Gene mutations and other genetic changes in neoplasia affect two broad groups of genes and both contribute to tumour formation and progression:

1. Gain-of-function (“dominant”) mutations occur in proto-oncogenes (they become activated and called oncogenes)
2. Loss-of-function (“recessive”) mutations occur in tumour suppressor genes (they become inactivated and usually both alleles become mutated or inactivated resulting in loss of the function of the encoded protein)

Oncogenes are alleles which, if mutated, act in a “dominant” or positive (gain-of-function) fashion.

Mutations usually affect one allele only.

Oncogenes are not ‘special’ cancer genes but normal genes that are important in growth control.

Background and molecular basis

Oncogenes were first discovered as the elements in certain tumour-producing retroviruses that were responsible for their carcinogenic properties.

Later, it was discovered that these viral oncogenes (v-onc) were close homologues of normal cellular genes (proto-oncogenes); these cellular genes could become carcinogenic if ‘activated’ (to become oncogenes or c-onc) by various means.

Excellent evidence shows that the viruses must have obtained their oncogenes from previous sojourns in mammalian cells, rather than the other way around.

Normal cellular genes, expressed as RNA, became recombined into the retroviral genome. In subsequent infections, these v-oncs were inappropriately over-expressed under the powerful viral promoters and enhancers.

In total, about 40 such oncogenic retroviruses are known.

Nomenclature

• Normal – proto-oncogenes
• Activated – oncogenes (or c-onc)
• Retroviral – v-onc

 

Oncogenes in the mammalian genome can become active through a number of different routes.

Retroviruses that carry oncogenes

Some tumour-producing retroviruses (that produce tumours quickly after infection – acute transforming retroviruses), such as the Rous Sarcoma Virus in chickens, were found to contain oncogenes (SRC in this case). These are expressed at high levels in the tumour cells.

Retrovirus promoter insertion

Some retroviruses are oncogenic but do not carry viral oncogenes. Instead, the provirus integrates beside a cellular proto-oncogene which is then under the control of the viral promoters (and/or enhancers) and is inappropriately overexpressed.

This is known as insertional mutagenesis and is seen in a leukaemia of chickens where c-MYC is overexpressed.

Point mutation

Genetic point mutations can occur in proto-oncogenes to activate them; the RAS gene is a good example.

RAS is a G protein that binds GTP. It is a key player in signal transduction.

The RAS gene undergoes point mutation, usually at either codon 12 or 13, such that the mutant protein cannot hydrolyse GTP. This leads to the RAS protein remaining locked in the Ras-GTP active configuration, and so the signalling pathway remains permanently activated.

Normal RAS (glycine at codon 12) -> mutated RAS (valine at codon 12)

Normal                            Mutant

 

GCT  GGT  GGC          GCT  GTT  GGC

 

Alanine                           Alanine

          Glycine                            Valine

                      Glycine                         Glycine

 

Normal K-RAS proto-oncogene is changed by a point mutation to an activated oncogene (or c-onc).

RAS family has 3 members: K-RAS, H-RAS, N-RAS

Oncogene amplification or truncation

C-ERB-B1 is a member of the EGF-R family that can be activated by:

• Amplification to many copies (sometimes 100s) resulting in marked over-expression of EGF-R
• Truncation (loss of the extracellular domain of the EGF-R protein) rendering the truncated (shortened) receptor constitutively active

Another member of the EGF-R family, C-ERB-B2 (HER2), is amplified in many breast cancers. Sometimes it can be successfully treated by blockade of its ligand (by the antibody drug Herceptin).

The N-MYC oncogene is also a target for gene amplification in the paediatric tumour, neuroblastoma, where the degree of amplification is proportional to the aggressiveness of the tumour.

Inappropriate regulation of expression

In Burkitt’s Lymphoma (a malignant cancer of B lymphocytes seen mostly in the malaria zone in Africa), a common finding is a chromosome translocation that moves part of chromosome 8 (containing the C-MYC gene) to join up with part of chromosome 14 (at the site / locus of the IgG antibody heavy chain genes) – known as a t(8;14) translocation.

This leads to aberrant high-level expression of C-MYC as it loses its normal regulatory sequences and now falls under the influence of the active IgG regulatory sequences (active in B lymphocytes).

