philadelphia chromosome, a translocation between chromosome 9 and 22, leads to fusion of the
gene BCR on chromosome 22 to Abl on chromosome 9. You get a funny fusion protein with aberrant
functions. These proteins have a active kinase domain but they don’t listen to the rules.
The BCR-Abl fusion proteins have a lot of activity in the signaling pathways, so it are really toxic
proteins. They play a role in leukemia but not in lung cancer at all.

Example oncogenic activation RAS

When RAS is mutated in amino acid 12 or 13 it loses its GTPase activity, so the signal can get on but
can’t get off so RAS stays active and there is continuous signaling in the cell to divide.

Example mutation RET oncogene

RET interacts with GDNF, this forms complexes. Then you have dimerization of the two receptor
kinases and phosphorylation.
This can happen with all 4 types of ligands, different types have different effects.
The complex has multiple tyrosines that can be phosphorylated, there are multiple molecules that can
give signals to the cell.
Mutations in RET oncogene causes neuroendocrine cancers. The type of mutations determines the
type of tumor. So, it’s very dependent on where the mutations are, that is not usually seen it is
specific for this oncogene.

Signaling in cancer
There are oncogenic pathways, growth stimulating pathways like RAS, and suppressing pathways,
growth inhibiting pathways like TGFb and p53. Oncogenes and tumor suppressor genes may act in
both. When RAS is mutated is has increased activity what causes cancer. PTEN is also an activating
pathway but it is a tumor suppressor gene because it inactivates the signal.

How do these genes cause the cell to be malignant;

o Overproduction of the growth factor
o Gain of function mutations in for example the receptors
 o Loss of function mutations, they inactivate inhibitors so the off-switch is inactivated
o Overexpression of transcription factors
Cells become addicted to is.

Signaling and treatment

When a tumor is found that is very dependent on for example Cetuximab or EGFR signaling there are
inhibitors/antibodies that bind to EGFR and inhibit the ligand for binding. These are antagonists, the
signaling is blocked. This can be used when a cancer normally uses this pathway. Another example is
Herceptine, an antibody that blocks Her2/Neu positive breast cancers.
Tyrosine kinase inhibitors like Gleevec, Iressa and Vermurafenib block kinases in the cell. If a tumor
activates a pathway by a mutation in a protein tyrosine kinase, this can be blocked by these inhibitors.

Example role of RAS in signaling melanoma

In melanoma, 85% has the V600E mutation in BRAF. BRAF is a serine/threonine kinase and there is
an inhibitor made against it, a TKI inhibitor called Vermurafenib, an ATP analog.

Lecture 4
Ch5 The cell cycle

The cell cycle

The cell cycle is the process that cells progress through in order to proliferate. It is key to organismal
development and homeostasis.
Dysfunction of the cell cycle can lead to;
o Uncontrolled cell proliferation
o Errors during the cell cycle: mutation

The key goal of the cell cycle is to generate to identical daughter cells out of a single progenitor cell.
Therefore, the cycle has to achieve two things;
o DNA replication
o Partitioning of the duplicated genome into two new cells
In eukaryotic cells these two events are very clearly separated.

The cycle can be divided in several different stages;

o S-phase: DNA replication
o G2-phase: gap phase for preparation and control, to make sure the genome is correctly
replicated
o M-phase: mitosis, cell partitioning
o G1-phase: gap phase for preparation and control, to make sure the cell is ready for replication
so that there are enough metabolites to make a new genome
A cell divides depending on its environment, so whether there are enough mitogens and metabolites
around the cell decide to immediately enter a new cell cycle in G1 or they go into a quiescent state in
G0 where they exit from the cycle and wait until there are favorable conditions (availability of
mitogens) so cells can re-enter into the cycle

Interphase is all phases of the cell cycle except mitosis, G1, S and G2. This is because cells look
different in mitotic cells in comparison to cells in these phases.

S-phase
Each chromosome needs to be duplicated. This happens in a semi-conservative way where the DNA is
unwound and then replication by a machinery.
After replication you end up with two identical sister chromatids. These are not floating around in the
nucleus. They are kept together by a complex called Cohesin, this holds the two sister chromatids
together.
Upon completion of the S-phase, cells contain replicated chromosomes. In a diploid cell, you then
have two pairs of each homolog and each have two sister chromatids so you have 4 copies of each
chromatids in your cell.

M-phase
In mitosis you start partition these chromatids and the cells so you end up with two cells with both one
of the copies. When mistakes occur, so when two of them ed up in one of the cells, the other cell that
misses a chromosome will not survive. The other cell might even do better or even other
characteristics.
How is chromosome partitioning achieved in mitosis?
Chromosome are moved around using the microtubules. Microtubules are made of polymerized
tubulin dimers and assemble into a structure called the mitotic spindle. They interact with the
chromosomes and move them around by making force and can eventually pull them apart.
So, the spindle microtubules form, interact with the microtubules and pull them apart. First, they make
sure all the chromosomes align in the middle. A state called metaphase. When they separate so when
you see the single sister chromatids go apart its called anaphase.
So, mitosis consists of multiple phases. You start with prophase where you see individual sister
chromatids appearing. During metaphase you see them aligning in the middle of the mitotic spindle.
In anaphase the segregation of the chromatids occurs and in telophase/cytokinesis new nuclei are
formed. At this point the cell decides to go to G1 and do the cell cycle again or to go to G0 and go into
quiescence.
Transition from metaphase to anaphase is nonreversible, the chromatids are pulled apart and you
cannot put them back. So this transition is where most control occurs.

