Abstract
Evidence on the association between greenness and telomere length (TL) is insufficient, particularly in middle-aged and elderly populations. Moreover, the modification effect of genetic susceptibility is unknown. We aimed to evaluate the association between residential greenness and TL and further assess the modification effect of genetic susceptibility. We conducted a cross-sectional study among participants in the UK Biobank (2006-2010). Leucocyte telomere length was assessed using quantitative polymerase chain reaction (qPCR). Residential greenness was measured by normalized difference vegetation index within 500 m buffer (NDVI500m) derived from spectral reflectance measurements with color infrared imagery data. Genetic risk score (GRS) was constructed using seven telomere-related SNPs data. Multivariate linear regression models were employed to conduct associational analyses. Among 344,081 participants, NDVI500m was associated with TL in a non-linear trend. In comparison to the lowest quintile, the 4th and 5th quintiles of NDVI500m were significantly associated with longer TL (quintile 4: effect value = 1.18%, 95 %CI: 0.13%, 2.24%; quintile 5: effect value = 2.77%, 95 %CI: 1.71%, 3.84%) after full adjustment. The association was partly mediated by PM2.5 (proportion mediated: 18.04 (3.28, 32.80)). When assessing the joint effects of NDVI500m and GRS, participants with low GRS and high NDVI500m were observed to have the longest TL, and an antagonistic interaction between the fourth quintile of NDVI500m and GRS was noticed as well. In conclusion, residential greenness was associated with longer TL in a non-linear trend. PM2.5 played an important role in mediating this association. The genetic analyses revealed potential antagonistic interaction between residential greenness and genetic susceptibility on TL.</p>