Oncogene function

Oncogenes encode oncoproteins that include components of growth signalling pathways:

• Growth factors – SIS (Simian sarcoma virus) encodes platelet-derived growth factor (PDGF)
• Receptors – ERB-B1 (Avian erythroblastosis virus) encodes epidermal growth factor receptor (EGFR)
• Signalling proteins (in growth signalling pathways) –
  •  ABL (Abelson mouse leukaemia virus) encodes a tyrosine kinase
  • RAS (Rat sarcoma virus) encodes a GTP-nucleotide binding molecular switch
• Transcription factor – MYC (Myelocytomatosis virus) encodes a transcription factor that binds DNA and stimulates proliferation and regulates apoptosis

Tumour suppressor genes are alleles that must be inactivated.

Both alleles must be mutated, deleted, functionally suppressed or lost in some way.

TSGs are said to be “recessive”* in mechanism to wild-type (meaning they have loss-of-function mutations).

They are often critical control and regulatory genes many of which restrain cell proliferation or are involved in DNA repair.

(*use of the word “recessive”, here indicating inactivation of both alleles, is in common usage, but strictly speaking this differs from the classical (and correct) use of recessive to indicate a pattern of Mendelian inheritance in a family tree/pedigree where 2 mutated alleles are inherited, 1 from each parent.)

Tumour suppressor genes were discovered largely through their co-segregation with inherited cancer susceptibility in certain human familial cancer syndromes, although their existence was predicted by cell hybridisation experiments.

Retinoblastoma (RB1) and TP53 are classic, very important TSGs.

Classical TSGs are genes which restrain proliferation or growth.

RB1 that controls the G1/S checkpoint, or R point, and plays a vital role in cell cycle control:

• Phosphorylation of RB1 is essential for the cell to progress through the G₁ restriction point (G1-S checkpoint) in the cell cycle, otherwise, the cycle is blocked at that point
• Unphosphorylated RB1 sequesters E2F
• E2F transcription factors coordinately regulate genes that combinatorially promote entry into S phase
• RB1/E2F binding on chromatin recruits histone de-acetylases and chromatin remodelling factors to the E2F responsive promoters repressing transcription
• RB1 phosphorylation releases E2F and initiates E2F dependent transcription

Loss or mutation of RB1 removes this block and the cell can cycle inappropriately.

Deregulation of the pathway controlling RB1 or pRb phosphorylation occurs in almost all cancers but can involve different players in the pathway in specific cancers. For example, mutation or deletion or silencing (by promoter methylation) of p16 or amplification of Cyclin D1 or mutation/deletion of RB1.

Function of p53

TP53 gene is mutated in ~ 50% of human cancers. The TP53 gene encodes the p53 protein.

The bioactive form of p53 is a tetramer complex of 4 p53 polypeptides. If TP53 is mutated at one allele then the probability of generating functional tetramers is significantly reduced.

Thus, a single mutation of TP53 results in defective p53 tetramer function.

INK4a gene

The INK4a gene encodes 2 potent tumour suppressor genes – p16^INK4a and p14^ARF.

ARF is important in activating p53 in response to abnormal proliferative signals such as inappropriate oncogene activation (e.g. RAS or MYC activation).

In such a situation, ARF protein accumulates and sequesters MDM2 (the protein that regulates p53 by targeting it for degradation), and

p53 is thus stabilised (p53 accumulates) and cell cycle arrest or apoptosis occurs.

P53 as the guardian of the genome

Mutating TP53 makes the cell genetically unstable as it permits survival and replication of cells that have sustained major DNA damage such as double-strand breaks.

There is now good evidence that such “undead” cells are at risk of surviving with:

1. Inappropriate recombination events that may activate oncogenes
2. Localised regions of gene amplification, that may promote growth or increase drug resistance

Rb and p53 in partnership

Deregulation of the pathways controlling Rb and p53 occurs in almost all cancers but can involve different players in the specific pathway in individual cancers.

RB1 pathway:

• Inactivate RB1 or p16^INK4a
• Activate cyclinD or cdk4/6

TP53 pathway:

• Inactivate p53 or ARF
• Activate MDM2

Deregulation of both RB1 and TP53 is essential to overcome replicative senescence: “immortality checkpoints”.

