Neighboring-nucleotide effects on the rates of germ-line single-base-pair substitution in human genes

Stability related bias in residues replacing glycines within the collagen triple helix (Gly-Xaa-Yaa) in inherited connective tissue disorders

A missense mutation leading to the replacement of one Gly in the (Gly-Xaa-Yaa)n repeat of the collagen triple helix can cause a range of heritable connective tissue disorders that depend on the gene in which the mutation occurs. Osteogenesis imperfecta results from mutations in type I collagen, Ehlers-Danlos syndrome type IV from mutations in type III collagen, Alport syndrome from mutations in type IV collagen, and dystrophic epidermolysis bullosa from mutations in type VII collagen. The predicted rates of substitutions by different amino acids for glycine in the alpha1(I), alpha2(I), alpha1(III), alpha5(IV), and alpha1(VII) chains (encoded by COL1A1, COL1A2, COL3A1, COL4A5, and COL7A1, respectively) were compared with missense mutations in those chains that have been observed to cause disease. The spectrum of amino acids replacing Gly was not significantly different from that expected for the alpha1(VII) chains, suggesting that any Gly replacement will cause disease. The distribution of residues replacing Gly was significantly different from that expected for all other collagen chains studied, with a particularly strong bias seen for alpha1(I) and alpha1(III) collagen chains. The bias did not correlate with the degree of chemical dissimilarity between Gly and the replacement residues, but in some cases a relationship was observed with the predicted extent of destabilization of the triple helix. For alpha1(III) collagen chains, the more destabilizing mutations were identified more often than expected. For alpha1(I), the most destabilizing residues, Val, Glu, and Asp, and the least destabilizing residue, Ala, were underrepresented. This bias supports the hypothesis that the level of triple-helix destabilization determines clinical outcome.

SNPNB: analyzing neighboring-nucleotide biases on single nucleotide polymorphisms (SNPs)

SNPNB is a user-friendly and platform-independent application for analyzing Single Nucleotide Polymorphism NeighBoring sequence context and nucleotide bias patterns, and subsequently evaluating the effective SNP size for the bias patterns observed from the whole data. It was implemented by Java and Perl. SNPNB can efficiently handle genome-wide or chromosome-wide SNP data analysis in a PC or a workstation. It provides visualizations of the bias patterns for SNPs or each type of SNPs.

AVAILABILITY

SNPNB and its full description are freely available at http://bioinfo.vipbg.vcu.edu/SNPNB/

Coding single-nucleotide polymorphisms associated with complex vs. Mendelian disease: evolutionary evidence for differences in molecular effects

Most Mendelian diseases studied to date arise from mutations that lead to a single amino acid change in an encoded protein. An increasing number of complex diseases have also been associated with amino acid-changing single-nucleotide polymorphisms (coding SNPs, cSNPs), suggesting potential similarities between Mendelian and complex diseases at the molecular level. Here, we use two different evolutionary analyses to compare Mendelian and complex disease-associated cSNPs. In the first, we estimate the likelihood that a specific amino acid substitution in a protein will affect the protein's function, by using amino acid substitution scores derived from an alignment of related protein sequences and statistics from hidden Markov models. In the second, we use standard Ka/Ks ratios to make comparisons at the gene, rather than the individual amino acid, level. We find that Mendelian disease cSNPs have a very strong tendency to occur at highly conserved amino acid positions in proteins, suggesting that they generally have a severe impact on the function of the protein. Perhaps surprisingly, the distribution of amino acid substitution scores for complex disease cSNPs is dramatically different from the distribution for Mendelian disease cSNPs, and is indistinguishable from the distribution for "normal" human variation. Further, the distributions of Ka/Ks ratios for human and mouse orthologs indicate greater positive selection (or less negative selection) pressure on complex disease-associated genes, on average. These findings suggest that caution should be exercised when using Mendelian disease as a model for complex disease, at least with respect to molecular effects on protein function.

Comparative genomics

The genomes from three mammals (human, mouse, and rat), two worms, and several yeasts have been sequenced, and more genomes will be completed in the near future for comparison with those of the major model organisms. Scientists have used various methods to align and compare the sequenced genomes to address critical issues in genome function and evolution. This review covers some of the major new insights about gene content, gene regulation, and the fraction of mammalian genomes that are under purifying selection and presumed functional. We review the evolutionary processes that shape genomes, with particular attention to variation in rates within genomes and along different lineages. Internet resources for accessing and analyzing the treasure trove of sequence alignments and annotations are reviewed, and we discuss critical problems to address in new bioinformatic developments in comparative genomics.

Bayesian Markov chain Monte Carlo sequence analysis reveals varying neutral substitution patterns in mammalian evolution

We describe a model of neutral DNA evolution that allows substitution rates at a site to depend on the two flanking nucleotides ("context"), the branch of the phylogenetic tree, and position within the sequence and implement it by using a flexible and computationally efficient Bayesian Markov chain Monte Carlo approach. We then apply this approach to characterize phylogenetic variation in context-dependent substitution patterns in a 1.7-megabase genomic region in 19 mammalian species. In contrast to other substitution types, CpG transition substitutions have accumulated in a relatively clock-like fashion. More broadly, our results support the notion that context-dependent DNA replication errors, cytosine deamination, and biased gene conversion are major sources of naturally occurring mutations whose relative contributions have varied in mammalian evolution as a result of changes in generation times, effective population sizes, and recombination rates.

Population differences in DNA sequence variation and linkage disequilibrium at the PON1 gene

Polymorphisms of the promoter region (-108C/T) and the coding region (192Q/R) of the paraoxonase 1 gene (PON1) showed differences in association with cardiovascular disease risk in various populations. To characterize the genetic variation underlying these important polymorphisms, we examined DNA sequence variation both in a 1.3-kb promoter region 16.5 kb from codon 192, and in a 1.7-kb region centered on the 192Q/R polymorphic site of the coding region of PON1, in 30 Africans, 30 Europeans and 64 Japanese. We found 10 polymorphic sites and 11 haplotypes in the 1.3-kb promoter region and 10 biallelic polymorphic sites and 10 haplotypes in the 1.7-kb region. From the PON1 sequences of chimpanzees and an orangutan, the ancestral type of codon 192 was found to be R. The number of pairs of polymorphic sites between the promoter and 1.7-kb regions that were in significant linkage disequilibrium was much higher in a Japanese population than in African and European populations. In addition, the pairs of polymorphic sites in linkage disequilibrium differed among the three populations. These results suggest that some of the population differences in association with risk for coronary heart disease can be explained by population differences in haplotype frequency of PON1 haplotypes.

