One rogue agent suffices for genomic chaos

Cancer is a disease of uncontrolled cell division that is fuelled by genetic instability a state in which cells acquire mutations at an abnormally high rate. When normal cells are transforming into cancer cells, a common early event is the acquisition of mutations in a type of gene called a tumour-suppressor gene. If both of the two copies of a tumour-suppressor gene are inactivated in a cell, this decreases genomic stability and aids the acquisition of other cancer-initiating mutations. Writing in Nature, Coelho et al.1 report their studies in budding yeast (Saccharomyces cerevisiae), which indicate that, frequently, the disruption of just one copy of certain genes can be sufficient to trigger genetic instability.

The existence of tumour-suppressor genes was first proposed about 50 years ago to explain why, in some families, there is a puzzling pattern of inheritance of a type of cancer called retinoblastoma2,3. Clinical observations suggested that this cancer is caused by a type of mutation, known as a recessive mutation, in the gene RB. Such a mutation has an effect only if both copies of the gene are mutated in a cell. The ‘two-hit’ hypothesis was proposed2 to explain the inheritance patterns of retinoblastoma. It suggested that if a cell inherits a recessive mutation in one copy of RB, it must also acquire a mutation in its other copy for cancer to develop. Subsequent research in mice4 revealed that, although a two-hit scenario is common, the presence of a mutation in only one of the two copies of some tumour-suppressor genes (a condition termed haploinsufficiency) can suffice to trigger cancer formation.

Many other tumour-suppressor genes have been identified as being haploinsufficient in mice5. Moreover, a mutation in one copy of the BRCA1 gene, which is associated with breast cancer, can cause genetic instability in the epithelial cells of human breast tissue grown in vitro6, suggesting that haploinsufficiency of tumour-suppressor genes can kick-start cancer formation in human cells. But determining how commonly genetic instability arises from inactivation of either one or both versions of a gene is challenging, and cannot be easily assessed using current mouse models of cancer or by analysing DNA sequences of tumours obtained from people who have cancer. This is because, if cell samples are obtained after the cancer has arisen, and both copies of a gene have mutations, it is difficult to know whether one or both of the mutations occurred before the cancer developed.

To address this, Coelho and colleagues devised a way of identifying mutations that cause genetic instability in yeast. They used a system in which there is an evolutionary selective pressure for the development of cells that can generate genetic changes enabling them to survive treatment with drugs that usually limit growth. A key advantage of this system is the ease with which characteristics that have a predictable, Mendelian pattern of genetic inheritance can be followed. This makes it easy to determine whether mutations occurred in one or both copies of a gene — states respectively known as heterozygosity and homozygosity. The authors found that most of the yeast cells in which genetic instability had occurred had inherited this capacity through a mutation in only one copy of the gene responsible (Fig. 1). Whole-genome sequencing of cells confirmed this and also revealed the identity of the mutations.

Figure 1 | Mutations that cause genetic instability. Coelho et al.1 report the identification of mutations in budding yeast (Saccharomyces cerevisiae) that cause genomic abnormalities called genetic instability. The authors induced mutations in yeast cells, and used growth-limiting conditions to positively select for surviving cells. Such cells might have acquired their capacity for survival in these conditions as a result of changes arising from genetic instability. Characterization of the mutations responsible for generating genetic instability revealed a high frequency of mutations in one of the two copies of a particular gene, which is termed a heterozygous mutation. Such a mutation can affect a cell if it causes a decrease in production of the protein encoded by the gene, a situation known as haploinsufficiency, or if it results in a version of the encoded protein that functions in an unusual way, called a gain-of-function mutation. Coelho and colleagues found that, of the mutations they identified as being responsible for genetic instability, homozygous mutations (which affect both copies of a particular gene) were less frequent than were heterozygous mutations. The type of homozygous mutations that have an effect only when both copies of a gene are mutated are termed recessive mutations, and some of these were found by the authors. This insight into the relative frequency of heterozygous and homozygous mutations associated with genetic instability in yeast might have relevance for understanding the mechanisms by which cancer cells acquire genetic instability.