How is the order established? How are events

controlled in order to prevent error?
The central players that determine the cell cycle are Cyclin-
dependent kinases (CDKs). They have a fluctuating
activity and this drives cell cycle transitions and processes.
The CDKs use ATP to position phosphates on target
proteins that then change behavior and drive cell cycle
processes. CDK activity defines where the cell cycle cells
are.
A major way how the activity of CDKs is controlled is by
binding to Cyclins, non-enzymatic cofactors of the CDKs
that allow the CDKs to become active, without these they
cannot become active.

There is different types of CDKs that drive different phases

of the cell cycle;
o G1: CDK-4 and -6
o S: CDK-2
o G2/M: CDK-1
There are also different Cyclins and their concentrations
fluctuate during the cell cycle. Cyclin availability sets cell
cycle stage. Certain cyclins work with certain CDKs.
o CDK-4 and -6: cyclin D
o CDK-2: cyclin E and A
o CDK-1: cyclin B
The cyclin levels are controlled in two ways. One is on the transcriptional level, as cells progress to
the cell cycle certain cyclins accumulate because their transcriptional activity is increased. The other
way of control is very abrupt. This is caused by active degradation by the cyclin by Ubiquitin-
mediated proteolysis.

In mitosis cyclin B levels increase and they go down during the metaphase to anaphase transition.
This is controlled by ubiquitin-mediated proteolysis by the APC/C complex, a ligase that specifically
ubiquitinates cyclin B and targets it for degradation. This only occurs during metaphase.
To prevent this they started to mutate this so cyclin B cannot be ubiquitinated anymore, there is made
non-destructible cyclin B. These cells where stuck is metaphase so destruction of cyclin B drives the
transition from metaphase to anaphase. This destruction leads to a rapid decrease in CDK1 activity,
the levels stay the same but the activity goes down.
Besides regulation of CDK by cyclin, there is also regulation by CDK inhibitors which associate with
CDK complexes to inhibit them and there is regulation that occurs on the cyclins. There are different

families of CDK inhibitors.

Checkpoints
Besides this basic controls that makes the cell cycle run, there is additional control that makes sure it
happens in a correct order. These are called checkpoints, biochemical signaling cascades that establish
order and make events contingent on the successful execution of previous events. So, for example
they make sure DNA replication is completely finished before cells enter mitosis and that the cell only
enters anaphase when all the chromosomes are properly segregated.
So, there are several moments in the cell cycle when checkpoints are active;
o G1-S: DNA integrity? controls if the DNA is in a proper state, so there is no damage
o G2-M: DNA integrity? is there complete DNA replication
 o Before S-phase to control if favorable conditions are present
o Mitosis: is there correct chromosome segregation

G1-S transition checkpoint

Transition from G1- S-phase is an important decision phase because if the DNA is damaged you don’t
want to go into the S-phase.
The transition depends on the TF family E2F which drives a program required for S-phase. The E2F
family is held in check by the protein retinoblastoma (Rb) which together with HDAC can bind to the
E2F complex and inactivate it. This prevents binding of the complex to E2F and recruits HDAC. So,
upon mitogen stimulation CDK4/6 and cyclin D, which drive the G1-S transition, can become active
and phosphorylate Rb. This leads to a partial release of the inactivation, which causes the start of
transcription of cyclin E. Cyclin E accumulates, this leads to formation and activation of CDK2-cyclin
E complexes. This phosphorylates another site of Rb, completely releasing the E2F complex from
inhibition, leading to full E2F activity. Cells can now progress into the S-phase.

1: 2:

3: 4:
DNA integrity influences this system, for example by inducing the protein p53. p53 is a tumor
suppressor that then induces the protein p21, a CDK inhibitor which can then for example inhibit
cyclin E to prevent S-phase entry.

G2-M transition checkpoint

In this transition it is important the DNA is replicated properly. When there is damage ATM and ATR
are activated. This are kinases that phosphorylate downstream factors which leads to inhibition of
Cyclin B and CDK activity and cells cannot enter mitosis.
ATM is also needed for the induction of p53 and p21 in G1.

Mitotic checkpoint
This transition makes sure correct chromosome segregation occurs. This happens during metaphase to
anaphase transition.
So, this monitors if all chromosomes are properly attached to the microtubules. If one of them is not
properly attached, you do not want to enter anaphase so you want to stop the cell cycle.
When there is incorrect attachment there is a cascade of signaling activated (aurora), partially
dependent on kinase activation, that eventually inhibits the protein APC/C. This is the ligase that
ubiquitinates cyclin B. So, as long as not all chromosomes are attached the cell cycle remains in
metaphase.
This mitotic checkpoint works in ubiquitin-mediated proteolysis.

Failure of checkpoints leads to problems;

o Propagation of mutated DNA
o Missegregation of chromosomes leading to aneuploidy.
Many cancers exhibit defaults in checkpoint function, mostly G1-S transition. Many cancers have
mutations in this pathway so they cannot stop the cell cycle in unfavorable conditions.
So, genes of the cell cycle that are often mutated in tumors are often involved in G1-S transition.
Often seen is an upregulation of cyclin D, loss of Rb, loss of p53, loss of p16.
So, there are alterations in Rb, cyclins CDK and CDK inhibitors.
Also other mutations occur which not control the G1-S transition, like in ATM kinase, p53 and aurora
kinase.