PU.1 is a suppressor of myeloid leukemia, inactivated in mice by gene deletion and mutation of its DNA binding domain

In most myeloid leukemias induced in mice by gamma-radiation, one copy of chromosome 2 has suffered a deletion. To search for a potential tumor suppressor gene in that region, we have delineated the deletions in a panel of these tumors. A commonly deleted region of 2 megabase pairs (Mbp) includes the gene encoding the PU.1 transcription factor, a powerful inducer of granulocytic/monocytic differentiation. Significantly, in 87% of these tumors the remaining PU.1 allele exhibited point mutations in the PU.1 DNA binding domain. Surprisingly, 86% of these mutations altered a single CpG, implicating deamination of deoxycytidine, a common mutational mechanism, as the origin of this lesion. The "hot spot" resides in the codon for a contact residue essential for DNA binding by PU.1. In keeping with a tumor suppressor role for PU.1, enforced expression of wild-type PU.1 in the promyelocytic leukemia cells inhibited their clonogenic growth, induced monocytic differentiation, and elicited apoptosis. The mutant PU.1 found in tumors retained only minimal growth suppressive function. The results suggest that PU.1 normally suppresses development of myeloid leukemia by promoting differentiation and that the combination of gene deletion and a point mutation that impairs its ability to bind DNA is particularly leukemogenic.

Context of deletions and insertions in human coding sequences

We studied the dependence of the rate of short deletions and insertions on their contexts using the data on mutations within coding exons at 19 human loci that cause mendelian diseases. We confirm that periodic sequences consisting of three to five or more nucleotides are mutagenic. Mutability of sequences with strongly biased nucleotide composition is also elevated, even when mutations within homonucleotide runs longer than three nucleotides are ignored. In contrast, no elevated mutation rates have been detected for imperfect direct or inverted repeats. Among known candidate contexts, the indel context GTAAGT and regions with purine-pyrimidine imbalance between the two DNA strands are mutagenic in our sample, and many others are not mutagenic. Data on mutation hot spots suggest two novel contexts that increase the deletion rate. Comprehensive analysis of mutability of all possible contexts of lengths four, six, and eight indicates a substantially elevated deletion rate within YYYTG and similar sequences, which is one of the two contexts revealed by the hot spots. Possible contexts that increase the insertion rate (AT(A/C)(A/C)GCC and TACCRC) and decrease deletion (TATCGC) or insertion (GCGG) rates have also been identified. Two-thirds of deletions remove a repeat, and over 80% of insertions create a repeat, i.e., they are duplications.

Missense mutations in transmembrane domains of proteins: Phenotypic propensity of polar residues for human disease

Previous experiments on the cystic fibrosis transmembrane conductance regulator suggested that non-native polar residues within membrane domains can compromise protein structure/function. However, depending on context, replacement of a native residue by a non-native residue can result either in genetic disease or in benign effects (e.g., polymorphisms). Knowledge of missense mutations that frequently cause protein malfunction and subsequent disease can accordingly reveal information as to the impact of these residues in local protein environments. We exploited this concept by performing a statistical comparison of disease-causing mutations in protein membrane-spanning domains versus soluble domains. Using the Human Gene Mutation Database of 240 proteins (including 80 membrane proteins) associated with human disease, we compared the relative phenotypic propensity to cause disease of the 20 naturally occurring amino acids when removed from-or inserted into-native protein sequences. We found that in transmembrane domains (TMDs), mutations involving polar residues, and ionizable residues in particular (notably arginine), are more often associated with protein malfunction than soluble proteins. To further test the hypothesis that interhelical cross-links formed by membrane-embedded polar residues stabilize TMDs, we compared the occurrence of such residues in the TMDs of mesophilic and thermophilic prokaryotes. Results showed a significantly higher proportion of ionizable residues in thermophilic organisms, reinforcing the notion that membrane-embedded electrostatic interactions play critical roles in TMD stability.

The amino-acid mutational spectrum of human genetic disease

The human disease spectrum is compared to the spectra of mutual amino-acid mutation frequencies, non-disease polymorphisms in human genes, and substitutions fixed between species.

Background

Nonsynonymous mutations in the coding regions of human genes are responsible for phenotypic differences between humans and for susceptibility to genetic disease. Computational methods were recently used to predict deleterious effects of nonsynonymous human mutations and polymorphisms. Here we focus on understanding the amino-acid mutation spectrum of human genetic disease. We compare the disease spectrum to the spectra of mutual amino-acid mutation frequencies, non-disease polymorphisms in human genes, and substitutions fixed between species.

Results

We find that the disease spectrum correlates well with the amino-acid mutation frequencies based on the genetic code. Normalized by the mutation frequencies, the spectrum can be rationalized in terms of chemical similarities between amino acids. The disease spectrum is almost identical for membrane and non-membrane proteins. Mutations at arginine and glycine residues are together responsible for about 30% of genetic diseases, whereas random mutations at tryptophan and cysteine have the highest probability of causing disease.

Conclusions

The overall disease spectrum mainly reflects the mutability of the genetic code. We corroborate earlier results that the probability of a nonsynonymous mutation causing a genetic disease increases monotonically with an increase in the degree of evolutionary conservation of the mutation site and a decrease in the solvent-accessibility of the site; opposite trends are observed for non-disease polymorphisms. We estimate that the rate of nonsynonymous mutations with a negative impact on human health is less than one per diploid genome per generation.

Patterns in interspecies similarity correlate with nucleotide composition in mammalian 3'UTRs

Post-transcriptional regulation and the formation of mRNA 3' ends are crucial for gene expression in eukaryotes. Interspecies conservation of many sequences within 3'UTRs reveals selective constraint due to similar function. To study the pattern of conservation within 3'UTRs, we compiled and aligned 50 sets of complete orthologous 3'UTRs from four orders of mammals. We observed a mosaic pattern of conservation, with alternating regions of high (phylogenetic footprints) and low similarity. Conservation in 3'UTRs correlates with their base composition and also with the synonymous substitution rate in corresponding coding regions. The non-uniform distribution of conservation is more pronounced for 3'UTRs with a moderate or low level of overall conservation, where invariant nucleotides are more numerous, and their runs of lengths 4-7 occur more frequently than if conservation were random. Many runs of invariant nucleotides are AU-rich or pyrimidine-rich. Some of these runs coincide with known functional cis- elements of eukaryotic mRNAs, such as the U-rich upstream element, polyadenylation signal and DICE regulatory signal. More divergent regions of multiple alignments of 3'UTRs are often more G- and/or C-rich. Our results provide evidence on the importance of moderately conserved regions in 3'UTRs and suggest that regulatory functions of 3'UTRs might utilize gene-specific information in these regions.

