Hypothesis: lab mice have more active transposons then wild mice
post by ChristianKl · 2021-06-05T12:36:06.664Z · LW · GW · 14 commentsAs johnswentworth recounts in Core Pathways of Aging [LW · GW], as an organism ages active transposons within it's stem cells duplicate and that mechanism might lead to increased average transposons count in stem cells. Those transposons then produce DNA damage which in turn leads to cell senescence.
If that hypothesis is true, there's evolutionary pressure to keep the count of active transposons low. That evolutionary pressure is greater in organism that reproduce at a later age then for organisms that reproduce at an earlier age.
As Bret Weinstein describes, breeding protocols for lab mice have lab mice reproducing at an earlier age then mice that live in the wild because it's economical to make the mice reproduce at a young age. Weinstein made the hypothesis that this leads to laboratory mice having elongated telomeres.
I hereby make the hypothesis that if we investigate the average amount of active transposons in laboratory mice and lab mice, we will find that the wild mice have less active transposons then the wild mice, because there's less evolutionary pressure in the laboratory mice to remove mutations that lead to increased active transposon count.
If investigation finds this hypothesis to be true, approaches to reduce transposon count should get more attention by antiaging researchers.
14 comments
Comments sorted by top scores.
comment by dkirmani · 2021-06-06T18:03:43.768Z · LW(p) · GW(p)
as an organism ages active transposons within it's stem cells duplicate and that mechanism might lead to increased average transposons count in stem cells
My model is that transposons duplicate in all somatic (non-reproductive) cells, not just stem cells.
If that hypothesis is true, there's evolutionary pressure to keep the count of active transposons low. That evolutionary pressure is greater in organism that reproduce at a later age then for organisms that reproduce at an earlier age.
The evolutionary basis of aging (and negligible senescence, like in hydras and naked mole rats) is still a total mystery to me. The arguments for why aging is adaptive all rely on group-selection, which I am wary of. The argument is basically that you grow old and die to benefit the tribe, just as your cells commit suicide when it is useful for you. I'm relatively unconvinced by this argument, as I believe that intra-tribal competition is a much more powerful selective force than inter-tribal competion, giving rise to (machiavellian) intelligence as well as extremely metabolically costly dominance competitions. Getting old doesn't make sense if you're the only one doing it.
Those who oppose the adaptive aging hypothesis generally fail to take into account the fact that naked mole rats exist, leaving us without an explanation as to why we have senescence and they don't.
As Bret Weinstein describes, breeding protocols for lab mice have lab mice reproducing at an earlier age then mice that live in the wild because it's economical to make the mice reproduce at a young age. Weinstein made the hypothesis that this leads to laboratory mice having elongated telomeres.
My model here is:
Longer teleomeres -> Higher Hayflick limit (number of times cell can divide before dying) -> Higher cancer risk, as tumor cells can divide more (decreases chance of being alive in late-life), as well as higher capacity for tissue damage repair, as existing cells can divide more to replace missing ones (greater chance of being alive in early-life)
This is consistent with longer telomeres being a reallocation from late-life health to early-life heath, and that tradeoff starts making sense when you reproduce at an earlier age. With respect to transposons, though, I don't understand what the trade-off is. With longer telomeres, you get an increased tissue regen rate at the expense of increased cancer risk. With transposons, you get senescence, but for what?
Replies from: ChristianKl↑ comment by ChristianKl · 2021-06-06T18:36:13.801Z · LW(p) · GW(p)
My model is that transposons duplicate in all somatic (non-reproductive) cells, not just stem cells.
They also duplicate in other somatic cells but in cells that have a low half life it doens't matter as much.
With transposons, you get senescence, but for what?
You don't get anything in return just like you don't get anything in return for getting infected with COVID-19.
You need evolutionary pressure to prevent transposons from constantly doublicating and accumulating in the DNA of your lineage.
Getting completely rid of transposons would be all upside but there's not strong enough evolutionary pressure for going to zero transposon count and even if a species would reach that for some amount of time, there seems to be horizontal transfer of transposons between species through viruses that reintroduce transposons.
While deciding whether or not tranposons are genes is a question of definition you can think of them in the framework of Dawkins selfish genes that care more about their own replication then the host organism.