In mistakes in the cell cycle point mutations are common but also changes in chromosome number,
aneuploidy. Aneuploidy is caused by errors in chromosome segregation so by mutations in the
metaphase to anaphase transition in the mitotic phase of the cell cycle. However, mutations in mitotic
checkpoint are rare, so how does this occur?
Without checkpoints, too many mistakes are made for the cell to live so even cancer cells will die.
You cannot live without the checkpoint. To a certain extend these mistakes are tolerated in cancer
cells, maybe even beneficial, but complete loss of the checkpoint is not tolerated.
There is still debate over the role of aneuploidy in cancer, so whether the cancer cells benefit from it
or if it is just a collateral effect of other problems instead of a driving force. There is evidence for
both.

Targeting the cell cycle

o Several of the most used ways to treat cancer are based on the cell cycle and targeting it. Many
approaches work by activating checkpoints, as such inhibiting proliferation;
- DNA damaging agents: mechanisms are used that generate DNA damage that stops
cells from going are that the mutation load gets so high the cells die.
- Microtubule drugs: drugs like Taxol target microtubules. It inactivates the function of
the spindle so chromosomes can’t be separated properly and they’re stuck in mitosis,
mitotic arrest. It has lots if side effects because it does not act only on cancer cells.
o In addition to generating damage, another way to target the cell cycle is to target the
machinery. Kinases control the cell cycle and are very mandible to develop small molecule
inhibitors for. So, treatment is then based on
- CDKs that drive the cycle
- Other kinases that control the checkpoints like ATR/ATM. Kinases are ATR/ATM are
kinases involved in G1-S transition checkpoint and ATM also in the G2-M transition.

There are now clinical trials to target these to treat cancer. What would be a good combinatorial
therapy with ATR/ATM inhibition?
It should be combined with drugs that damage the DNA like radiotherapy. Unfortunately, this causes
also damage in no-tumor cells.
Also CDKs are used as targets. For example inhibiting CDK4 stops cell cycle entry so this stops
cancer cells from dividing. However, this is quite effective in some tumors but not in all because some
tumors don’t respond to the drug efficiently and often resistance is developed. What might lead to
resistance non-responsive behavior after treatment with CDK4 inhibitors?
The kinase that you try to inhibit, controls Rb what controls E2F, which are both also controlled by
cyclin E. Mutations in these genes can trigger resistance. One of the factors by which CDK4 drives
G1-S transition is by phosphorylating Rb what releases the inhibition Rb normally has on E2F. But, if
there is a mutation in Rb, so you don’t have any normal Rb, it is not needed to inhibit CDK4 but cells
don’t care because there is no inhibition by Rb that is normally performed.
This is the same with E2F and cyclin E. Upregulation of E2F and cyclin E is a cause of resistance or
pore response.
So, it is important to know what the statis of Rb and cyclin E is because then you might know whether
the patient might respond to the drug. You don’t want to give the drugs if you already know it won’t
work because there always are side effects.
Are there genes that affect sensitivity to the CDK inhibitor (Palbociclib)?
Cells were treated with Palbociclib and individual genes that might influence sensitivity were
manipulated (knock-out) using CRISPR. So, in cells individual knockouts were made, and part of
cells was thereafter treated with the inhibitor. When knocking out a gene that causes high sensitivity
only after treatment with the inhibitor, these cells should die and in the wildtype not. Eventually you
find genes that cause sensitivity or resistance to the drug.
In the figure, on the left genes are causing extreme sensitivity to CDK inhibitors and on the right that
cause resistance. In the right you see Rb, Rb mutations are resistant to CDK inhibition. On the left you
cyclin E and CDK2, if you reduce the levels if cyclin E and CDK2 the cells become extremely
sensitive because CDK4 then becomes really important. AMBRA1 has the strongest hit. When there is
a loss in AMBRA1, cyclin D goes up because AMBRA1 ubiquitinates cyclin D so it gets degraded.
This allows the CDK4 complex to be present. In wildtype cells, when using CDK4 inhibitors, the
cyclin E and CDK4 levels can be inhibited, inhibiting cell cycle entry. In cells without AMBRA1,
there are high levels of cyclin D which causes CDK4 inhibitors to be not as effective because cyclin D
start to associate with other cyclins and CDKs that cannot be inhibited by this CDK4 inhibitor and this
then drives entry into S-phase. So, AMBRA1 mutations occur in cancer and might impact resistance
to therapy.

In a subset of cancers there is amplification of cyclin E. PKMYT1 leads to sensitivity, only when
there is amplification of cyclin E.
CDKs are phosphorylated and dephosphorylated by different proteins. Wee1/Myt1 are kinases that
phosphorylate CDK and can lead to inactivation. For some reason, tumor with more cyclin E very
sensitive when Myt1 is inhibited.
This might not be problematic for cells without cyclin E overexpression (normal cells) what makes it
an attractive way to target specifically tumors.
 Lecture 5
Ch6 Growth inhibition and tumor suppressor genes
The tumor suppressor gene PTEN plays a role the EGFR pathway. In normal cells EGFR is activated
when the EGF growth factor binds on it on the outside of the cell. This triggers a cascade of kinase
activations what in the end results in proliferation of the cell and its survival.
You want to be able to stop this when proliferation is not needed anymore. When EGF is gone the
signal stops and the cell will stop proliferating. However, if you want to do it more quickly you need a
brake on the system. It are all kinase steps and one of the steps in mediated by the enzyme PI 3-kinase
which adds a phosphate group on PIP2 making it PIP3. The tumor suppressor protein PTEN is a
phosphatase that removes the phosphate group, turning it back into PIP2. This is the brake on this
pathway.
If you lose PTEN not so much will go wrong. Phosphate groups can only be added, they can’t be
removed again so the cell will proliferate a bit more and it will take longer to stop when there is no
EGF. This is not a problem in itself, it becomes a problem when the EGF receptor is mutant and has
become a oncogene (EGFR is a proto-oncogene). Now it is constituently active and is driving the
pathway also in absence of EGF. In this situation you really need the brake. So, the combination of
activation of this oncogene with the loss of the tumor suppressor gene causes the cell to turn into a
cancer cell.