Dependence of mutational asymmetry on gene-expression levels in the human genome

A great deal of effort has been devoted to measuring the rates of different types of nucleotide substitutions. Mutation rates are known to depend on factors such as methylation status and nearest-neighbor nucleotide effects. However, until recently, in eukaryotes, the rates have not been considered to be strand specific. In a recent analysis of mammalian lineages, Green et al. (2003) uncovered an asymmetry in the frequencies of substitutions on the coding and noncoding strands of genes and showed that this resulted in a nucleotide-content asymmetry within most genes. The authors argue that this bias may be caused by the mammalian transcription-coupled repair in germ cells, but they did not demonstrate an association with germ-cell gene expression. In this work, I analyze nucleotide contents in genes with known expression patterns and levels and provide evidence that the observed asymmetry in mutation rates is, in fact, caused by transcription. The results also imply that germline transcription may occur in a large percentage, 71%-91%, of all human genes.

Theoretical analysis of mutation hotspots and their DNA sequence context specificity

Mutation frequencies vary significantly along nucleotide sequences such that mutations often concentrate at certain positions called hotspots. Mutation hotspots in DNA reflect intrinsic properties of the mutation process, such as sequence specificity, that manifests itself at the level of interaction between mutagens, DNA, and the action of the repair and replication machineries. The hotspots might also reflect structural and functional features of the respective DNA sequences. When mutations in a gene are identified using a particular experimental system, resulting hotspots could reflect the properties of the gene product and the mutant selection scheme. Analysis of the nucleotide sequence context of hotspots can provide information on the molecular mechanisms of mutagenesis. However, the determinants of mutation frequency and specificity are complex, and there are many analytical methods for their study. Here we review computational approaches for analyzing mutation spectra (distribution of mutations along the target genes) that include many mutable (detectable) positions. The following methods are reviewed: derivation of a consensus sequence, application of regression approaches to correlate nucleotide sequence features with mutation frequency, mutation hotspot prediction, analysis of oligonucleotide composition of regions containing mutations, pairwise comparison of mutation spectra, analysis of multiple spectra, and analysis of "context-free" characteristics. The advantages and pitfalls of these methods are discussed and illustrated by examples from the literature. The most reliable analyses were obtained when several methods were combined and information from theoretical analysis and experimental observations was considered simultaneously. Simple, robust approaches should be used with small samples of mutations, whereas combinations of simple and complex approaches may be required for large samples. We discuss several well-documented studies where analysis of mutation spectra has substantially contributed to the current understanding of molecular mechanisms of mutagenesis. The nucleotide sequence context of mutational hotspots is a fingerprint of interactions between DNA and DNA repair, replication, and modification enzymes, and the analysis of hotspot context provides evidence of such interactions.

Stop codon suppression via inhibition of eRF1 expression

In humans, recognition of a stop codon by protein release factor eRF1 leads to release of the nascent peptide from the ribosome. Although efficient eRF1 activity is usually desirable, numerous pathologies result from eRF1 recognition of premature stop mutations in essential genes. In these cases, decreased eRF1 activity could increase readthrough of the premature stop codon, thereby making full-length protein. To broaden the means available to beneficially decrease eRF1 activity, we have targeted eRF1 mRNA using siRNAs and antisense oligonucleotides. We show that both eRF1-targeted siRNA and antisense oligonucleotides decrease eRF1 mRNA and eRF1 protein concentrations, and increase UAG readthrough in cultured human cells.

Neutral substitutions occur at a faster rate in exons than in noncoding DNA in primate genomes

Point mutation rates in exons (synonymous sites) and noncoding (introns and intergenic) regions are generally assumed to be the same. However, comparative sequence analyses of synonymous substitutions in exons (81 genes) and that of long intergenic fragments (141.3 kbp) of human and chimpanzee genomes reveal a 30%-60% higher mutation rate in exons than in noncoding DNA. We propose a differential CpG content hypothesis to explain this fundamental, and seemingly unintuitive, pattern. We find that the increased exonic rate is the result of the relative overabundance of synonymous sites involved in CpG dinucleotides, as the evolutionary divergence in non-CpG sites is similar in noncoding DNA and synonymous sites of exons. Expectations and predictions of our hypothesis are confirmed in comparisons involving more distantly related species, including human-orangutan, human-baboon, and human-macaque. Our results suggest an underlying mechanism for higher mutation rate in GC-rich genomic regions, predict nonlinear accumulation of mutations in pseudogenes over time, and provide a possible explanation for the observed higher diversity of single nucleotide polymorphisms (SNPs) in the synonymous sites of exons compared to the noncoding regions.

The methyl-CpG binding domain and the evolving role of DNA methylation in animals

DNA methylation occurs in bacteria, fungi, plants and animals, however its role varies widely among different organisms. Even within animal genomes, methylation patterns vary substantially from undetectable in nematodes, to global methylation in vertebrate genomes. The number and variety of proteins containing methyl-CpG binding domains (MBDs) that are encoded in animal genomes also varies, with a general correlation between the extent of genomic methylation and the number of MBD proteins. We describe here the evolution of the MBD proteins and argue that the vertebrate MBD complement evolved to exploit the benefits and protect against the dangers of a globally methylated genome.

