The schedule for the next few lectures will be slightly different than the one in the syllabus:
Friday February 16 - sex, recombination, Muller's rachet (Mary)
Monday February 19 - Holiday
Wednesday February 21 - Speciation mechanisms (Peter)
Friday February 23 - Gradual versus punctuated evolution (Peter)
Self-fertilization versus cloning. It's important to know that an organism can be asexual in two different ways. It can self- fertilize, making eggs and sperm and then combining them. This is a severe form of inbreeding and rapidly eliminates all heterozygosity. Alternatively, it can clone itself, making offspring with an exact copy of its genome. This method does not reduce heterozygosity, it just leads to a lot of (possibly heterozygous) individuals who are all exactly alike except for new mutations.
Twofold cost of sexual reproduction. Consider a population of
normal diploid individuals with gender. One exceptional female arises
who is homozygous for a mutation that makes her clone herself asexually.
Suppose that each sexual pair produces, on average, two successful offspring and so the population is stable. Assume also that the asexual female can produce two offspring. (This is the same as saying that the male doesn't contribute anything to the female's ability to raise her offspring. This is true in most plants and many animals. For example, a tree which receives a pollen grain gets no further help from the father, and has to produce the expensive seed all by itself.)
Every generation the number of asexual individuals doubles. The number of copies of the asexuality gene doubles too. Eventually it should crowd out the other form; it has a twofold advantage.
Given this reasoning, why do species without paternal care ever reproduce sexually?
No reshuffling. One theory is that asexuals are at a disadvantage because they inherit all their genes as a single block and can't reshuffle. This theory can be expressed either positively or negatively; the negative form was put forward by Muller and is called Muller's ratchet. (A ratchet is a device that keeps a gear from slipping backward. Muller's ratchet keeps a population from slipping backward toward higher fitness.)
Imagine a pool of chromosome copies. They have different numbers of mildly harmful mutations that have not been wiped out by selection yet. Some chromosomes have more harmful mutations, and some have fewer. By chance, the group with the fewest harmful mutations could be lost by drift or by additional mutations. Then the overall fitness would decrease permanently, because there is no way to reduce the number of harmful mutations (except waiting for a rare back mutation). You can't select for a type of chromosome if there are no copies of it around.
Each loss of a less-mutant type of chromosome is one click of the ratchet.
The other way to state the same basic idea is to think about favorable mutations. When a favorable mutation arises in an asexual, it will increase (unless lost to drift) until all individuals are descended from it. This has two effects. Any other favorable mutations that arise on different copies (not descended from the first one) will either disappear or cause the first one to disappear. There is no way to get both favorable mutations on the same chromosome unless they happen to occur there. Also, any mildly bad mutations on the same chromsome as a highly favorable mutation will hitchhike along with it and become fixed.
Recombination allows a species to escape from Muller's ratchet. Two chromosomes with unfavorable mutations can recombine and produce an offspring with no unfavorable mutations. It also allows a species to combine favorable mutations together even if they occured on different chromosome copies. The argument, then, is that sex is preserved because over the long term asexuality leads to deterioration of the genome.
This theory predicts that asexuals should be young species (presumably with no future) which is mainly true, but bdelloid rotifers are an exception. The group has a long fossil record and many living species, and no one has ever caught them having sex. Their genomes show no sign of recombination at all. John Maynard Smith called them an evolutionary scandal.
Quicker adaptation. A second argument is that sexual reproduction allows the rapid generation of new forms. You don't have to wait for mutation; you can get new phenotypes by new gene combinations. This may enable organisms to adapt faster, especially to evolving diseases and parasites. An asexual organism is gambling that her genotype will be a success for her children. A sexual organism will have many diverse children; perhaps one of them will be successful even if the parent's genome is poorly suited to new conditions.
This argument is supported by the fact that organisms which have sex only occasionally often do it when they are under stress. Aphids have males only in the fall, when they are about to face the challenge of winter. Some fungi have sex only to make spores, which are used to survive harsh conditions and spread over long distances.
On the other hand, this theory predicts that asexuals should dominate in very stable environments and sexuals in more disturbed ones; but the plants that first invade a forest fire or human disturbed area are often asexuals, whereas climax forest plants are usually sexual. The asexual advantage of not needing to find a mate may be one reason why they do well in disturbed areas. It may also be that while disturbed areas seem difficult, they have less disease and predation because predators haven't yet adapted.
Repair of damage. A more recent view is that recombination (and thus sex) evolved mainly so that DNA could be proofread and corrected. Having two copies allows correction of problems such as stuck-together nucleotides or gaps. This could explain why bacteria, which are normally haploid, have sex-it gives a chance to proofread. In diploid organisms, it doesn't immediately explain sex: presumably a diploid can proofread its two chromosomes against each other. However, in the long term it may be difficult to remain diploid without sex. Genes which are only needed in one copy will tend to lose the other copy. Eventually the unused copies will be so badly deteriorated that they won't be good for proofreading any more. In a sexual diploid this is not as likely because the occasional production of homozygotes will clean out the bad copies.
However, it's not clear that this sort of long-term selection is strong enough to maintain sex either. The question of sexual reproduction is still a very hot debate in the field.
Strange forms of sexual reproduction. The range of diversity is surprising.
In some lizards, salamanders, and fishes, there are species which are all female. The lizards are parthenogenetic and no males are needed, but the females in salamanders and fishes do need to mate, because sperm penetration is needed to start the egg developing; but they discard the male genome, and use a complete copy of the female genome instead. They rely completely on males of related sexual species. There is a tricky evolutionary balance here-if the asexuals are too successful they will drive out the sexuals and then die for lack of males. They can also be wiped out if the sexual males evolve an ability to detect them (there is no advantage to the male in mating with one of these females). It seems like a rather disadvantagous system but it is apparently easy to evolve and has happened several times. These species probably have short lifespans.
Nematodes have males and hermaphrodites. Males are produced by rare development of an egg with only one set of chromosomes instead of two. Nematodes are highly inbred because the hermaphrodites self-fertilize (instead of making exact copies of themselves).
Ferns and some other plants have an alternation of generations between haploid and diploid. The diploid plant produces spores which develop into haploid individuals (without fertilization). The haploid individuals produce sexual gametes which combine and form diploid individuals. Both kinds grow into decent-sized plants, but they look different and were not easy to identify as the same species.
Some yeasts do the same, except that both haploids and diploids can also reproduce asexually by cell division. So the yeast strain may be haploid for many generations, then mate and become diploid and stay that way for many generations. Yeasts also have gender, but many are capable of changing gender as required. A stock that starts out all female will have half the individuals switch to male.
Some fishes are male in one part of the life cycle and female in another; usually male when they are small (producing sperm is less energy-intensive than producing eggs) but not always.
Flowering plants often have male and female on the same flower or at least on the same plant, but use genetic methods to prevent self-fertilization. For example, they may have one or more self- incompatibility loci, genes which prevent fertilization unless the pollen genotype is different from the mother's genotype. (These genes, like immune-system genes, are among the most diverse known.) On the other hand, some flowering plants are partial or complete self-fertilizers; some go so far as to develop unopened flowers so that the pollen cannot escape.
Questions to think about. For many of these questions there are no solid answers yet.