Minireview
Experimental investigations of carcinogen-induced mutation spectra: Innovation, challenges and future directions

https://doi.org/10.1016/j.mrgentox.2020.503195Get rights and content

Highlights

  • Mutational signatures elucidate the roles of mutagenic factors in cancer formation.

  • Experimental modelling of mutational signatures can explain origins of cancer.

  • Signature modelling relies on innovative exposure models and genomic sequencing.

  • Mechanistic data from signature modeling can support cancer prevention efforts.

Abstract

Recent years have witnessed an expansion of mutagenesis research focusing on experimentally modeled genome-scale mutational signatures of carcinogens and of endogenous processes. Experimental mutational signatures can explain etiologic links to patterns found in human tumors that may be linked to same exposures, and can serve as biomarkers of exposure history and may even provide insights on causality. A number of innovative exposure models have been employed and reported, based on cells cultured in monolayers or in 3-D, on organoids, induced pluripotent stem cells, non-mammalian organisms, microorganisms and rodent bioassays. Here we discuss some of the latest developments and pros and cons of these experimental systems used in mutational signature analysis. Integrative designs that bring together multiple exposure systems (in vitro, in vivo and in silico pan-cancer data mining) started emerging as powerful tools to identify robust mutational signatures of the tested cancer risk agents. We further propose that devising a new generation of cell-based models is warranted to streamline systematic testing of carcinogen effects on the cell genomes, while seeking to increasingly supplant animal with non-animal systems to address relevant ethical issues and accentuate the 3R principles. We conclude that the knowledge accumulating from the growing body of signature modelling investigations has considerable power to advance cancer etiology studies and to support cancer prevention efforts through streamlined characterization of cancer-causing agents and the recognition of their specific effects.

Introduction

The previous decade has witnessed a broad implementation of massively parallel sequencing technologies into biomedical sciences, and associated in-depth investigations of human cancer genomes have generated vast, publicly available data collections [1]. This work has substantially expanded our knowledge about the accumulation and functional impact of somatic mutations in tumor genomes. A special type of computational analysis of large sets of mutated genomes has focused on the extraction of mutational signatures, i.e. characteristic somatic DNA mutation patterns attributable to particular mutagenic sources, reflecting their contributions to cancer development, and often shaped further by the activity of the DNA repair machinery [2,3]. The current catalog of mutational signatures reflects an expansion from the initially considered patterns of single-base substitutions (SBS) (from 30 to ∼50), and contains also 11 signatures based on doublet-base substitutions (DBS) and 17 based on small insertions-deletions (indels) [2]. The original as well as the updated patterns can be accessed on the COSMIC web site (https://cancer.sanger.ac.uk/cosmic/signatures). Recently genome-scale sequencing studies also revealed extensive mutagenesis occurring in cancer-associated genes and mutational signatures in various normal tissues in the absence of overt neoplastic disease as well as in the normal cells of cancer patients [[4], [5], [6], [7], [8], [9], [10], [11]]. These studies present a paradigm shift that poses new challenges and opportunities with regards to the precise identification of the causal events underlying cancer development.

The power of mutational signature analysis lies in its ability to point to the effects of specific etiologic agents and to reveal exposure history imprinted into the exposed cell genome. These mutagenic factors can be of endogenous nature, such as DNA repair deficiency or deregulation of endogenous mutagenic enzymes, or of exogenous nature, originating from environmental, life-style or therapeutic exposures to mutagenic agents, with both types of activities often acting in cooperation to leave mixed mutation fingerprints. A number of the known mutational signatures have already been associated with particular etiologies, although a large proportion of signatures remain unexplained. Furthermore, as highly similar mutation spectra can arise in a given tumor from various convergent mutagenic processes, or the results of certain background mutagenic processes do not yield a robust signal, it remains a challenge to disentangle the resulting signatures by the currently applied de-convolution procedures such as non-negative matrix factorization (NMF), the most commonly used method of unscrambling mutational signatures from the somatic mutation mixtures of complex origins [12,13]. It is plausible that a number of signatures may have eluded discovery at the level of the pan-cancer genome signature extraction analyses conducted to-date [2], while novel mutational signatures are being uncovered within more focused studies including also experimental modeling [6,11,14–16].

