Heavy metal and junk DNA

Far from being discarded nucleic acid material, “Junk DNA” i.e. transposable elements (TEs) are a powerful genetic force impacting both genome stability and human disease. Because of the genetic damage potential of TEs, it is of particular interest to understand the effect of environmental exposures on TE-containing genomes. The past years, our lab has explored TE-mediated damage as an additional mechanism contributing to heavy metal carcinogenesis (reviewed in¹). Our recent data demonstrate that heavy metal exposure not only increases TE activity, but favors mutagenic repair of double strand breaks near Alu elements.² This commentary puts our recent published observations into context of the role of TEs in heavy metal disease risk (e.g. cancer), specifically highlighting the importance of understanding the complexity of polymorphic TE variation within human populations and exposure outcomes.

The dangers of heavy metal exposure have been known for centuries. Arsenic has been a favorite poison for eliminating many kings and noblemen^3,4). Fatalities due to mercury exposure and cadmium have been recorded as early as the 1800s.^5,6 History shows that childhood lead poisoning is an ongoing 100-year old epidemic in US history⁷ with the most recent cases reported from the community of Flint, Michigan resulting from lead contamination of the drinking water supply.⁸ Epidemiological data associate many heavy metal exposures with a large variety of diseases.^9,10 In fact, the data associating heavy metal exposure and cancer provided the basis for the classification of several heavy metals including cadmium, nickel and arsenic as carcinogens.¹¹ Despite their well-known toxic effects, heavy metal exposure remains a common event. Worldwide estimates predict that millions people are exposed to arsenic and lead through drinking water and food. Exposure to heavy metals frequently occurs due to their wide use in industry and long-term environmental persistence. Most humans have likely been exposed to heavy metals sometime in their life. Unfortunately, many exposures occur due to unawareness of the presence of the contaminant and are likely not recorded. Interestingly, not all exposures are unintentional. For example, therapeutic exposure of arsenic has been used in the treatment of syphilis¹² and some hematological malignancies.¹³ Furthermore, cigarette smoking is one of the most common sources of exposure to nickel and cadmium creating a large population of chronically exposed individuals.^14–17 In addition, the contribution of e-cigarettes to heavy metal exposure requires further investigation.¹⁸ However, not all exposed individuals are affected. So, why is it that not all exposed individuals show disease or develop cancer? Assuming that time and dose of exposure are equivalent, the genetic variation between individuals likely accounts for differences in outcome. Although there are many genes (e.g., metabolic enzymes, DNA repair, etc.) known to either decrease or increase risk of a negative outcome, this commentary highlights how polymorphic TEs can also contribute to heavy metal disease risk susceptibility.

TE-mediated genomic damage is well documented to occur as: 1) a consequence of TE activity by contributing to insertion associated mutagenesis, and 2) a source of repetitive sequences that serve as nuclei for mutagenic genomic recombination (e.g. non-allelic homologous recombination: NAHR).¹ However, to determine the contribution of TEs to an individual's risk to disease becomes complex, as different individuals can vary in their TE content. The continued activity of TEs provides a steady source of genetic variation between individuals, of which a significant portion of the population will harbor a new polymorphic TE insert unique to them (private insertion). Data from the 1000 Genome Project (Phase3 November 2014) reported that a typical genome contains an estimated 2,100 to 2,500 structural variants including ∼915 Alu insertions, ∼128 L1 insertions and ∼51 SVA insertions.¹⁹ Because, human polymorphic TE insertions are found at very low frequencies,²⁰ it is expected that 2 unrelated individuals will contain different sets of polymorphic TE inserts, of which many will not be annotated in the human reference genome (Fig. 1A). For example, one study demonstrated that a typical individual has ∼100 non-reference, polymorphic full-length L1 elements.²¹ Importantly, due to their recent insertion, polymorphic TEs are more likely to retain sequence features needed for activity.²² Thus, not only the number of TEs can vary, but the number of active TEs in any given individual can also differ. In other words, an individual with a highly active polymorphic TE (also referred to as a “hot” element) may be more susceptible to TE-mediated diseases. Data show that highly active TEs may significantly contribute to cancers. Studies demonstrate that some cancers contain multiple somatic inserts that can be traced back to one individual “hot” element.^23,24 Another layer of complexity regarding somatic disease risk, such as cancer, derives from the evidence that TE content can also vary between different tissues due to somatic activity^25–27 or activity during early embryonic developmental stages.²⁸ Sequencing studies from a variety of cancer and matched normal tissues suggest that somatic retrotransposition occurs with varying efficiency,^29–32 which could reflect even another component contributing to more variation between individuals. However, the specific contribution of polymorphic somatic TE insertions to disease risk remains undefined. In addition to the damage caused by active insertions, the presence of inactive polymorphic TEs can also contribute to increased disease risk. For example, the presence of additional Alu inserts may locally increase the Alu density making a locus more susceptible to TE-mediated genomic rearrangements such as those reported to recurrently occur in the MLL and VHL loci.³³ Thus, similar to SNPs, polymorphic TEs may associate with a disease, where the presence of a TE increases risk (Fig. 1B).