Fas-associated death domain protein interacts with methyl-CpG binding domain protein 4: a potential link between genome surveillance and apoptosis

Fas-associated death domain protein (FADD) is an adaptor protein bridging death receptors with initiator caspases. Thus, its function and localization are assumed to be cytoplasmic, although the localization of endogenous FADD has not been reported. Surprisingly, the data presented here demonstrate that FADD is mainly nuclear in several adherent cell lines. Its accumulation in the nucleus and export to the cytoplasm required the phosphorylation site Ser-194, which was also required for its interaction with the nucleocytoplasmic shuttling protein exportin-5. Within the nucleus, FADD interacted with the methyl-CpG binding domain protein 4 (MBD4), which excises thymine from GT mismatches in methylated regions of chromatin. The MBD4-interacting mismatch repair factor MLH1 was also found in a complex with FADD. The FADD-MBD4 interaction involved the death effector domain of FADD and a region of MBD4 adjacent to the glycosylase domain. The FADD-binding region of MBD4 was downstream of a frameshift mutation that occurs in a significant fraction of human colorectal carcinomas. Consistent with the idea that MBD4 can signal to an apoptotic effector, MBD4 regulated DNA damage-, Fas ligand-, and cell detachment-induced apoptosis. The nuclear localization of FADD and its interaction with a genome surveillance/DNA repair protein that can regulate apoptosis suggests a novel function of FADD distinct from direct participation in death receptor signaling complexes.

Transcription-associated mutational asymmetry in mammalian evolution

Although mutation is commonly thought of as a random process, evolutionary studies show that different types of nucleotide substitution occur with widely varying rates that presumably reflect biases intrinsic to mutation and repair mechanisms. A strand asymmetry, the occurrence of particular substitution types at higher rates than their complementary types, that is associated with DNA replication has been found in bacteria and mitochondria. A strand asymmetry that is associated with transcription and attributable to higher rates of cytosine deamination on the coding strand has been observed in enterobacteria. Here, we describe a qualitatively different transcription-associated strand asymmetry in mammals, which may be a byproduct of transcription-coupled repair in germline cells. This mutational asymmetry has acted over long periods of time to produce a compositional asymmetry, an excess of G+T over A+C on the coding strand, in most genes. The mutational and compositional asymmetries can be used to detect the orientations and approximate extents of transcribed regions.

Sequence context at human single nucleotide polymorphisms: overrepresentation of CpG dinucleotide at polymorphic sites and suppression of variation in CpG islands

Human polymorphisms originate as mutations, and the influence of context on mutagenesis should be reflected in the distribution of sequences surrounding single nucleotide polymorphisms (SNPs). We have performed a computational survey of nearly two million human SNPs to determine if sequence-dependent hotspots for polymorphism exist in the human genome. Here we show that sequences containing CpG dinucleotides, which occur at low frequencies in the human genome, are 6.7-fold more abundant at polymorphic sites than expected. In contrast, polymorphisms in CpG sequences located within CpG islands, important regulatory regions that modulate gene expression, are 6.8-fold less prevalent than expected. The distribution of polymorphic alleles at CpGs in CpG islands is also significantly different from that in non-island regions. These data strongly support a role for 5-methylcytosine deamination in the generation of human variation, and suggest that variation at CpGs in islands is suppressed.

Discovering genotypes underlying human phenotypes: past successes for mendelian disease, future approaches for complex disease

The past two decades have witnessed an explosion in the identification, largely by positional cloning, of genes associated with mendelian diseases. The roughly 1,200 genes that have been characterized have clarified our understanding of the molecular basis of human genetic disease. The principles derived from these successes should be applied now to strategies aimed at finding the considerably more elusive genes that underlie complex disease phenotypes. The distribution of types of mutation in mendelian disease genes argues for serious consideration of the early application of a genomic-scale sequence-based approach to association studies and against complete reliance on a positional cloning approach based on a map of anonymous single nucleotide polymorphism haplotypes.

Amino acid substitutions in the human genome: evolutionary implications of single nucleotide polymorphisms

Functional differences between amino acids have long been of interest in understanding protein evolution. Several indices exist for comparing residues on the basis of their physicochemical properties and frequencies of occurrence in conserved protein alignments. Here we present a residue dissimilarity index based on coding single nucleotide polymorphisms (SNPs) in the human genome. The index represents an average, organism-wide set of differences between residues and provides important insight into evolutionary restraints on residue substitutions in the human genome. Unlike previous models, it is not restricted to highly conserved protein structures, nor confounded by evolutionary differences between species. Our results confirm earlier observations regarding residue mutabilities but also suggest that in addition to the established key properties, such as size and polarity, charge conservation may be an important and currently underestimated factor in protein evolution. We also estimate that less than 51% of amino acid substitutions occurring in the human genome are evolutionarily neutral.

A low rate of simultaneous double-nucleotide mutations in primates

The occurrence of double-nucleotide (doublet) mutations is contrary to the normal assumption that point mutations affect single nucleotides. Here we develop a new method for estimating the doublet mutation rate and apply it to more than a megabase of human-chimpanzee-baboon genomic DNA alignments and more than a million human single-nucleotide polymorphisms. The new method accounts for the effect of regional variation in evolutionary rates, which may be a confounding factor in previous estimates of the doublet mutation rate. Furthermore we determine sequence context effects by using sequence comparisons over a variety of lineage lengths. This approach yields a new estimate of the doublet mutation rate of 0.3% of the singleton rate, indicating that doublet mutations are far rarer than previously thought. Our results suggest that doublet mutations are unlikely to have caused the correlation between synonymous and nonsynonymous substitution rates in mammals, and also show that regional variation and sequence context effects play an important role in primate DNA sequence evolution.

Meta-analysis of indels causing human genetic disease: mechanisms of mutagenesis and the role of local DNA sequence complexity

A relatively rare type of mutation causing human genetic disease is the indel, a complex lesion that appears to represent a combination of micro-deletion and micro-insertion. In the absence of meta-analytical studies of indels, the mutational mechanisms underlying indel formation remain unclear. Data from the Human Gene Mutation Database (HGMD) were therefore used to compare and contrast 211 different indels underlying genetic disease in an attempt to deduce the processes responsible for their genesis. Each indel was treated as if it were the result of a two-step insertion/deletion process and was assessed in the context of 10 base-pairs DNA sequence flanking the lesion on either side. Several indel hotspots were noted and a GTAAGT motif was found to be significantly over-represented in the vicinity of the indels studied. Previously postulated mechanisms underlying micro-deletions and micro-insertions were initially explored in terms of local DNA sequence regularity as measured by its complexity. The change in complexity consequent to a mutation was found to be indicative of the type of repeat sequence involved in mediating the event, thereby providing clues as to the underlying mutational mechanism. Complexity analysis was then employed to examine the possible intermediates through which each indel could have occurred and to propose likely mechanisms and pathways for indel generation on an individual basis. Manual analysis served to confirm that the majority of indels (>90%) are explicable in terms of a two-step process involving established mutational mechanisms. Indels equivalent to double base-pair substitutions (22% of the total) were found to be mechanistically indistinguishable from the remainder and may therefore be regarded as a special type of indel. The observed correspondence between changes in local DNA sequence complexity and the involvement of specific mutational mechanisms in the insertion/deletion process, and the ability of generated models to account for both the number and identity of the bases deleted and/or inserted, makes this approach invaluable not only for the analysis of indel formation, but also for the study of other types of complex lesion.