Section snippets

Experimental approaches to mutational signature modeling

Experimental models of chemical exposure and genetic perturbation provide highly controlled approaches allowing to systematically address some of the challenges associated with the discovery of mutational signatures linked to specific causal factors or agents. In the last few years these invaluable models have been combined with massively-parallel sequencing at the exome and genome-scale levels to generate large amounts of experimental data using various in vitro and in vivo exposure systems,

Clonal expansion systems based on primary cells and cell lines

Current experimental exposure mutagenesis models have been developed based on the combination of clonally expanded cell lines coupled with genome sequencing technologies, allowing modeling of exposure-specific mutational signatures. Primary mouse embryonic fibroblasts harboring a knocked-in human TP53 gene (Hupki MEFs), thanks to the general inherent propensity of MEF cells to bypass a biological barrier (senescence) and immortalize in a clonal manner, have emerged as a robust and suitable in

Induced pluripotent stem cells

A recent study [35] based on human induced pluripotent stem cells (iPSC) and their intrinsic clonal expansion stands out for addressing for the first time the mutagenic effects of a larger number (∼80) of environmental and therapeutic exposures by whole-genome sequencing. About half of the exposure experiments yielded SBS signatures, with some corroborating the previous cell line-based studies (platinum complexes [15], aflatoxin [28] or aristolochic acid [36]), while other exposures generated

3-D exposure cell systems

Numerous 3-D cell culture and organoid-based models have been developed over the past two decades. These developments are driven by the promise that such 3-D models resemble and/or mimic real tissue arrangements, and they may thus significantly accelerate translational research in cancer biology, tissue engineering, and regenerative medicine. In most cases, however, the 3-D cell culture models represent solely the epithelial cell layer and lack the native microenvironment and the usual

Retrospective use of rodent bioassay material

In contrast to the cell-based model systems discussed above, in vivo animal bioassays permit studies of mutation spectra at an organismal level. However, since such studies are based on exposure-mediated tumor induction and cancer burden in animals, they have been linked to ethical concerns. Apart from de novo cancer-formation rodent models used to derive mutational signatures of real-life or research carcinogens and their matches with human cancer signature data [28,37,47,48], additional

Non-mammalian exposure systems

The analyses of genetic and environmental signatures using simpler model organisms including bacteria, yeast, Drosophila, and the nematode C. elegans were thoroughly summarized in the beginnings of the experimental signature modeling research field [54,55]. A recent analysis of close to 3000 genomes of more than 50 C. elegans strains, each deficient for a specific DNA repair gene, compared the derived mutation profiles to those of wild-type counterparts in the context of a dozen of distinct

Prioritization of cancer risk agents for mutagenicity testing

A key aspect for the characterization of mutation spectra associated with exogenous exposures using any of the discussed experimental models is the selection of the tested compounds. The agents prioritized for systematic mutational signature analysis should be selected based on evidence for DNA damage induction and mutagenicity, following the recently established framework for key characteristics of carcinogens [63,64]. Ideally they would consist of chemicals introducing different types of

Future directions

At the current stage of development of the experimental signature modeling field, choosing a reliable, uniform, one-size-fits-all approach remains a challenge. However, the field of experimental signature modeling has solidified in the last several years, providing extensive insights. Summary frameworks are being put in place, recommending experimental designs, protocols and analytical procedures, and discussing advantages and limitations associated with various methods and systems [20,27,34].

Role of funding sources

Experimental studies discussed here and co-authored by M.K. and J.Z. were possible thanks to the funding support from ITMO CANCER INCa – INSERM, Plan Cancer 2015 Grant No. ENV201507 to J.Z.

Disclosure

Views expressed in this review are those of the authors and not necessarily reflect the views, the decisions, policy of the International Agency for Research on Cancer/World Health Organization.