The evolutionary basis of aging (and negligible senescence, like in hydras and naked mole rats) is still a total mystery to me.
Given that naked mole rats also happen to be less transposons it starts getting less of a mystery to me.
To register another hypothesis, I would expect hydras also to have less transposons.
That doesn't mean that the transposons might be the only reason for the negligible senescence but it might be one of the ways the evolutionary pressures according to which negligible senescence is advantagous for naked mole rats worked in practice to reduce the senescence.
Replies from: dkirmani↑ comment by dkirmani · 2021-06-06T18:47:13.839Z · LW(p) · GW(p)
Given that naked mole rats also happen to be less transposons it starts getting less of a mystery to me. To register another hypothesis, I would expect hydras also to have less transposons.
I agree! It makes sense that NMRs have fewer transposons and also age less. As for hydras, they do have less transposon activity in the soma, as the PIWI/piRNA pathway is active in their somatic tissue. (From The Mechanism of Ageing: Primary Role of Transposable Elements in Genome Disintegration, which is a (very informative) paper that asserts the transposon hypothesis)
What I was wondering was why NMRs / hydras have less transposon activity than us, and what selective pressures caused this to come about.
Replies from: ChristianKl↑ comment by ChristianKl · 2021-06-06T19:50:34.167Z · LW(p) · GW(p)
I agree! It makes sense that NMRs have fewer transposons and also age less. As for hydras, they do have less transposon activity in the soma, as the PIWI/piRNA pathway is active in their somatic tissue.
It seems I was slightly of and it might be that the hydra's simply have more processes to block transposon doublication which is another way to reduce problems with transposons compared to reducing their count.
What I was wondering was why NMRs / hydras have less transposon activity than us, and what selective pressures caused this to come about.
I remember Brad Weinstein talking about how some biologist (maybe it was his professor?) predicted that if there's a species with negligible senescence it would have a bunch of characteristics that are true for naked mole rats. He didn't know about naked mole rats at the time and it turned out to be one of the successful theoretic predictions based on modern evolutionary theory.
If I look at hydras, the way there sexual reproduction works differently then for a lot of other animals. The male and female don't meet but the male releases their gamete into the water and thus it doesn't stop reproducing.
Hydras also have asexual reproduction where new Hydras seem to be created out of somatic cells and not via the sexual production. If hydras would allow for transposon activity in their somatic cells then children they prodcue with asexual reproduction would also have a higher transposon count which would be very inconvenient.
Replies from: dkirmani↑ comment by dkirmani · 2021-06-06T20:39:10.916Z · LW(p) · GW(p)
I remember Brad Weinstein talking about how some biologist (maybe it was his professor?) predicted that if there's a species with negligible senescence it would have a bunch of characteristics that are true for naked mole rats. He didn't know about naked mole rats at the time and it turned out to be one of the successful theoretic predictions based on modern evolutionary theory.
I remember this too (vaguely)! I think it was from this episode of The Portal podcast with Bret and Eric Weinstein.
If I look at hydras, the way there sexual reproduction works differently then for a lot of other animals. The male and female don't meet but the male releases their gamete into the water and thus it doesn't stop reproducing.
Fish do this too, and I don't recall anything about fish not aging, so I don't think this is useful.
Hydras also have asexual reproduction where new Hydras seem to be created out of somatic cells and not via the sexual production. If hydras would allow for transposon activity in their somatic cells then children they prodcue with asexual reproduction would also have a higher transposon count which would be very inconvenient.
I think this is the key factor. In humans (and most animals), transposons are suppressed only in the germline cells, but if a hydra can reproduce via budding with its somatic cells, then all of its 'somatic' cells are in fact potential germline cells, and it is important that they have transposon suppression mechanisms.
However, naked mole rats don't reproduce by budding, so the mystery of NMR negligible senescence is still unresolved. I'm going to give that podcast another listen, and keep an ear open for that section you mentioned.
(Edit: 43:30 - 1:06:00 is the relevant section of the podcast)
Replies from: dkirmani↑ comment by dkirmani · 2021-06-07T01:14:18.146Z · LW(p) · GW(p)
Okay, so I think I realize what's going on.