Propionic acidemia: identification of twenty-four novel mutations in Europe and North America

Propionic acidemia is an inherited metabolic disease caused by the deficiency of the mitochondrial protein propionyl-CoA carboxylase (PCC), one of the four biotin-dependent enzymes. PCC is a multimeric protein composed of two different alpha- and beta-PCC subunits, nuclearly encoded by the PCCA and PCCB genes, respectively. Mutations in either gene cause the clinically heterogeneous disease propionic acidemia. In this work we describe the mutational analysis of PCCA and PCCB deficient patients from different European countries (Spain, Italy, Belgium, Croatia, and Austria) and from America (mainly USA). We report 24 novel PA mutations, nine affecting the PCCA gene and 15 affecting the PCCB gene. They include six missense mutations, one nonsense mutation, one point exonic mutation affecting splicing, seven splicing mutations affecting splice sequences, and nine short insertions or deletions, only two in-frame. We have found a highly heterogenous spectrum of PCCA mutations, most of the PCCA deficient patients are homozygous carrying a unique genotype. The PCCA mutational spectrum includes a high proportion of short insertions or deletions affecting one nucleotide. In the PCCA mutant alleles analyzed we have also found one single nucleotide change, a novel nonsynonymous SNP. On the other hand, the PCCB deficient patients carry a more reduced spectrum of mutations, 50% of them are missense. This work represents an extensive update of the mutational study of propionic acidemia providing important information about the worldwide distribution of PA mutations and representing another essential part in the study of the phenotype-genotype correlations for the prediction of the metabolic outcome and for the implementation of treatments tailored to each PA patient.

Patterns in spontaneous mutation revealed by human-baboon sequence comparison

We have analyzed the alignment of a long homologous region of the human and baboon genomes (approximately 1.5 Mb). We show that the frequency of gaps between aligned segments decreases slowly with gap length, indicating that several successive nucleotides are often deleted or inserted in one event. By contrast, runs of consecutive mismatches decrease rapidly in frequency with increasing length, following an exponential distribution, indicating that nucleotides are mostly substituted one at a time. Nucleotide substitutions are clumped at the scales of <10 and 1000-10,000 nucleotides, but show almost no aggregation at the scales of <10-100 and over approximately 50,000 nucleotides. Apparently, two rather different factors make the substitution rate not exactly uniform along the DNA sequence. Comparison of regions of very similar genomes that are approximately selectively neutral makes it possible to study spontaneous mutation at a new level of resolution.

Neighboring-nucleotide effects on single nucleotide polymorphisms: a study of 2.6 million polymorphisms across the human genome

We investigated substitution patterns and neighboring-nucleotide effects for 2,576,903 single nucleotide polymorphisms (SNPs) publicly available through the National Center for Biotechnology Information (NCBI). The proportions of substitutions were A/G, 32.77%; C/T, 32.81%; A/C, 8.98%; G/T, 9.06%; A/T, 7.46%; and C/G, 8.92%. The two nucleotides immediately neighboring the variable site showed major deviation from genome-wide and chromosome-specific expectations, although lesser biases extended as far as 200 bp. On the 5' side, the biases for A, C, G, and T were 1.43%, 4.91%, -1.70%, and -4.62%, respectively. These biases were -4.44%, -1.59%, 5.05%, and 0.99%, respectively, on the 3' side. The neighboring-nucleotide patterns for transitions were dominated by the hypermutability effects of CpG dinucleotides. Transitions were more common than transversions, and the probability of a transversion increased with increasing A + T content at the two adjacent sites. Neighboring-nucleotide biases were not consistent among chromosomes, with Chromosomes 19 and 22 standing out as different from the others. These data provide genome-wide information about the effects of neighboring nucleotides on mutational and evolutionary processes giving rise to contemporary patterns of nucleotide occurrence surrounding SNPs.

The functional importance of disease-associated mutation

BACKGROUND

For many years, scientists believed that point mutations in genes are the genetic switches for somatic and inherited diseases such as cystic fibrosis, phenylketonuria and cancer. Some of these mutations likely alter a protein's function in a manner that is deleterious, and they should occur in functionally important regions of the protein products of genes. Here we show that disease-associated mutations occur in regions of genes that are conserved, and can identify likely disease-causing mutations.

RESULTS

To show this, we have determined conservation patterns for 6185 non-synonymous and heritable disease-associated mutations in 231 genes. We define a parameter, the conservation ratio, as the ratio of average negative entropy of analyzable positions with reported mutations to that of every analyzable position in the gene sequence. We found that 84.0% of the 231 genes have conservation ratios less than one. 139 genes had eleven or more analyzable mutations and 88.0% of those had conservation ratios less than one.

CONCLUSIONS

These results indicate that phylogenetic information is a powerful tool for the study of disease-associated mutations. Our alignments and analysis has been made available as part of the database at http://cancer.stanford.edu/mut-paper/. Within this dataset, each position is annotated with the analysis, so the most likely disease-causing mutations can be identified.

Assessing the relative importance of the biophysical properties of amino acid substitutions associated with human genetic disease

The inclusion of a mutation in a pathology-based database such as the Human Gene Mutation Database (HGMD) is a two-stage process: first, the mutation must occur at the DNA level, then it must cause a clinically detectable disease state. The likelihood of the latter step, termed the relative clinical observation likelihood (RCOL), can be regarded as a function of the structural/functional consequences of a mutation at the protein level. Following this paradigm, we modeled in silico all amino acid replacements that could potentially have arisen from an inherited single base pair substitution in five human genes encoding arylsulphatase A (ARSA), antithrombin III (SERPINC1), protein C (PROC), phenylalanine hydroxylase (PAH), and transthyretin (TTR). These proteins were chosen on the basis of 1) the availability of a crystallographic structure, and 2) a sufficiently large number of amino acid replacements being logged in HGMD. A total of 9,795 possible mutant structures were modeled and 20 different biophysical parameters assessed. Together with the HGMD-derived spectra of clinically detected mutations, these data allowed maximum likelihood estimation of RCOL profiles for the 20 parameters studied. Nine parameters (including energy difference between wild-type and mutant structures, accessibility of the mutated residue, and distance from the binding/active site) exhibited statistically significant variability in their RCOL profiles, indicating that mutation-associated changes affected protein function. As yet, however, a biological meaning could only be attributed to the RCOL profiles of solvent accessibility and, for three proteins, local energy change, disturbed geometry, and distance from the active center. The limited ability of the biophysical properties of mutations to explain clinical consequences is probably due to our current lack of understanding as to which amino acid residues are critical for protein folding. However, since the proteins examined here were unrelated, and our findings consistent, it may nevertheless prove possible to extrapolate to other proteins whose dysfunction underlies inherited disease.

