Faculty Bibliography

Humans are frequently exposed to acrylamide, a probable human carcinogen found in commonplace sources such as most heated starchy foods or tobacco smoke. Prior evidence has shown that acrylamide causes cancer in rodents, yet epidemiological studies conducted to date are limited and, thus far, have yielded inconclusive data on association of human cancers with acrylamide exposure. In this study, we experimentally identify a novel and unique mutational signature imprinted by acrylamide through the effects of its reactive metabolite glycidamide. We next show that the glycidamide mutational signature is found in a full one-third of approximately 1600 tumor genomes corresponding to 19 human tumor types from 14 organs. The highest enrichment of the glycidamide signature was observed in the cancers of the lung (88% of the interrogated tumors), liver (73%), kidney (>70%), bile duct (57%), cervix (50%), and, to a lesser extent, additional cancer types. Overall, our study reveals an unexpectedly extensive contribution of acrylamide-associated mutagenesis to human cancers.

Purpose: Acrylamide, a probable human carcinogen (Group 2A), is ubiquitously present in the human environment. The compound is found in heated starchy foods, coffee, as well as cigarette smoke. Humans are also exposed to acrylamide occupationally. The carcinogenicity of acrylamide has been attributed to the effects of glycidamide, its reactive and mutagenic epoxide metabolite shown to be an inducer of rodent tumors at various anatomical sites. Covalent adducts with haemoglobin were reported in humans exposed to acrylamide, and DNA adducts in experimental systems. Methods: In order to characterize the pre-mutagenic DNA lesions and global mutation spectra induced by acrylamide and glycidamide, we combined DNA-adduct and wholeexome sequencing analyses in an established exposureclonal immortalization system based on mouse embryonic fibroblasts. Results: Sequencing and computational analysis of cell clones immortalized after exposure revealed a unique mutational signature of glycidamide, characterized by predominant T:A>A:T transversions, T:A>C:G transitions and C:G>A:T transversions in specific trinucleotide contexts, and with significant transcription strand bias. Computational interrogation of the human pan-cancer genome sequencing data (PCAWG data set) using experimentally derived glycidamide signature revealed the presence of the glycidamide signature in lung adenocarcinomas, lung squamous cell carcinomas and liver cancer, manifesting as a sub-component of the COSMIC tobacco smoking-related signature 4, as well as possibly associated with dietary intake of acrylamide. Conclusions: Our study demonstrates how the controlled experimental characterization of specific genetic damage associated with glycidamide exposure facilitates identifying corresponding patterns in cancer genome data, thereby underscoring how mutational signature laboratory experimentation contributes to the elucidation of cancer causation

Introduction The analysis of somatic mutational signatures in a single tumour can reveal clues on the natural history of this tumour, including past exposures to carcinogens. Characterising the mutational signatures of potential carcinogens used in industry may thus help identify cancers that are due to occupational exposure. Here, we used experimental models to define the mutational signatures of dichloromethane (DCM) and 1,2- dichloropropane (DCP), two solvents suspected to cause cholangiocarcinomas in occupational settings. Material and methods Experimentally induced tumours and matching normal tissues from mice exposed to DCP (by gavage), DCM (by gavage or inhalation) or DCP +DCM (by gavage), and spontaneous tumours from vehicle/sham-exposed mice were analysed by whole-exome sequencing. Somatic mutation calling was performed to define exome-wide mutation patterns induced by DCP and DCM in these assays. Mutation data obtained in biliary tract cancers of workers in the printing industry who have been exposed to DCP and DCM as single agents or as mixtures, as well as with public somatic mutation data from biliary tract cancers were mined for the presence of the experimentally defined DCP and DCM mutational signatures. Results and discussions Liver tumours from DCP and DCM exposed mice had distinct somatic mutations patterns compared to spontaneous liver tumours from vehicle/sham-exposed mice. While mutations in DCP-exposed mice were dominated by C:G>T:A transitions, the most frequent types of mutations in DCM-exposed mice were T:A>A:T and T:A>C:G substitutions. An average of 10.3 somatic mutations was observed per Mb in tumours from DCM-exposed mice, approximately 3- fold higher than in DCP-exposed or sham-treated mice. The mutation patterns found in DCP-exposed mice, but not DCMexposed mice, presented some similarities with the mutational signature observed in cholangiocarcinomas of Japanese patients with occupational DCP/DCM exposure history. Conclusion These results show that the analysis of tumour genomes from mouse carcinogenicity assays can support the characterisation of mutational signatures of carcinogenic compounds relevant to human exposures

