February 24th, 2010
Bioe 109
Winter 2010
Lecture 18
Life history evolution
What do mean by “life history” characters?
- from a “fitness” perspective, there are only two important events in life: reproduction and death.
- the traits that determine the timing and details of these two events are termed “life-history” characters, or life-history traits.
- these include the following traits:
1. age at first reproduction
2. total life span
3. mode and frequency of reproduction
4. fecundity
5. parental care
- these life-history characters exhibit enormous variation not only among species, but also among individuals of the same species.
- if much of this variation among individuals of the same species has a genetic basis (which experiments have generally shown to be the case), then this provides natural selection with an important pool of raw material to direct evolutionary change.
- fundamentally, differences in life-history traits involve differences in the allocation of energy.
- for example, a female could stop allocating energy to growth and mature at a smaller size.
- this strategy involves a trade-off since by maturing at a smaller size she will likely produce smaller litters.
- another female could, upon reaching sexual maturity allocate less energy to reproduction and more to repair and maintenance thereby keeping her in better condition and live longer.
- again, there is a trade-off: allocating less energy to reproduction means having smaller litters but this could be counter-balanced by living longer to produce additional litters.
- there is a also catch – she has to live to produce those additional litters.
- trade-offs between life-history traits make it impossible to maximize all components of fitness simultaneously.
- today, I will focus on three questions that illustrate some of these trade-offs:
1. Why do individuals age and die?
2. How many offspring should an individual produce in any given year?
3. How large should each offspring be?
Why do organisms age and die?
- the variation in life expectancies among plant and animal groups is astounding.
- annual plants complete a single life cycle in less than a year whereas some trees live for thousands of years.
- the American Bristlecone pine, for example, is believed to live for some 5,000 years.
- the variation in life expectancies among animal species is also substantial but not as extreme as seen among plants.
- at the low end of the spectrum, we see many species of insects and crustaceans that live only for a few days.
- at the other extreme are tortoises that may live for hundreds of years.
- last fall, the oldest animal was reported – a 405 year-old clam dredged up off the coast of Iceland.
- last spring some deep-water corals living off Hawaii were estimated to be about 2,700 and 4,300 years old.
- what is the reason for this enormous variation in longevities?
- all else being equal, aging reduces an individual’s fitness.
- because it reduces fitness, aging should be opposed by natural selection.
- why then do organisms age and die?
- there are two main theories for why individuals age and die.
- these are the “rate-of-living” theory and the evolutionary theory.
The “rate-of-living theory” (ROL)
- the “rate-of-living” theory holds that aging is caused by the accumulation of irreparable damage to cells and tissues.
- perhaps we can also call this theory the “live fast die young theory”.
- all species are driven by natural selection to resist and repair cell damage to the maximum extent that is physiologically possible.
- they have thus reached the limit of biologically possible repair.
- in other words, populations would have exhausted the genetic variation that could enable them to repair biological damage.
- the ROL theory makes two important predictions:
1. lifespan should correlate negatively with metabolic rate.
- this is because cell and tissue damage is caused in part by the by-products of metabolism.
2. longevity should not respond to selection.
- species should not be capable of evolving longer life spans when subjected to artificial selection.
- this is because no genetic variation should exist for genes that could increase longevity.
- the first prediction has been tested with data from all 14 orders of mammals.
- if the theory is correct, then all mammals should expend roughly the same amount of energy per gram of body tissue per lifetime.
- they can burn this slowly over a long lifetime or fast over a shorter period.
- the data from different mammals clearly refute this prediction - within orders of closely related species metabolic rate varies dramatically.
- for example in bats the range is 325 to 1,102 kcal/g/lifetime.
- this is far too variable.
- as a group bats have metabolic rates similar to other mammals of the same size but they live 3 times as long.
- in marsupials very different patterns are seen.
- energy expenditure ranges from 43 to 406 kcal/g/lifetime, well below other mammals of similar size.
- one may predict from this pattern that marsupials would be longer lived but they are not - lifespans are significantly shorter than other mammals of the same size.
- these patterns are inconsistent with the rate of living theory.
- the second prediction of the ROL theory has been tested in several species, most notably D. melanogaster.
- in a study published in 1984, Luckinbill et al. selected for both shorter and greater longevity.
- they were not able to get a response to selection for shorter lifespans (this may indicate that there is no additive genetic variation for the trait , or that natural selection was opposing the attempt to manipulate this trait).