Enhanced CpG mutability and tumorigenesis in MBD4-deficient mice

The mammalian protein MBD4 contains a methyl-CpG binding domain and can enzymatically remove thymine (T) or uracil (U) from a mismatched CpG site in vitro. These properties suggest that MBD4 might function in vivo to minimize the mutability of 5-methylcytosine by removing its deamination product from DNA. We tested this hypothesis by analyzing Mbd4-/- mice and found that the frequency of of C --> T transitions at CpG sites was increased by a factor of three. On a cancer-susceptible Apc(Min/+) background, Mbd4-/- mice showed accelerated tumor formation with CpG --> TpG mutations in the Apc gene. Thus MBD4 suppresses CpG mutability and tumorigenesis in vivo.

An approach for global scanning of single nucleotide variations

Efficient global scanning of single nucleotide variations in DNA sequences between related, complex DNA samples remains a challenge. In the present article we present an approach to this problem. We have used immobilized thymidine DNA glycosylases to capture and enrich DNA fragments containing internal mismatched base pairs and separate these fragments as a pool from perfectly base-paired fragments as another pool. Enrichments of up to several hundredfold were obtained with one cycle of treatment, and all of the four groups of single nucleotide mismatches were fully covered by combining use of two thymine DNA glycosylases generated here. We have used a heterohybrid-orientating strategy for selective amplification of duplexes with one strand derived from each of two input DNA samples, which can also be used for selective amplification of duplexes with both strands derived from one of two input samples when desired. By combining these methods, the single nucleotide variations either between two DNA pools or within one DNA pool can be obtained in one process. This approach has been applied to the total cDNA from a human cell line and has several potential applications in mapping genetic variations, particularly global scanning of cDNA single nucleotide variations or polymorphisms, and finally high-throughput mapping of complex genetic traits.

Human gene mutation in pathology and evolution

Mutations in human gene pathology and evolution represent two sides of the same coin in that the same mechanisms that have frequently been implicated in disease-associated mutagenesis appear also to have been involved in potentiating evolutionary change. Indeed, the mutational spectra of germline mutations responsible for inherited disease, somatic mutations underlying tumorigenesis, polymorphisms (either neutral or functionally significant) and differences between orthologous gene sequences exhibit remarkable similarities, implying that they may have causal mechanisms in common. Since these different categories of mutation share multiple unifying characteristics, they should no longer be viewed as distinct entities but rather as portions of a continuum of genetic change that links population genetics and molecular medicine with molecular evolution.

Formation of a fairly stable diazoate intermediate of 5-methyl-2'-deoxycytidine by HNO2 and NO, and its implication to a novel mutation mechanism in CpG site

The intermediate produced from 5-methyl-2'-deoxycytidine ((5me)dCyd) by HNO2 and NO treatments was isolated and characterized. When 10mM (5me)dCyd was incubated with 100mM NaNO2 at pH 3.7 and 37 degrees C, a previously unidentified product was formed. The product was identified as a diazoate derivative of (5me)dCyd, 1-(beta-D-2'-deoxyribofuranosyl)-5-methyl-2-oxopyrimidine-4-diazoate ((5me)dCyd-diazoate), on the bases of several measurements including LC/MS. The time course of the concentration change of the diazoate showed a characteristic profile of a reaction intermediate, and the steady state concentration was 2.3 microM (0.023% yield). When an aqueous solution of 10mM (5me)dCyd (10 mL) was bubbled by NO at 37 degrees C under aerobic conditions holding the pH around 7.4, the diazoate was also generated. The yield of the diazoate was 0.041 micromol (0.041% yield) at 20 mmol of NO absorption. At physiological pH and temperature (pH 7.4, 37 degrees C), the diazoate was converted to dThd exclusively with a first order rate constant k=9.1x10(-6) x s(-1) (t(1/2)=21 h). These results show that the diazoate is generated as a relatively stable intermediate in the reactions of (5me)dCyd with HNO2 and NO and further suggest that the diazoate can be formed in cellular DNA with biologically relevant doses of HNO2 and NO.

Mutations in the factor IX gene (F9) during the past 150 years have relative rates similar to ancient mutations

Pollutants and dietary mutagens have been associated with somatic mutation and cancer, but the extent of their influence on germline mutation is not clear. Since deleterious germline mutations can be transmitted for thousands of years, any influence on germline mutation from the vast increase in man-made chemicals of the past 150 years would be an important public health issue. Observed disease causing mutations in the X-linked factor IX gene (F9) of hemophilia B patients originated predominantly in the past 150 years, since the half-life of these mutations in human populations had been about two generations before effective treatment became available about a generation ago. Recent changes in germline mutational processes may be detected by comparison of the observed hemophilia B causing mutation pattern in F9 with the pattern of neutral polymorphisms which occurred over a much longer period of time. By scanning a total of 1.5 megabases of deep intronic regions of F9 in the genomic DNA from 84 individuals, 42 neutral polymorphisms were found in 23 haplotypes that differed by at least 11 mutations from the ancestral primate haplotype. By sequencing F9 in seven non-human primates, 39 of these polymorphisms were characterized as ancient mutations relative to a unanimous ancestral primate allele. This ancient mutation pattern was compared to the recent pattern of hemophilia B causing mutations. Remarkably, no significant difference was found (P=0.5), suggesting that the vast increase in man-made chemicals during the past 150 years has not had a major impact on the pattern of human germline mutation. This result is consistent with the hypothesis that endogenous processes dominate germline mutation.

