Chapter 12 THE GENETICAL THEORY OF NATURAL SELECTION Important points to remember about natural selection: 1. Natural selection is not the same as evolution. Evolution requires the origin of variation by mutation or recombination, followed by the change in gene frequencies. natural selection and genetic drift do not account for the origin of variation. 2. Natural selection is different from evolution by natural selection. 3. Natural selection can have no evolutionary effect unless phenotypes differ in genotype. 4. Natural selection is variation in average reproductive success; a feature cannot evolve by natural selection unless it makes a positive contribution to the reproduction or survival of individuals that bear it. 5. In some instances, genotypes differ in survival and fecundity but the proportion of genotypes and alleles stay the same from one generation to another. Genetic drift, inbreeding and gene flow act at the same rate in all loci in a genome. The allele frequency changes caused by natural selection in a sexually reproducing species proceed largely independently at different loci. FITNESS The relationship between phenotype and fitness, and genotype and phenotype causes a relationship between fitness and genotype, which determines whether or not evolutionary change occurs. Modes of selection The relationship between phenotype and fitness can be described as one of three modes of selection. For a continuously varying trait selection is 1. Directional if one extreme phenotype is fittest 2. Stabilizing or normalizing if an intermediate phenotype is fittest 3. Diversifying or disruptive if two or more phenotypes are fitter than the intermediate Which genotype has the greatest fitness depends on the relationship between phenotype and genotype. If the homozygous at either extreme was the fittest, the population will become fixed for one of the homozygous, but that would not be the case if the heterozygous was the fittest.
If the heterozygote has greatest fitness, it is said to be overdominant; if it has lowest fitness, it is said to be underdominant. The disadvantageous allele is lower in frequency and may be entirely eliminated. That is called purifying or negative selection. The relationship between phenotype and fitness can depend on the environment, because different environmental conditions can favor different phenotypes. Defining fitness A genotype is likely to have different phenotypic expressions as a result of environmental influences on development, so the fitness of genotype is the mean of the fitness of its several phenotypes. The fitness of a genotype is the average lifetime contribution of individuals of that genotype to the population after one or more generations measured at the same stage in the life history. "Fitness is a property, not of an individual, but of a class of individuals for example homozygous for allele A at a particular locus. Thus the phrase expected number of offspring means the average number, not the number produced by some one individual. If the first human infant with a gene for levitation were struck by lightning in its pram, this would not prove the new genotype to have low fitness, but only that the particular child was unlucky." Maynard-Smith, J. (1989) Evolutionary Genetics. Reproductive success includes not only the number of offspring produced but also survival since survival is a prerequisite for reproduction. Fecundity is the potential reproductive capacity of an individual or population, e.g. number of eggs produced by female. Fecund individuals may not contribute to the growth of the population due lack of opportunities to mate or other external factor. Fertility is the actual reproductive rate of an organism or population. A measure of the number of individuals produced. It is usually describe in terms of the number of individuals per female. Genotype absolute fitness: see example on page 306, second paragraph. (proportion surviving) x (average fecundity) = R, per capita replacement rate or biotic potential a population made up of this genotype will grow by a factor of R at every generation. The more fit genotype will increase in numbers faster. In a population, the genotypes with the higher R will increase faster than those with lower R, per capita growth rate. The relative fitness of a genotype is its value of R relative to that of some reference genotype. Relative fitness is quantified as the average number of surviving progeny of a particular genotype compared with average number of surviving progeny of competing genotypes after a single generation, i.e. one genotype is normalized at w = 1, and the fitnesses of other genotypes are measured with respect to that genotype. By convention, the reference genotype is often the one with the highest value, and it is given the relative fitness of 1.0.
