The cancer genome Flashcards
What needs to be known to understand the gene changes in cancer
Cancer is due to alteration of the genes in a clone of cells, so must be able to identify the genes concerned and how they’re altered. Need to ask - how many genes are altered? which genes? what do the genes do in the cell? in what way are they altered? how does the alteration of the gene contribute to cancer development
Which genes are commonly affected and what do they do in the cell?
a) control of transcription
b) chromatin modifiers
a) i) Tcf proteins are targets of the Wnt pathway and are occasionally mutated in colon cancer instead of APC of β-catenin ii) MYC is an oncogene and transcription factor that seems to powerfully upregulate many genes involved in proliferation and is widely activated iii) ERG is a transcription factor that may control differentiation and is fused in prostate cancer
b) Include proteins that modify histones to turn genes on and off eg i) MLL is a histone methylase that is fused to other genes by chromosome translocations in leukaemias ii) chromatin is also modified by protein complexes, such as the BAF complex, that move nucleosomes. Several components of these complexes are mutated eg ARIB1A. Overall, maybe 20% of human cancers have mutations in BAF components.
Which genes are commonly affected and what do they do in the cell?
a) Cell adhesion molecules
b) Carbohydrate metabolism
a) eg E-cadherin, which is inactivated in some breast cancers, often by epigenetic change (DNA methylation)
b) Mutations in isocitrate dehydrogenase IDH1 or IDH2 (an enzyme in the Krebs cycle) are detected in some brain tumours and leukaemias. The mutation changes the enzyme’s specificity to make hydroxyglutarate, which accumulates and interferes with DNA demethylation (which may block differentiation) and histone acetylation (which may block DNA damage signalling)
In what way are genes altered?
a) overview of why this question is important
b) Sequence-level changes
c) structural changes
a) ‘mutation’ covers and DNA sequence change, from changes to one or a few bp, up to large chunks of DNA being swapped, deleted or inserted - the most dramatic being chromosome translocation, in which pieces of two chromosomes are joined to eachother following breakage of the DNA or an accident during DNA replication
b) Include those to single bp, and ‘indels’ (In - small insertions) or deletions (del) of one or more bp. Indels usually result in the transcription and translation of truncated proteins because most of them cause frameshifts that usually lead to a stop codon.
c) these are many and varied, including deletions, duplications and amplifications as well as gene fusions at translocation breakpoints. Amplifications - a cell acquires a large number of copies of an oncogene by repeated duplication of a segment of the genome. Gene fusion - rearrangement of large chunks of DNA can create a new gene by joining two genes together to create a gene fusion, potentially the most powerful kind of cancer mutation. Chromosome translocations are famous for doing this, particularly in leukaemias where they were first discorvered.
How are mutations studied?
a) overview
b) Structural DNA rearrangements
a) sequencing of DNA used to be done by PCR, amplifying a region of interest such as an exon and sequencing it on a Sanger sequencer. This has been replaced by Illumina sequencing in which millions of DNA fragments, are attacged to a glass slide and sequenced simultaneously
b) Identified by finding sequence fragments that cross rearrangement junctions, eg for the BCR-ABL fusion, finding a DNA fragment with sequence from the BCR gene at one end, and sequence from the ABL gene at the other. However, there is no way to distinguish driver from passenger rearrangements. Rearrangements, particularly chromosome translocations in leukaemias, were first found by classical cytogenetics (the study of chromosomes by microscopy) and this remains an important clinical tool: cells are arrested in metaphase and their chromosomes spread on a slide and stained. This can show chromosome translocations and some large deletions or inversions.
How are mutations studied?
a) FISH
b) Copy number measurement
a) Fluorescence in situ hybridization - cytogenetics can be enhanced by FISH whereby DNA is labelled with fluorescence and hybridised to metaphase chromosomes. DNA from a chromosome can be labelled, or a small segment of the genome can be labelled. FISH is uced clinically to detect amplification of HER2 (ERBB2), the target of Herceptin therapy in breast cancer. Labelling each chromosome in a different colour gives a rough karyotype showing many translocations
b) large deletions and amplifications have mainly been detected by measuring the relative loss or increase in the amount of DNA in different regions of the genome. For example (see image), y-axis shows the relative amount of tumour DNA on a log to base scale 2, with absolute number of copies marked in dark red - this tumour shows amplification of the EGF receptor (green line) and deletion of CDKN2A (red line). The amount of DNA in each region of the genome is measured by sequencing the whole genome and counting the number of DNA fragments in each bit of the genome.
Examples of genes mutated by point mutation - eg signalling pathway downstream of tyrosine kinases like ERBB/HER-1 and ERBB2/HER2
a) RAS family
b) BRAF
c) PIK3CA
a) Activation of the RAS family is a classic example of an activating point mutation. RAS is a G-protein that is activated by GTP binding and inactivated by hydrolysis of GTP to GDP. Mutations block hydrolysis by preventing GTPase-activating proteins that complete the active site. L-RAS in particular is frequently mutated in a range of human cancers
b) often aso activated in cancers that have mutations in the RAS-MAP kinase pathway, notably melanomas (60% cases) and colorectal carcinomas. BRAF and RAS mutations seem to be alternatives with rather similar effects on teh cell. The commonest mutation changes valine to a glutamic acid at position 600. This is adjacent to an activating phosphorylation site, and the negative charge of the glutamic acid presumably mimics phosphorylation and so activates the protein
c) mutated in about 1/3 breast cancers, the most frequent mutation yet. One of the most commonly observed mutation changes one particular negatively charged aa to a positively charged one, a dramatic change
Examples of genes mutated by point mutation - eg signalling pathway downstream of tyrosine kinases like ERBB/HER-1 and ERBB2/HER2
How to these examples show how a point mutation can activate a gene/protein
Mutations may stabilise the protein by blocking degradation, stabilising an active conformation (like in RAS), or mimic an activating phosphorylation event (BRAF).
