Population genetics of malaria resistance in humans.

The high mortality and widespread impact of malaria have resulted in this disease being the strongest evolutionary selective force in recent human history, and genes that confer resistance to malaria provide some of the best-known case studies of strong positive selection in modern humans. I begin by reviewing JBS Haldane's initial contribution to the potential of malaria genetic resistance in humans. Further, I discuss the population genetics aspects of many of the variants, including globin, G6PD deficiency, Duffy, ovalocytosis, ABO and human leukocyte antigen variants. Many of the variants conferring resistance to malaria are 'loss-of-function' mutants and appear to be recent polymorphisms from the last 5000-10 000 years or less. I discuss estimation of selection coefficients from case-control data and make predictions about the change for S, C and G6PD-deficiency variants. In addition, I consider the predicted joint changes when the two ß-globin alleles S and C are both variable in the same population and when there is a variation for α-thalassemia and S, two unlinked, but epistatic variants. As more becomes known about genes conferring genetic resistance to malaria in humans, population genetics approaches can contribute both to investigating past selection and predicting the consequences in future generations for these variants.

Isolation by distance among California sea lion populations in Mexico: redefining management stocks.

Understanding the spatial structure of a population is critical for effective assessment and management. However, direct observation of spatial dynamics is generally difficult, particularly for marine mammals. California sea lions (Zalophus californianus) are polygynous pinnipeds distributed along the Pacific coast of North America. The species' range has been subdivided into three management stocks based on differences in mitochondrial DNA, but to date no studies have considered nuclear genetic variation, and thus we lack a comprehensive understanding of gene flow patterns among sea lion colonies. In light of recent population declines in the Gulf of California, Mexico, it is important to understand spatial structure to determine if declining sea lion colonies are genetically isolated from others. To define population subdivision and identify sex biases in gene flow, we analysed a 355-bp sequence of the mitochondrial DNA control region and 10 polymorphic microsatellite loci from 355 tissue samples collected from six colonies distributed along Mexican waters. Using a novel approach to estimate sex biases in gene flow, we found that male sea lions disperse on average 6.75 times more frequently than females. Analyses of population subdivision strongly suggest a pattern of isolation by distance among colonies and challenge current stock definitions. Based on these results, we propose an alternative classification that identifies three Mexican management units: Upper Gulf of California, Southern Baja Peninsula, and Upper Pacific Coast of Baja. This revised classification should be considered in future assessment and management of California sea lion populations in Mexican waters.

Premating, not postmating, barriers drive genetic dynamics in experimental hybrid populations of the endangered Sonoran topminnow.

The timing and pattern of reproductive barrier formation in allopatric populations has received much less attention than the accumulation of reproductive barriers in sympatry. The theory of allopatric speciation suggests that reproductive barriers evolve simply as by-products of overall genetic divergence. However, observations of enhanced premating barriers in allopatric populations suggest that sexual selection driven by intraspecific competition for mates may enhance species-specific signals and accelerate the speciation process. In a previous series of laboratory trials, we examined the strength of premating and postmating barriers in an allopatric species pair of the endangered Sonoran topminnow, Poeciliopsis occidentalis and P. sonoriensis. Behavioral observations provided evidence of asymmetrical assortative mating, while reduced brood sizes and male-biased F(1) sex ratios suggest postmating incompatibilities. Here we examine the combined effects of premating and postmating barriers on the genetic makeup of mixed populations, using cytonuclear genotype frequencies of first- and second-generation offspring. Observed genotype frequencies strongly reflect the directional assortative mating observed in behavioral trials, illustrating how isolating barriers that act earlier in the reproductive cycle will have a greater effect on total reproductive isolation and may be more important to speciation than subsequent postmating reproductive barriers.

Genetic variation in a heterogeneous environment. V. Spatial heterogeneity in finite populations.

The spatial model of Levene (1953) was examined in a finite population and compared to a temporal model. The spatial model was much more effective in retaining genetic variation in a finite population. Furthermore, the haploid spatial model was more effective in retaining genetic variation than the analogous diploid absolute dominance model. This is the opposite from that found for the temporal model, where the diploid model was more effective than the haploid. Here is an example where diploidy (sexual reproduction) may be disadvantageous. A model that permitted both spatial and temporal variation to act in concert gave retention of genetic variation in situations where either spatial or temporal variation, separately, did not. The relationship between the amount of heterozygosity and the retardation factor was discussed. An example of how spatial or temporal variation affects the proportion of populations fixed after a certain number of generations was given. It seems that these models have biological analogues, several examples of which are mentioned.

Hitchhiking: a comparison of linkage and partial selfing.