The mean fitness (w) is the average fitness of individuals in a population relative to the fittest genotype. Frequency of genotype A x R A + frequency of genotype B x R B = mean fitness Coefficient of selection is the amount by which the fitness of one genotype is reduced relative to the reference genotype. If the relative fitness of a genotype is 0.75, then the coefficient of selection is 0.25 The coefficient of selection measures the selective advantage of the fitter genotype, or the intensity of selection against the less fit genotype. The rate of genetic change under selection depends on the relative, not the absolute, fitnesses of genotypes. Components of fitness Survival and female fecundity are two components of fitness. Rate of population increase: When generations overlap, the usual way of measuring the absolute fitness of a genotype is expressed as the per capita rate of population increase per unit time. The rate at which the population is increasing or decreasing in a given year expressed as a percentage of the base population size. It takes into consideration all the components of population growth, namely births, deaths and migration. http://new.hst.org.za/indic/indic.php/18/ This rate depends on the proportion of individuals surviving to each age class and on the fecundity of each age class. The age at which female have offspring is important and not only the number of females should be considered. A genotype may have superior survival and another genotype may have superior fecundity, but the overall fitness determines the outcome of natural selection. COMPONENTS OF SELECTION IN SEXUALLY REPRODUCING ORGANISMS Biotic potential (r) is capacity of a population to increase. It is the maximum rate of increase of a population living under ideal conditions. Ideal conditions exist when conditions are uncrowded. It usually is potential and not actual. It is also called the per capita rate of increase or intrinsic rate of natural increase. It is a measure of the instantaneous (unit of time) rate of change of population size per individual. It is the difference between the instantaneous birthrate and the instantaneous death rate.
When generations overlap, the absolute fitness of a genotype may be measured by its per capita rate of population increase per unit of time, r. This rate depends on the proportions of individuals surviving to each class and on the fecundity of the age class. The per capita rate of increase, r, is strongly affected by the age at which females have offspring, not just by their number. The genotypes may different in the length of a generation, e.g. genotype A produces 5 offspring at the age of 6 months and genotype B produces 5 offspring at the age of 12 months, the rate of increase, fitness, of A is about twice that of B, because A will have twice as many generations of descendants per unit of time. In sexually reproducing species, genotypes transmit haploid gametes. The genotype frequency depends on the allele frequencies of the uniting gametes. These allele frequencies are affected by several components of selection ZYGOTIC SELECTION A. Viability. The probability of survival of the genotype through each of the ages at which reproduction can occur. B. Mating success. The number of mates obtained by an individual affects the individual s progeny, as is often the case for males, but less often for females whose eggs may all be fertilized by a single male. C. Fecundity. The average number of viable offspring per female. GAMETIC SELECTION A. Segregation advantage. And allele has an advantage if it segregates into more than half the gametes of a heterozygote. B. Gamete viability. Dependence of a gamete s viability on the allele it carries. C. Fertilization success. An allele may affect the gamete s ability to fertilize an ovum. See Fig. 12.3 and Table 12.1 on page 307. MODELS OF SELECTION DIRECTIONAL SELECTION The replacement of relatively disadvantageous alleles by more advantageous ones is the fundamental basis of adaptive evolution.
Replacement occurs when the homozygote for an advantageous allele has fitness equal to or greater than that of a heterozygote or any other genotype in the population. Advantageous alleles may be very common if under previous environmental conditions it was selectively neutral or was maintained by on of several forms of balancing selection. An advantageous allele may be initially very rare if it is newly arisen by mutation or if it was disadvantageous before an environmental change made it advantageous. An advantageous allele that increases from a very low frequency is said to invade the population. Unless an allele can increase in frequency when it is very rare, it is unlikely to become fixed in the population. The mystery of the brilliant pattern of the coral snake. A character state with even a minuscule advantage will be fixed by natural selection as long as no other evolutionary factors intervene. Purifying selection reduces the frequency of a deleterious allele. The number of generations required for an advantageous allele to replace one that is deleterious depends on the initial allele frequencies, the selection coefficient and the degree of dominance. An advantageous allele will take very long time to increase in frequency if it is very rare because very few homozygous are formed. Once established, and advantageous recessive reaches fixation very rapidly as the deleterious dominant is eliminated. An advantageous allele can increase in frequency more rapidly if it is dominant. After a dominant advantageous allele attains high frequency, the deleterious recessive allele is eliminated very slowly because a rare recessive allele occurs mostly in heterozygous form, and is thus shielded from selection. Deleterious alleles in natural populations Directional selection fixes advantageous alleles but deleterious alleles often persist because they are repeatedly reintroduced either by mutation or by gene flow from other populations in which they are favored by a different environment. Selection and mutation The frequency of the deleterious allele moves toward a stable equilibrium that is a balance between the rate at which it is eliminated by selection and the rate at which it is introduced by mutation or gene flow. The equilibrium frequency of a deleterious recessive allele is directly proportional to the mutation rate and inversely proportional to the strength of selection.