Acknowledgments

The authors gratefully acknowledge the relevant contributions of all team-mates and collaborators and colleagues in the field of experimental modeling of mutational signatures.

References (69)

  • S.F. Brunner et al.

    Somatic mutations and clonal dynamics in healthy and cirrhotic human liver

    Nature

    (2019)
  • P.E. Garcia-Nieto et al.

    The somatic mutation landscape of the human body

    Genome Biol.

    (2019)
  • H. Lee-Six et al.

    The landscape of somatic mutation in normal colorectal epithelial cells

    Nature

    (2019)
  • I. Martincorena et al.

    Somatic mutant clones colonize the human esophagus with age

    Science

    (2018)
  • I. Martincorena et al.

    Tumor evolution. High burden and pervasive positive selection of somatic mutations in normal human skin

    Science

    (2015)
  • K. Yizhak et al.

    RNA sequence analysis reveals macroscopic somatic clonal expansion across normal tissues

    Science

    (2019)
  • A. Yokoyama et al.

    Age-related remodelling of oesophageal epithelia by mutated cancer drivers

    Nature

    (2019)
  • K. Yoshida et al.

    Tobacco smoking and somatic mutations in human bronchial epithelium

    Nature

    (2020)
  • M. Hollstein et al.

    Base changes in tumour DNA have the power to reveal the causes and evolution of cancer

    Oncogene

    (2017)
  • M. Zhivagui et al.

    Experimental and pan-cancer genome analyses reveal widespread contribution of acrylamide exposure to carcinogenesis in humans

    Genome Res.

    (2019)
  • A. Boot et al.

    In-depth characterization of the cisplatin mutational signature in human cell lines and in esophageal and liver tumors

    Genome Res.

    (2018)
  • C. Pleguezuelos-Manzano et al.

    Mutational signature in colorectal cancer caused by genotoxic pks(+) E. coli

    Nature

    (2020)
  • Z. Liu et al.

    Human tumor p53 mutations are selected for in mouse embryonic fibroblasts harboring a humanized p53 gene

    Proc. Natl. Acad. Sci. U. S. A.

    (2004)
  • M. Olivier et al.

    Modelling mutational landscapes of human cancers in vitro

    Sci. Rep.

    (2014)
  • S. Nik-Zainal et al.

    The genome as a record of environmental exposure

    Mutagenesis

    (2015)
  • L. Holzl-Armstrong et al.

    Characterising mutational spectra of carcinogens in the tumour suppressor gene TP53 using human TP53 knock-in (Hupki) mouse embryo fibroblasts

    Methods Protoc.

    (2019)
  • Z. Li et al.

    Genome-wide allelotyping of a new in vitro model system reveals early events in breast cancer progression

    Cancer Res.

    (2002)
  • M.R. Stampfer et al.

    Induction of transformation and continuous cell lines from normal human mammary epithelial cells after exposure to benzo[a]pyrene

    Proc. Natl. Acad. Sci. U. S. A.

    (1985)
  • D.E. Wazer et al.

    Loss of p53 protein during radiation transformation of primary human mammary epithelial cells

    Mol. Cell. Biol.

    (1994)
  • N. Saini et al.

    The impact of environmental and endogenous damage on somatic mutation load in human skin fibroblasts

    PLoS Genet.

    (2016)
  • M. Petljak et al.

    Characterizing mutational signatures in human cancer cell lines reveals episodic APOBEC mutagenesis

    Cell

    (2019)
  • J. Zavadil et al.

    Experimental delineation of mutational signatures is an essential tool in cancer epidemiology and prevention

    Chem. Res. Toxicol.

    (2019)
  • M.N. Huang et al.

    Genome-scale mutational signatures of aflatoxin in cells, mice, and human tumors

    Genome Res.

    (2017)
  • B. Szikriszt et al.

    A comprehensive survey of the mutagenic impact of common cancer cytotoxics

    Genome Biol.

    (2016)
  • 1

    Present address: Department of Pharmacology and Toxicology, American University of Beirut, Beirut 1107 2020, Lebanon.

    View full text