The authors individually introduced DNA sequences corresponding to 16 of these mutations into wild-type yeast cells, and found that this was sufficient to cause genetic instability. Further testing revealed that ten of these were loss-of-function mutations in which the encoded protein has reduced function, doesn’t function at all or isn’t made, whereas the other six had a profile consistent with gain-of-function mutations, in which the protein encoded by the mutant gene functions in an abnormal way. Of the genes identified that caused genetic instability when mutated, 57 have related versions in humans, and 10 of these have previously been linked to cancer. The other 47 were mainly genes that encode proteins involved in processes such as protein quality control and cellular transport mechanisms. These genes have not previously been connected to the generation of genetic instability.

The authors tested the effect of inactivating the human versions of six of these genes in vitro in haploid cancer cells — cancer cells that contain only one copy of each gene. They found that the mutations caused an increase in genetic instability, confirming the power of the approach and suggesting that such mutations might underlie cancer development. A follow-up study could test diploid versions of the cells — those that have two copies of each gene — to confirm the role of heterozygous mutations in driving genetic instability.

Another future direction for research might be to investigate the mechanisms that link genetic instability to mutations in genes involved in protein quality-control processes. One possibility is that mutations in such genes cause a decrease in the degradation of key enzymes that affect DNA synthesis, modification or repair, or that affect proteins important for cell division. If the pattern of degradation changes for such enzymes, the resulting interference with protein turnover might have negative consequences for processes in which the enzymes function. For example, mutations in genes that encode components of a protein complex called the proteasome, which has a key role in protein degradation, cause genetic instability in yeast by affecting the turnover of an enzyme that repairs breaks in DNA7. Another possibility is that genes involved in quality-control processes have a more direct role than previously appreciated in regulating DNA metabolism or cell division.

It is also interesting to consider whether, during cancer development, heterozygous mutations conferring genetic instability might subsequently be lost as the cancer evolves. Although instability might fuel evolutionary processes and increase tumour-cell fitness under adverse circumstances, it could also cause mutations that decrease the overall fitness of the cancer-cell population in the absence of stress. Depending on the evolutionary selective pressure faced by cancer cells, the gain or loss of genetic instability might be selected for according to circumstances, either to increase the variation in the cancer-cell population when the cells are facing an evolutionary selective pressure, or to ensure that useful variants are retained.

One way of quantifying such changes would be to take samples of cancer cells from a tumour at different time points, and to track genomic evolution during the acquisition of resistance to anticancer drug treatment. Heterozygous mutations conferring genetic instability would be predicted to be enriched at earlier stages of an evolutionary selection process, but there might be a reversion to a homozygous wild-type condition once other mutations have arisen that provide the adaptive features the tumour needs. Elucidating such dynamics will be a challenging task, but the outcome should be rewarding because it will increase our understanding of the competition that occurs between different cancer-cell lineages as a tumour develops.

Nature 566, 188-189 (2019)


  1. 1.

    Coelho, M. C., Pinto, R. M. & Murray, A. W. Nature 566, 275–278 (2019).

    Article  Google Scholar 

  2. 2.

    Knudson, A. G. Jr Proc. Natl Acad. Sci. USA 68, 820–823 (1971).

    PubMed  Article  Google Scholar 

  3. 3.

    Nordling, C. O. Br. J. Cancer 7, 68–72 (1953).

    Article  Google Scholar 

  4. 4.

    Fero, M. L., Randel, E., Gurley, K. E., Roberts, J. M. & Kemp, C. J. Nature 396, 177–180 (1998).

    PubMed  Article  Google Scholar 

  5. 5.

    Inoue, K. & Fry, E. A. Adv. Med. Biol. 118, 83–122 (2017).

    PubMed  Google Scholar 

  6. 6.

    Konishi, H. et al. Proc. Natl Acad. Sci. USA 108, 17773–17778 (2011).

    PubMed  Article  Google Scholar 

  7. 7.

    Ben-Aroya, S. et al. PLoS Genet. 6, e1000852 (2010).

    PubMed  Article  Google Scholar 

Download references

Nature Briefing

An essential round-up of science news, opinion and analysis, delivered to your inbox every weekday.

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing