UV-induced G:C-->A:T transitions at the APRT locus of Chinese hamster ovary cells cluster at frequently damaged 5'-TCC-3' sequences

Genomic analysis identifies new drivers and progression pathways in skin basal cell carcinoma

Basal cell carcinoma (BCC) of the skin is the most common malignant neoplasm in humans. BCC is primarily driven by the Sonic Hedgehog (Hh) pathway. However, its phenotypic variation remains unexplained. Our genetic profiling of 293 BCCs found the highest mutation rate in cancer (65 mutations/Mb). Eighty-five percent of the BCCs harbored mutations in Hh pathway genes (PTCH1, 73% or SMO, 20% (P = 6.6 × 10(-8)) and SUFU, 8%) and in TP53 (61%). However, 85% of the BCCs also harbored additional driver mutations in other cancer-related genes. We observed recurrent mutations in MYCN (30%), PPP6C (15%), STK19 (10%), LATS1 (8%), ERBB2 (4%), PIK3CA (2%), and NRAS, KRAS or HRAS (2%), and loss-of-function and deleterious missense mutations were present in PTPN14 (23%), RB1 (8%) and FBXW7 (5%). Consistent with the mutational profiles, N-Myc and Hippo-YAP pathway target genes were upregulated. Functional analysis of the mutations in MYCN, PTPN14 and LATS1 suggested their potential relevance in BCC tumorigenesis.

Evidence for recent, population-specific evolution of the human mutation rate

As humans dispersed out of Africa they adapted to new environmental challenges, including changes in exposure to mutagenic solar radiation. Humans in temperate latitudes have acquired light skin that is relatively transparent to UV light, and some evidence suggests that their DNA damage response pathways have also experienced local adaptation. This raises the possibility that different populations have experienced different selective pressures affecting genome integrity. Here, I present evidence that the rate of a particular mutation type has recently increased in the European population, rising in frequency by 50% during the 40,000-80,000 y since Europeans began diverging from Asians. A comparison of SNPs private to Africa, Asia, and Europe in the 1000 Genomes data reveals that private European variation is enriched for the transition 5'-TCC-3' → 5'-TTC-3'. Although it is not clear whether UV played a causal role in changing the European mutational spectrum, 5'-TCC-3' → 5'-TTC-3' is known to be the most common somatic mutation present in melanoma skin cancers, as well as the mutation most frequently induced in vitro by UV. Regardless of its causality, this change indicates that DNA replication fidelity has not remained stable even since the origin of modern humans and might have changed numerous times during our recent evolutionary history.

Cytosine containing dipyrimidine sites can be hotspots of cyclobutane pyrimidine dimer formation after UVB exposure

Exposure to the UV component of sunlight is the principal factor leading to skin cancer development. Cyclobutane pyrimidine dimers (CPD) are considered to be the most important pre-mutagenic type of DNA damage involved in skin carcinogenesis. To better understand the biological mechanisms of UV carcinogenesis, it is critical to understand the CPD distribution between the four types of dipyrimidine sites. Most of our knowledge regarding CPD distribution comes from in vitro studies or from investigations using UVC, even though we are not naturally exposed to these UV wavelengths. We exposed normal human fibroblasts and purified DNA to UVB. Using ligation-mediated PCR, we quantified the CPD formation at 952 dipyrimidine sites among the PGK1 (phosphoglycerate kinase 1), JUN, HRAS, KRAS, NRAS and TP53 genes. In cellulo, we found a CPD distribution of 27 : 27 : 25 : 21 for TT : CC : TC : CT. This distribution is similar to that observed in vitro. In the analysed genes, we observed some extremely frequently damaged dipyrimidine sites and many of these occurred at potentially frequently mutated sites, i.e. at dipyrimidine sites containing cytosine. Also, most of the frequently damaged dipyrimidine sites in cellulo that are not frequently damaged in vitro are found on TP53 and NRAS. This indicates that many of the frequently damaged dipyrimidine sites in cellulo are on genes frequently mutated in skin cancer. All these results support the view that CPD are the main UVB-induced mutagenic photoproducts and provide evidence of the importance of CPD formation at sites containing cytosine.

