Evolution of the mutation rate at a heterotic locus.

A diffusion model of the modification of mutation rates at a heterotic locus in a finite population is examined. An asymptotic analysis assuming strong selection and weak linkage shows that selection can operate on mutation rates in this setting. There exists a favored mutation rate which is a function only of the equilibrium allele frequency of the heterotic locus and the population size. It is independent of the strength of selection at the heterotoxic locus. Computer simulations are also provided to show that this form of natural selection can occur.     

Prenatal and postnatal diagnostic difficulties in a family with rare alleles of the galactose-1-phosphate uridyl transferase locus

Whole-blood galactose-1-phosphate uridyl transferase (Gal-PUT) (EC 2.7.7.12) activity was absent in a newborn boy with galactosaemic symptoms. The symptoms disappeared on a galactose free diet.In the next pregnancy prenatal diagnosis was performed. Gal-PUT activity was measured by isotope technique and Gal-PUT genotype was determined by gel electrophoresis. The mother was shown to be heterozygous Duarte/heterozygous Los Angeles, the father heterozygous Duarte/heterozygous galactosaemia. The fetus had the same genotype as the father. A normal girl without galactosaemic symptoms was born. Reinvestigation of the index case showed that he was also heterozygous Duarte/heterozygous galactosaemia.It is concluded that the activity of Gal-PUT should always be measured by isotope technique to prove the diagnosis of hereditary galactosaemia. Furthermore, Gal-PUT-genotyping in families with rare alleles is essential for safe prenatal diagnosis.     

Mutation rate varies among alleles at a microsatellite locus: Phylogenetic evidence

The understanding of the mutational mechanism that generates high levels of variation at microsatellite loci lags far behind the application of these genetic markers. A phylogenetic approach was developed to study the pattern and rate of mutations at a dinucleotide microsatellite locus tightly linked to HLA-DQB1 (DQCAR). A random Japanese population (n = 129) and a collection of multiethnic samples (n = 941) were typed at the DQB1 and DQCAR loci. The phylogeny of DQB1 alleles was then reconstructed and DQCAR alleles were superimposed onto the phylogeny. This approach allowed us to group DQCAR alleles that share a common ancestor. The results indicated that the DQCAR mutation rate varies drastically among alleles within this single microsatellite locus. Some DQCAR alleles never mutated during a long period of evolutionary time. Sequencing of representative DQCAR alleles showed that these alleles lost their ability to mutate because of nucleotide substitutions that shorten the length of uninterrupted CA repeat arrays; in contrast, all mutating alleles had relatively longer perfect CA repeat sequences.     

Confidence intervals for frequencies of rare alleles

Letters

Confidence intervals for frequencies of rare alleles     

Common and rare alleles as causes of complex phenotypes

A full understanding of the molecular basis for genetically determined human traits, including susceptibility to disease, appears to be within reach following recent breakthroughs. How fully this promise will be realized, and by which combination of study designs, will depend to a large extent on the allelic architecture of each trait, which is still unknown in most cases. The prevailing belief that traits common in the general population must depend on common variants is challenged by theoretical predictions based on the mutation-selection model. This model states that if disease variants are subject to even weak purifying selection, their presence can be maintained only by new mutations, resulting in a multitude of rare alleles at each locus. Predictions favoring each scenario have relied on biased evidence and unverifiable assumptions, respectively. However, unbiased factual testing of them may soon be possible, as data accumulate from genome-wide association studies and high-throughput resequencing. Because the models are not mutually exclusive, the question should be not which model is correct, but rather what is the relative contribution of each, which is something that may vary dramatically among traits.     

Drift and mutation with a finite number of allelic states

The frequencies with which alleles are alike within (Q) and between (q) populations are formulated for monoecious populations undergoing drift and mutation with unequal mutation rates among alleles for a finite number (k) of allelic states. The effective number of alleles and an application to Nei's measure of genetic distance [Nei, M. (1972) Am. Nat. 106, 283-292] are also considered for this model. The equilibrium values of Q and q increase as k decreases. Unequal mutation rates further increase the equilibrium values and reduce the rates of approach of Q and q to these values. The transitional values of Q and q are very dependent on the initial population frequency composition when mutation rates are unequal. Reducing k, of course, reduces the effective number of alleles, which is further reduced by unequal mutation rates. Complications introduced by initial population composition, unequal mutation rates, and number of allelic states, coupled with data limitations for long-term measures of genetic distance or population differentiation, with mutation as the main driving force, are discussed.     