Knudson’s observations
There are families with a high incidence of cancer, also with an earlier onset of the cancer.
Usually, people get retinoblastoma later in their life and in one eye, but there are families that have a
lot of retinoblastomas developing early in life and in both eyes.
Knudson’s two-hit hypothesis: a tumor suppressor gene is a gene in which a germline mutation
predisposes an individual to cancer.
In the left you see two chromosomes with alleles of the tumor suppressor gene, in the wildtype
situation. At some point you get a mutation in one of the copies of this gene. This is not a problem
because still have a correct copy on the other allele so you can still produce functional tumor
suppressor protein. But if you later also lose the second copy of the gene you have cannot produce the
protein anymore.
This usually happens in sporadic cancers. It takes a long time to get the two hits and the chance that it
will happen is very small because the mutation has to be in the same gene in the same cell.
Sometimes in families there is already a mutation in one of the two copies. This means the mutation is
in the germline so you have it in every cell. So, in any cell a hit in the other allele is already causing
loss of the tumor suppressor protein. This means the chance is way higher. So, this is not sporadic but
familial cancer.

In oncogenes activation of a gene is a dominant effect, it only has to happen in one of the alleles. In
tumor suppressor genes it is recessive so you need to lose both copies. There are exceptions, called
haploinsufficiency where one mutation reduces the amount of the protein so you don’t have enough
for proper function.

Tumor suppressor genes and their associated familiar predisposition syndromes;

o Rb1: retinoblastoma
o p53: LI-Fraumeni
o APC: colorectal cancer
o BRCA: breast/ovarian cancer

Retinoblastoma gene
Retinoblastoma is a pocket protein that binds the transcription factor E2F. The release of E2F is
regulated by phosphorylation of Rb mediated by two complexes of a cyclin and a CDK. This are
CDK4/6 with cyclin D and CDK2 with cyclin E.

Mutations in Rb pathway;
o Loss of Rb protein: if there is no functional Rb protein in the cell E2F is freely available and
will constantly trigger the cell cycle. This is caused by deletion, frame shit or nonsense
mutations)
o Missense mutation: there is Rb protein but it has a mutation in the pocket domain because of
which it cannot bind E2F anymore.
o Hyperphosphorylation through upstream mutation: the Rb protein itself is functional but an
upstream mutation leads to hyperphosphorylation, so Rb is constantly on a state where it does
not bind E2F.
o Sequestration pRb by tumor virus protein: Rb is still wildtype but it is being inactivated.
All these mutations lead to more cell cycles.

p53 pathway
p53 is a tumor suppressor that is often called ‘the guardian of the genome’.
Upstream there are all kind of deregulation that a cell can have, for example, DNA damage, oncogene
activation and stresses. This results in different cascades that cause a signal that will activate p53. p53
is a decision point for what will happen as reaction of the cell to overcome this deregulation. This is
then mediated by certain effector proteins. If the cell fails to overcome the deregulation p53 decides
the cell can better die because it otherwise can be starting point for cancer.

Structure of p53
The p53 protein consists of different domains; the tetramerization. DNA binding, MDM2 binding and
regulatory domain. The tetramerization domain is a recognition sequence for p53 proteins to be able to
bind to each other. p53 is only functional as a tetramer. This tetramer needs to bind to certain
recognitions sequences in the promotor of target of p53. Therefor it has a DNA binding domain that
binds very specifically to sequence elements that are only present in the promotors of genes that are
driven by p53. So, the tetramer can bind to this sequence to drive the transcription of this gene, but
more is needed. On the N-terminal site of the protein it has a binding domain for the protein MDM2.
When MDM2 is bound to p53 it is still inactive. So, it sits on the promotor if its target genes but is
still not doing anything as long MDM2 is bound. To active p53 you have to get rid of the binding of
this MDM2. This can be achieved in different ways. After getting rid of it transcription can start. How
much transcription will start depends on post-translational modification of the p53 protein by
ubiquitination/acetylation/phosphorylation on different sites of the protein and by the binding of
cofactors. These cofactors then bind to the regulatory domain. Only when positive acting cofactors are
bound to p53 there is strong transactivation of the promotors.
Regulation of p53 by MDM2
MDM2 is the critical regulator of p53. p53 is regulated at the post-transcriptional level. The
transcription of p53 level is constant, it does not increase when more is needed. So, there is constant
production but the activity of p53 is mediated by binding to MDM2. When it is bound it is inactive,
when it is free it is active.
The MDM2 gene is one of the target genes of p53 so p53 stimulates production of its own inhibitor.
This keeps balance.
MDM2 has different function. One is stimulating the export of p53 from the nucleus to the cytoplasm,
then it also ubiquitinates it to drive it into degradation. This causes the p53 activity to be low so
transactivation of the MDM2 promotor is low what leads to less MdM2 is produced. Because there is
les MDM2 there is less brake so there will be more p53. So, in a healthy cell there is constantly low
amount of p53 available in active and inactive form.
When the cell meets a stress signal the brake is released, p53 is immediately active and the cell can
respond immediately on the stress signal. There is also more production of MDM2 what helps the cell
to stop the reaction when the stress is gone.

How to inhibit the binding between MDM2 and p52 depends on what kind of stress signal is received.
There are two different ways. One is when DNA damage or cell stress is sensed. The sensors are all
kinases. In the end this results in phosphorylation of p53 on sites where also MDM2 can bind.