Cutaneous beta human papillomavirus (HPV) types are suspected to be involved, together with ultraviolet (UV) radiation, in the development of non-melanoma skin cancer (NMSC). Studies in in vitro and in vivo experimental models have highlighted the transforming properties of beta HPV E6 and E7 oncoproteins. However, epidemiological findings indicate that beta HPV types may be required only at an initial stage of carcinogenesis, and may become dispensable after full establishment of NMSC. Here, we further investigate the potential role of beta HPVs in NMSC using a Cre-loxP-based transgenic (Tg) mouse model that expresses beta HPV38 E6 and E7 oncogenes in the basal layer of the skin epidermis and is highly susceptible to UV-induced carcinogenesis. Using whole-exome sequencing, we show that, in contrast to WT animals, when exposed to chronic UV irradiation K14 HPV38 E6/E7 Tg mice accumulate a large number of UV-induced DNA mutations, which increase proportionally with the severity of the skin lesions. The mutation pattern detected in the Tg skin lesions closely resembles that detected in human NMSC, with the highest mutation rate in p53 and Notch genes. Using the Cre-lox recombination system, we observed that deletion of the viral oncogenes after development of UV-induced skin lesions did not affect the tumour growth. Together, these findings support the concept that beta HPV types act only at an initial stage of carcinogenesis, by potentiating the deleterious effects of UV radiation.

The information on candidate cancer driver alterations available from public databases is often descriptive and of limited mechanistic insight, which poses difficulties for reliable distinction between true driver and passenger events. To address this challenge, we performed in-depth analysis of whole-exome sequencing data from cell lines generated by a barrier bypass-clonal expansion (BBCE) protocol. The employed strategy is based on carcinogen-driven immortalization of primary mouse embryonic fibroblasts and recapitulates early steps of cell transformation. Among the mutated genes were almost 200 COSMIC Cancer Gene Census genes, many of which were recurrently affected in the set of 25 immortalized cell lines. The alterations affected pathways regulating DNA damage response and repair, transcription and chromatin structure, cell cycle and cell death, as well as developmental pathways. The functional impact of the mutations was strongly supported by the manifestation of several known cancer hotspot mutations among the identified alterations. We identified a new set of genes encoding subunits of the BAF chromatin remodeling complex that exhibited Ras-mediated dependence on PRC2 histone methyltransferase activity, a finding that is similar to what has been observed for other BAF subunits in cancer cells. Among the affected BAF complex subunits, we determined Smarcd2 and Smarcc1 as putative driver candidates not yet fully identified by large-scale cancer genome sequencing projects. In addition, Ep400 displayed characteristics of a driver gene in that it showed a mutually exclusive mutation pattern when compared with mutations in the Trrap subunit of the TIP60 complex, both in the cell line panel and in a human tumor data set. We propose that the information generated by deep sequencing of the BBCE cell lines coupled with phenotypic analysis of the mutant cells can yield mechanistic insights into driver events relevant to human cancer development.Oncogene advance online publication, 10 July 2017; doi:10.1038/onc.2017.215.

Aflatoxin B1 (AFB1) is a mutagen and IARC (International Agency for Research on Cancer) Group 1 carcinogen that causes hepatocellular carcinoma (HCC). Here, we present the first whole-genome data on the mutational signatures of AFB1 exposure from a total of >40,000 mutations in four experimental systems: two different human cell lines, in liver tumors in wild-type mice, and in mice that carried a hepatitis B surface antigen transgene-this to model the multiplicative effects of aflatoxin exposure and hepatitis B in causing HCC. AFB1 mutational signatures from all four experimental systems were remarkably similar. We integrated the experimental mutational signatures with data from newly sequenced HCCs from Qidong County, China, a region of well-studied aflatoxin exposure. This indicated that COSMIC mutational signature 24, previously hypothesized to stem from aflatoxin exposure, indeed likely represents AFB1 exposure, possibly combined with other exposures. Among published somatic mutation data, we found evidence of AFB1 exposure in 0.7% of HCCs treated in North America, 1% of HCCs from Japan, but 16% of HCCs from Hong Kong. Thus, aflatoxin exposure apparently remains a substantial public health issue in some areas. This aspect of our study exemplifies the promise of future widespread resequencing of tumor genomes in providing new insights into the contribution of mutagenic exposures to cancer incidence.

