Q: Dear Professor Bennett, you just published a very fascinating study in Circulation Research on DNA repair impairment and atherosclerosis. Would you mind pointing out the highlights of your study?
A: DNA damage, including mitochondrial DNA damage, has been found in atherosclerosis for many years, both in circulating cells and the plaques themselves. Similarly, DNA damage has been found in organs associated with the metabolic syndrome. However, it has not been known whether such damage, particularly mitochondrial DNA damage, is causal, and whether the same underlying processes might promote both atherosclerosis and the metabolic syndrome.
Q: Indeed, very fascinating findings. Going through these one by one, let’s focus on the extent of atherosclerosis of the aortic root and the descending aorta in fat fed ApoE double knockout mice with versus without single knockout of ATM, a kinase involved in DNA repair. What did you find, were these two regions affected differently?
A: We found that ATM heterozygous ApoE null mice had a 1.6-1.7 fold increase in atherosclerosis in both vascular beds, so we can be sure that this is not just a localized effect.
Q: Literally, going deeper into the matter, the fraction of vascular smooth muscle cells (VSMC) and macrophages was only marginally smaller in aortic plaques of ATM-deficient mice. This is intriguing in view of the finding that the fraction of cells with evidence of activation of the apoptotic pathway was significantly smaller in the ATM-deficient animals and yet there seemed to be a definite trend for less proliferating cells as well. Would you say then that the apoptosis-proliferation balance was even or in favor of the former? Intriguingly, the necrotic core area did not differ and one may wonder if dynamics were shifted to a particular plaque type or. In other words, did ATM deficiency lead to the development of a more matrix tissue-rich, “stable” type of plaque rather than a cell- and lipid core-rich “unstable” plaque? Is there any element of a successful or unsuccessful “scarring” response pattern? Would you mind commenting on these aspects and how you would characterize the observed plaque dynamics?
A: The plaques were reduced in size, without any specific change in composition to a more/less ‘stable’ phenotype. The effect of changes in the apoptosis proliferation/balance is hard to predict. For example, the simplistic explanation that cell proliferation is ‘balanced’ by cell death in the plaque is untrue, and the effects depend upon when and where in atherogenesis these processes occur, and the cell types involved. For example, in our previous studies vascular smooth muscle cell (VSMC) apoptosis promotes atherogenesis whereas macrophage apoptosis reduces atherogenesis. In this study, ATM+/- VSMCs show impaired apoptosis, but apoptosis would also be promoted by the pro-atherogenic metabolic disturbances we found.
Q: Next, you performed elegant bone marrow transplant studies and were able to show that restoring ATM in bone-marrow derived cells increased the proliferative index of the atherosclerotic plaques and aborted the development of accelerated atherosclerosis. Would you say then that this is more of a bone marrow-driven disease process and/or a disease of a particular cell type stemming from the bone marrow?
A: Certainly, bone marrow-derived cells such as monocyte/macrophages appear to be important in determining the local plaque response to ATM heterozygosity. However, we also found that ATM+/- VSMCs show defective growth arrest and apoptosis signalling.
Q: Looking at the cells that populate the atherosclerotic plaque in culture, you found that vascular smooth muscle cells from ATM-deficient mice had increased rates of proliferation and reduced rates of apoptosis, recapturing the atherosclerotic plaque data. Moreover, VSMCs showed evidence of genomic instability and impaired DNA repair. Macrophages, on the other hand, seemed to be not as sensitive. What is the explanation and how does this connect with the bone marrow transplant results?
A: The regulatory pathways that control apoptosis and cell proliferation in macrophages and VSMCs are different; for example we have previously shown that these cells have different sensitivities to p53-induced apoptosis. Whilst the in vitro effects of ATM heterozygosity in macrophages appears to be less impressive, the bone marrow experiments clearly show that ATM is important in bone marrow-derived cells to suppress atherosclerosis.
Q: The high fat diet did increase the scale of lipid values. However, the ATM-deficient ApoE knochout mice already had lipid abnormalities to begin with, which were not truly corrected by an ATM-competent bone marrow. What is the underlying mechanism?
A: We think that the metabolic effects of ATM heterozygosity / DNA damage are already present in ApoE null mice, and are just amplified by the high fat diet. As these were not corrected by the bone marrow transplant, they are likely to be due to ATM effects on metabolically important organs, such as the liver and pancreas.
Q: These lipid abnormalities lead over to what was termed “features of the metabolic syndrome” in your paper. Interestingly, some of these features were present before but others required fat feeding to become apparent. Would you mind elaborating on this?