In oncogene activation, p14 is produced. P14 can bind to MDM2 and sequester it away at different
locations in the nucleus. Now MDM2 is not capable to bind p53 because it is sequestered by p14.
Downstream there are a lot of reactions among which cell cycle inhibition and apoptosis induction;
o Inhibition of the cell cycle: one of the most important target genes of p53 is p21, the CDK
inhibitor. So, p21 inhibits the CDK/cyclin complexes that phosphorylate Rb. Now Rb cannot
be phosphorylated anymore so E2F remains bound to Rb. This stops the cell cycle.

>>>
o Apoptosis: there are different ways in which p53 an activate apoptosis;
- Intrinsic apoptosis via mitochondria: one is by transactivating of different target genes.
Many of the targets of p53 are encoding proteins that are involved on the intrinsic
apoptosis via the mitochondria. These proteins form channels in the outer membrane of the
mitochondria. This results in the release of the molecule Cytochrome C in the cytoplasm.
This triggers the intrinsic apoptosis cascade. The process of apoptosis is very well
controlled, there are really strong brakes among which is the expression of BCL2 that
prevents the release of cytochrome C. On top of activating genes, p53 suppresses the
expression of BCL2. So it takes away the brake, inducing apoptosis.
- Transcription-independent pro-apoptotic effect of p53: p53 as a protein can go to the
mitochondria to assist in apoptosis induction. For this it doesn’t need its transcriptional
activity so this is transcription independent.
- Extrinsic apoptosis via FASR

p53 does not only activate all these processes by transcriptional activation of certain genes, it also
promotes production of miRNAs that inhibit other genes and in this way contribute to cell cycle
inhibition. So, the different processes involved in cancer are also inhibited by p53 by promoting
certain miRNAs.

p53: factors that decide downstream outcome

When stress signals come in, p53 has to decide what will be the response of the cell to overcome the
deregulation, if so cause cell cycle arrest, DNA repair or apoptosis. The first thing p53 will try to do is
to stop the cell cycle. This gives the cell some time to repair DNA damage. Then it does other kind of
things. Only if p53 is not able to direct downstream processes so that the cell has restored its damage
and has overcome the deregulation, p53 decides to drive the cell in apoptosis.
Different factors that determine what p53 will do, so to go in cell cycle arrest or into apoptosis are;
o Amount of p53 available: normally there is not much p53 in a cell because it is constantly
being degraded by MDM2 but if this inhibition is stopped p53 activity can increase.
o Modification (acetylation and phosphorylation etc) determines on which promotor it works
better
o Co-factor present at the promotor
o Co-activators binding to p53, this changes it preference for binding to different promotor.

1 Difference in promotor binding affinity

There are different affinities for binding to different sequence recognitions motifs in different
promotors. The promotor of the genes involved in cell cycle arrest like p21 is very high so only a little
bit of p53 is needed for it to immediately bind with high affinity t the promotor of p21.
The affinity for the promotors that drive apoptosis is low, binding only occurs if there is a lot of p53.

2 Co-factors determine affinity of p53 for effector promoters; MIZ-1

One of the co-factors is MIZ-1, a protein that binds to the promotor of p21. Thereby, it attracts p53 to
this promotor so it further enhances p53 to go to this promotor instead of others. So, first p53
stimulates cell cycle arrest, the cell tries to overcome its troubles. If not, the stress signal continues
which causes p53 to accumulate and then it can also go to the lower affinity promotors to cause
apoptosis. So, only when there is a lot of -53 available there is sufficient p53 left to also go there.

2 Co-factors determine affinity of p53 for effector promoters; ASPP

ASPP binds to p53, thereby changes its affinity so also a lower amount of p53 can already go to the
apoptotic promotors.

3 Competition with other proteins: myc

Other proteins can compete with p53. In cancers that are expressing the myc protein, myc also binds
with high affinity to MIZ-1, sheeling the binding site of p53. p53 cannot bind to p21 and stop the
deregulation of the cell by inhibiting cell cycle. Now p53 has no other choice then to bind to other
promotors and cause apoptosis.

>>

4 Post-translational modification
There are different modifications sites of the protein. Some of these modifications promote cell cycle
and others promote apoptosis. So, the affinity to different promotors is chanced by different
modifications to the protein that can occur.

Mutations in the p53 pathway in cancer

 o Mutations in the p53 gene, this can lead to:
- No expression, so complete loss of p53
- Missense mutation: most of the missense mutations found in cancer are in the DNA
binding domain so they prevent binding of p53 to the different promotors. Within this
DNA binding domain there are hotspots that together represent 30% of all cancers.
All mutations in the p53 gene together are only present in 50% of the cancers. The other 50%
have wildtype p53 so something else is wrong.
o Inhibition of p53 function: there is overexpression of p53 inhibitors, for example
overexpression of MDM2 caused by an amplification of the MDM2 gene. Because of the high
amount of MDM2 p53 is constantly degraded so these cancers don’t respond by activation of
p53.
There are also viral encoded genes.