- in lines selected for increased longevity they achieved a strong response.
- after only 15 generations, mean lifespan had increased from about 35 days to about 60 days.
- can the ROL theory explain this result?
- yes, if metabolic rate in these lines had decreased.
- this was tested and found to be true, but only for the first 15 days of life.
- thus, the ROL theory cannot account for the dramatic response to selection found by Luckinbill et al. (1984).
The evolutionary theory
- the results presented so far present a paradox.
- if selection can favor longer life spans as it has in bats, then why has it not done so in other species?
- if additive genetic variation exists for increasing life span in Drosophila then why doesn’t longevity increase?
- the evolutionary theory aims to explain this paradox.
- this theory states that aging is not caused by cell and tissue damage per se but by the inability of natural selection to prevent this damage from occurring later in life after the peak reproductive period has passed.
- the failure to repair such damage could be the result of either:
1. deleterious mutations, or
2. trade-offs between repair and reproduction.
- there is support for both possibilities.
- first, there are mutations in human that are manifest later in life (Huntington’s and hereditary non-polyposis colon cancer) that would be highly deleterious if expressed early in life, but are only mildly deleterious later in life.
- these mutations persist and cause individuals to die because they experience very low selection pressures.
- the explanation for this result is that mutations having a beneficial effect early in life have a detrimental effect later in life.
- this is a form of pleiotropy.
- pleiotropy is a type of genetic correlation that results from a gene having more than one phenotypic effect.
- in the case considered here the effect is antagonistic - good early but bad later.
- it has thus been called “antagonistic pleiotropy”.
- selection experiments in Drosophila have found that there is a trade-off between reproduction and longevity.
- similar trade-offs have been described in many species (collared flycatchers, opossums, giant lobelias).
- furthermore, a growing number of genes acting to increase longevity at the expense of early fecundity have been characterized (two discussed in the textbook are the age-1 locus in C. elegans, and the methuselah locus in Drosophila).
- so, the best answer for why species differ age and die is first, that natural selection is incapable of preventing damage and death to organisms late in their lives and second, some inevitable trade-offs exist between fecundity and total life span.
How many offspring should an individual produce in a given year?
- two patterns can be described in species strategies of reproduction.
- the first occurs when a species reproduces only once and then dies.
- this is called semelparity.
- the second, and more common, pattern is for a species to reproduce several times throughout life.
- this is termed iteroparity.
- one of the best known examples of semelparity involves the five species of Pacific salmon belonging to the genus Oncorhynchus.
- these species are anadromous - they spawn in freshwater, spend as little as a month to two years in freshwater before heading out to sea to grow and mature.
- they return to river of origin to spawn, hang around the stream trying to defend their redds where they have eggs deposited and then die.
- pink salmon live for two years, both spent in saltwater - the chinook or king salmon may live to 7 years of age.
- what are the selective pressures that have favored this semelparous lifestyle?
- other species of salmon like the Atlantic salmon, Salmo salar, are not entirely semelparous.
- typically, about 25% of the Atlantic salmon that spawn are capable of surviving the winter in their natal streams, return to the sea the following spring and then spawn again later that same year.
- it is usually the largest individuals that are capable repeated spawning.
- why don’t Pacific salmon exhibit the same strategy? It would seem that a mutation that allowed a Pacific salmon to spawn even twice would have an enormous selective advantage.
- unusual examples of semelparity are also found in plants.
- many large and long-lived plants such as trees reproduce continuously throughout their lives.
- oaks, for example, produce acorn sets every year and commonly live to 150 years or older.
- a long life expectancy in plants does not necessarily guarantee iteroparity.
- for example, one species of bamboo is known that commonly lives to an age of 120 years without reproducing even once.
- at the end of this long period, this bamboo reproduces once with an enormous reproductive effort and then dies!
- how did this life-history evolve?
- once again, it appears that natural selection is rarely capable of optimizing these two fitness components - longevity and fecundity - simultaneously.
- how do these trade-offs come about?
- life-history theory suggests that the best strategy for an individual would be to reproduce early in life rather than later, and to reproduce as soon after birth as possible.
- in species that have high early mortalities reproduction should occur early.
- however, in species that have lower mortality rates (particularly in the juvenile and early adult stages), individuals can afford to delay their reproductive effort until they are older and larger.
- in this situation, the relative fitness of a genotype that delays reproduction to a later age may well end up being greater than a genotype that chooses to reproduce at a smaller size.
- how can this happen?
