‘Latent drivers’ expand the cancer mutational landscape

https://doi.org/10.1016/j.sbi.2015.01.004Get rights and content

Highlights

  • Currently mutations are classified into drivers and passengers.

  • We suggest an additional ‘latent driver’ classification.

  • Latent driver mutations behave as passengers and do not confer a cancer hallmark.

  • Latent driver mutations emerge prior to and during cancer evolution.

  • Coupled with other emerging mutations, they drive cancer.

A major challenge facing the community involves identification of mutations that drive cancer. Analyses of cancer genomes to detect, and distinguish, ‘driver’ from ‘passenger’ mutations are daunting tasks. Here we suggest that there is a third ‘latent driver’ category. ‘Latent driver’ mutations behave as passengers, and do not confer a cancer hallmark. However, coupled with other emerging mutations, they drive cancer development and drug resistance. ‘Latent drivers’ emerge prior to and during cancer evolution. These allosteric mutations can work through ‘AND’ all-or-none or incremental ‘Graded’ logic gate mechanisms. Current diagnostic platforms generally assume that actionable ‘driver’ mutations are those appearing most frequently in cancer. We propose that ‘latent driver’ detection may help forecast cancer progression and modify personalized drug regimes.

Introduction

All cancer cells carry mutations in their genomes; some are inherited, others gained. Most mutations appear during normal cell growth [1••, 2••]. Pre-neoplastic mutations have been postulated to be ‘passengers’ as are most of those that emerge during cancer development. Driver mutations have been causally implicated in oncogenesis by conferring clonal selective advantage with the cell acquiring some of the hallmarks of cancer [3, 4, 5, 6]. Driver mutations also emerge in drug resistance. Cancer is an evolutionary process [7, 8], driven by somatic-cell mutational events with sequential, subclonal selection [9]. Mutation rates vary, and depend on the cancer type and its environment [10, 11, 12]. Applying the evolutionary clock model to tumor mutational data suggested that the vast majority of the mutations in metastatic lesions were already present in a large number of cells in the primary tumor [1••]. Similar to organ development [7], the long time period which is usually required for cancer symptoms to emerge reflects not only the evolutionary time, but the complexity of the mutational landscape. In contrast, drug resistance typically emerges on short timescales, begging the question of why.

The important question of why the development of the primary cancer can take a long period of time while drug resistance typically arises very rapidly is puzzling. Many resistance mechanisms may contribute. These include genomic rearrangements, deletions, as for example observed in Raf's deletion of the Ras binding domain, alternative signaling venues, for example other receptors taking over following EGFR's inhibition [13], or other mechanisms side-stepping regulatory controls [14, 15, 16]. Below we hypothesize that among these the many overlooked ‘latent driver’ mutations may bear on this conundrum. Our ‘latent driver’ hypothesis rests on a conformational view of protein molecules where mutations may incrementally shift the protein ensembles toward populating a constitutively active (or inactive for a repressor protein) state [17••]. While the extent of the shift by a pre-existing mutation may be small  which could be the reason why it escaped discovery  combinatorial emergence of a cooperative second may switch the protein and the cell to a constitutively active (or inactive) state bearing a cancer hallmark. Figure 1 presents a schematic diagram and its legend explains the differences between a driver mutation, a passenger mutation and a latent driver mutation, and how a latent mutation can work, summarizing the main concept of this paper.

Section snippets

The cancer mutational landscape

Cancers vary [9]. Most mutations are passengers and do not confer selective advantage; a few can be functionally relevant drivers. They can be point mutations, or other genetic aberrations, such as higher gene copy number, deletions, inversions, and translocations [18]. They can lead to over-expression or under-expression, or dysfunctional gene products. Most translocations appear to be passengers. There are ∼10 times fewer genes affected by chromosomal changes than by point mutations, and the

Driver and passenger mutations

A driver mutation confers growth advantage; a passenger mutation does not. Passenger mutations populate cancer genomes prior to the emergence of driver mutations [12]. Age-related statistics suggested 5–7 driver mutations in epithelial cancers such as breast, colorectal and prostate [19]; more recent analyses indicated that the number could be much higher [20].

Driver mutations can affect recruitment or catalysis. In recruitment, driver point mutations can be at the interface [21] and

The concept of ‘latent driver’ mutations

Currently, the classification is binary: a mutation is designated as either a driver or a passenger. This classification is important since it provides the genetic basis for cancer treatment decisions [17••]. Statistics is a primary factor in this classification [3, 4]. However, low statistics and lack of observable cancer hallmark effects do not necessarily imply that a passenger mutation does not harbor potential oncogenic effects. Here we suggest a modification of this assignment model.