Normal cells shorten their telomeres (black ovals getting smaller over time) every time that they undergo cell division (until they run out of repetitive telomeric DNA sequence and undergo cell cycle arrest and senescence).

In neoplastic cells that become immortal, there is usually dysfunction or inactivation of:

1. The RB1 pathway (often through p16 dysfunction) allowing the cell to continue to cycle through the “CRISIS” when the telomeres run out
2. The TP53 pathway allowing cells to survive DNA damage (including complete loss of telomeric sequences such that chromosome ends appear as double-stranded DNA breaks)

Eventually, a cell emerges from this CRISIS (and self-selects) that has aberrant expression of the telomerase enzyme and can re-synthesise its telomeric sequences and protect the chromosome ends, thus becoming immortal and with the potential to evolve into a neoplastic clone.

Cells are social animals.  They talk to each other and they know their place in the social hierarchy.  This is a consequence of signals from other cells and from the extracellular matrix.

Decisions about when to divide and when to move depend upon effective processing of these signals. These signalling pathways can go wrong during the development of neoplasms. APC and Wnt are examples of this.

Adenomatous Polyposis Coli (APC) protein

The APC gene was identified by studying Familial Adenomatous Polyposis Coli; this is a condition where patients have hundreds of adenomatous polyps in the large intestine. Patients inherit one mutant APC an one wild-type APC allele.

Pathogenesis of APC

• The APC protein is mostly cytoplasmic. It is mutated in most (~80%) colonic carcinomas and adenomas
• APC regulates the WNT signalling pathway, a crucial pathway both in development and the maintenance of tissue organization. It is important in cell-to-cell signalling in bowel epithelium
• APC controls β-catenin levels (the key signalling intermediate in the WNT pathway)
• β-catenin is a bi-functional molecule –
  • at the plasma membrane, it forms a complex with E-cadherin at adherens junctions
  • in the cytoplasm, it’s a “free” protein that is phosphorylated (by GSK3B) and rapidly degraded by the proteasome
  • APC acts as a binding platform to promote this phosphorylation and degradation of β-catenin

Mutant APC allows accumulation of β-catenin (via reduced degradation).

WNT signalling pathway

Wnt ligands (paracrine growth factors) binds to receptors to allow intermediates to inhibit GSK3B. Unphosphorylated β-catenin then accumulates and translocates to the nucleus. In the nucleus, it binds to a binding partner (LEF/TCF) and mediates transcription of growth-promoting genes.

The activity of the β-cat/TCF complex is a master switch in intestinal crypts controlling proliferation versus differentiation (it upregulates c-MYC expression, cyclinD1 and others).

Pathway deregulation occurs in many cancers, involving different players in individual cancers (e.g. inactivating APC mutation or activating β-catenin mutation).

Dying at the right time, and in the right way, is critical for cellular society, but cancer cells survive long after their “die by” date. Why don’t cancers die by apoptosis-like normal cells?

• TP53 inactivation
• Telomerase activation
• BCL2 overexpression

However, the physiological response of cells is to die by apoptosis when they are:

• Damaged
• Severely stressed
• No longer needed
• In an inappropriate environment

Apoptosis or Programmed Cell Death

Apoptosis is the physiological way of cell death; a simplified outline of the apoptosis death pathways is shown below.

Understanding these pathways, and how tumours evade the death signals, is important because restoring effective death signals could result in novel therapies.

It is also becoming increasingly evident that cancer therapies, radiation and chemotherapy work by inducing cancer cells to undergo apoptosis.

In tumour cells, at least one death pathway seems to remain intact and the apoptosis evasion mutations are targeted to certain genes (not the caspases):

• Inactivation of genes encoding the receptors (FAS, TNFR)
• Activation of anti-apoptotic proteins such as BCL-2
• Downregulation of pro-apoptotic proteins (BAX)
• p53 inactivation (leading to reduced activation of  BAX, PUMA, NOXA)

Why is it important?