Site-specific analysis of UV-induced cyclobutane pyrimidine dimers in nucleotide excision repair-proficient and -deficient hamster cells: Lack of correlation with mutational spectra

Irradiation of cells with UVC light induces two types of mutagenic DNA photoproducts, i.e. cyclobutane pyrimidine dimers (CPD) and pyrimidine (6-4) pyrimidone photoproducts (6-4 PP). To investigate the relationship between the frequency of UV-induced photolesions at specific sites and their ability to induce mutations, we quantified CPD formation at the nucleotide level along exons 3 and 8 of the hprt gene using ligation-mediated PCR, and determined the mutational spectrum of 132 UV-induced hprt mutants in the AA8 hamster cell line and of 165 mutants in its nucleotide excision repair-defective derivative UV5. In AA8 cells, transversions predominated with a strong strand bias towards thymine-containing photolesions in the non-transcribed strand. As hamster AA8 cells are proficient in global genome repair of 6-4 PP but selectively repair CPD from the transcribed strand of active genes, most mutations probably resulted from erroneous bypass of CPD in the non-transcribed strand. However, the relative incidence of CPD and the positions where mutations most frequently arose do not correlate. In fact some major damage sites hardly gave rise to the formation of mutations. In the repair-defective UV5 cells, mutations were almost exclusively C>T transitions caused by photoproducts at PyC sites in the transcribed strand. Even though CPD were formed at high frequencies at some TT sites in UV5, these photoproducts did not contribute to mutation induction at all. We conclude that, even in the absence of repair, large variations in the level of induction of CPD at different sites throughout the two exons do not correspond to frequencies of mutation induction.

RNA Is More UV Resistant than DNA: The Formation of UV-Induced DNA Lesions is Strongly Sequence and Conformation Dependent

DNA and RNA hairpins, which represent well-folded oligonucleotide structures, were irradiated and the amount of damaged hairpins was directly quantified by using ion-exchange HPLC. The types of photoproducts formed in the hairpins were determined by ESI-HPLC-MS/MS experiments. Irradiation of hairpins with systematically varied sequences and conformations (A versus B) revealed remarkable differences regarding the amount of photolesions formed. UV-damage formation is, therefore, a strongly sequence and conformation dependent process.

H-NMR studies of duplex DNA decamer containing a uracil cyclobutane dimer: implications regarding the high UV mutagenecity of CC photolesions

To determine the origin of the UV-specific CC to TT tandem mutation at the CC site, we made a duplex DNA decamer containing a uracil cis-syn cyclobutane dimer (CBD) as the deaminated model of a cytosine dimer. Two-dimensional 1H-NMR spectroscopy studies were performed on this sequence where two adenines (Ade) were opposite to the uracil dimer. Two imino protons of the uracil dimer were found to retain Watson-Crick hydrogen bonding with the opposite Ade, although the 5'-U(NH) of the dimer site showed unusual upfield shift like that of the 5'-T(NH) of the TT dimer, which seemed to be associated with deshielding by the flanking base rather than with reduced hydrogen bonding. (McAteer et al. 1998, J. Mol. Biol. 282:1013-1032). Hydrogen bondings at the dimer site were also supported by detecting typical strong nuclear Overhauser effects (NOE) between two imino protons and the opposite Ade H2 or NH2. But sequential NOE interactions of base protons with sugar protons were absent at the two flanking nucleotides of the 5' side of the uracil dimer and at the intradimer site, contrasting with its thymine analog where sequential NOE was absent only at the A4-T5 step. In addition, NOE cross peak for U5(NH) <--> A4(H2) was detected, although the NOE interactions of U6(NH) with A7(H2) and A17(H2) were not observed in contrast to the thymine dimer duplex. This different local structural alteration may be affected by the induced right-hand twisted puckering mode of cis-syn cyclobutane ring of the uracil dimer in the B-DNA duplex, even though the isolated uracil dimer had left-hand twisted puckering rigidly. In parallel, these observations may be correlated with observed differences in mutagenic properties between cis-syn UU dimer and cis-syn TT dimer.