Estimation of the mtDNA mutation rate in aging mice by proteome analysis and mathematical modeling

The accumulation of mitochondria containing mutated genomes was proposed to be an important factor involved in aging. Although the level of mutated mtDNA has shown to increase over time, it is currently not possible to directly measure the mtDNA mutation rate within living cells. The combination of mathematical modeling and controlled experiments is an alternative approach to obtain an estimate for the mutation rate in a well-defined system. In order to judge the relevance of mitochondrial mutations for the aging process, we used a mouse model to study age-related alterations of the mitochondrial proteins. Based on these experimental data we constructed a mathematical model of the mitochondrial population dynamics to estimate mtDNA mutation rates. Mitochondria were isolated from mouse brain and liver at six different ages (newborn to 24-months). A large-gel 2D-electrophoresis-based proteomics approach was used to analyze the mitochondrial proteins. The expression of two respiratory chain complex I subunits and one complex IV subunit decreased significantly with age. One subunit of complex III and one subunit of complex V increased in expression during aging. Together, these data indicate that complex I and IV deficiency in aged tissues might be accompanied by feedback regulation of other protein complexes in the respiratory chain. When we fitted our experimental data to the mathematical model, mtDNA mutation rate was estimated to be 2.7x10(-8) per mtDNA per day for brain and 3.2x10(-9) per mtDNA per day for liver. According to our model and in agreement with the mitochondrial theory of aging, mtDNA mutations could cause the detrimental changes seen in mitochondrial populations during the normal lifespan of mice, while at the same time ensure that the mitochondrial population remains functional during the developmental and reproductive period of mice.     

Human leptin locus (LEP) alleles and BMI in Samoans

OBJECTIVES: Because of their location in known candidate gene regions for obesity the associations between six microsatellite markers (D2S2170, D2S144, D2S1268, D2S1788, D2S1348 and a tetranucleotide repeat in the 3' UTR of the LEP locus) and body mass index (BMI) were studied in adult Samoans. DESIGN: The study was designed to detect differences in the proportion of alleles at the six microsatellite markers between two groups of adult Samoans at the extremes of the longitudinal BMI distribution. SUBJECTS AND MEASUREMENTS: The 181 unrelated Samoan participants were 25-55 y of age, reported that all four grandparents were Samoan, resided in American Samoa (AS) or Samoa (S) and were without diagnosed hypertension or type 2 diabetes. Initial statistical analysis was based on chi(2) tests of independence between marker allele frequencies and BMI status at each marker. The association of individual alleles with BMI status was tested by aggregating a marker's allelic data into a two-by-two contingency table and applying a two-tailed version of Fisher's exact test, with a Bonferroni correction to account for the multiple testing implicit in the procedure. RESULTS: There were no significant differences in allele frequencies at any of the markers between AS and S, as expected from our prior population genetic analyses. Only the LEP gene 3'-tetranucelotide repeat was associated (P<0.006) with BMI status. The distribution of the marker alleles at the LEP locus was significantly associated with the BMI groups (P<0.01), due to the low frequency of allele 226 in the high BMI group. The same pattern of association was found in sub-group analyses with low BMI individuals from AS and high BMI individuals from S. CONCLUSION: These findings indicate that the leptin 3'-tetranucleotide repeat is associated with high BMI in adult Samoans, with allele 226 having a low frequency in the high BMI group.     

Comparison of somatic mutation in a transgenic versus host locus.

Somatic mutations can now be quantified in almost any cell type in mice carrying bacterial genes in a lambda phage shuttle vector. Mutations induced in vivo are detectable ex vivo, after packaging host-cell DNA into phage that are grown on suitable bacteria. However, the transgenic DNA differs from many host loci in several ways: it (i) is prokaryotic DNA, (ii) is present in multiple tandem copies, and (iii) is heavily methylated and probably not expressed. Thus, mutation of a transgene may not be a suitable model of the host loci, which are eukaryotic, unique, and expressed. To test the relevance of the transgene mutation model, the frequencies of the bacterial lacI+ to lacI- mutations induced in half of the small intestine were compared with the frequencies of the host Dlb-1b to Dlb-1a mutations induced in the other half. The loci responded similarly to ethyl nitrosourea (ENU) with respect to the animal's age and sex, sex of the parent transmitting the transgene, and expression time. ENU dose-response curves were similar. Furthermore, no difference was found at the Dlb-1 locus between transgenic and nontransgenic siblings. In contrast, x-rays induced few lacI mutations but many Dlb-1 mutations. Probably few large deletions are detectable at lacI, but many are detectable at Dlb-1. If so, an important class of mutation is not readily detected in these transgenic mice. With this exception, the transgene and host gene responded similarly in this somewhat limited trial, as is necessary if the transgenic mice are to be a useful model.     