(Epistemic status: Totally made up. A lot of the 'evidence' here is remembered from papers I read a month ago. The third-last and second-last paragraphs are mostly just conjecture.)
Transposons are present in basically every genome. Animals can choose to invest in transposon suppression mechanisms (the PIWI/piRNA pathway), but doing so comes with a metabolic cost that increases proportionally to the amount of tissue in which these mechanisms are active.
Humans have (to a first approximation) transposon suppression active in the germline cells of the gonads, and nowhere else. If a human gets a mutation that expresses transposon-supressors in the soma, he/she would have (in the ancestral environment) suffered a small metabolic cost, but would enjoy reduced aging. However, the additional fitness conferred by the reduced aging rate would not offset the metabolic costs enough to be worth it. The mutation disappears.
Naked mole rats live in groups with a size of, on average, 75 individuals. There is one queen, three fertile males, and everyone else is an infertile worker. To the worker, the chance of being the next queen / a mating male is vanishingly small. Therefore, the worker can best optimize their fitness by serving the needs of the collective. Some workers tunnel for food (they eat tubers), others raise the pups, and the 'soldiers' defend the tunnels from predators.
For humans (in the ancestral environment), the critical period in determining the majority of one's reproductive success is between the ages of 15 and 35. That's why it makes sense to sacrifice one's welfare in middle/late age in order to better compete for mates during the physical prime of one's life. So, no transposon suppression for us.
Naked mole rat workers take food from the collective, and provide labor to the collective. Old naked mole rats are less efficient at turning food into labor than young naked mole rats. So why not go the route of bees and ants, and give the workers short, but productive, lifespans? For insects, reproduction is relatively cheap: just lay a small egg. For mammals, like the Queen Naked Mole Rat, there is a more restrictive bottleneck to how many pups may be produced in a year. Additionally, gestating a mammalian pup comes at a higher relative marginal metabolic cost than producing an insectoid egg. Therefore, in order to best utilize resources to serve the collective, workers should both be long-lived and also not get frail.
Investing in transposon-suppression mechanisms is worth it to the naked mole rat's genes, but not to the genes of the human.
Replies from: ChristianKl↑ comment by ChristianKl · 2021-06-07T10:31:55.403Z · LW(p) · GW(p)
If the difference would be just about transposon-supressors in the soma in mole rats, we wouldn't see them to have less transposon-derived repeats in their genome.
The fact that we do see less transposon-derived repeats in the genome of mole rats suggests that having active transposons is a higher cost to their reproductive success then happens to be the case for other mammels.
If naked mole rats would have a system that completely shuts down tranposon activity in soma cells while other mammels don't, we would expect that the count of active transposons matters less for naked mole rats and therefore we would expect more active transposons in naked mole rats then in other mammels.
In contrast to mole rats hydras do have ~57% of it's genome made up of transposable elements (I'm just copying terms from the paper and not 100% sure whether transposable elements == transposon-derived repeats).
Alternative Animal Models of Aging Research is an interesting overview article
According to it planaria which also have asexual reproduction do the same with PIWI/piRNA as the hydras.
Interestingly, according to the article naked mole rats do have phenotypical aging. It just virtually absent for 80% of their lifespan.
According to the article naked mole rats have "efficient DNA damage repair". Efficient DNA repair in germline tells means that transposons will doublicate less in the germline cells and it's therefore easier for evolution to reduce the transposon count.
It's also worth noting that the breeders do have a higher average lifespan then the workers.
In summary it's plausible that while hydras/planaria actually have no phenotypical aging due to having PIWI/piRNA in their soma, naked mole rats just have a lot less aging due to having found ways to reduce the active transposon count but still do have some aging due to the low remaining active transposon count.
Replies from: dkirmani↑ comment by dkirmani · 2021-06-07T16:19:42.200Z · LW(p) · GW(p)
If the difference would be just about transposon-supressors in the soma in mole rats, we wouldn't see them to have less transposon-derived repeats in their genome.
That's a good point, my previous idea about NMR somatic transposon suppression is probably incorrect.
According to the article naked mole rats have "efficient DNA damage repair". Efficient DNA repair in germline tells means that transposons will doublicate less in the germline cells and it's therefore easier for evolution to reduce the transposon count.