Exceptions on Knudson’s two hit hypothesis

There are exceptions to Knudson’s two-hit hypothesis, p53 is one of them.
In a tumor suppressor gene that fulfils this hypothesis, if there is only one mutation there is still
production of the wildtype protein so there is still tumor suppressor function. In p53 this is different,
there are two differences;
o Dominant negative mutation: these occur in the p53 gene which causes that the p53 activity of
p53 is lost but the protein has a intact tetramerization domain so it can still form tetramers. It
can also form tetramers with the wildtype p53 so now there are 5 different types. Only one of
these is fully wildtype and can activate promotors etc, the other four can still bind to the
promotor but are transcriptionally inactive so they compete with the wildtype for the binding
sites. So, if you have 50% mutant protein, only 20% of the tetramers is transcriptionally active,
so loss of 50% tumor suppressor protein results in 80% loss of tumor suppressor activity.

o Gain of function mutations: p53 is still fully active but the protein is mutated such that it can
also bind to promotors of genes that are normally not target genes for p53. So, it has its normal
activities but on top of that it has gained extra function and can transactivate genes that should
not be transactivated.
Interaction of DNA viral protein products with Rb and p53
Rb binds E2F, is E2F is released what induces the cell cycle. E2F is also a transcription factor that
promotes the expression of p14, an inhibitor of MDM2. So, more production of p14 leads to less
MDM2, more p53 causing cell cycle arrest.

Viruses use host cells to proliferate, so they need to replicate their genome. Viruses bring minimal
information so they use as much as possible of factors produced by the host cell. When a DNA virus
infects a healthy cell that is not actively proliferating it needs the proteins of the host cell to produce
DNA so it wants to drive the cell into S-phase because then things needed for replication (like
polymerase) are produced that the virus needs for the synthesis of its DNA. So, the virus produces
proteins that bind to Rb stimulating the cell cycle. However, this also stimulates the inhibitory
pathway so in order to prevent this the virus produces a second protein that binds to p53 so there is no
cell cycle arrest. This causes the cell to be triggered into S-phase, this allows the virus to produce its
DNA. If these proteins, after viral infections, remain present they result in deregulation of the cell
because Rb and p53 are tumor suppressor genes and only the loss of these is not enough to develop
cancer, also activation of the oncogene is needed. Later in life, you do have an increased chance to get
cancer if you are still carrying the proteins around.

Several viruses do this;

o HPV: produces E7 that binds P53 and E6 that binds Rb
o Adenovirus: produces E1B that binds P53 and E1A that binds Rb
o SV40: produces a single protein (a large T antigen) that can bind both Rb and p53

Targeting of the p53 pathway

For tumor maintenance you need sustained p53 inactivation. The loss of p53 is associated with the
oncogenesis, it’s one of the steps you need for cancer to start, but it’s also needed to maintain the
tumor. Reactivating p53 can stop the tumor growth.
This means activation of p53 can be a therapeutic approach to treat cancers. This can be done in
different ways;
o p53 gene therapy: introducing a functional copy of the p53 gene. This idea is simple but
difficult to make it work because it is difficult to deliver a gene to every cancer cell. Even if
you get it in 90%, the other 10% will still proliferate so this only works of the p53 activity is
restored in every cancer cell and this is impossible.
o Restore aberrant protein conformation: if there is mutant p53 that is folded different which
makes it hard to bind the target genes, p53 can be modified/refolded into its proper structure
using small molecules PRIMA-1 or APR-246. So, the mutant p53 is restored to it can act as
wildtype p53.
o Oncolytic adenovirus: the adenovirus makes proteins that bind to p53 and Rb. You can make a
mutant virus that lacks the gene for this protein that binds to p53. This virus then starts to
replicate, only when p53 is already deficient in these cells. If this virus enters a healthy cell,
there is p53 activity and the virus is incapable of inactivating p53 so p53 will inhibit the DNA
synthesis so there will be no virus replication. when the virus enters a cancers cell there is
already inactivated p53, the virus can replicate and kill the cell. So, the virus is selectively
replicating in cancer cell. What the book says is not entirely correct. The virus makes use of
the fact that the cancer cell p53 deficient, in order to only kill the cancer cells. Besides this, the
virus has other reasons for killing cancers cells, not following this mechanism.
Adenovirus expresses E1A and E1B proteins, when this infects a normal cell it will induce
DNA synthesis by inhibiting Rb and p53.
An oncolytic virus that lacks E1B can only inhibit Rb and not p53 anymore so hardly any
DNA synthesis is possible and it will not replicate. However, in a cancer cell that already does
not have p53, E1B is not needed so only inhibiting Rb already drives the DNA synthesis and
the virus can replicate. In this process of virus replication it will in the end kill the cell so new
virus particles are released that can go find new cancer cells to replicate this process. However,
here is a limitation, these viruses are immunogenic so in two weeks the process stops because
the immune system kicks out the virus, so it’s a temporary treatment.

Strategies that aim to active endogenous p53

This only works if you have endogenous wildtype p53, so p53 is not mutated but its inhibited by
another protein that binds to it and degrades it. For this, molecules called Nutlins are made that bind to
MDM2 on the place that normally p53 would bind. MDM2 cannot bind p53 anymore, this releases
p53 and reactivates it also in presence of a lot of MDM2.

Strategies that aim to suppress endogenous p53

This is not a treatment based on p53 deficiency in cancer cells but this is based on inhibiting the
wildtype p53 in your healthy cells. Chemotherapy and radiotherapy cause side effects. Part of these
side effects on the normal tissue are mediated by p53. So, by using Pifithrin-a, an inhibitor of the
production of p53, the production of p53 is temporarily inhibited in the healthy cells. (This also
happens in the cancer cells but these already p53 deficient so it does not change anything). So,
Pifithrin-a is given before treatment with chemo- or radiotherapy to diminish side effects.

Molecular What does a clinical scientist in molecular pathology do?

Diagnostics in o Responsible for using
Pathology
Molecular
Diagnostics in
pathology-NGS-
based mutation
In the molecular diagnostic lab tests are used to see changes in nucleotides. There is qualitative and
quantitative testing and on DNA as well as on RNA level. So, there are many different kinds of tests.

