A Mouse Model of Accelerated Mitochondrial Deletion Mutations that Doesn't Exhibit Signs of Accelerated Aging

Few roads in the life sciences are straight and broad, and the way forward to prove and quantify the contribution of mitochondrial DNA damage to aging is turning out to be particularly winding. The open access paper noted below is the latest in a series of attempts to engineer mice that generate specific forms of mitochondrial mutation at an accelerated rate. The hope here is that this sort of study will, even if carried out for other reasons, help to clarify contradictory results obtained from prior lineages of mitochondrial mutator mice, but I'm not sure that any such goal has been achieved in this case. When compared with the theory of what is expected to happen as the result of a greater number of mitochondrial mutations, the results here are more of an additional puzzle than an answer to outstanding questions.

There is a herd of mitochondria in every cell, replicating like bacteria, and each carrying their own small circular genome - mitochondrial DNA. One important view of mitochondrial DNA mutation in aging is summarized in the SENS research proposals. In short, deletion mutations eliminate important mitochondrial genes, and an affected mitochondrion malfunctions in a way that causes it to either replicate more efficiently or resist cellular quality control mechanisms more effectively than its undamaged peers. The cell is overtaken by the descendants of this broken mitochondria, and as a consequence enters a dysfunctional state that exports harmful reactive molecules into the environment, contributing to the aging process.

In this view, point mutations are not thought to be anywhere near as important - but they are much more common. Indeed, some thought has to go into explaining how deletion mutations can be significant in aging given their rarity. In the course of investigating these questions, mice have been engineered to have abnormally high levels of mitochondrial mutations. There are mice with enormous numbers of point mutations in mitochondria that exhibit no signs of accelerated aging, and there are the later mitochondrial mutator mice with both greatly increased point mutations and deletions that do exhibit accelerated aging. A reasonable conclusion on this basis is that the deletions are the important factor.

Now, however, we have this new lineage of mice exhibiting extra deletions and no point mutations, but that also show no signs of accelerated aging. At this point, I think we're forced to concede that the implementation details matter greatly, and every one of these studies and models is going to have to be picked over with a fine comb in order to figure out what to try next. It is perhaps time to give up on building a model of accelerated aging, and put time and effort into engineering a mouse with fewer mitochondrial mutations to see if more can be learned by trying to slow aging.

On this front, it isn't clear that the SENS program of allotopic expression has progressed far enough to make an attempt to gain data in mice. There are thirteen mitochondrial genes to protect, and only protecting the three that can so far be protected might not be enough of a difference to obtain reliable data for outcomes on aging. Mitochondrial damage is only one of seven classes of damage that cause aging, and what is the effect size of a quarter of a seventh? How comfortably would anyone feel trying to find an adjustment in aging rate of a few percentage points in mice? Smaller effects are very hard to reliably identify in animal studies, in which 10% effect sizes typically come and go at random and should really be treated as noise. Up to a certain point, it is more cost effective to put resources towards protecting more mitochondrial genes.

Mice lacking the mitochondrial exonuclease MGME1 accumulate mtDNA deletions without developing progeria

Mutations in nuclear genes can cause mitochondrial DNA (mtDNA) instability resulting in mtDNA depletion or accumulation of deletions and/or point mutations, ultimately leading to impaired oxidative phosphorylation (OXPHOS). The vast majority of mutations causing human mtDNA instability map to genes encoding proteins involved in mtDNA replication. Extensive in vitro work has led to significant progress in our understanding of the biochemical processes underlying mtDNA maintenance disorders, but animal models are nevertheless essential to understand the wide range of phenotypes and secondary metabolic consequences of mtDNA instability in different tissues.

To gain further insight into diseases of defective mtDNA replication, we created a knockout mouse model for the recently described disease gene encoding MGME1 (also known as Ddk1). Loss-of-function mutations in MGME1 were reported to cause a severe multisystem mitochondrial disorder in humans with depletion and rearrangements of mtDNA. Loss of MGME1 expression, either in siRNA treated cells or in patient fibroblasts, leads to an accumulation of 7S DNA, which is the single-stranded DNA species formed by premature replication termination at the end of the control region of mtDNA, thus suggesting a role for MGME1 in repressing formation or increasing turnover of these molecules.

