The success of multi-cellular life depends on most individual cells agreeing to limit
their reproductive capacity; however, sometimes some cells break this
reproductively repressive agreement and go their own way. We call this cancer.
We don’t like it!
1. four most common types of cancer in Canada
a. Breast cancer
b. Colon cancer
c. Lung cancer
d. Prostate cancer
2. likely factors contributing to cancer incidence in Canada
Note: men are at higher risk cancer than women (45% vs 40 %)!
Rates among youth are rising
high environmental component
rapid cell division during embryo synthesis
b. low heritability (0.27-0.42)
Embryogenesis rapidly diving cell:
Embryogenesis of volvox is simpler but giraffe is more complicated, however, their cells both divide
Describe the roles of checkpoints, cyclin, Cdk, and MPF in the cell cycle control system?
-> MPF is a complex of cyclin and Cdk. cyclin levels rise and fall throughout the cell cycle; cdk levels
remain constant. when a checkpoint approaches, cyclin levels rise until there is enough cyclin to make
many cyclin-cdk complexes (MPF). this rise in MPF triggers the cell to pass through the checkpoints
(providing nothing blocks progression, e.g. unrepaired DNA damage) to the next stage of the cell cycle.
different cyclins are involved in progression through different checkpoints, e.g cyclins D, B, E and H.
3. role of cyclin/CDK complexes in cell cycle regulation
a. G1 to S check point- prevents cell from replicating their DNA if its damaged. It will
resume until everything is repaired. this check point is monitored by CDK (Cyclin
Dependent Kinase) complex.
b. CDK: complexes that are master control switches from cell checkpoints
c. CDK phosphorylate other target proteins.
d. they are produced all the time (constitutively)
e. they are only active if they bound to their respective cyclin proteins
f. this drives the cell cycle through cell checkpoints. g. CYCLE:
i. G1/S cyclin binds to CDK2 (cyclin)
ii. G1/S-CDK2 complex activates CDK2 which phosphorylates target proteins to
G1/S transition (post translation regulation)- commits cells to DNA replication
iii. Cyclin then is degraded
iv. then another cycle of cyclin
v. S cyclin binds to CDK2 which initiates DNA replication and progression thru S
vi. Cyclin then is degraded
vii. then another cycle of cyclin
viii. M cyclin binds to CDK1. it activates CDK1 and phosphorylates target proteins for
G2/M transition and progression thru M
ix. Cyclin then is degraded
4. role of proto-oncogenes, tumor suppressor genes and oncomirs in caner
a. Expression of proto-oncogene promotes cell cycling.
i. cells need to be responsive to environment, they need to know when to divide
ii. EGFR-Epidermal Growth Factor/protein Receptor is one of proteins that does
that. it is a trans membrane domain that has a receptor on the outside and a
transmitter on the inside.
iii. So when there is EGF in envi, it will bind on this receptor (EGFR). there is an
internal signal domain and membrane domain.
iv. Tumor: EGFR is on all the time - divide all the time - mutation.
v. Oncogene is not a cancer gene, it is a normal gene, it is embryo gene, it is
needed for cell division. when they mutate, it can become tumour cell.
vi. before they activate it - it is called proto - oncogenes
vii. All gene can be tumor gene if they are deregulated. no wonder so many possible
ways to have cancer.
b. Expression of Tumor suppressor genes slows cell cycling. (p.53)
i. Shouldn’t be called tumor suppressors, they should be called embryo
ii. They tend to slow down/shut down cell division.
iii. TP53 is the master tumor suppressor, coding a transcription factor whose
activity can result in:
1. Recognize DNA mutation and call for repair (increased DNA repair)
2. Blocks the cyclin/CDK at first checkpoint (G1-S check point). it prevents
that cell from passing G1 S checkpoint if they have too much DNA
3. Tells the cell to die-apoptosis
c. Inappropriate expression of miRNA can promote cycling (e.g. oncomirs) i. microRNA turns out very integral in cell division and regulation, researcher call
them oncomir: cancer causing RNA.
ii. Normal tissue: micro RNA tends to have high expression
iii. Tumor tissue: micro RNA tends to have a lower expression
1. There is a sharp extinction in microRNA expression
2. miRNA expression profiles classify human cancers....diagnostic for
particular tumours "mRNA is where the action is"
3. Heat map: blue=mRNA under expressed (a lot in tumour)
red= mRNA over expressed ( a lot in normal tissue)
4. each tumour has its own finger print
5. role of p53 gene
a. a gene that codes for a protein that regulates the cell cycle and hence functions as a
6. explanation for why increased cancer risk can be inherited
a. BRCA1 - normal cell growth from mom and dad. it can divide out of control over time.
but it is RARE
b. BRCA1 mutation from dad (Familial cancer requires loss of function mutation in one
allele) and normal BRCA 1 from mom; you still develop tumour (higher risk)
7. explanation for why cancer incidence tends to increase with age
a. cancer is progressive - single mutation do not cause cancer. cancer requires multiple
mutation over time. this is why at 50 years old, the medical community stresses cancer
8. role of stem cells in tumor growth
a. cancer may begin as alteration to gene expression in stem cells
i. 3 hypotheses of how a cancer stem cell may arise:
(1) A stem cell undergoes a mutation,
(2) A progenitor cell undergoes two or more mutations, or
(3) A fully differentiated cell undergoes several mutations that drive it back to a
In all 3 scenarios, the resultant cancer stem cell has lost the ability to regulate its
own cell division.
ii. pleuri potent is able to differentiate into a wide range of tissue type. they know
how to be different kind of tissue. stem cell niche -> progenitor cells -> fully
differentiated cell ---------> cancer stem cell.