Genetic hitchhiking occurs when alleles at unselected loci are changed in frequency because of an association with alleles at a selected locus. This association may be mediated either by linkage or partial selfing (inbreeding) and can affect the gene frequency and gametic disequilibrium at the neutral loci. Hitchhiking from partial selfing (unlinked loci) occurs more quickly than linkage hitchhiking and generally has a greater effect. In addition, partial-selfing hitchhiking can cause increases or changes in sign in gametic disequilibrium between neutral loci. The effects of the two types of hitchhiking with different levels of dominance, zygotic frequencies and number of selected loci are also examined. The general conditions for linkage and partial-selfing hitchhiking are outlined and the implications of hitchhiking are discussed for marker or electrophoretic loci.

Gametic disequilibrium measures: proceed with caution.

Five different measures of gametic disequilibrium in current use and a new one based on R. C. Lewontin's D', are examined and compared. All of them, except the measure based on Lewontin's D', are highly dependent upon allelic frequencies, including four measures that are normalized in some manner. In addition, the measures suggested by A. H. D. Brown, M. F. Feldman and E. Nevo, and T. Ohta can have negative values when there is maximum disequilibrium and have rates of decay in infinite populations that are a function of the initial gametic array. The variances were large for all the measures in samples taken from populations at equilibrium under neutrality, with the measure based on D' having the lowest variance. In these samples, three of the measures were highly correlated, D2, D (equal to the correlation coefficient when there are two alleles at each locus) and the measure X(2) of Brown et al. Using frequency-dependent measures may result in mistaken conclusions, a fact illustrated by discussion of studies inferring recombinational hot spots and the effects of population bottlenecks from disequilibrium values.

Genetic variation in a heterogeneous environment. I. Temporal heterogeneity and the absolute dominance model.

The conditions for a stable polymorphism and the equilibrium gene frequency in an infinite population are compared when there is spatial or temporal environmental heterogeneity for the absolute dominance model. For temporal variation the conditions for stability are more restrictive and the equilibrium gene frequency is often at a low gene frequency. In a finite population, temporal environmental heterogeneity for the absolute dominance model was found to be quite ineffective in maintaining genetic variation and is often less effective than no selection at all. For comparison, the maximum maintenance for temporal variation is related to the overdominant model. In general, cyclic environmental variation was found to be more effective at maintaining genetic variation than where the environment varies stochastically. The importance of temporal environmental variation and the maintenance of genetic variation is discussed.

HLA-sharing, recurrent spontaneous abortion, and the genetic hypothesis.

A number of studies indicates that there is a high sharing of HLA antigens in couples having recurrent spontaneous abortions. The genetic hypothesis to explain this phenomenon suggests that this fetal loss results from homozygosity of recessive lethal or deleterious alleles in gametic disequilibrium with HLA antigens. Theory predicting the lethality rate is derived when antigens are shared at one, two or three loci, given that the disequilibrium is absolute. In addition, the effects of partial disequilibrium, inbreeding, and segregation distortion on the lethal proportion are examined.

Female choice and variation in the major histocompatibility complex.

The cause of the high genetic variability in the major histocompatibility complex (MHC) is not entirely clear. Recently, two reports suggest that female mice prefer to mate with males different from them at the MHC. A model of female choice appropriate for those observations is developed here. Female choice can in fact reduce the observed proportions of homozygotes, maintain genetic polymorphism, influence mating-type frequencies and generate gametic disequilibrium.

Heterozygosity and fitness: no association in Scots pine.

The association of six quantitative traits related to fitness with heterozygosity at 12 allozyme loci has been examined in three populations of Scots pine, Pinus sylvestris. Because of several characteristics of this organism and of this extensive data set, it appeared that this study would show a positive association between heterozygosity and these traits if indeed heterozygotes had higher values for these quantitative traits. Using several different statistical techniques including analysis of variance, regression with the scaling recommended from the adaptive distance model, and multiple regression, no evidence of an association was found. For example, only between 7 and 8% of the regression tests were significant at the 5% level and half of these showed a positive association and half showed a negative association. Further, the multiple regression analysis explained on average only 5.8% of the variation observed in the six different traits and only 1.5% of this variation was explained by a positive association. Power analysis was carried out (for the first time on these type of data), both for the single locus heterozygous advantage and the association of individual multiple locus heterozygosity and the quantitative traits. For diameter and height, two traits often used in similar studies, the average power to detect a single locus heterozygous advantage of 0.10 was 0.737 and the average power to detect a mean heterozygote advantage of 0.05 per locus for multiple loci was 0.797. As a result of this study and an examination of the published results from other studies, it appears that what positive associations have been observed are probably not, in large part, due to the presence of intrinsic heterozygote advantage.

Genetic variation in a heterogeneous environment. II. Temporal heterogeneity and directional selection.