- the answer appears to be that in many species, body size is positively correlated with fecundity - particularly in females.
- therefore, females that delay reproduction to more advanced ages have a greater reproductive success (and thus higher relative fitness) than those that chose to reproduce at an earlier age.
- an obvious complication arises in iteroparous species, namely, how much energy should be expended on reproducing in any given year?
- so, for an iteroparous species how many offspring should an individual produce in a given year?
- this question has been addressed most thoroughly in birds.
- the reason for this is that the number of eggs laid by any female can be easily recorded and also easily manipulated (by either adding or removing eggs).
- assuming that the size of individual eggs is fixed, how many eggs should a bird lay in any one clutch?
- the simplest answer, first put forward by David Lack in 1947 is that selection will favor the clutch size that produces the largest number of surviving offspring.
- Lack’s model assumes a trade-off in which the probability that an individual survives declines with increasing clutch size.
- the number of surviving offspring per clutch is thus equal to the number of eggs laid times the probability that any one offspring will survive.
- many researchers have tested Lack’s hypothesis.
- surprisingly, the vast majority have found that birds lay fewer eggs than predicted by Lack’s hypothesis.
- let’s consider one example.
- one large study was conducted in Wytham Wood near Oxford, England by Boyce and Perrins involving the great tit (Parus major).
- they had a data set involving 4,489 clutches extending over a period of 23 years (1960 to 1982).
- the range in clutch size was 1 to 17 eggs.
- the mean clutch size was found to 8.53 eggs.
- Boyce and Perrins also recorded the survivorship of clutches of different sizes.
- they found the highest survivorship was for clutches of 12 eggs.
- this is a rather dramatic difference between the number predicted by Lack’s hypothesis and that actually observed.
- how can we explain these results?
- Lack’s hypothesis must make one, or more, assumptions that are incorrect.
- let’s consider three of these assumptions.
- Lack’s hypothesis assumes that:
1. no trade-offs between reproductive efforts across years.
- clearly, reproduction has costs and borne this out.
- many studies added extra eggs to nests and found the extra effort expended by parents on rearing the larger number of young has a negative impact on subsequent clutches.
- therefore, when reproduction is costly selection favors some “restraint” in effort such that the optimal clutch may be less than the most “productive” clutch.
2. clutch size only affects viability.
- being part of a large clutch may have other costs that must be considered as well.
- one study that examined these extra costs was by Schluter and Gustafsson (1993).
- they manipulated clutches of collared flycatchers and monitored the chick’s subsequent life histories.
- they found a negative relationship between the size of a clutch the young produced and how extensively the clutch in which they had reared had been manipulated.
- this result shows that clutch size influences not only survival but also offspring reproductive success.
- when larger clutches give rise to reduced offspring reproductive success, the optimal clutch size will be less than the most numerically productive clutch size.
3. no year-to-year variation in optimal clutch size.
- when there is variation among years, the best measure of a genotype’s fitness is not the arithmetic mean fitness but the geometric mean fitness.
- this has the effect of reducing the optimal clutch size to a lower value than predicted by the arithmetic mean fitness.
- in summary, Lack’s hypothesis is not upheld in natural populations because of a number of factors that act to favor a smaller clutch size than predicted by his model.
How big should each offspring be?
- given that an organism can invest a certain amount of energy in a reproductive period, we can ask whether that energy is better invested in many small or a few large offspring.
- a trade-off between size and number of offspring should be fundamental.
- as expected, therefore, many researchers have found a correlation between size and number of eggs.
- what is the optimal compromise?
- the analysis of Smith and Fretwell (1974) has led to an interesting conclusion
- this is presented in the textbook on pages 509-510.
- their model assumes that:
1. a trade-off exists between size and number of offspring.
- see figure 13.25(a)
2. a minimum size exists below which individual offspring do not survive.
- see figure 13.25(b)
- above this size, the probability of survival is an increasing function of size.
- with these two assumptions, we can predict the parental fitness gained from a single clutch of offspring of a given size.
- this is simply the number of offspring produced multiplied by the probability that any individual offspring will survive.
- this is shown in figure 13.25(c).
- in many cases, the optimal offspring size predicted by this model is intermediate.
- why is this result interesting?
- it is interesting because selection on parental fitness often favors offspring that are smaller in size than that favored by selection acting on offspring fitness (which should be as large as possible since this maximizes their probability of survival).
- this identifies a conflict between parents and offspring - one of a large number that have been identified and studied.