9. Evidence that epigenetic regulation may be relevant in cancer
Some tumour cell nuclei can be re-programmed!
these mice are heterozygous for a tumor suppressor gene along their life, they suffer a mutation in the normal allele... now they have no functioning
tumour suppressor. lost a tumour suppressor gene.
the normal cells in mouse are (ts+/ts-) while in the tumour the cells are (ts-/ts-)
you can take the nucleus out of the egg cell and replace the nucleus from tumour cells.
if you give this egg cell a little shock, it has been fertilized. then it starts dividing out of control
if you implant this egg into a mouse it will develop an embryo.
if you look at section of the skin .... it is normal! no tumour
this is because the eggs cell "reprograms" the tumour nucleus
even though it got this mutation, you see normal development
this nucleus is being epigenetically re-programmed.
Cancer is deregulation - uncontrolled growth can arise from upsetting the balance between the activity
of gene products that promote cell cycling vs. those products that suppress cell cycling.
HPV causes cancer, women needs a vaccine to prevent HPV. Lecture 19
Determining if two structures are homologous is of central importance to
evolutionary biology – how does one prove homology?
1. Strategies for determining if features are homologous?
Note: it is the same gene in Volvox and in Chlamy. this gene involved in asymmetrical
cell division in volvox is homologous to the one to Chlamy.
a. sequence genomes: compare genome because they are the same
b. genome annotation: a bunch of A, T, C, G.
we want to attach biological meaning to sequence:
gene prediction (is it code for protein or RNA gene?), regulatory elements, biological
function through similarity searches, automated.
c. protein-coding gene prediction: 15 thousand prediction. GlsA codes for protein.
protein coding gene: computer algorithms, detect promoter elements, intron/exon
boundaries, other conserved DNA motifs. But here I do not know the orientation.
For this sequence, there are 6 possible reading frames. so out of 6 which one is the most
probable? you can do this by finding the longest ORF (the largest number of amino acid
from the start to the stop)
d. align sequences: compare GlsA sequences and look for the similar sequences
you can look at the similar sequences (KEY: it can suggest the structure and funtional
similarity as well as evolutionary relatedness)
i. you can do the sequence alignment (DNA or protein)
ii. Arrange sequences to show regions of similarity.
one of those 1500 sequence is similar to volvox sequence by arranging the
sequencing using the computer.
iii. several method: BLAST (local), CLUSTAL (global)
1. Global vs. Local Alignment
2. CLUSTAL looks for the global similarities. they start from the beginning
of the protein coding gene( the right sequences) and try to align them. if
they are not align you will create a gap.
3. BLAST looks for the small region of high similarities and align them;
unlike CLUSTAL, it doesnt try to force two sequences to align perfectly.
it is common and faster.
4. Example when LOCAL is strong but the GLOBAL is weak???
e. determine homology
2. Sequences detected by annotation programs to detect open reading frames (ORF)
a. For this sequence, there are 6 possible reading frames. so out of 6 which one is the most
probable? you can do this by finding the longest ORF (the largest number of amino acid
from the start to the stop) 3. Characteristics that are, and are not, common between homologous genes
a. the sequences are not the same. there are fair amount of similarities in sequences, but
not the identical nucleotide sequences. (similar not identical)
b. not have identical protein sequence, but high similarities because their DNA sequences
c. the amino acid-protein is not the same length
d. not the same funstion - just high level of similarities
4. Usefulness of BLAST analysis of sequences in Genbank at NCBI
a. BLAST looks for the small region of high similarities and align them; unlike CLUSTAL, it
doesnt try to force two sequences to align perfectly. it is common and faster.
b. it is found the small region that has high similarities in chlamy and volvox eg GlsA
5. Reasons why amino acid sequence comparisons are more informative than nucleotide sequence
a. more information in an amino acid sequence of same length. it is found that there are
4.32 bits in a single amino acid..
b. the genetic code is redundant:
i. replace proteins with different word. you have 64 possible triplets, but you only
have 20 possible amino acids. thus more than one codon can specify the same
ii. amino acid sequence more highly conserved because
1) redundancy in DNA allows for more than one codon to code for an amino
2) protein interactions with substrates are very specific, even small changes can
result in a loss of function to the protein and can lead to the organism's death,
so these changes are mostly selected against
3) differences observed at the DNA level often do not appear at the protein
c. DNA databases are much larger. DNA sequences have junk DNA, repetitive junk DNA
which is hard to search. but the predicted protein database is more specific, no protein is a
junk protein, they are all important.
in the picture, there are 10 nucleotide differences in here, a lot of polymorphism, mutation
that accounts for these differences. this is why their sequences are not identical.
there are 10 mismatches, BUT when we translate, one of the strands codes for the same
exact section of protein.
there are a lot of difference in the level of nucleotide, it just does not express in the level
of protein 6. Mathematical relationship among total information, # of symbols, #letters in the alphabet
a. l is the total information in a message with G (how long the nucleotide is )symbols
written in an alphabet of n(how many characters in this alphabet) letters.