The maintenance of genetic variation is investigated in a finite population where selection at an autosomal locus with two alleles varies temporally between two environments and the heterozygote has an intermediate fitness value. When there is additive gene action and equal selection in both environments, the autocorrelation between subsequent environments must be negative for more maintenance of genetic variation than for neutrality. The maximum maintenance occurs when there is equal selection in the two environments and the autocorrelation approaches -1.0 (for a stochastic model), or when there is short repeating cycle such as one related to seasons. Also comparison of the effects of stochastic variation in selection in finite ans infinite populations is made by using Monte Carlo simulation. One situation was found where temporal environmental variation maintains genetic variation very effectively even in a small population and that is when there is evolution of dominance, i.e., the heterozygote is closer in fitness to the favored homozygote than the other homozygote. An important conclusion is that in a finite population genetic tracing of environmental change, particularly when there is a positive autocorrelation between environments or a long environmental cycle, leads to an increased loss of genetic variation making such a response undesirable in the long term, a result different from that in infinite populations.

Simulation of X-linked selection in Drosophila.

The change in gene frequency for two X-linked mutants, gamma and omega, in a number of experiments was compared to that predicted from a genetic simulation program which utilized estimated differences in relative mating ability, fecundity, and viability. The simulation gave excellent predictions of gene frequency change even when experiments were started with different initial gene frequencies in the males and females or when the two loci were segregating simultaneously. The rate of elimination was slower when there were unequal initial gene frequencies than when males and females had equal initial gene frequencies. Simulation demonstrated that this was a general phenomenon when there is strong selection but that the opposite is true for weak selection. In two other experiments, the mating advantage of wild-type males was balanced by a fecundity advantage in mutant females. In all four replicates of both experiments, the mutant was maintained for several generations at the high initial frequency but then decreased quickly and was eliminated. Results obtained restarting one of these experiments with flies from a generation after the decline in gene frequency indicated that a linked gene and not frequency-dependent selection was responsible for the unpredictable gene-frequency change in the mutant. Using a least squares technique, it was found that a recessive fecundity locus 15 map units from the omega locus gave the best fit for both experiments.

Maintenance of genetic variation with a frequency-dependent selection model as compared to the overdominant model.

A frequency-dependent selection model proposed by Huang, Singh and Kojima (1971) was found to be more effective at maintaining genetic variation in a finite population than the overdominant model. The fourth moment parameter of the distribution of unfixed states showed that there was a more platykurtic distribution for the frequency-dependent model. This agreed well with the expected gene frequency change found for an infinite population.

Selection in Finite Populations. III. an Experimental Examination.

The theory determining the probability of fixation of new mutants in a population that allows increased adaptation has been developed by Kimura. Two experimental systems were used to examine this theory. First, alleles were introduced at low frequencies into populations and their fate observed. The second approach was to follow populations with closely linked mutants that were introduced on different chromosomes and to follow the fate of favorable recombinants. These experiments allowed the investigation of the appropriateness of the mathematical model for the particular biological system. In the experiments reported here using two X-linked Drosophila mutants, y and w, the predictions of the theory were reasonably fulfilled in the first experimental system (introduction of new alleles), but not in the second system (favorable recombinants). However, the theoretical framework seems quite robust in that it allowed a satisfactory explanation of the experimental results for the second system as well. The probability of, and the time until, production of a favorable recombinant is discussed.

A two-locus neutrality test: applications to humans, E. coli and lodgepole pine.

The expected disequilibrium between two loci with k alleles at one locus and l alleles at the other is given for a sample of size n drawn from a population under neutrality equilibrium. Three different measures of disequilibrium with 95% intervals are tabulated for combinations of n, k, l and 4Nc, where N is the effective population size and c is the amount of recombination between the loci. The extent and pattern of disequilibrium are strongly dependent upon 4Nc and are somewhat dependent on n, k and l. The 95% intervals are large, particularly for low numbers of alleles and low values of 4Nc. As examples, observed disequilibrium from histocompatibility loci in humans (HLA) and electrophoretic data in E. coli and lodgepole pine were compared to these theoretical values. Using information about recombination rates, the HLA data showed more disequilibrium than neutrality expectations, whereas electrophoretic data from E. coli and lodgepole pine had somewhat less disequilibrium than neutrality expectations.

Maternal-fetal interactions and the maintenance of HLA polymorphism.

There is some empirical evidence that a fetus with an HLA antigen not present in its mother has a higher survival than a fetus sharing antigens with its mother. We have developed both single locus and two-locus theoretical models to examine this mode of selection. First, this immunologically based model appears to have the potential to maintain many alleles at a single locus and to result in an excess of heterozygotes when selection is strong. Second, substantial gametic disequilibrium is maintained between alleles at two loci for this selection mode when recombination is that observed between HLA loci A, B, and DR. Overall, it appears that this mode of selection has the potential to strongly affect genetic variation in the HLA region.

Evidence for balancing selection at HLA.

HLA data from the A and B loci for 22 populations were compared with the neutrality expectations from Ewens' sampling theory. In 25 of 44 cases, there was significantly less homozygosity than expected. Although a number of factors can affect homozygosity in this manner, upon close examination only symmetrical balancing selection appears to be consistent with these data.