We have studied the in vivo mtDNA replication phenotypes associated with MGME1 deficiency in various mouse tissues of knockout mice. Although MGME1 is not essential for embryonic development, its loss leads to accumulation of multiple deletions and depletion of mtDNA in a range of different mouse tissues.

A hallmark of MGME1 deficiency in patient fibroblasts and mice is an 11 kb linear mtDNA fragment spanning the entire major arc of the mtDNA, which has been previously described in mtDNA mutator mice and flies. Numerous studies suggest that mtDNA mutations and deletions contribute to the ageing phenotypes in experimental animals and in humans. Indeed, mtDNA mutator mice develop progressive premature ageing syndrome phenotypes. In addition to the presence of the above mentioned 11 kb subgenomic mtDNA species, the mtDNA mutator mice also accumulate an increased number of point mutations, that most likely drive the ageing phenotype. Consistent with this hypothesis, Mgme1-/- mice do not accumulate point mutations and do not display a progeroid phenotype. In line with this finding, mtDNA subgenomic fragments have not been detected in tissues from aging mammals further indicating that this lesion on its own does not induce ageing.


Although this paper does find large numbers of deletions in these mice, these mutations are quite unlike the 'common deletion' and other mtDNA mutations that accumulate in aging cells: they originate on the minor arc, whereas age-related deletions occur on the major arc, also with breakpoints located between the origin of light and heavy strands - and, as Kowald and Kirkwood emphasize (PMIDs 24569805 & 24569805), very largely overlap with the genes for ND4 and possibly ND5.

Additionally: an early and critical observation about cells and muscle fibers overtaken by mutant mitochondria is that they simultaneously show very high levels of activity of succinate dehydrogenase (which participates in both the citric acid cycle and the electron transport system (ETS)) and no activity of cytochrome oxidase (COX). This is a sign (along with other evidence) that such cells are unable to produce energy via the most efficient route in the mitochondria (oxidative phosphorylation (OXPHOS)), and are instead leaning very heavily on glycolysis.

Now unlike these age-related mtDNA deletions, the deletions in these mice have no effect on OXPHOS proteins. More surprisingly - and I have not the foggiest how this can be - these mutations increase cells with no COX activity in the heart, yet not in skeletal muscle.

And on the flip side, their claim of 'no premature aging' is based on pretty skimpy grounds: "normal testis weight, normal sperm count, normal sperm motility and unaltered testis morphology (Fig. 3a-e). ... At the age of 70 weeks [late middle age or early seniority -MR], Mgme1-/- mice have moderate anemia without reticulocytosis accompanied by moderate splenomegaly, whereas mtDNA mutator mice at 40 weeks of age have profound anemia, marked reticulocytosis and massive splenomegaly with extramedullary hematopoiesis."

Dr. de Grey additionally has some prima facie narrowly technical concerns about how some of the data were determined, and is reading the paper in detail to get more clarity on these.

So I don't think this is any kind of challenge to the proposition that age-related mtDNA deletions contribute to degenerative aging.

Posted by: Michael at April 10th, 2018 5:59 PM

Mouse models are notoriously a bad emulation of human aging. I would prefer to see results in human tissues in-vitro. Of course, cultured tissues would need to be aged faster. Probably by radiation or chemical stress. Could it be easier to control for different mutation types in a cell culture then a mouse body?

Posted by: Cuberat at April 10th, 2018 9:13 PM

Post a comment; thoughtful, considered opinions are valued. New comments can be edited for a few minutes following submission. Comments incorporating ad hominem attacks, advertising, and other forms of inappropriate behavior are likely to be deleted.

Note that there is a comment feed for those who like to keep up with conversations.