Tenspace

If anyone listed to Talk of the Nation, Science Friday on NPR this past week, you heard a couple of scientists talking about genes related to aging. Really amazing were the results of Dr. Cynthia Kenyon at UCSF, who showed that by blocking certain protein receptors related to the insulin and IGF-1, she could increase the age of the worm by 70%!

It doesn't lengthen life, as we normally think... it slows down our overall growth, so, theoretically, at age 10, you're equivalent to a normal 7 year old. Weird. Also, she said that others had begun work in the same direction on mice.

Tenspace

Wounded King

The insulin pathway work on mice and C. elegans is related to a growing body of work on Caloric Restriction as a factor producing extended lifespans.

Shimokawa I, Higami Y, Tsuchiya T, Otani H, Komatsu T, Chiba T, Yamaza H. Lifespan extension by reduction of the growth hormone-insulin-like growth factor-1 axis: relation to caloric restriction.
FASEB J. 2003 Apr 8

Doubting Didymus

This is a standard group selection model, but I am not sure that Lobstrosities hypothetical is quite the same thing. It might not even be 'selection' in the sense we usually use it. This is the scenario I envisioned: Note that founding rates are not an important ingredient, and they are assumed to be equal.

ten species of about 10,000 blaps each exist

mortal blaps live one year, generations are discrete

there is a 1/1,000,000 chance that the immortality gene will appear in a species each year

There is a relatively large scale extinction event every, say, 100 000 years.

(imagine that mortal blaps have smooth boom/bust cycles, spending about as much time with a high population as they do with a low one, while immortal blaps have a peak population for 10% of the time, and spend the rest of their existance recovering, which is not an uncommon ecological pattern)

Mortal blaps have a 50% chance of surviving the extinction event (this is obviously oversimplified)

Immortal blaps have a 10% chance of surviving the extinction event.

Assuming mortality as an original state, the chances are that one of the species will become immortal within 100 000 years. This species is more successful than the other 9 for the time being, but when the extinction event rolls around, it only has a 1 in 10 chance of pulling through. Most probably, we will end up with 5 species of blap, all mortal. Every 100 000 years we get the same screening, immortal species that happen to survive mass extinctions are still screwed when the next extinction rolls round. This means that, while senescence is still just as likely to appear and be selected for in modern species, it is not likely to be seen to be common in the current set of species, having been screened out historically by extinction events.

Now, let me think about this... What if I imagine one hundred species of blap, which means that 10 of them are probably immortal before the extinction event. We should end up with one immortal species of blap and 45 mortal species of blap. If we assume a founding rate just enough to replace the lost species, then we end up with about 90 mortal species and only 2 immortal ones. The immortality trait has little chance of surviving the next extinction event, and even if it does, it is still at the same disadvantage it was to start with. In order for a trait to overcome this pattern, it must have either a probability of arising in a population that outweighs the time between extinction events (in the case of a hundred species: one in 10 thousand years), or to have a speciation rate that guarantees at least enough new species for the deficit to be outweighed. In this case, where a trait allows only a 10% probability of survival, at least ten daughter species must be founded every extinction period. If a trait does not fulfill either of those criteria, then we should not expect to see it in many modern species, regardless of how much you play with your numbers.

The question remains: "is senescence such a trait?". I, of course, have no idea.

Peez

I would argue that founding rates are a critically important factor in real life, but I can accept the assumption here.Fair enough, though I would still like someone to explain why immortal populations would behave this way any more than mortal populations.This is pretty much the same result as I got, I think.Speciation rate in this model is the equivalent of founding rate in my model.By the same argument, mortal species would have to generate two daughter species, so immortal species would have to do about five times better (depending on how ones quantifies it).You say "either of those criteria" as if to imply that there are just two, but I assume that you are including the other assumptions of your model. Certainly changing the time between extinction events, for example, would alter the picture. Other things could potentially change the outcome, such as a more complex model that did not consider immortality as a trait conferred by a single allele. But even with your result we would expect to find immortal species, perhaps about 5% of all species.We do know that there are genetic trade-offs between early fitness and senescence, which suggests that aging can evolve by simple within-population selection.

Peez

Doubting Didymus

I think I agree with you on all points bar just this one:

That depends on almost all the assumed figures from our scenarios. What if the evolution of the trait is simply hugely unlikely? What if only a small amound of time has passed since the most recent extinction event? I don't think we could confidently say we should find immortal species (I assume we both mean metazoans), unless we had some fairly accurate figures for such things. Besides, I don't think either of the scenarios we have proposed would apply to this specific case, for reasons you have already made clear.

... which does not present ageing as an adaptive trait at all, which is something I can live with.

Peez

Sorry for not making myself clear, I based that on the assumptions that you proposed.That would be my position.

Peez

scombrid

Many anadromous salmonids are fully semelparous.