The key finding of our study is that in a postulated chain of events, a hierarchical relationship was outlined between different PAH exposures (diet, indoor, outdoor, traffic, residential) as well as smoking that, acting through anti-B[a]PDE–DNA adduct formation, affected LTL and LmtDNAcn. This opens up the possibility that carcinogenic PAHs may also be gerontogenic for the general population (in particular males), by speeding up biological indicators of aging. In fact, LTL and LmtDNAcn decreased with anti-B[a]PDE–DNA formation, particularly in males. In accordance with our previous findings, we found that anti-B[a]PDE–DNA was significantly increased with PAH exposure and active smoking, whereas the presence of the detoxifying GSTM1 decreased adduct levels. Lastly, male gender was also associated with decreased GSTM1 levels, to a lesser extent.
The direct negative relationship between LTL and DNA adducts is in line with our study in coke oven workers, who are extremely exposed to work-related PAH carcinogens [17], and thus would suggest that adduct formation might have a direct role in LTL erosion. In fact, anti-B[a]PDE–DNA adducts verified by means of HPLC-Fluorescence, are the consequences of the stereo-selective link between B[a]P and the exocyclic N2 of guanine, which is considered to be the crucial step in B[a]P carcinogenicity [9]. Furthermore, telomeres (as triple-G-holding chains) are a susceptible point for damage by such BPDE genotoxic metabolites. In fact, double-strand breakdowns and interference with the replication fork, which occur with extremely damaged telomeric bases, might directly generate telomere attrition [25]. Furthermore, anti-B[a]PDE–DNA creation and subsequent telomere erosion may be regulated by a reduction in B[a]P detoxification, attributable to the presence of a specific GSTM1 detoxifying polymorphism. The formation of adducts within the proteins of the telomere-protecting complexes (including those regulated by GSTM1), can also be thought of as an extra outcome accounting for shorter LTL. In fact, altered mRNA expression levels of end-binding proteins associated with telomere injury were recently found in individuals with PAH exposure [26].
-
In this study, smoking was found to significantly increase LTL. Several studies, including large surveys such as those conducted by Bischoff et al. [27]and Cassidy et al. [28] were unable to substantiate the negative association between LTL and smoking discovered by others [29]. These inconsistent results seem to detect a moderate effect of smoking on LTL, if any, which may simply not be measurable. In-vitro experiments have shown that telomere length increases in younger inflammatory T cells [30] over the course of inflammation, which is a key process in mediating smoking-related health effects [31]. Therefore, the higher LTL in smokers that we discovered in the current study could be ascribed to the recall of younger inflammatory cells from the bone marrow to the blood circulation, in response to inflammatory signals [30]. Likewise, short-term exposure to PM was associated with a quick rise in blood LTL [32, 33]. Taken together, these observations indicate that short-term smoking exposure may generate a rapid increase in LTL, which may then contribute to sustaining the inflammatory mechanisms related to adverse health effects.
In the current study we also found that LmtDNAcn significantly decreased with anti-B[a]PDE–DNA. As for LTL, PAH exposure appeared to have no direct effect on decreased mtDNAcn but an indirect effect through the mediation of anti-B[a]PDE–DNA. Our results agree with those by Pieters et al. [34]who found a reduction of LmtDNAcn associated with low chronic exposure to PAHs in house dust during wintertime. The ability of B[a]P to reduce mtDNA content was also validated in vitro in human TK6 cells [34]. In view of the genotoxic capability of B[a]P, this carcinogen may also have an impact on mtDNA dynamics. The lipophilic nature of B[a]P and its metabolite anti-B[a]PDE, combined with the very high ratio of lipid/DNA in mitochondria, expedites the rate at which these molecules enter mitochondria. Moreover, anti-B[a]PDE possesses 40 to 90-fold higher affinity for mtDNA than for nuclear DNA [12]. Compared with nDNA, mtDNA has fewer protective histones and DNA repair ability, and is therefore extremely vulnerable to DNA injury. Consequently, the role of mtDNA-bound anti-BPDE could be a valuable fraction of the total cellular cargo of DNA adducts. Our results are consistent with a study that reported a significant reduction in LmtDNAcn in the lungs of long-term heavy cigarette smokers. Conversely, our previous study in non-smoking male workers with high exposure to PAHs (> 3 µmol 1-pyrenol/mol creatinine) detected significantly higher LmtDNAcn in these subjects, compared with controls [7]. It has been proposed that the oxidative stress, triggered by exposure to PAHs, has a dual impact on mitochondrial DNA content. Indeed, mild stress can promote mitochondrial DNA production and increased mitochondria number in order to sustain the higher respiratory needs of the cell and, as such, maintain cell viability [35]. However, excessive oxidative stress spawned by tobacco smoke might instead lead to reduced (or no) synthesis of mtDNA, as tobacco smoke contains many toxic, carcinogenic and mutagenic compounds, as well as stable and unstable free radicals and reactive oxygen species (ROS), with the potential for oxidative DNA damage. Such damage may then eventually lead to cell senescence or cell death.
