Ambient air pollution and lung cancer risk among never-smokers in the Women's Health Initiative.

BACKGROUND: Ambient air pollution is classified as a human carcinogen by the International Agency for Research on Cancer (IARC). However, epidemiologic studies supporting this classification have focused on lung cancer mortality rather than incidence, and spatial and temporal resolutions of exposure estimates have varied considerably across studies. METHODS: We evaluated the association of outdoor air pollution and lung cancer incidence among never-smoking participants of the Women's Health Initiative (WHI) study, a large, US-based cohort of postmenopausal women (N = 65,419; 265 cases). We used geospatial models to estimate exposures to fine particulate matter (PM2.5) and nitrogen dioxide (NO2) based on residential addresses at baseline and throughout follow-up. We also characterized exposures to traffic-related air pollution by proximity to major roadways. We estimated hazard ratios (HRs) for the risk of lung cancer in association with these exposure metrics using Cox proportional hazards regression models. RESULTS: No compelling associations of PM2.5 and NO2 exposures with lung cancer risk were observed. An increased risk of lung cancer was observed when comparing those individuals with residences <50 versus ≥200 meters from a primary limited access highway (HR = 5.23; 95% confidence interval = 1.94, 14.13). CONCLUSIONS: Our results do not exclude lung cancer risk estimates observed in association with PM2.5 and NO2 exposures identified in previous studies. Our results suggest that residential proximity to major roadways may be a proxy for carcinogenic exposures not correlated with PM2.5 or NO2 levels. New studies of air pollution and lung cancer incidence should characterize additional aspects of proximity to major roadways.

DNA methylation supports intrinsic epigenetic memory in mammalian cells.

We have investigated the role of DNA methylation in the initiation and maintenance of silenced chromatin in somatic mammalian cells. We found that a mutated transgene, in which all the CpG dinucleotides have been eliminated, underwent transcriptional silencing to the same extent as the unmodified transgene. These observations demonstrate that DNA methylation is not required for silencing. The silenced CpG-free transgene exhibited all the features of heterochromatin, including silencing of transcriptional activity, delayed DNA replication, lack of histone H3 and H4 acetylation, lack of H3-K4 methylation, and enrichment in tri-methyl-H3-K9. In contrast, when we tested for transgene reactivation using a Cre recombinase-mediated inversion assay, we observed a marked difference between a CpG-free and an unmodified transgene: the CpG-free transgene resumed transcription and did not exhibit markers of heterochromatin whereas the unmodified transgene remained silenced. These data indicate that methylation of CpG residues conferred epigenetic memory in this system. These results also suggest that replication delay, lack of histone H3 and H4 acetylation, H3-K4 methylation, and enrichment in tri-methyl-H3-K9 are not sufficient to confer epigenetic memory. We propose that DNA methylation within transgenes serves as an intrinsic epigenetic memory to permanently silence transgenes and prevent their reactivation.