• The transition from normal growth regulated cell through to a malignant cancer cell requires several mutations. Much evidence suggests that at least 6 are needed and common solid tumours have probably sustained many more
• The probability of spontaneously acquiring >6 mutations in a single cell, several of which have to be in tumour suppressor genes and therefore both alleles of the same gene, can be computed on the basis of “background” mutation rates and would require a lifespan of several 100 years
• Something has to speed up the mutation rate and that seems to be the acquisition of genetic instability called by some the mutator phenotype

What is the evidence?

• Most solid tumours are aneuploid – they have abnormal chromosome numbers and chromosome rearrangements
• Most cancers have deregulated the TP53 pathway that arrests the cell cycle to allow DNA repair
• Mutations in genes encoding DNA repair proteins gives an increased susceptibility to cancer

 

Major DNA repair activities include:

• DNA mismatch repair
• DNA nucleotide excision repair
• DNA strand break repair
• Fanconi anaemia DNA crosslink repair
• Chromothripsis
• Kataegis

DNA mismatch repair (MMR)

MMR corrects mismatched bases (eg C-T instead of C-G).

It also corrects insertion/deletion loops that most commonly occur where short sequences are repeated e.g. AAAAAAA or CACACACA – microsatellite sequences.  Such repeats are common in the human genome.

In the absence of mismatch repair, mutation rate increases 100x-1000x. If mutations occur in a coding region this results in a mutant protein.

Mutations in mismatch repair genes (MLH1, MSH2) occur in Hereditary Non-Polyposis Colorectal Cancer (HNPCC) syndrome (or Lynch Syndrome), which accounts for ~2-5% of all colorectal cancers (showing defective mismatch repair with microsatellite instability).

In sporadic colorectal cancer, MLH1 silencing (by promoter methylation) is found in ~15% of lesions (and these show defective mismatch repair with microsatellite instability).

DNA nucleotide excision repair

UV radiation can induce adjacent thymine bases to become linked to form thymidine dimers. These (and some other forms of DNA damage such as certain carcinogens attached to bases) are removed/repaired by the Nucleotide Excision Repair (NER) pathway that excises the damaged bases and replaces them with normal bases.

Xeroderma pigmentosum is a disease in which mutated NER genes are inherited. The NER pathway is defective, making the sufferer predisposed to UV-induced skin tumours.

DNA strand break repair

BRCA1 and BRCA2 genes encode proteins that are involved in the double-strand DNA break repair pathway. Inherited BRCA1/2 mutations confer susceptibility to both breast cancer and ovarian cancer.

Fanconi anaemia DNA crosslink repair

DNA interstrand crosslinks that covalently link one DNA strand to the opposite DNA strand (caused by acetaldehyde, formaldehyde, platin drugs, etc) block both transcription and replication at the site of the crosslink.

Such interstrand cross-links are repaired by the Fanconi anaemia (FA) repair pathway.

Fanconi anaemia patients inherit two mutated FA genes and are defective for this type of repair. They experience aplastic anaemia (due to bone marrow haematopoiesis failure), congenital defects and cancer susceptibility.

Chromothripsis

Chromothripsis is clustered chromosomal rearrangements (hundreds or thousands) occurring in a single event, in localised and confined genomic regions, in one (or a few) chromosomes (thripsis = shattering).

Kataegis (thunder)

Kataegis is a pattern of localized hypermutation identified in some cancer genomes, almost exclusively C>T (in the context of TpC dinucleotides).

Enzymes of the APOBEC deaminase family are responsible for the process of kataegistic clusters of mutations (e.g. APOBEC3).

Chromosomal rearrangements in solid tumours, particularly carcinomas, are extensive.

They are not characteristic of a specific cancer type, and the rearrangements often involve tumour suppressor genes (by deleting them).

The mechanisms by which aneuploidy arises in cells are still poorly understood but deregulation or loss of functional TP53 pathway is usually involved.

• Genetic changes often affect signalling pathways controlling proliferation, apoptosis, DNA integrity
• Defective DNA Repair – Genetic instability -> increases mutation rate and aneuploidy
• Proto-oncogenes -> activated oncogenes
• Tumour suppressor genes -> inactivated TSGs
• Cancer is a multi-step process, reflecting an accumulation of multiple genetic changes to oncogenes and TSGs