Statistical inference of sequence-dependent mutation rates

Several lines of research are now converging towards an integrated understanding of mutational mechanisms and their evolutionary implications. Experimentally, crystal structures reveal the effect of sequence context on polymerase fidelity; large-scale sequencing projects generate vast amounts of sequence polymorphism data; and locus-specific databases are being constructed. Computationally, software and analytical tools have been developed to analyze mutational data, to identify mutational hot spots, and to compare the signatures of mutagenic agents.

Automated mutation screening using dideoxy fingerprinting and capillary array electrophoresis

The rapid progress in the isolation of genes associated with human disease has resulted in an increasing demand for mutation screening methods. The molecular diagnosis of the long QT syndrome (LQTS), a cardiac disorder characterized by prolongation of the QT(c) interval in the ECG, syncopes, and sudden death, requires mutation screening of all exons in at least five genes, encoding cardiac Na(+) and K(+) channel subunits. A method for automated dideoxy fingerprinting (ddF) using capillary array electrophoresis (CAE) was developed and the efficiency of the method was tested by analyzing 24 DNA samples with mutations in one of the genes KCNQ1 and KCNH2, which are involved in 50% of LQTS cases. One of these mutations, 362insQK in KCNQ1, is novel. The sensitivity was 100% using a single electrophoresis temperature of 18 degrees C or 25 degrees C. However, analysis of the samples in both the sense and anti-sense direction were required for high sensitivity. Analysis in a single direction resulted in a decrease of the sensitivity to 74% and 70%, respectively. The throughput of the ddF method, if performed with a 16 capillary CAE instrument, is 288 samples per seven hr if each sample is analyzed on both strands.

High-resolution haplotype structure in the human genome

Linkage disequilibrium (LD) analysis is traditionally based on individual genetic markers and often yields an erratic, non-monotonic picture, because the power to detect allelic associations depends on specific properties of each marker, such as frequency and population history. Ideally, LD analysis should be based directly on the underlying haplotype structure of the human genome, but this structure has remained poorly understood. Here we report a high-resolution analysis of the haplotype structure across 500 kilobases on chromosome 5q31 using 103 single-nucleotide polymorphisms (SNPs) in a European-derived population. The results show a picture of discrete haplotype blocks (of tens to hundreds of kilobases), each with limited diversity punctuated by apparent sites of recombination. In addition, we develop an analytical model for LD mapping based on such haplotype blocks. If our observed structure is general (and published data suggest that it may be), it offers a coherent framework for creating a haplotype map of the human genome.

Gene conversion and different population histories may explain the contrast between polymorphism and linkage disequilibrium levels

To characterize linkage disequilibrium (LD) levels in human populations, we have analyzed 10 independent noncoding segments in three population samples from the major ethnic groups--that is, Africans, Asians, and Europeans. Descriptive statistics show that LD decays much faster in the African samples than in the non-African ones. With the assumption of an equilibrium model, we estimated the population crossing-over parameter (4N(e)r(bp), where N(e) is the effective population size and r(bp) is the crossing-over rate per generation between adjacent base pairs) in the presence of gene conversion. In the African sample, LD and polymorphism levels lead to similar estimates of effective population size, as expected under an equilibrium model. Conversely, in both non-African samples, LD levels suggest a smaller effective population size than that implied by polymorphism levels. This observation is paralleled by significant departures from an equilibrium model in the spectrum of allele frequencies of the non-African samples. Besides ruling out the possibility that non-African populations are at equilibrium, these results suggest different demographic history (temporal and spatial) of these groups. Interestingly, the African sample fits the expectations of an equilibrium model based on polymorphism and divergence levels and on frequency spectrum. For this sample, the estimated ratio of gene conversion to crossing-over rates is 7.3 for a mean tract length of 500 bp, suggesting that gene conversion may be more frequent than previously thought. These findings imply that disease-association studies will require a much denser map of polymorphic sites in African than in non-African populations.

DHPLC analysis of the MECP2 gene in Italian Rett patients

Rett Syndrome (RTT) is an X-linked dominant neurodevelopmental disorder, which almost exclusively affects girls, with an estimated prevalence of one in 10,000-15,000 female births. Mutations in the methyl CpG binding protein 2 gene (MECP2) have been identified in roughly 75% of classical Rett girls. The vast majority of Rett cases (99%) are sporadic in origin, and are due to de novo mutations. We collected DNA samples from 50 Italian classical Rett girls, and screened the MECP2 coding region for mutations by denaturing high-performance liquid chromatography (DHPLC) and subsequent direct sequencing. DHPLC is a recently developed method for mutation screening which identifies heteroduplexes formed in DNA samples containing mismatches between wild type and mutant DNA strands, combining high sensitivity, reduced cost per run, and high throughput. In our series, 19 different de novo MECP2 mutations, eight of which were previously unreported, were found in 35 out of 50 Rett girls (70%). Seven recurrent mutations were characterized in a total of 22 unrelated cases. Initial DHPLC screening allowed the identification of 17 out of 19 different mutations (90%); after optimal conditions were established, this figure increased to 100%, with all recurrent MECP2 mutations generating a characteristic chromatographic profile. Detailed clinical data were available for 27 out of 35 mutation carrying Rett girls. Milder disease was detectable in patients carrying nonsense mutation as compared to patients carrying missense mutations, although this difference was not statistically significant (P = 0.077).

PCR candidate region mismatch scanning: adaptation to quantitative, high-throughput genotyping

Linkage and association analyses were performed to identify loci affecting disease susceptibility by scoring previously characterized sequence variations such as microsatellites and single nucleotide polymorphisms. Lack of markers in regions of interest, as well as difficulty in adapting various methods to high-throughput settings, often limits the effectiveness of the analyses. We have adapted the Escherichia coli mismatch detection system, employing the factors MutS, MutL and MutH, for use in PCR-based, automated, high-throughput genotyping and mutation detection of genomic DNA. Optimal sensitivity and signal-to-noise ratios were obtained in a straightforward fashion because the detection reaction proved to be principally dependent upon monovalent cation concentration and MutL concentration. Quantitative relationships of the optimal values of these parameters with length of the DNA test fragment were demonstrated, in support of the translocation model for the mechanism of action of these enzymes, rather than the molecular switch model. Thus, rapid, sequence-independent optimization was possible for each new genomic target region. Other factors potentially limiting the flexibility of mismatch scanning, such as positioning of dam recognition sites within the target fragment, have also been investigated. We developed several strategies, which can be easily adapted to automation, for limiting the analysis to intersample heteroduplexes. Thus, the principal barriers to the use of this methodology, which we have designated PCR candidate region mismatch scanning, in cost-effective, high-throughput settings have been removed.