The expected equilibrium of the CpG dinucleotide in vertebrate genomes under a mutation model.

The CpG dinucleotide is present at approximately 20% of its expected frequency in vertebrate genomes, a deficiency thought due to a high mutation rate from the methylated form of CpG to TpG and CpA. We examine the hypothesis that the 20% frequency represents an equilibrium between rate of creation of new CpGs and accelerated rate of CpG loss from methylation. Using this model, we calculate the expected reduction in the equilibrium frequency of the CpG dinucleotide and find that the observed CpG deficiency can be explained by mutation from methylated CpG to TpG/CpA at approximately 12 times the normal transition rate, the exact rate depending on the ratio of transitions to transversions. The observed rate of CpG dinucleotide loss in a human alpha-globin nonprocessed pseudogene, psi alpha 1, and the apparent replenishment of the CpG pool in this sequence by new mutations, agree with the above parameters. These calculations indicate that it would take 25 million years or less, a small fraction of the time for vertebrate evolution, for CpG frequency to be reduced from undepleted levels to the current depleted levels.     

Transgene-induced mutation of the murine steel locus.

The product of the steel locus is essential for normal development of three distinct populations of stem cells--the neural crest-derived melanoblasts, germ cells, and blood cell precursors. Many mutant alleles at steel are lethal in homozygotes and produce coat color dilution in heterozygotes. We have identified a transgenic mouse with diluted pigmentation that closely resembles that of steel heterozygotes. We have demonstrated that the site of transgene insertion is genetically linked to the phenylalanine hydroxylase locus on mouse chromosome 10. In addition, the chromosome carrying the transgene fails to complement the recessive lethality of the Sl allele of steel and the pigmentation defect of the Slpan allele. The data indicate that the inserted transgene has disrupted the steel locus. The resulting allele, designated Sltg, provides a molecular tag for isolation of the steel gene, as well as a new allele for characterization of this developmentally important locus.     

cricklet: A locus regulating a number of adult functions of Drosophila melanogaster

During a screen for mutations in trans-acting genes regulating yolk protein synthesis in Drosophila melanogaster, we have isolated a mutant (cricklet, clt) that is defective in yolk protein synthesis, histolysis of the larval fat body, vitellogenesis, and synthesis of larval serum protein 2 in the adult, larval synthesis occurring normally. Larval serum protein 2 was previously thought to be synthesized only in the third-instar larva. We suggest that the clt locus may encode a protein essential for mediating the response of adult tissues to juvenile hormone.     

Isolation of two alleles of the b locus of Ustilago maydis.

The basidiomycete Ustilago maydis is the causative agent of the disease corn smut. To be pathogenic, strains of U. maydis must be heterozygous for a locus called "b," which appears to control both pathogenicity and sexual development. Two alleles of the b locus of U. maydis were isolated by complementation and hybridization. The clones have the specificities of b1 and b2 alleles as demonstrated by their effects on the colony morphology and pathogenicity of haploid and diploid strains upon transformation. For example, nonpathogenic haploid and diploid strains of U. maydis carrying b2 alleles became pathogenic when transformed with the cloned b1 allele. Furthermore, an a2 b2 haploid strain could be transformed to an a2 b1 genotype by gene replacement using a DNA fragment containing a b1 allele and a selectable marker. The isolation of b alleles represents an important step toward understanding the control of dicaryon formation, dicaryon maintenance, and pathogenicity in U. maydis.     

Population genetic theory of kin selection: Multiple alleles at one locus

Exact population genetic models of one-locus sib-to-sib kin selection with an arbitrary number of alleles are studied. First, a natural additive scaling is established for the genotypic value associated with probabilities of performance of altruism. Two classes of polymorphic equilibria are possible, one corresponding to the usual one-locus viability equilibria and the other reflecting the kin-selection assumptions of the model. At both, the covariance between additive genotypic value and genotypic fitness vanish. Further, the sign of this covariance determines the fate of rare alleles introduced near the first class of equilibria. In addition, the covariance explains the differences between Hamilton's rule, which results from Hardy-Weinberg assumptions, and exact initial increase conditions.