I = (G ln n)
b. How many bits of information in a single amino acid?
n=20, G=1, I=?
7. Relative number of bits of information in a single nucleotide vs. single amino acid
a. 4 letters/nucleotide: A00 G01 C10 T11
8. Relationship between E-value and likelihood homology
Homology determination is based on probability.
a conclusion that two or more genes are homologous is not proven experimentally. it is because
we do not know the MRCA.
a. decision about homology are made based on similarity-numerical and correlated with
b. the higher the similarity between two sequences, the lower the probably that they
originated independently of each other and became similar just by chance.
c. the lower the E-value, the less likely that they didn't share MRCA/the greater the
9. What’s the information content in a sequence that consist of a single amino acids
10. Think about underlying reasons to explain that two proteins have weak global aligment
(CLUSTAL) but strong local alignment (BLAST)
Note: molecular homology
-molecular evolution: study of evolution at the level of nucleic acid and amino acid.
-gene evolution: how the gene change over evolutionary time
mutation-> duplication (novel function)->rearrangement(novel funtion)->loss
all of these will affect the phenotype, but these changes can have no effect
these phenotype subject to SELECTION. but selection acts on phenotype NOT gene!
Homology means similar. Bad definition!
Homology means sharing common ancestor. the flipper of dolphin and the wing of the bat are
1. if GlsA in Volvox and GlsA in Chlamy are homologous - what do they have in common?
-> it's NOT identical nucleotide seq, it's not identical amino acid seq, it's not identical length of
polypeptide, and it is not because of their same function. Lecture 20
Proteins do not have to homologous to have similarities in structure/function
1. Synonymous vs. nonsynonymous mutations
a. synonymous mutation= change in the sequence in nucleic acid but no change in amino
acid. (NEUTRAL mutation).
b. non synonymous mutation= change in the sequence in nucleic acid and also change the
amino acid ex. from arg to sir.
Since selection acts on the phenotype, it got to affect the amino acid
2. Characteristics of the neutral theory of molecular evolution
a. Selection theory vs neutral theory
b. in selection theory: all mutation would affect fitness. most mutation may be deleterious
(harmful) and be selected against out of population. small will be advantageous.
c. in neutral theory: many mutations have no effect in amino acid. few will be
advantageous but mostly neutral.
3. Relationship between frequency of amino acid substitution in given proteins vs. time since
a. The number of differences between protein sequences of different species are roughly
proportional to the time since those species diverged.
b. proportional means if they diverge long time ago then you expect more differences
amino acid (ex yeast and human) and if they diverge less time 5 millions years then you
expect fewer amino acid differences) (ex. chimp and human)
c. they are linear graph. change of the number of AA substitution pers 100 residues vs.
time since divergence = rate
d. Mutation occurs at the constant rate and it's called molecular clock.
4. Relative rates of accumulation of synonymous vs non-synonymous mutations
a. the rate of change of synonymous-silent substitution is faster (slope is higher) that the
rate of nonsynonymous-replacement substitution
5. why the rates between proteins (histone vs alpha globin) are different?
a. histone is more sensitive to changes (highly constraint). if you changes the protein, it
will affect the structure and the function.
b. alpha globin is not so sensitive to changes (weakly constraint). no effect on the function.
you dont need to maintain 3D structure. 6. Deduce time of divergence given number of amino acid changes in particular protein
a. if the rate of the mutation is the same, the molecular clock is the same, then we can use
the number of the differences which occurs in evolutionary time to figure out the time
of divergence given number of amino acids changes in particular protein.
7. Characteristics of the “molecular” clock
a. phonotype can be similar regarding the homology and regarding the convergence.
8. Regions of two unrelated proteins that would be expected to be similar if they were the
products of convergent evolution
why is similarity between sequences considered to be evidence of homology?
phenotype can be similar because of homology and also because of convergence.
a. there is no reason for two proteins to share high identity across their entire sequence
b. you would not expect if two proteins were distantly related yet converge and share
similar structure or function phenotype. BUT you would expect "localized region of
similar" ex. catalyzing a similar reaction, binding to certain region of DNA which is
similar or receptor binding sites, location of cycteins (S-S bonding). but you cannot
expect the entire protein to be similar.
c. the entire sequence from beginning to end is highly similar due to homology not based
9. Function of lysozyme
a. a small enzyme- antibacterial activity