If the deleterious allele is dominant or partially dominant, its equilibrium frequency will be lower than if it was recessive because of the selection against both the homozygous and heterozygous carriers. The mutation-selection balance explains why many chromosomes carry rare mutations that slightly reduce fitness in the heterozygous condition and are strongly deleterious or even lethal when homozygous. Selection and gene flow Environmental or extrinsic selection is done by the environment. Very often, different environmental conditions favor different alleles in different populations of a species. Frequency of deleterious allele will arrive at an equilibrium between the gene flow of the deleterious allele and the elimination by selection. A smooth cline in allele frequencies may be established if there is gene flow among populations along an environmental gradient over which the fitness of different genotypes changes. See Fig. 12.11. In the absence of gene flow, there will be an abrupt change in gene frequencies among populations of the gradient. The width of the cline over which gene changes is directly proportional to the distance that genes disperse, and inversely proportional to the strength of selection against the allele. If selection is strong relative to gene flow, a steep cline in allele frequency will result, and the populations are strongly differentiated. Stronger selection narrower cline; greater dispersal distance, broader cline. POLYMORPHISM MAINTAINED BY BALANCING SELECTION A wealth of variation is maintained by several means: 1. Recurrent mutation producing deleterious alleles, which are subject to only weak selection. 2. Gene flow of locally deleterious alleles from other populations in which they are favored by selection. 3. Selective neutrality (e. g. genetic drift) 4. Maintenance of polymorphism by natural selection. Balancing selection is selection that maintains polymorphism. Polymorphism refers to a variety of alleles in a population that usually represent different phenotypes. Balancing selection maintains two or more stable frequencies of phenotypic forms. Example: Human birth weight. 1. Heterozygote advantage
Heterozygote advantage occurs when the heterozygote has greater fitness than either homozygous. If the heterozygote is more fit, both alleles will be passed on to successive generations producing the three genotypes among the zygotes. Overdominance and single-locus heterosis are other names. Heterozygous advantage depends on the balance between the fitness values of the two homozygotes. Balancing selection is maintained by heterozygous advantage. Example: Heterozygous individuals carriers of the sickle cell anemia have the advantage of showing little anemia and having the resistance to malaria, while both homozygous are either anemic or vulnerable to malaria. 2. Antagonistic and varying selection The opposing forces acting on the sickle-cell polymorphism are an example of antagonistic selection. Polymorphism is maintained because the heterozygote has greater fitness than both homozygotes. Temporal fluctuation occurs when a changing environment favors one genotype over another in different generations. Spatial fluctuation occurs when different genotypes are best adapted to different microhabitats or resources. Spatial variation in the environment is most likely to maintain polymorphism if different homozygotes with a single population are best adapted to different microhabitats or resources. This is sometimes called multiple niche polymorphism. Example: the bimodal distribution of bill width in populations of the African finch black-bellied seedcracker. Soft selection occurs when the number of survivors in a population is determined by competition for a limiting factor such space or food, and the relatively superior genotype has a higher probability of survival, e.g. depends on the number of competitors for a single resource, such nesting sites or food. Hard selection occurs when the survival of an individual depends on its absolute fitness, not on the density of its competitors, e.g. resistance to insecticides depends on having the resistance gene and not on the number of individuals in the population. FREQUENCY DEPENDENT SELECTION The fitness of a genotype depends on the genotype frequencies in the population. The population then undergoes frequency-dependent selection. Inverse frequency-dependent selection
Inverse frequency-dependent selection occurs when the more rare phenotype is in the population, the greater its fitness. o Genotype fitness declines with its frequencies. The self-incompatibility alleles of many plant species are an example. The S locus in the cabbage family: 1. The "S locus" consists in reality of three loci. 2. There are multiple alleles of these genes, up to 50. 3. The proteins coded by these loci are located one in the membrane of the stigma cells, another in the cell wall of the stigma cells, and the third is secreted by mature pollen grains. 4. If the proteins secreted by the pollen are the same as one or both of the proteins in the cell membrane and wall of the stigma, the pollen grain does not form a pollen tube. Similarity of alleles means that they are probable from the same plant. 5. If the proteins secreted by the pollen tube are different from both of those in the cell wall and the cell membrane of the stigma, the pollen tube forms. The pollen comes from a different plant. Also see the example of the cichlid Perissodus microlepis on page 319, fig. 12.16. The more intense competition among individuals of the same genotype would then impose inverse frequency-dependent selection. THE EVOLUTION OF THE SEX RATIO Individuals that vary in the sex ratio of their progeny can differ in the number of their grandchildren and later descendants, and thus differ in fitness if this is measured over two or more generations. The average per capita reproductive success of the minority sex is greater than the majority sex. Selection favors genotypes whose individual sex ratios are biased toward that sex that is in the minority in the population as a whole. Fisher s principle explains why for most species, the sex ratio is approximately 1:1. William Hamilton expounded Fisher s argument in his 1967 paper on Extraordinary sex ratios as follows, given the assumption of equal parental expenditure on offspring of both sexes. 1. Suppose male births are less common than female. 2. A newborn male then has better mating prospects than a newborn female, and therefore can expect to have more offspring. 3. Therefore parents genetically disposed to produce males tend to have more than average numbers of grandchildren born to them. 4. Therefore the genes for male-producing tendencies spread, and male births become more common. 5. As the 1:1 sex ratio is approached, the advantage associated with producing males dies away.
