Tumor suppressor p53 (TP53) at the crossroads of the exposome and the cancer genome.

T umor suppressor p53 (TP53) was discovered more than two decades ago to be frequently mutated in diverse types of human cancer (1, 2), and the mutation frequency was shown to be even higher by whole-exome sequencing of the cancer genome (3, 4). In PNAS (5), Chen et al. show a dose dependency between exposure to aristolochic acid (AA), a known human carcinogen and nephrotoxin, and the TP53 mutational spectrum in the cancer genome of upper urinary tract urothelial carcinomas. AA is metabolically activated by enzymatic nitroreduction to form aristolactam-DNA (AL-DNA) adducts. These adducts are poorly removed by nucleotide excision repair, persist for many years in human tissues, and cause mutations in TP53 and other cancer-related genes, including HRAS and FGFR3. Different carcinogens are known to cause specific types of mutations in TP53 and these are collectively referred to as the TP53 mutation signature (2, 6). For example, ingesting aflatoxin B1, a toxin produced by the mold Aspergilla flavus, causes G:C to T:A transversion mutations at codon 249 of TP53 in hepatocellular carcinoma in geographic areas with endemic hepatitis B virus infection (7, 8). Ultraviolent light from sun exposure causes tandem CC to TT transitions in TP53 in nonmelanoma skin carcinoma (9). G:C to T:A transversions in TP53 are caused by tobacco smoke in human lung cancer (10). The results of the molecular epidemiologic study by Chen et al. demonstrate that exposure to AA causes A:T to T:A transversions in TP53 and exposure to AA may account for over 50% of urothelial carcinomas in Taiwan. Most TP53 missense mutations are loss-of-function variants, but some are gain-of-function variants (11). The TP53 mutation signatures are mainly determined by DNA base selectivity of the electrophillic metabolite of the carcinogen, which can also be influenced by the surrounding DNA bases; e.g., AA has a preference for 5′AG (acceptor) splice sites, by DNA repair selectivity and finally by clonal cell selection by the biological activity of gain-offunction TP53 mutant proteins. AA is produced by Aristolochia and contaminates grains used for food and is frequently included in herbal remedies that were initially discovered to be toxic in Belgian women who developed renal failure and urothelial carcinoma after taking herbal remedies for weight loss and later linked to Balkan endemic nephropathy (reviewed in ref. 5). Chinese medicines using Aristolochia herbs are used internationally and the Food and Drug Administration in the United States has warned both manufacturers and consumers of the public health hazards. Nevertheless, these herbal remedies are readily available via the internet. A major concern is that AA-induced cancer will occur from past exposures to AA and the persistence of its promutagenic DNA adducts due to the long tumor latency. However, less than 5% of the AA-exposed population develops cancer or renal insufficiency, so that susceptibility factors may play a role. Effective biomarkers of cancer risk and development are needed to identify individuals at high risk of renal failure and urothelial carcinoma. Tumor-specific DNA mutations have been shown to be cancer biomarkers (12, 13); therefore DNA with TP53-related AA mutations quantitatively measured in urine or plasma of AA-exposed individuals may have promise as biomarkers of cancer risk. In addition to differences in exposure to carcinogens, germ-line polymorphisms in the genes encoding the enzymes that metabolize chemical carcinogens can cause interindividual variation in formation of DNA adducts and cancer risk (14). Analysis of the cancer genome of AA-induced upper urinary tract urothelial carcinomas may identify cancer-initiating and -progressing “driver” genes with the A→T transversions that may be found in other cancer types that contain AL-DNA adducts and may be induced by AA. The exposome has recently joined the “omic” vocabulary of the scientific community and is the ability to precisely measure exposure, exposure biomarkers, and biological effects of exposures increasing the risk of disease (15, 16). Both the external and the internal environment is in the domain of the exposome (Fig. 1). The measurement of carcinogen–DNA adducts and the TP53 mutation signature, Fig. 1. Analysis of the exposome (external and internal environment) and the cancer genome (somatic and germ-line mutations and epigenetic changes, e.g., DNA methylation and noncoding RNAs) will improve the understanding carcinogenesis, cancer therapy, and cancer prevention.