Mutation spectra in cancer genomes provide information on the disease aetiology and the causality underlying the evolution and progression of cancer. Genome-wide mutation patterns reflect the effects of mutagenic insults and can thus reveal past carcinogen-specific exposures and inform hypotheses on the causative factors for specific cancer types. To identify mutation profiles in human cancers, single-gene studies were first employed, focusing mainly on the tumour suppressor gene TP53. Furthermore, experimental studies had been developed in model organisms. They allowed the characterization of the mutation patterns specific to known human carcinogens, such as polycyclic aromatic hydrocarbons or ultraviolet light. With the advent of massively parallel sequencing, mutation landscapes become revealed on a large scale, in human primary tumours and in experimental models, enabling deeper investigations of the functional and structural impact of mutations on the genome, including exposure-specific base-change fingerprints known as mutational signatures. These studies can now accelerate the identification of aetiological factors, contribute to carcinogen evaluation and classification and ultimately inform cancer prevention measures.

Exposure to aristolochic acid (AA, IARC Group 1 carcinogen) present in some traditional herbal medicines leads to aristolochic acid nephropathy (AAN), often complicated by development of multiple urothelial carcinomas of sequential onset. We used genome-scale mutational signature analysis of multiple urinary tract tumors of AAN cases from a unique patient group to determine the relationships of the patients' late-onset second cancers to the AA exposure as well as to the first cancers. Aristolactam-DNA adduct-positive AAN patients (n=4) who developed cancer within 8 years following the initial exposure to AA were chosen for analysis of their first cancers (upper tract urothelial carcinomas, UTUC) and second cancers of delayed onset (1-9 years after first-cancer diagnosis, involving the bladder or ureteral meatus). All patients had received a kidney transplant before developing the second cancers and had a functional renal graft prior to prophylactic nephroureterectomy. Genomic DNAs were isolated from FFPE sections of the renal cortex and from the upper and lower tract tumors of each patient using laser capture micro-dissection or macrodissection of the tumor areas. Low-coverage (~15x) exome 100-bp paired-end sequencing was performed using Illumina HiSeq2500. Customized variant calling identified somatic variants absent in non-tumor tissues. Non-negative matrix factorization was applied to extract mutational signatures in the tumor tissues. In all cases, we established the mutational signature of AA (the COSMIC signature 22) in the first UTUC as well as second cancers involving the bladder or lower ureter (meatus). Additionally, the first and second cancers harbored considerable overlaps in exposure-specific (A>T) somatic mutations. This finding provides evidence that the delayed onset of bladder urothelial carcinomas in AAN patients is likely due to distal seeding of cancer cells originating from the primary UTUC tumors. Our first-of-its-kind study addresses the risk as well as mechanistic factors leading to the second, lateonset bladder urothelial carcinomas following kidney transplantation and primary UTUC development. Our results underline the importance of long-term bladder follow-up in high-risk populations with established or suspected AA exposure

Next-generation sequencing (NGS) technology has demonstrated that the cancer genomes are peppered with mutations. Although most somatic tumour mutations are unlikely to have any role in the cancer process per se, the spectra of DNA sequence changes in tumour mutation catalogues have the potential to identify the mutagens, and to reveal the mutagenic processes responsible for human cancer. Very recently, a novel approach for data mining of the vast compilations of tumour NGS data succeeded in separating and precisely defining at least 30 distinct patterns of sequence change hidden in mutation databases. At least half of these mutational signatures can be readily assigned to known human carcinogenic exposures or endogenous mechanisms of mutagenesis. A quantum leap in our knowledge of mutagenesis in human cancers has resulted, stimulating a flurry of research activity. We trace here the major findings leading first to the hypothesis that carcinogenic insults leave characteristic imprints on the DNA sequence of tumours, and culminating in empirical evidence from NGS data that well-defined carcinogen mutational signatures are indeed present in tumour genomic DNA from a variety of cancer types. The notion that tumour DNAs can divulge environmental sources of mutation is now a well-accepted fact. This approach to cancer aetiology has also incriminated various endogenous, enzyme-driven processes that increase the somatic mutation load in sporadic cancers. The tasks now confronting the field of molecular epidemiology are to assign mutagenic processes to orphan and newly discovered tumour mutation patterns, and to determine whether avoidable cancer risk factors influence signatures produced by endogenous enzymatic mechanisms. Innovative research with experimental models and exploitation of the geographical heterogeneity in cancer incidence can address these challenges.Oncogene advance online publication, 6 June 2016; doi:10.1038/onc.2016.192.