MUTATIONS IN CANCER GENES
Tumours result from the subversion of the processes that control the growth, location and mortality of cells. This loss of normal control mechanisms arises from the acquisition of mutations in three broad categories of genes:
- Proto-oncogenes. These encode components of signalling pathways that regulate proliferation. In their mutated form, they can become dominant ‘oncogenes’.
- Tumour suppressor genes. These encode proteins whose loss of function leads to de-regulated control of cell cycle progression, protein degradation, cellular adhesion and motility. They generally exhibit recessive behaviour.
- DNA repair enzymes. These proteins maintain genomic integrity and mutations causing loss of function, therefore, attenuate the repair processes we have discussed and promoted genetic instability.
Mutations in these genes are also presumed to produce changes in cell surface protein expression, protein secretion and cell motility that contribute to metastasis, no mutations having been specifically associated with metastatic development. Genes are shown to have a functional association with specific cancers now number some 500 oncogenes and about 100 tumour suppressor genes.
Oncogenes were first identified in viruses capable of transforming cells in culture or inducing tumours in animals. The classic experiment of Peyton Rous in the early years of the twentieth century showed that cell-free filtrates from chicken tumours can give rise to sarcomas (tumours of connective tissue) when inoculated into normal chickens. The infectious agent was identified by electron microscopy in the 1950s as a virus – the Rous sarcoma virus – eventually revealed to be a retrovirus, i.e. having an RNA genome. Studies of temperature-sensitive mutants of the virus showed that sustained expression of a viral protein was required for transformation. The gene encoding this protein had been picked up by the virus from the DNA of its host during the normal life cycle of the retrovirus in which viral RNA is converted to DNA and inserted in the host genome. The cellular gene that had been captured by the virus was Src, which encodes a tyrosine kinase highly conserved throughout the animal kingdom. Many other oncogenes have now been identified both in retroviruses and by other means, including, most recently, whole genome sequencing. They have in common their derivation from normal genes (proto-oncogenes) that are highly conserved across all species, they function mainly in growth factor signalling pathways and they act in a dominant manner (e.g. a single allele mutation is activating). Mutant forms of the virus were also found that multiplied normally in infected cells but did not cause transformation. The non-transforming mutants had lost all or part of the gene that had been acquired from the host genome in the generation of the transforming virus (Fig. 1).
1. Acquisition of a cellular gene by a retrovirus.
Although many retroviruses can cause tumours in animals, particularly the feline and bovine leukaemia viruses, they are rarely associated with human cancers. The RNA human immunodeficiency virus (HIV) destroys cells of the immune system so that victims become susceptible to infection. Infection by the Kaposi sarcoma-associated herpesvirus (KSHV or HHV8) causes the gradual development of Kaposi’s sarcoma (which is not really a sarcoma because it starts in the lymphatic system). This is because KSHV encodes two proteins that act directly on cell signalling pathways to override normal controls of proliferation and survival. HIV is a member of the human T-cell lymphotropic virus (HTLV) family (it’s HTLV-III). HTLV-I weakens the immune system and so increases the risk of opportunistic infection by bacteria, fungi, viruses or protozoa. The HTLV-I virus is thought to cause the rare adult T-cell leukaemia, a form of non-Hodgkin’s lymphoma, but this cancer takes a very long time to develop and only a small fraction (1 in 1,500) of those infected with HTLV-I develop the disease. HTLV-II and HTLV-IV have not been specifically linked to any disease.
The key experiment showing that the human version of a gene that had been acquired by an oncogenic retrovirus can cause cancer, independent of any virus, came in 1983. In principle it couldn’t have been simpler: extract DNA from a human tumour, fragment and transfect into cells in culture (Fig. 2).
2. Transformation of an immortal cell line by DNA from a human tumour.
Pick out the cells that become transformed, into mice, excise a tumour that develops and isolate the gene responsible from the tumour cells. If the transfection is into mouse cells, the cancer-promoting gene can be identified as human and thus derived from an original tumour. Through this kind of experiment Robert Weinberg and colleagues identified the first human oncogene, RAS – so named because a retroviral form causes rat sarcomas. We now know that there are three closely related human RAS genes (NRAS, HRAS1 and KRAS2), each encoding a GTPase implicated in the control of cell proliferation. Mutations that convert them from proto-oncogenes to oncogenes occur in about 20% of human tumours.