In situ repair of cyclobutane pyrimidine dimers in skin and melanocytic nevi of cutaneous melanoma patients

The development of cutaneous malignant melanoma (CMM) and its precursor lesions, melanocytic nevi, has been linked to sun exposure. Cyclobutane pyrimidine dimers (CPDs) are the majority of DNA lesions induced by sun exposure. In our study, we investigated if CMM patients have impaired ability to repair CPDs in skin as well as in melanocytic nevi. The repair kinetics were followed up to 3 weeks after exposure to 40 mJ/cm(2) of solar simulating radiation. Altogether 12 CMM patients and 10 healthy controls were included in our study. Buttock skin biopsies were taken at 0 hr, 48 hr and 3 weeks after UV exposure, whereas melanocytic nevi and surrounding skin biopsies were taken only at 0 hr and 3 weeks. The CPD levels were measured by a (32)P-postlabeling method. The results showed that the repair rate of CPDs in neither the skin nor the nevi was significantly different between the CMM patients and the control group. For both groups, the repair rate of TT = C was faster than that for TT = T. The important finding is that about 10% of the initial TT = T damage remained unrepaired after 3 weeks, and was detectable in normal epidermis as well as in nevi of all subjects. We also found that the amount of TT = C and TT = T at 0 hr in nevi was significantly lower than that in surrounding skin (Wilcoxon rank sum test, p < 0.05).

Repair of the main UV-induced thymine dimeric lesions within Arabidopsis thaliana DNA: evidence for the major involvement of photoreactivation pathways

The UV-B induced formation of thymine cis-syn cyclobutane dimer and related (6-4) photoproduct was monitored within DNA of cultured cells and plants of Arabidopsis thaliana. This was achieved using a sensitive and accurate HPLC-tandem mass spectrometry assay. It was found that the cyclobutane pyrimidine dimer was formed in a ninefold higher yield than the (6-4) photoproduct. The removal of the lesions was then studied by incubating irradiated cells either in the darkness, under visible light or upon exposure to UV-A radiation. Dark repair of both cyclobutane dimers and (6-4) photoproducts was found to be very ineffective. In contrast, a rapid decrease in the level of photoproducts was observed when UV-B-irradiated cells were exposed to UV-A and, to a lesser extent, to visible light. The removal of (6-4) adducts was found to occur more efficiently. These results strongly suggest that repair of UV-induced photolesions in plants is mainly mediated by photolyases.

Direct and indirect effects of UV radiation on DNA and its components

In this survey, emphasis was placed on the main photoreactions of nucleic acid components, involving both direct and indirect effects. The main UVB- and UVA-induced DNA photoproducts, together with the mechanisms of their formation, are described. Information on the photoproduct distribution within cellular DNA is also provided, taking into account the limitations of the different analytical methods applied to monitor the formation of the DNA damage. Thus, the formation of the main DNA dimeric pyrimidine lesions produced by direct absorption of UVB photons was assessed using a powerful HPLC-tandem mass spectrometry assay. In addition, it was found that UVA photooxidation damage mostly involves the guanine residues of cellular DNA as the result of singlet oxygen generation by still unknown endogenous photosensitizers.

Effect of a dCTP:dTTP pool imbalance on DNA replication fidelity in Friend murine erythroleukemia cells