6. The same reasoning holds if females are substituted for males throughout. Therefore 1:1 is the equilibrium ratio. Hamilton, W. 1967. Science 156 (3774): 477 88. A genotype that produces a sex ratio of 0.5 represents an evolutionarily stable strategy of ESS. MULTIPLE OUTCOMES OF EVOLUTIONARY CHANGE - DISRUPTIVE SELECTION The initial genetic conditions usually determine the path of genetic change a population will follow. In other words, the evolution of a population often depends on its previous evolutionary history. POSITIVE FREQUENCY-DEPENDENT SELECTION In positive frequency-dependent selection, the fitness of a genotype is greater the more frequent it is in a population. Whichever allele is initially more frequent will be fixed. See the example of the Heliconius butterflies on page 321. Heliconius melpomene and H. erato butterflies have very distinct races in Central and South America. Both species are distasteful to predators, which tend to avoid both species after an unpleasant experience by attempting to eat them. When one race butterflies are released in the area of another race, they are selected against by strong bird predation that does not recognize the introduced butterflies with the unusual color pattern. The phenotypes common to the area are recognized by predators, but the introduced phenotypes are not. The recapture of both phenotypes showed a strong selection against the introduced race, against their genotypes. Gene flow is countered by positive frequency-dependent selection. This is an example of Mullerian mimicry. 1. Heterozygote disadvantage Underdominance is another name for heterozygote disadvantage. If monomorphism is the stable solution, heterozygotes will be selected against. If a mutant gene appears or an allele is introduced in a monomorphic population, it will be carried mostly by heterozygotes. If these heterozygotes are less fit than the homozygotes, it will eventually be eliminated. Underdominance causes disruptive selection. 2. Adaptive landscape
Population size affects natural selection: Advantageous mutations may be lost by chance rather than become fixed; deleterious mutation may become fixed for the same reason. Population bottleneck provides the opportunity to increase the frequency of deleterious alleles When the heterozygote is less fit than either homozygote, genetic drift is necessary to initiate a shift form one homozygous equilibrium to another. See Fig. 12.7D. Adaptive landscapes are graphs that visualize the relationship of genotypes and reproductive success. The graphs plots fitness (y axis) against genotype or mutants (x axis). Each point of the curve represents the average fitness of individuals. An evolving population typically climbs uphill to an adaptive peak in the fitness landscape, by a series of small genetic changes, until a local optimum is reached (Fig. 1). There it remains, unless a rare mutation (or migration or genetic drift) opens a path to a new, higher fitness peak. Note, however, that at high mutation rates this picture is somewhat simplistic. A population may not be able to climb a very sharp peak if the mutation rate is too high, or it may drift away from a peak it had already found; consequently, reducing the fitness of the system. The process of drifting away from a peak is often referred to as Muller's ratchet. Muller s Ratchet refers to the process by which asexual organisms accumulate deleterious mutations. See for more information: http://en.wikipedia.org/wiki/muller%27s_ratchet Figure 1: Sketch of a fitness landscape. The arrows indicate the preferred flow of a population on the landscape, and the points A, B, and C are local optima. The red ball indicates a population that moves from a very low fitness value to the top of a peak. Illustration by C.O. Wilke, 2001. Source of Adaptive Landscape: http://en.wikipedia.org/wiki/fitness_landscape 3. Interaction of selection and genetic drift
In finite populations, allele frequencies are simultaneously affected by both selection and chance. The effective size of a population, N e, and the strength of selection both affect changes in allele frequencies. Random genetic drift is negligible if selection on a locus is strong relative to the population size. If selection is so weak that the allele frequencies change mostly by genetic drift: the alleles are nearly neutral. Selection determines the outcome of evolution if the population is large. In a sufficiently small population, genetic drift is more powerful than selection. Selection alone cannot move a population down the slope of one peak in the adaptive landscape and across a valley to the slope of another peak, even if the second peak is higher: a population does not first become poorly adapted so that it can then become better adapted. In a very low population size, allele frequencies may fluctuate so far by genetic drift that they cross the adaptive valley, after which selection can move the population uphill to the other better adapted peak. Effects of population size on the efficacy of selection have several important consequences: 1. If the population is large, the equilibrium allele frequency predicted from its genotypes fitnesses would likely wander by genetic drift in the vicinity of the equilibrium frequency. 