Such oncogene-activating mutations have to be precise, occuring at particular aas. In contrast, inactivating tumour suppressor genes is much earier - anything that interferes with the protein’s function suffices. eg PTEN (reverses the action of PIK3CA) is a known tumour suppressor gene and is generally affected by inactivating point mutations or deletions. Less common are activating mutations of AKT and the kinases ERBB and ERBB2 themselves.
Indels are though to be particularly important as inactivators of tumour suppressors as they usually trunctate, with many APC mutations being indels.
Examples of genes affected by structural mutations
a) chromosome translocations
b) deletion
c) duplication
d) amplified genes
a) There are many chromosome translocations in leukaemias. BCR-ABL is the oconic fusion found in most cases of chronic myeloid leukaemia, formed by he reciprocal t(9:22) translocation. ABL is a tyrosine kinase, normally held in check by a regulatory domain at the N terminus and is activated by dimerisation. The N terminus of the BCR protein naturally forms dimers or oligomerises. The fusion does several things - i) the inhibitory N terminus of ABL is removed ii) BCR holds the ABL kinase in dimers/oligomers which activates it iii) the BCR promotor may be stronger than that of ABL
b) Fusion gene in a carcinoma is TMPRSS2-ERG, found in ~50% of prostate cancers, and is the most prevalent known fusion gene (as prostate cancer so common). Formed by deletion between the two genes TMPSSR2 and ERG, rather than by chromosome translocation. PTEN is also often deleted, along with Rb and p16/INK4A.
c) Tandem duplicatoon of a segment of DNA causes a fusion of BRAF to a neighbouring gene KIAA1549. In paediatric brain tumours
d) Amplified genes include the EGFR/ERBB/HER1 in brain cancers and ERBB2/HER2 in breast cancers (10-20%). Also CyclinD1 in breast cancer and MYC family gene in various cancers.
Other kinds of gene change
a) epigenetics
b) viruses
a) Some important examples of gener inactivated by DNA methylation: MLH1 (involved in mismatch repiar) and E-cadherin (involved in cell adhesion) in colon cancers and lobular breast cancers respectively. Some cancers, including 30-50% colon cancers, appear to show ‘epigenetic instability’ analogous to genetic instability (a high rate of aberrant DNA methylation)
b) Some viruses like HPV bring genes into cells. Several viruses have proteins that inactivate p53 and RB1 in this manner
a) how many genes are altered in cancer
b) are all mutations productive
a) Not certain, but adult cancers have over 10. The ‘Hallmarks’ idea suggest a reasonably large number. The current Vogelstein model includes 6 point mutations, and it’s incomplete. Estimated that there are about 8 mutations per tumour in breast and ovarian cancer, 75% of which are rearrangements
b) Current sequencing shows thousands of mutations in the cancer genome, but most are irrelevant, random ‘passenger’ mutations that accumulated in between the key ‘driver’ mutations (those that give the cell a selective advantage). One way to estimated the proportion of driver mutations is to compare the relative proportions of mutations that do or do not alter amino acids, as most that don’t will be random.
How do we find out what mutating the genes does to the cells and tissues
a) In vitro systems
b) Animal models examples
a) Often, in vitro cell line models are first assessed for the impact of gene mutations, but these are generally already transformed and so don’t represent the human scenario. Immortalised cell lines can alternatively be used, such as fibroblasts which on expression of an oncogene begin to outgrow their wild-type neighbours losing ‘contact inhibition’, forming layers of cells growing on top of eachother. However, this model doesn’t account for the effect of whole-body systems on the dynamics of tumour growth, so animal models are employed to dissect apart the roles of various mutant proteins
b) 1) the genomic region around CCND1 gene (encodes CyclinD1) is amplified in many breast cancers, but this doesn’t prove CCND1 is an oncogene as the neighbouring gene could be the actual problem. A transgenic mouse was made that expresses CyclinD1 from multiple copies of its encoding gene, driven by the MMTV promoter, which is specifically active in mammary glands. These mice develop mammary hyperplasias that develop into tumours, so CCND1 is the oncogene
2) Inactivation of APC in the crypts of the colon in mice is was discovered as removing APC profoundly alters the differentiation pattern of the crypt, so that the dividing cell compartment expand and the cells are prevented from migrating up the villus and maturing. To delete APC expression in cells of the crypt, a mouse expressing a Cre recombinase driven by a crypt-specific promoter is backcrossed to a mouse in which the APC gene is flanked by loxP sites. Recombination of the loxP sites by Cre recombinase in the cells of interest leads to deletion of the APC gene.