Nucleotide pool imbalances have been reported to affect the fidelity of DNA replication and repair in prokaryotic and eukaryotic cells. We have reported previously that the mutagen-hypersensitive thymidine kinase (TK)-deficient Friend erythroleukemia (FEL) cells (subclones 707BUF and 707BUE), have a more than sixfold increase in the dCTP:dTTP pool ratio when compared to that of wild-type, TK-positive (TK(+)) clone 707 cells. In this study we present the results of an investigation of the effect of the dCTP:dTTP pool imbalance on the accuracy of DNA replication within 707BUF cells. We examined the spontaneous mutation spectra occurring at the adenine phosphoribosyltransferase (aprt) locus within clone 707 (TK(+)) and 707BUF (TK(-)) FEL cells. Mutations recovered at the aprt locus in FEL cells comprised: base substitutions (43:73), frameshifts (14:13.5), and deletions (43:13.5) [clone 707 (TK(+)):707BUF (TK(-)), respectively, expressed as percentages]. A comparison of the mutation spectra obtained for the two cell lines did not reveal any significant increase in misincorporation of dCTP, the nucleotide in excess, in 707BUF (TK(-)) cells, during DNA replication synthesis. These data suggest that the dCTP:dTTP pool imbalance does not alter the fidelity of DNA replication synthesis in 707BUF (TK(-)) FEL cells. Rather, the predominance of GC --> AT transitions (53%) in the 707BUF (TK(-)) spectrum may reflect a reduced efficiency of repair by uracil DNA glycosylase of uracil residues within these cells.

Mutation induction by different types of radiation at the Hprt locus

Mutation induction at the Hprt locus in Chinese hamster cells was studied after exposure to ultraviolet light, X-rays and alpha particles. While mutant frequency as a function of dose or fluence followed a linear-quadratic relationship with UV and X-rays, it showed a linear dependence for alpha particles. If mutant frequency is plotted vs. the logarithm of surviving fraction, a linear relationship is found in all cases although with different slopes. These are about equal with the two types of ionising radiations but about 10 times larger for UV. They can be used as a measure of mutagenic potential and are termed mutagenicity. It is shown that this parameter is correlated with the maximum of mutant yield, i.e., the number of mutants per cell at risk. It is concluded from this analysis that the maximum mutant yield is always found at doses or fluences which lead to 37% survival irrespective of the kind of radiation. If mutation induction is measured in X-irradiated cells after pre-exposure to UV, mutant frequency is higher than expected on the basis of independent action of the two radiations. Deletion spectra were determined by using multiplex polymerase chain reaction. It was found that the background of spontaneous mutants varied considerably and showed frequently repetitive patterns, presumably because of clonal expansion of pre-formed mutants. UV-induced mutants did not contain any deletions, while those with both X-rays and alpha particles the majority displayed partial and total deletions. Based on a total number of 134 X-ray- and 192 alpha-induced mutants, it is concluded that the total fraction of mutant clones without deletions (partial or total) is about 40% for X-rays and only about 20% for alpha-particles.

Mutation spectra of complex mixtures

Mutation spectra of complex mixtures or simple binary mixtures have been determined thus far only in Salmonella. This review summarizes these studies, which have involved a variety of complex mixtures, chemical fractions of the mixtures, and single compounds that are representative of the primary chemical classes within the mixtures. For the particulate organics from urban air and municipal waste incinerator emissions, cigarette smoke condensate, and organic extracts of chlorinated drinking water, the mutation spectrum of the mixture reflected the dominance of one or a few chemical classes within the mixture. The mutation spectra of sunlight and cigarette smoke in Salmonella were similar to those found in the p53 gene of skin or lung tumors in people associated with exposure to these two mutagens. These data provide possible mechanistic links between the types of mutations induced by complex environmental mutagens in an experimental organism (Salmonella) and the types of mutations found in humans exposed to the same environmental mutagens. The primary class of mutation produced by many mutagens within a chemical class (and even among different classes) is frequently the same. These studies indicate that, to some extent, complex mixture-induced mutation spectra can be modeled by the predominant mutagenic chemical class in that mixture and even by a single model compound in that chemical class. Such information should be useful in the interpretation of mutation spectra in molecular epidemiological studies where the exposure is primarily to a complex mixture.