Molecular diagnostics is done to get more knowledge about the molecular defects involved in the
pathogenesis of diseases. It is contributing to;
o Early detection using screening
o Tumor classification
o Prognostic marker
o Predictive marker
o Disease/treatment monitoring marker

DNA mutations in cancer

Targeting agents have been developed against upregulated proteins due to DNA mutations. For
example, lung cancer can be split up in squamous cell carcinoma and adenocarcinoma. Looking at
molecular profiling you see these can be subdivide on many more subtypes, based on the presence on
certain mutations. Against these mutations, specific types of treatments are developed.

Why molecular diagnostics in pathology

Molecular methods can provide very important extra information from cells, tissues and liquid
biopsies. Liquid biopsies are used when patients are old, have stress, or it is not possible to become
close to the tumor.

Molecular diagnostic flow

First there is tissue collection. This will be checked by the pathologist. When performing a test it is
important the tissue contains enough cancer cells, so the pathologist looks at morphology, selects the
tumor area and there the technicians isolate the tumor cells. This is done mechanically with a needle
under the microscope, this improves the number of tumor cells in the sample. Technicians are busy
doing administration and check quality. We are doing for instance a NGS analysis which provides a
lot of data. The bioinformatic department obtains the data and looks together with the pathologist if it
is a important tumor and gives a final conclusion for the clinic.

Test used in molecular diagnostics

o NGS: some labs have one big panel with 500 genes that are tested with NGS sequencing and
look for specific genes based on what they think are present in the tumor. In this lab there are
custom made panels, specific for all kind of tumors.
Besides mutation analysis in tissues, also liquid biopsies are done. Besides this, we look for
fusions for translocation because patients who get resistance to a therapy can have a
translocation. Also, translocation can help in diagnosis of the tumor.
o MLPA: promotor methylation and copy number variation
o Fragment analysis: clonality and differences in tandem repeats
o FISH
o HPV
o sWGS: to look for copy number variations

NGS
o DNA mutation analysis: there are custom made panels have hot spot mutations specific for
certain cancers. So, there are panels for different cancers and there are some specific mutations
for diagnosis.
For example:
o MYD88 in Waldenstrom disease
o BRAF and TERT in melanoma
 o EGFR in lung cancer
o DNA mutation analysis liquid biopsy: this is especially in lung cancer. For example a patients
with a EGFR deletion is treated with tyrosine-kinase inhibitors. Then you want to follow up to
see whether the patient is responding. In a few months you get blood and check for the EGFR
deletion. In this way you can see if the patient is responding to the therapy.
o Archer fusioplex analysis: the detection of fusions/translocations. This is not only important
for diagnosis but also for therapy. This is at RNA-level. For example in lung cancer you can
have a ETV6-NTRK3 fusion which leads to upregulation of the NTRK3 protein. an NTRK
inhibitor is developed. So, patients can be treated based on the profile of fusion.

NGS workflow
Is about 5 days.
Day 1: Material of patients collected, slides of the tissue made, tumor field selected, hematoxylin
eosin staying performed. Together with the pathologist the technicians is looking where cancer cells
are present. Then, the technicians will then under the microscope select the specific area with the
tumor cells.
Day 2: DNA isolated, library is made, formation of the template
Day 3: sequencing
Day 4: analyze the data
Day 5: report which kind of mutation is present
Then it is also important to look whether the mutation is pathogenic, benign etc, This is not mentioned
to the clinic, the clinic only gets results of the pathogenic variance that is observed.

Panels
So, we have all kind of panels. The AmsterdamUMC unco panel is for when a patients has no
treatment options anymore. All genes that have targeting agents are in that panel. Most of the time,
this is the last option for a patient.

Oncogene/tumor suppressor gene

U should know whether the gene you found is an oncogene or tumor suppressor gene.
Oncogene: can cause cancer, overactivity. Only one copy has to have the mutation gene gain function
and activate certain pathways.
Tumor suppressor gene: inhibits, there is loss of the protein, underactivity due to mutation. A second
mutation is needed for loss of function.

In oncogenes only one of the copies needs the mutation, this already leads to gain of function and
overexpression of the protein.
In tumor suppressor genes one mutation does not yet have a effect on the protein. For inactivation a
second mutation is needed.

Missense vs nonsense mutation

Missense: point mutation, change in the codon which causes it to code for a different amino acid.
Nonsense: knock out of the function of the gene, point mutation that leads to stopcodons. This is
especially found in tumor suppressor genes.

Nomenclature
DNA:
First a c for coding DNA
2573 is position, where in the DNA the change is
T for nucleotide that was already present
G for what it is changed into
So, at position 2573 from a T into a G

Protein:
First a p for protein
First amino acid that it was used to be
Position of amino acid
Amino acid where it is changed into

Variant allele frequency

The Variant allele frequency (VAF) is the frequency of an allele, how many tumor cells have the
mutation, in a specific.
VAF = the relative frequency of an allele on a specific locus in a specific population expressed as a
fraction of a percentage.
KRAS is oncogene so only 1 mutation is needed. So, when tumor percentage is 70%, we see most of
the time that KRAS is present in 34% of tumor cells so VAF is 34%.
TP53 is tumor suppressor gene so two hits are needed so VAF is doubled.

You can have all kind of changes.

Substation; change where you compare to a reference sequence, one nucleotide is replaced by another
Deletion; one or more nucleotides are not present. If it is one nucleotide is missing you have a change
of the reading frame and this most of the time results in nonsense mutation.
Insertion; one or more nucleotides is inserted. This is not a copy added on the 5’end side that’s
duplication.
Splicesite mutation; normal splicing pattern is changed due to mutation in the intron close to the exon.