Following the article's citation for the 'efficient DNA damage repair' claim, we get this study, that analyzes NMR, human and mouse liver tissue (not germline), and finds that, compared to mice, humans and NMRs have higher expression of genes central to DNA repair pathways. The paper then reminds the reader that both humans and NMRs live longer than mice, so maybe DNA repair makes organisms live longer.
This is consistent with johnswentworth's model of:
Transposon activity -> (DNA damage <-> mitochondrial ROS feedback loop) -> age-associated cellular dysfunction
where DNA repair pathways slow the progress of the DNA damage / mitochondrial ROS feedback loop, making aging progress at a slower rate.
So it looks like:
- NMRs have DNA repair pathways in the soma that are abnormally active for a rodent their size
- NMRs have fewer transposon-derived repeats in their genome, indicating vigilant transposon suppression in the germline
- These mechanisms counter aging, but come with some hidden cost to fitness, and that's why humans don't have even higher DNA repair activity, or more vigilant germline transposon suppression
But what are these hidden costs? Surely there is variation among humans w.r.t rates of somatic DNA repair, and effectiveness of germline transposon suppression. Why aren't people with beneficial versions of these traits aging less, living longer, and having more children?
Edit: I think I might have (partially) answered my own question:
https://en.wikipedia.org/wiki/Antagonistic_pleiotropy_hypothesis#DNA_Damage_Theory_of_Aging
Replies from: ChristianKl↑ comment by ChristianKl · 2021-06-07T19:42:37.857Z · LW(p) · GW(p)
Why aren't people with beneficial versions of these traits aging less, living longer, and having more children?
Even if people with less active transposons have slightly more children that's only one part of the evolutionary equation.
The frame I was taught is evolution= natural selection + mutation + gene drift. Mutations are a force for increased transposon count (and if there's more DNA repair there are less mutations). Natural selection is a force for decreased transposon count. You would expect those two forces to find an equilibrium so that the amount of transposons isn't a problem at the age where an animal bears children. If you have very effecitve germline transposon suppression transposon also evolve to escape the transposon suppression.
When it comes to DNA repair I would expect that it's simply a cost for a cell to put resources into DNA repair.
more vigilant germline transposon suppression
Without evolutionary pressure that prevents individuals with more active transposon to reproduce less that just gets the transposons to express themselves more and you have +/- zero.
Replies from: dkirmani↑ comment by dkirmani · 2021-06-08T21:08:06.046Z · LW(p) · GW(p)
Sure, but wouldn't a hunter that stayed physically 25 until the age of 45 have a higher inclusive genetic fitness, all else being equal? (edit circa 2022-07-02: Wow, I was dumb. Red Queen hypothesis, baby.)
Replies from: ChristianKl↑ comment by ChristianKl · 2021-06-08T21:38:23.575Z · LW(p) · GW(p)
All else isn't equal. Transposons replicate. That's what transposons are about.
The genetic fitness of the hunter doesn't really matter (that's basically the insight Dawkins wrote about in the selfish gene). You have to look at the fitness of the transposon and what's good for the transposon.
If a transposon copies itself into another chromosome it can increase the chance of being inherited to a child from 50% to 100%. That's highly useful to the fitness of the transposon and worth slightly reduced lifespan of the child.
Both of those forces are at an equilibirum. Transposons copy themselves till the reduced lifespan/health isn't worth it to double the chance of being inherited.
comment by ChristianKl · 2021-06-05T22:03:27.086Z · LW(p) · GW(p)
While reading more on transposons I found in Ten things you should know about transposable elements:
Transposition events are also common and mutagenic in laboratory mice, where ongoing activity of several families of LTR elements are responsible for 10–15% of all inherited mutant phenotypes [36].
comment by Randomized, Controlled (BossSleepy) · 2021-06-05T12:54:56.244Z · LW(p) · GW(p)
I hereby make the hypothesis that if we investigate the average amount of active transposons in laboratory mice and wild mice, we will find that the wild mice have less active transposons then the wild mice
Think the last pair should be "lab mice", no?
Replies from: ChristianKl↑ comment by ChristianKl · 2021-06-05T13:59:54.851Z · LW(p) · GW(p)
Yes, I fixed it.