Figure 1.

Potential effect of TE variation (polymorphic inserts) in a population and disease risk. A. Polymorphic TE inserts (represented as stars) vary throughout individuals and populations. Studies used to identify disease risk factors sometimes identify chromosomal regions that show as gene deserts in the human reference genome. Is it possible that the identified association reflects the presence of a non-reference (i.e., unannotated) polymorphic TE insert at this site? B. The presence of a polymorphic TE can increase genetic risk. The presence of polymorphic Alu inserts (yellow arrow) would increase the genomic repeat density increasing the locus susceptibility to mutagenic recombination events. Alternatively, the presence of polymorphic full-length active “hot” L1s (green arrow) would increase the genomic load of active elements increasing the potential for L1-mediated mutagenic events. Both of these scenarios can contribute to an increased risk to disease. In contrast, the absence of the TE (lower Alu density, lower number of active L1s) or the presence of an inactive L1 would be considered as “lower” risk genetic loci.

Our review Morales et al., 2015,¹ summarized the available data on how a variety of heavy metals influence L1 activity. Overall, the data indicate that the ability of heavy metals to affect the function of many different proteins and pathways serves to potentiate TE-mediated damage. Thus, individuals with high TE content or containing “hot” polymorphic TEs may be more susceptible to detrimental effects from the environmental exposure of heavy metals. In this case, heavy metal exposure can inhibit multiple regulatory pathways to induce a cellular environment more permissive to insertional activity. For example, many heavy metals induce genomic demethylation.³⁴ Thus, exposure may promote higher expression and activity of a “hot” L1 through the demethylation of its promoter (Fig. 2A). However, the effect of heavy metal exposure on TE-mediated damage is not just limited to increasing TE activity. In our recent paper, Morales et al. 2016,² we demonstrated that some heavy metals favor mutagenic DNA repair through Alu-mediated recombination. The data show that the exposure to the heavy metal alters the choice of DNA repair pathway increasing mutagenic repair outcomes near repetitive elements (e.g., Alus). In this scenario, individuals containing loci with higher TE density due to the presence of polymorphic TEs would be more susceptible to heavy metal induced TE-mediated recombination (Fig. 2B). Overall, heavy metal exposure can contribute to a genomic mutagenesis through TE-mediated damaging events.

Figure 2.

The potential role of TE variation (polymorphic inserts) on differential effects of heavy metal exposure. A. The role of polymorphic hot L1s. Heavy metal exposure (cadmium is used as a representative example), may concomitantly induce demethylation of the promoter of polymorphic young active “hot” L1 elements (green arrow) to increase L1 expression and stimulating L1 insertion by inhibiting cellular controls (e.g. inhibition of APOBECs or DNA repair factors). B. The role of polymorphic Alu insertions. Heavy metal exposure increases reactive oxygen species that may cause double strand breaks (DSBs) near Alus; where the presence of polymorphic Alu inserts (yellow arrow) can contribute to an increased density increasing risk of Alu-mediated genomic rearrangements. In addition, we have demonstrated that heavy metal exposure alters DSB repair near Alu elements and promotes mutagenic outcomes.

In summary, TE-mediated damage is one of many genetic events that should be considered when evaluating effects of exposure to environmental toxicants such as heavy metals. The current available data just begin to show the significant effect that environmental exposures have on TE-mediated damage. However, much more work needs to be done to determine how TE variation in combination with environmental exposures contribute to complex diseases, such as cancer, before we can get a truly accurate assessment of an individual's disease risk.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Funding

The published research from the manuscripts discussed in this commentary was supported by National Institutes of Health (NIH) R01GM079709A (AMR-E), R01GM045668 (PLD), P20GM103518/P20RR020152 (AMR-E and PLD) and R01CA120954 to JMS.