Multiplex ligation-dependent probe amplification (MLPA)

You are looking for promotor methylation and copy number variations.
Promotor methylation: for example used for mismatch repair deficiency, then DNA repair is not
working anymore. This can be caused by one of the important mismatch repair proteins MLH1, when
this gene is methylated.
CNV variations: for example to see of there are gains/loses of chromosomes, break points in the
chromosomes and the number of CNV.

Fragment analysis
o B/T cell clonality example: fragment analysis is looking at for instance lymphomas. Most of
the time these grow from one cancer stem cell. Then you can look and B- and T-cell receptor
rearrangement because if a lymphoma is present then only one cancer stem cell is growing out
with the same B- or T-cell receptor rearrangement. Then with fragment analysis can be
detected whether there is one clone present, an indication for the presence of a lymphoma, or if
it is a whole subsets of B- or T-cells with all kinds of different clones.
o Powerplex analysis: every person has tandem repeats. These are different in size and number.
15 places in the genome are selected that are unique for every person so this is some kind of
passport. When it is not clear whether a tumor is from a donor kidney or from a metastases that
the patient already has from itself we can look at these tandem repeats.

FISH analysis
Detection of fusions/translocation in tissue. In fusion, the signal is split up so you see red and green
instead of just yellow.

Introduction to Practical Assignment Molecular diagnostics in

pathology
EGFR
Wilde type: ligand binding to egf-receptor leads to receptor dimerization ,formatuin of atp binding sit
at intracellular tyrosine kinase domein, which then leads to phosphorylation of the intracellular part of
the receptor and activation of downstream signalling pathways which leads to proliferation and
migration.
Mutation in tyrosin kinase domains lead to conformational changes near atp binding sit, causing
receptor dimerization independent of ligand binding zso leading to consistent activation of
downsignalling oathways leading to uncontrolled cell proliferation.
These mutations not only stabilize interaction with atp but also confere a sensitivity to small molecule
TKIs. EGFR-TKIs compete with atp for binding thereby blocking the intracellular cascade and
downsignalling. Treatment with this improves progression survival of pople with EGFR mutation
positive patients than normal chemotherapy.
The tyrosine kinase domain Is mainly transcribed from exons 18 till 21. Thre are a number of
mutations in these exons which caus the constent activation of the EGFR pathway. All these mutation
are associated with sensitivity to EGFR-TKI treatment.
Tumor cells try to invade the treatment by ontroducing drug resistant mutation, mainly in exon 19 and
20.
EGFR plays a role in several types of cancer.
It can also be inhibited by using anti- EGFR mAb which target the extracellular part of the receptor
thereby blocking ligand dependent signal transduction. This is used in colorectal cancer. For selection
of this therapy, KRAS can be used as a predictive marker. KRAS is located to intracellular surface of
the cell membrane. ligand binding to EGFR induces dimerization of the receptor, autophosporylation
of intracellular domain and transient activation of KRAS. Axctivatin mutation in KRAS bypass the
need for ligand binding, can resut in constituative activity of downstream singanlling pathways,
independent of ligand bindng. Activating KRAS mutation support cancer cell survival, independent of
EGFR inhibition.
Study: treated with standard of care (yellow) and with ab in combination with standard care (blue). In
wildtype the progressive free surval increased when using combination thrapy while in mutant there is
no difference in survival.
NGS data
At the top in green you see the reference seuqnce, the sequence of the xon in an healthy individual. In
light green is sequence that comes out of the tumor DNA. Here you see a single nucelotde change
from a C to a G.
Also excel file with all sequenced genes and notes when a variance is found. The excel is build up of
many columns;
Amplicon; name of sequeced region. Multiple amplicaons may cover one gene because it exits to 100
bp.
Depth; the amount of sequence reads generated from the amplicon. Should be over 100 to ensure you
have accurate data.
QUAL; qualtity score for identified variance to make sure you have accurate data it should be over
900
Codon change; nucleotide change
Aminoacid change; predicted aminoacid change, predicted cause we can only make sure what the
codon change is because we sequence on DNA level
VAF; Proportion of mutation-containning reads compared to toal number of reeds.
Chrom; chromosomal position.

Base pare transitions that are found by ngs also conatin germline matations (snp’s) and artefacts.
1 select amplicons with a reported variant.
2 from this list of variances, whe only interestend in ones who are clinically relevant so associated
with cancer and predictive for treatment response. So, all common data is removed.
SNPs; positions in the genome where some individuals have a different nucelotides. There are many
forms. Snps can cause both a transition in the amino acid but multiple codons code for the same
amnioacid so it is possible the mani acid remains the same.
Snps are identified by looking at the amnioacid change. You look them up in database by entering the
id of the variance

3 there re sequence artefacts because DNA is derived for FFPE tissues because this preserves the
architecture of the tissue and makes sure composition of the cell is rserved.
It may lead to sequencing artefacts.
Formaldehyde, part of pormalin, crosslinks the DNA which results in DNA fragmentation and it
causes deamination of the C-nucleotide which leads to artifactual changes from a C T on the sense
strand and G to A on the anti-sense strand.
So, formaldehyde is highly reactive with DNA bases and proteins so generates damage to the DNA.
Next to the artefacts, the NGS technique itself can also cause errors in the DNA sequence so there is a
risk of finding false positives.
Important to look at quality score, this should be over 990. Also look at VAF which should be over
5%.
4 in remaining variances you should check for cancer associations and clinical relevance for drugs.
p.notation is easy.. For c.notation of single nucleotide change you have to use the coding DNA
sequencing and navigate through it.
c-1 starts at A of ATG translation inition codon.