Bias in the introduction of variation as an orienting factor in evolution

According to New Synthesis doctrine, the direction of evolution is determined by selection and not by "internal causes" that act by way of propensities of variation. This doctrine rests on the theoretical claim that because mutation rates are small in comparison to selection coefficients, mutation is powerless to overcome opposing selection. Using a simple population-genetic model, this claim is shown to depend on assuming the prior availability of variation, so that mutation may act only as a "pressure" on the frequencies of existing alleles, and not as the evolutionary process that introduces novelty. As shown here, mutational bias in the introduction of novelty can strongly influence the course of evolution, even when mutation rates are small in comparison to selection coefficients. Recognizing this mode of causation provides a distinct mechanistic basis for an "internalist" approach to determining the contribution of mutational and developmental factors to evolutionary phenomena such as homoplasy, parallelism, and directionality.

MECP2 mutations in Danish patients with Rett syndrome: high frequency of mutations but no consistent correlations with clinical severity or with the X chromosome inactivation pattern

Rett syndrome (RTT) is a neurodevelopmental disorder, which almost exclusively affects girls, who, after an initial period of apparently normal development, display gradual loss of speech and purposeful hand use, gait abnormalities and stereotypical hand movements. In the year 2000, mutations in the gene for the methyl CpG binding protein 2, MECP2, have been identified in 35-80% of the patients in three different studies. We have identified 15 different MECP2 mutations in 26 of 30 Danish RTT patients. The mutations included five novel mutations (one point mutation, three smaller deletions involving identical regions in the gene, and one duplication). In contrast to the point mutations and the duplication, which all affect the methyl binding domain or the transcriptional repressing domain, the three overlapping deletions are clustered in the 3' end of the gene. We found no consistent correlation between the type of mutation and the clinical presentation of the patient or the X-inactivation pattern in peripheral blood. Our high mutation detection rate, compared to two of the previous studies, underscores the importance of the inclusion criteria of the patients and supports that MECP2 is the most important, if not the only, gene responsible for RTT.

Use of mutation spectra analysis software

The study and comparison of mutation(al) spectra is an important problem in molecular biology, because these spectra often reflect on important features of mutations and their fixation. Such features include the interaction of DNA with various mutagens, the function of repair/replication enzymes, and properties of target proteins. It is known that mutability varies significantly along nucleotide sequences, such that mutations often concentrate at certain positions, called "hotspots," in a sequence. In this paper, we discuss in detail two approaches for mutation spectra analysis: the comparison of mutation spectra with a HG-PUBL program, (FTP: sunsite.unc.edu/pub/academic/biology/dna-mutations/hyperg) and hotspot prediction with the CLUSTERM program (www.itba.mi.cnr.it/webmutation; ftp.bionet.nsc.ru/pub/biology/dbms/clusterm.zip). Several other approaches for mutational spectra analysis, such as the analysis of a target protein structure, hotspot context revealing, multiple spectra comparisons, as well as a number of mutation databases are briefly described. Mutation spectra in the lacI gene of E. coli and the human p53 gene are used for illustration of various difficulties of such analysis.

A highly variable segment of human subterminal 16p reveals a history of population growth for modern humans outstide Africa

We have sequenced a highly polymorphic subterminal noncoding region from human chromosome 16p13.3 flanking the 5' end of the hypervariable minisatellite MS205, in 100 chromosomes sampled from different African and Euroasiatic populations. Coalescence analysis indicates that the time to the most recent common ancestor (approximately 1 million years) predates the appearance of anatomically modern human forms. The root of the network describing this variability lies in Africa. African populations show a greater level of diversity and deeper branches. Most Euroasiatic variability seems to have been generated after a recent out-of-Africa range expansion. A history of population growth is the most likely scenario for the Euroasiatic populations. This pattern of nuclear variability can be reconciled with inferences based on mitochondrial DNA.

A highly variable segment of human subterminal 16p reveals a history of population growth for modern humans outside Africa

We have sequenced a highly polymorphic subterminal noncoding region from human chromosome 16p13.3 flanking the 5' end of the hypervariable minisatellite MS205, in 100 chromosomes sampled from different African and Euroasiatic populations. Coalescence analysis indicates that the time to the most recent common ancestor (approximately 1 million years) predates the appearance of anatomically modern human forms. The root of the network describing this variability lies in Africa. African populations show a greater level of diversity and deeper branches. Most Euroasiatic variability seems to have been generated after a recent out-of-Africa range expansion. A history of population growth is the most likely scenario for the Euroasiatic populations. This pattern of nuclear variability can be reconciled with inferences based on mitochondrial DNA.

A SNP resource for human chromosome 22: extracting dense clusters of SNPs from the genomic sequence

The recent publication of the complete sequence of human chromosome 22 provides a platform from which to investigate genomic sequence variation. We report the identification and characterization of 12,267 potential variants (SNPs and other small insertions/deletions) of human chromosome 22, discovered in the overlaps of 460 clones used for the chromosome sequencing. We found, on average, 1 potential variant every 1.07 kb and approximately 18% of the potential variants involve insertions/deletions. The SNPs have been positioned both relative to each other, and to genes, predicted genes, repeat sequences, other genetic markers, and the 2730 SNPs previously identified on the chromosome. A subset of the SNPs were verified experimentally using either PCR-RFLP or genomic Invader assays. These experiments confirmed 92% of the potential variants in a panel of 92 individuals. [Details of the SNPs and RFLP assays can be found at http://www.sanger.ac.uk and in dbSNP.]

Thymine-DNA glycosylase and G to A transition mutations at CpG sites

About 23% of mutations in hereditary human diseases and 24% of mutations in p53 in human cancers are G to A transitions at sites of cytosine methylation suggesting that these sites are either foci for DNA damage, or foci for damage that is poorly repaired. Thymine produced at these sites by the hydrolytic deamination of 5-methylcytosine is removed by thymine-DNA glycosylase. Thymine-DNA glycosylase will also remove 3,N(4)-ethenocytosine and uracil from DNA. The action of this enzyme is limited by its very low k(cat) and by tight binding to the apurinic site produced when the thymine is removed. These properties of the enzyme suggest that the inefficiency of the base excision repair pathway that it initiates may be the underlying cause of the prevalence of these mutations.