Latent driver mutations can be understood through allosteric integration mechanisms

Signal transduction rarely reflects a single signaling source [34, 35]. Multiple signaling events act on a protein node, including binding and post-translational modifications, and the signal integration mechanisms govern the output response. In electronics, there are three types of logic gate mechanisms. We adopt these to describe a protein switching from an inactive to an active state (or vice versa): OR, AND or Graded. For the case of two allosteric events, in an OR gate one allosteric event

Approaches to differentiate between driver and passenger mutations

The catalog of somatic mutations establishes the record of mutations that accumulated in the cancer cell throughout the lifetime of the patient. Databases compiling these, such as COSMIC and TCGA have led to important discoveries [37••, 45, 46•, 47, 48, 49]; however, they are still predominantly unmined. Broadly, cancer genome analysis aims to identify cancer genes that carry driver mutations; thus, a key challenge is to distinguish oncogenic driver mutations from their neutral passenger

Conclusions

The number of observed driver mutations in solid tumors is smaller than that expected for tumor development and progression. Several explanations have been offered to explain the ‘missing mutations’, including technical issues, statistical problems, exclusion of the tumor stromal environment, accounting only for coding regions, and overlooking epigenetic effects [1••]. Latent drivers which are counted as passengers can be added to this list. Latent drivers’ actions are triggered by changes in

References and recommended reading

Papers of particular interest, published within the period of review, have been highlighted as:

  • • of special interest

  • •• of outstanding interest

Acknowledgements

This project has been funded in whole or in part with Federal funds from the National Cancer Institute, National Institutes of Health, under contract number HHSN261200800001E. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government. This research was supported (in part) by the Intramural Research Program of the

References (64)

  • G. Ciriello et al.

    Emerging landscape of oncogenic signatures across human cancers

    Nat Genet

    (2013)
  • G.M. Wang et al.

    Single copies of mutant KRAS and mutant PIK3CA cooperate in immortalized human epithelial cells to induce tumor formation

    Cancer Res

    (2013)
  • K. Hashimoto et al.

    Oncogenic potential is related to activating effect of cancer single and double somatic mutations in receptor tyrosine kinases

    Hum Mutat

    (2012)
  • M.D. Onken et al.

    A surprising cross-species conservation in the genomic landscape of mouse and human oral cancer identifies a transcriptional signature predicting metastatic disease

    Clin Cancer Res

    (2014)
  • J.A. Sparano et al.

    Translating the TCGA breast cancer results into clinical practice: searching for therapeutic clues

    Oncology

    (2013)
  • C.X. Ma et al.

    The cancer genome atlas: clinical applications for breast cancer

    Oncology

    (2013)
  • G. Smith et al.

    Activating K-Ras mutations outwith ‘hotspot’ codons in sporadic colorectal tumours – implications for personalised cancer medicine

    Br J Cancer

    (2010)
  • B. Vogelstein et al.

    Cancer genome landscapes

    Science

    (2013)
  • J.S. Welch et al.

    The origin and evolution of mutations in acute myeloid leukemia

    Cell

    (2012)
  • D. Hanahan et al.

    Hallmarks of cancer: the next generation

    Cell

    (2011)
  • M. Egeblad et al.

    Tumors as organs: complex tissues that interface with the entire organism

    Dev Cell

    (2010)
  • A. Youn et al.

    Identifying cancer driver genes in tumor genome sequencing studies

    Bioinformatics

    (2011)
  • G.R. Bignell et al.

    Signatures of mutation and selection in the cancer genome

    Nature

    (2010)
  • M.R. Stratton et al.

    The cancer genome

    Nature

    (2009)
  • R. Roskoski

    The ErbB/HER family of protein-tyrosine kinases and cancer

    Pharmacol Res

    (2014)
  • P.I. Poulikakos et al.

    RAF inhibitors transactivate RAF dimers and ERK signalling in cells with wild-type BRAF

    Nature

    (2010)
  • J. Knievel et al.

    Multiple mechanisms mediate resistance to sorafenib in urothelial cancer

    Int J Mol Sci

    (2014)
  • R. Nussinov et al.

    The structural basis for cancer treatment decisions

    Oncotarget

    (2014)
  • M. Pelizzola et al.

    The DNA methylome

    FEBS Lett

    (2011)
  • D.G. Miller

    On the nature of susceptibility to cancer – the presidential-address

    Cancer

    (1980)
  • N. Beerenwinkel et al.

    Genetic progression and the waiting time to cancer

    PLoS Comp Biol

    (2007)
  • N. Tuncbag et al.

    Towards inferring time dimensionality in protein–protein interaction networks by integrating structures: the p53 example

    Mol Biosyst

    (2009)
  • Cited by (0)

    View full text