Correlation of UV-induced mutational spectra and the in vitro damage distribution at the human hprt gene

We have determined the in vitro DNA damage distribution induced by 254 mm UV in the human hprt gene. The sequence-specific nature of the DNA damage for both main classes of UV-induced photoproducts, i.e., cyclobutane pyrimidine dimers (CPDs) and the pyrimidine <6-4> pyrimidone photoproducts (64PyPy), was evaluated. Utilizing an automated DNA sequencer plus auxiliary software, semi-automated analyses were performed for peak quantitation and retention-time to sequence-position correlation. 64PyPy were predominantly formed at 5'-YTC-3' sites (p < 0.02; where Y = C,T). The effect of the 3'flanking nucleotide on the 64PyPy formation at 5'-TC-3' sites was 64PyPy at 5'-TCT-3' sites were induced at lower frequencies compared to 5'-TCM-3' sites (where M = A or C; p < 0.03). No effect of flanking nucleotides was detected for CPDs recovered at 5'-TT-3' sites. Sites of mutations in the hprt gene were compared to the sites of DNA damage. Two regions of frequently mutated nucleotides corresponded to sites of high deposition of damage. The two sites either had a high frequency of CPDs or 64PyPy, which implicated both types of photoproducts as premutagenic lesions.

A role for ultraviolet A in solar mutagenesis

It is well established that exposure to solar UVB (290-320 nm) gives rise to mutations in oncogenes and tumor suppressor genes that initiate the molecular cascade toward skin cancer. Although UVA (320-400 nm) has also been implicated in multistage photocarcinogenesis, its potential contribution to sunlight mutagenesis remains poorly characterized. We have determined the DNA sequence specificity of mutations induced by UVB (lambda > 290 nm), and by UVA (lambda > 350 nm), at the adenine phosphoribosyltransferase locus of Chinese hamster ovary cells. This has been compared to results previously obtained for stimulated sunlight (lambda > or = 310 nm) and 254-nm UVC in the same gene. We demonstrate that T-->G transversions, a generally rare class of mutation, are induced at high frequency (up to 50%) in UVA-exposed cells. Furthermore, this event comprises a substantial proportion of the simulated sunlight-induced mutant collection (25%) but is significantly less frequent (P < 0.05) in cells irradiated with either UVB (9%) or UVC (5%). We conclude that the mutagenic specificity of broad-spectrum solar light in rodent cells is not determined entirely by the UVB component and that UVA also plays an important role.

Mutation spectra in Salmonella of sunlight, white fluorescent light, and light from tanning salon beds: induction of tandem mutations and role of DNA repair

We evaluated the mutagenicity of sunlight (SUN), uncovered coolwhite fluorescent light (FLR), and light from a tanning salon bed (TAN) at the base-substitution allele hisG46 of Salmonella in four DNA repair backgrounds (wild type, uvrB, pKM101, and uvrB + pKM101). Approximately 80% of the radiation emitted by TAN was within the ultraviolet (UV) range, whereas only approximately 10% of the SUN and approximately 1% of the FLR radiation was UV. TAN emitted similar amounts of UVA and UVB, whereas SUN emitted 50-60x and FLR emitted 5-10x more UVA relative to UVB. Based on total dose (UV + visible), the mutagenic potency ranking was TAN > FLR > SUN. Using colony probe hybridization and PCR/DNA sequence analysis, approximately 3000 revertants were analyzed to determine the mutational specificity of the three light sources. The mutation spectra and those induced by 254-nm UV had common features. The uvrB mutation enhanced the mutagenicity of the environmental UV sources more (20-216x) than did the pKM101 plasmid (approximately 20x) relative to wild type DNA repair. All light sources induced equal proportions of transitions and transversions in excision repair-proficient strains, but they induced more transitions relative to transversions in uvrB-containing strains. The majority of the mutations were G.C-->A.T transitions that were induced equally frequently at the first or second position of the CCC codon of the hisG46 allele in all strains except TA1535 (uvrB), where SUN and FLR induced transitions preferentially at the first position, and TAN induced them preferentially at the second position. Identified or presumptive multiple mutations, which constituted the only mutational class enhanced by all three light sources in the presence of uvrB and pKM101 either alone or together, accounted for 3-5% of the induced mutations in the plasmid-containing strains, and their increases (38-82-fold) in TA100 (uvrB, pKM101) were the highest of any mutational class. Of the TAN-induced multiple mutations, 83% (19/23) were CC-->TT tandem transitions. These results show that exposures to the nonsolar environmental UV sources FLR and TAN produce mutations similar to those produced by SUN, a known carcinogen.