2. A slightly advantageous mutation is less likely to be fixed by selection if the population is small than if it is large, because it is more likely to be lost simply by chance. Conversely deleterious mutations can become fixed by genetic drift, especially if selection is weak and the population is small. 3. Population bottlenecks provide temporary conditions under which genetic drift may counteract selection so that a deleterious allele may increase in frequency. When the heterozygote is less fit than either homozygote, genetic drift is necessary to initiate the shift from one homozygous equilibrium state to the other. Genetic drift and selection may act in concert to accomplish what selection alone cannot, moving a population from one adaptive peak to another. Example: Some chromosome rearrangements such translocation and pericentric inversion are thought to conform to the model of underdominance because heterozygotes have lower fertility than either homozygote. THE STRENGTH OF NATURAL SELECTION Selection through both survival and reproduction can be very strong. Studies have shown that natural selection can be very strong in some circumstances, and, therefore, a strong force of evolution. See example of Biston betularia, the peppered moth in England, page 324, fig. 12.22.
MOLECULAR SIGNATURES OF NATURAL SELECTION Originally, studies of natural selection were done in the field or in the laboratory focusing on phenotypic characteristics. These efforts have been more recently complemented by inferring selection from variation in DNA sequences. Without further biological and ecological information, this approach does not reveal the causes of selection or even the function of the gene. Theoretical expectations Selection on DNA sequences may be inferred if the pattern of variation differs from patterns expected under the neutral theory of molecular evolution. If a new mutation occurs in a gene than increases fitness of the carrier versus other members of the population, natural selection will favor this individual and its descendants that inherit the gene. With time the newly mutated variant of the gene will increase in frequency relative to other variants (alleles) of the gene. As the frequency of this allele increases, neutral or slightly deleterious mutations linked to this gene will also increase; this is called genetic hitchhiking. In this manner, positive directional selection for an advantageous mutation reduces variation at closely linked sites. A strong selective sweep results in a region of the genome where the positively selected haplotype (the mutated allele and its neighbours) are essentially the only ones that exist in the population, resulting in a large reduction of the total genetic variation in that chromosome region. Whether a selective sweep has occurred or not can be investigated by measuring linkage disequilibrium, i.e., whether a given haplotype is overrepresented in the population. Under neutral evolution, genetic recombination will result in the reshuffling of the different alleles within a haplotype, and no single haplotype will dominate the population. However, during a selective sweep, selection for a positively selected gene variant will also result in selection of neighbouring alleles and less opportunity for recombination. Therefore, the presence of strong linkage disequilibrium might indicate that there has been a 'recent' selective sweep, and can be used to identify sites recently under selection. A study of genetic variation among 269 humans found evidence for selective sweeps on chromosomes 1, 2, 4, 8, 12, and 22 [1]. 1 "A haplotype map of the human genome" (2005) by the International HapMap Consortium in Nature Volume 437, pages 1299-1320. http://en.wikipedia.org/wiki/selective_sweep A selective sweep resembles a bottleneck in population size in that it reduces variation and increases the genealogical relatedness among gene copies, but a bottleneck will affect the entire genome, not just the portion surrounding an advantageous mutation. Purifying selection against deleterious mutations reduces neutral polymorphism at closely linked sites. When deleterious mutations are eliminated from a population, selectively neutral mutations linked to it are eliminated as well. This is called background selection.
