Different traits age at different rates

The question of why we age, though at first seemingly difficult to answer, has a relatively simple evolutionary explanation.  As the vast majority of organisms get older, things start to go wrong: physiological function is impaired and they are more likely to die of diseases and disorders and less able to reproduce. Why then, has evolution not weeded out fast-ageing genes and left us with long-lived organisms? The answer is that natural selection weakens with age. This is because there are fewer individuals left at older ages: the rest have caught diseases, been eaten by predators, starved to death, fallen off cliffs or died in any number of unfortunate or just plain stupid ways. Since selection is much stronger in early life, genes which improve the chances of young individuals (e.g. by accelerating growth) but damage old individuals (e.g. by causing accumulation of harmful metabolites) with be favoured by natural selection. Hence, organisms age.

 A more difficult problem is explaining variation in ageing rates, though we may notice that birds age slower than mammals, flying species age slower than those that can’t fly and females often age slower than males, for sound evolutionary reasons. We also know that ageing may be accelerated by adverse conditions in early life. But what about different aspects of the same individual organism: should different functions, such as reproductive capacity and survival, age at different rates, or should everything age at same rate?

We investigated this question in a new paper in the journal Experimental Geronotology. We wanted to know whether different aspects (what I’m going to call ‘traits’) of an organism showed different rates of ageing, or whether related groups of traits aged in similar ways. We collected data on 20 traits (below) from the wild population of Soay sheep on St Kilda and asked a simple question: do all traits age at the same rate, or do some traits age at different rates?

(4) Sheep traits

What was our expectation? When we looked at the evolutionary theory, we didn’t find a lot to help us. In fact, the only theory we could find was a verbal argument made by John Maynard Smith dating back to 1962. He asked us to imagine a situation where the circulatory system (in red) ages faster than the nervous system (in grey). This means that the circulatory system reaches a low threshold at which death occurs (dotted line) earlier than the nervous system, so everyone dies of circulatory failure. Individuals therefore never get old enough to die of nervous system failure.

(4) JMS 1

This means that there is no selection against mutations which make the nervous system age faster- individuals still never die of a failed nervous system. This means that nervous system ageing should evolve to a faster rate (below).

(4) JMS 2

People are still dying of circulatory system failure, however. This means that there should be strong selection for a slower ageing of the circulatory system. Ultimately then, the nervous and circulatory systems should age at the same rate. The only existing evolutionary hypothesis therefore predicts that ageing should occur at the same rate in all traits.

(4) JMS 3

Of course, this model is too simple: for example changes in traits to a threshold, beyond which we die, are extremely rare.  Studies in Drosophila fruit flies show that ageing in different traits, like ability to withstand heat stress, or ability to move quickly, occurs at different times and different rates. We don’t know how true this is in wild animals, though. We might also expect related traits (for example whether or not you reproduce, how many offspring you have, and how large they are) to age at the same rate. We analysed ageing in our 20 different Soay sheep traits, testing different statistical models which grouped the phenotypes in different ways. And the winner was……

Complexity. The winner was definitely complexity. All the traits, measured separately in the two sexes, had a different ageing pattern. Or to be more precise, none of the patterns we could have imagined (such as the sexes ageing differently from each other, or reproductive traits ageing differently from size traits), were apparent. But don’t despair, because we did find some very cool results, including the five shown below. On all the plots below, males are in blue and females are in red. Also, the value at age 4 (when ageing starts, roughly speaking) is set to zero, and everything else given relative to that.

First, males and females show totally different changes in annual breeding success (below, right). Males get better at fathering lambs as they get older, while females get less good at giving birth to them. Female senescence is expected, because we know the reproductive system deteriorates with age, but male lack of senescence is not. Potentially, the oldest males are simply the biggest and best and therefore able to keep fathering offspring, despite the fact that…

(4) ABS

Important male reproductive traits, like the thickness of their horns, and the size of their testicles, declines in older males. So….despite the fact that they are potentially declining in their fighting ability and level of sperm production, they are still improving their reproductive success! This might be because…

(4) Male traits

Male get bigger, better home ranges as they age, while females move around a smaller area with less good-quality grass. So *perhaps* big old males are either ranging more or accessing more resources, enabling them to sire lots of offspring, despite their apparent physiological decline.

(4) Home range

Excitingly, body weight is not determined by how old you are, but how long it is until you die. In other words, sheep lose lots of weight in their final year of life, no matter whether they are five years old or ten years old. This suggests some kind of loss of function from which they never recover fully.

(4) Body weight YTD

The date on which females give birth, the weight of their lamb, and the survival of their lambs age in different ways. Meanwhile, whether they produce a single lamb or a twin doesn’t change with age at all. These are four closely-linked traits and yet they show very different ageing patterns.

(4) Female traits

In conclusion then, there is lots of variation in ageing rates between different systems, between the sexes, and even within the same system in the same sex! This really emphasizes that ageing is a very complex process acting in different ways across the body, which current evolutionary theory is not equipped to predict. To create better evolutionary predictions, we really need to do two things. First, we need to determine how natural selection on different traits changes with age: phenotypes on which selection weakens with age should show faster senescence. Second, we need to work out how the relationships between different traits change with age, especially at the genetic level: this will enable us to predict how selection on one phenotype will affect other phenotypes and therefore how evolution has shaped ageing in multiple traits.

So lots to do, and this is just the start of some very interesting studies to come- so watch this space!

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s