A: Again, ApoE null mice already are predisposed to features of the metabolic syndrome, including hyperlipidaemia, hepatic steatosis, and adiposity. The high fat diet amplifies these features, as does ATM heterozygosity.
Q: You also undertook an extensive metabolomic screening effort in key tissues to identify further differences. Increase in beta-hydroxybutyrate seems to be a recurrent theme in this. Does this indicate altered glucose metabolism and possibly then altered insulin metabolism?
A: We think that ATM+/-/ApoE-/- mice have changes in both lipid and glucose metabolism. -hydroxybutyrate is a ketone body produced in the liver from incomplete oxidation of long chain fatty acids, particularly when glucose cannot be effectively metabolized, for example in diabetes or when oxidative phosphorylation is compromised. We couldn’t demonstrate defects in insulin secretion or abnormalities on insulin tolerance testing, but we suspect DNA damage is also likely to occur in the pancreas.
Q: Ultimately, your studies scale down on the mitochondria and you performed a number of elegant studies to define damage to mitochondrial DNA (mtDNA) further. Would mind taking a moment to recapitulate for us what you did and what you found?
A: The metabolomics studies suggested that mitochondrial DNA damage / dysfunction may also be present in these mice. We found higher levels of mitochondrial DNA adducts, and the ‘common’ mitochondrial DNA deletion, higher levels of reactive oxygen species, and reduced Complex I activity in livers of ATM+/-/ApoE-/- mice.
Q: How did these changes relate to changes in expression and activity of complexes of the respiratory chain?
A: Expression of all complexes was unchanged, but Complex I activity was reduced in ATM+/-/ApoE-/- mice. As Complex I activity is a rate-limiting step in mitochondrial electron transport, this would be predicted to affect activity of the other complexes.
Q:Did these respiratory chain changes translate into increased ROS production? How does oxidative stress factor in?
A: Yes we found increased ROS in ATM+/-/ApoE-/- mice. However, its not clear whether mitochondrial dysfunction comes first, which increases ROS, or ROS induces mitochondrial dysfunction / damage. I suspect that both occur. If true, reducing ROS in these mice may reduce mitochondrial dysfunction and may correct many of the metabolic features we observe.
Q: How would you put it all together? What is the first effect of ATM deficiency on cellular level and what is the response? Does it start out with natural “wear and tear” of the DNA in a biological environment, which without repair results in miscoded information and further generation of cellular stress? An editorial on your paper mentioned the “chicken and the egg” question regarding oxidative DNA damage and disease. However, you must have a theory based on what you observed in your study.
A: We are using ATM+/-/ApoE-/- mice as a model of DNA damage in atherosclerosis. We think that ROS, for example produced locally in response to oxidized lipids, induces DNA damage in cells comprising the plaque. This generates further ROS, resulting in both genomic and mitochondrial DNA damage, affecting cell proliferation, senescence and apoptosis of cells comprising plaques. Thus, there is a local effect of DNA damage in plaques. Systemic hyperlipidaemia may have the same effect on other organs, including liver and pancreas. DNA damage, including mitochondrial damage and dysfunction, results in defects in glucose and lipid metabolism, resulting in adipose tissue deposition, and multiple features of the metabolic syndrome.
Q: What are the therapeutic implications? The antioxidant trials did not pan out but what might?
A: The studies suggest that we need to target mitochondrial ROS production and mitochondrial DNA damage and dysfunction to correct the metabolic features shown here. Antioxidants may be one way to do this, although it is not clear whether antioxidants would be able to penetrate the plaque, and correct the genomic and mitochondrial DNA damage in plaque cells. We have also previously found that ‘Statins’ can directly accelerate DNA repair, and it would be interesting to know whether these drugs reduce mitochondrial DNA damage and dysfunction.
Q: What is that we have to look at next in this area?
A: The study shows that genomic and mitochondrial DNA damage and dysfunction occurs in both atherosclerosis and the metabolic syndrome in these mice. However, we still haven’t proven causality (the chicken and egg). If we can reduce genomic/ mitochondrial DNA damage and ROS in these mice, we can show that DNA damage is causal in these pathologies, which sets the stage for new therapeutics.
Q: This truly is a fascinating study and we are so thankful for your time and this interview. Surely, there is more to come on this topic in the future and we are eagerly looking forward to more.
A: Both atherosclerosis and the metabolic syndrome are widely prevalent in human populations. We hope that the study will stimulate interest in processes that may underlie both atherosclerosis and the metabolic syndrome.
European Society of Cardiology
European Heart HouseLes Templiers2035 Route des CollesCS 80179 Biot
06903Sophia Antipolis, FR
Our mission: To reduce the burden of cardiovascular disease
© 2018 European Society of Cardiology. All rights reserved