Background mutations reduce neutral polymorphism. For this region of DNA, the effective population size is reduced to the proportion of gametes that are free of deleterious mutations. The effects of balancing selection (e.g. heterozygote advantage or frequency-dependent selection) are opposite to those of positive directional selection. Balancing selection favors polymorphism (multiple alleles): two or more stable frequencies. Directional selection favors a single allele. Genes subjected to balancing selection will display elevated variation in the vicinity of the selected site. Because of long-lasting balancing selection, a polymorphism may be inherited by two or more species from their common ancestor, and certain haplotypes in each species may be most closely related (has more similar nucleotide sequence) to haplotypes in the other species. Example: some chimpanzee alleles in the MHC are more closely related (has more similar nucleotide sequence) to a human allele than to other chimpanzee alleles. See Fig. 12.27. Signatures of selection Selection can leave several imprints or signatures in genomes. Departures from the patterns expected under the neutral theory will show selection. 1. Proportion of functional changes If selection has operated over a long period of time, the proportion of nonsynonymous to synonymous mutations is expected to equal 1 if all the nonsynonymous differences are neutral. But Whereas d n = nonsynonymous mutations and d s = synonymous mutations. d n/d s < 1 implies that many possible mutations are deleterious and have been purged by purifying selection. d n/d s > implies a preponderance of fixed advantageous mutations. 2. Reduced sequence diversity The principal reasoning here is that a selective sweep, by increasing the frequency of one haplotype at the expense of others, reduces heterozygosity at linked sites, so the frequencies of linked variants become more uneven. This condition may reflect adaptation to a new situation. 3. Long haplotypes and linkage disequilibrium One indication of a selective sweep is an allele that has high frequency and is strongly associated by linkage disequilibrium with distant genetic markers, thus forming a long haplotype that has high frequency.
High frequency and long haplotype associated with distant genetic markers is a sign of young alleles. High frequency is typical of old alleles but long distance LD in a chromosome region with normal recombination is likely to indicate a young allele. Recombination dissociates increasingly close alleles as time passes. Long distance LD in a chromosome region with normal recombination is likely to indicate a young allele, because recombination dissociates increasingly close alleles as time passes. Chromosome with shorter distances for LD is the result of recombination. The strength of selection can be inferred from the length and frequency of the haplotype by comparison with alleles at various distances, and its pattern of decay. Linkage disequilibrium marks selection for lactase persistence in human populations. Lactase persistence refers to the continuance of the lactase enzyme (ability to digest lactose) into adulthood. See example in fig. 12.29. Haplotypes in European, Asian and African populations are longer in sequence with lactasepersistence than in nonpersistence sequences, as expected if the lactase-persistence allele has recently increased in a partial selective sweep. Nonpersistence is the ancestral allele. Adaptive evolution across the genome There are tests for selection to thousands of genes using increasing databases on human sequence variation and other organisms. Bustamante and colleagues (2005) found a strong correlation between the strength of purifying selection of a gene and likelihood that the gene is known to display disease-causing mutations in humans. Hawks and collaborators (2007) found after studying 3.9 million SNP dataset using linkage disequilibrium decay to distinguish haplotypes that signify positive selection in African, European and East Asian population, more than 3000 selective events in the Yoruba (African), and more than 2000 in one European and two Asian populations. These sweeps occurred about 8000 years ago in Africa and 5000 years ago in Europe. The rate at which genes have undergone evolution by natural selection has greatly increased in the last 40,000 years. Hawks attributes this acceleration to the great increase of human population, which has greatly increased the supply of new mutations, and to the massive changes humans have created in their ecological and cultural environment. This site summarizes many of the main concepts presented in this chapter, and gives examples of each. http://www.d.umn.edu/~jetterso/2012lecture19-chapter12-formsofselection.htm http://www.ucl.ac.uk/~ucbhdjm/courses/b242/hz/hzpp.pdf