Next to active smoking, the most significant determinants of PAH exposure leading to increased anti-B[a]PDE–DNA, were “diet” and “indoor” exposure, while the least significant was “outdoor” exposure. Interestingly, much attention has been paid to the detrimental effects of outdoor pollution, even though the majority of individuals spend most of their time inside, with indoor air that could be even more contaminated than outdoor air. Indoor air pollution was actually listed as one of the ten leading risk factors for the worldwide burden of disease [36]. Similarly, Delgado et al. [37] clearly showed that exposure to wood smoke (without radon, which was the confounding factor in the Chinese studies) is a risk for lung cancer in non-smokers. Our findings show that indoor pollution (mainly due to wood burning) could be a considerable font of B[a]P intake via inhalation that, by anti-B[a]PDE–DNA adduct formation, affects LTL and LmtDNAcn and thus should be thought of as a potential risk factor for lung cancer. Conversely, less attention has been given to risk analysis of gastrointestinal exposure to PAHs route. One study has shown that human exposure to environmental carcinogens such as PAHs occurs primarily from dietary sources [38], and epidemiological studies have revealed that a large proportion of human malignances is attributable, at least in part, to nutritional factors [39]. Furthermore, a clear relationship between aromatic-DNA adducts and food (e.g., grilled hamburgers) was found in subjects with low professional exposure to PAH, where chargrilled meat intake was verified by a questionnaire [40]. In two planned eating studies (involving grilled meat) with a small number of participants (n = 4 and 21), a significant enhancement of DNA adduct level was reported in a few subjects [41, 42]. These results, together with ours, would imply that PAH intake by means of diet is another crucial source of carcinogenetic exposure to consider, and is detectable by DNA adduct dosimetry in blood leukocytes.
Interestingly, both LTL and LmtDNAcn adduct-related erosion are evident in males, while females had much less. The majority of studies examining differences in LTL between women and men found that women have longer telomeres than men [43]. Several plausible biological arguments can be formulated to explain this. These include the action of estrogens [44], which can stimulate the production of telomerase [45] and protect against reactive oxygen species damage [46]. Furthermore, the rate of LTL shortening was slower in women than men [43], as detected in cross-sectional and longitudinal studies [47]. This same mechanism may also result in lower susceptibility to genotoxic damage in women, via anti-B[a]PDE-DNA adducts in the telomeric region. This is in agreement with our observation on LmtDNAcn in a study on 1,088 subjects of European descent, in which higher LmtDNAcn content was detected in women compared with men [48, 49]. Our findings are also consistent with previous studies that pointed to sex-specific differences in the effect of air pollution on cord blood LmtDNAcn, with boys being the most highly affected [34, 50]. On the other hand, some published reports have failed to demonstrate differences in LmtDNAcn content between men and women [51]. Taken together, our data therefore supports the concept that males could be more vulnerable to the effect of PAH exposure on LTL and LmtDNAcn.
In our opinion, statistical analysis with SEM is a strength of the present research. In medicine and natural sciences, a given outcome is often affected or influenced by more than one factor simultaneously. Multivariate techniques try to statistically account for these differences, adjusting an outcome measure Y to a 1-unit change in X, holding all other variables constant. However, it may be that other variables are not likely to remain constant; a change in X can produce a change in Z (direct effect) which in turn produces a change in Y (indirect effect). Both the direct and indirect effects of X on Y must be considered if we want to know what effect a change in X will have on Y. This can be done mathematically and statistically only using SEM. The procedure decomposes a correlation between two variables into their component parts: direct effects, indirect effects, common causes (X affects both Y and Z; this is a spurious association) and correlated causes (X is a cause of Z and X is correlated with Y). The user is required to state, often using a path diagram, the way that they believe the variables are inter-related. Via complex internal rules, SEM decides which model fits the data better. This method is more suitable for the analysis of complex interrelationships because it tests causal relationships rather than mere correlations.