DNA restriction fragment length polymorphisms and heterozygosity in the human genome

Results of a study of genetic factors predisposing to the development of psoriasis among the population of the Russian Federation

Goal. To assess the incidence of genes predisposing to the development of psoriasis based on selected molecular and genetic study methods among the Russian population. Materials and methods. By using allele specific hybridization methods in the form of real-time PCR and RFLP assay (Restriction Fragment Length Polymorphism), single-nucleotide polymorphisms of fragments of 13 genes associated with predisposition to the development of psoriasis were studied in whole blood samples obtained from 45 psoriatic patients: rs4649203 (gene IL-28RA), rs11209026 (gene IL-23R), rs610604 (gene TNFAIP3), rs514315 (gene SERPINB8), rs9304742 (gene ZNF816A), rs17728338 (gene TNIP1), rs13190932 (gene TRAF3IP2), rs2235617 (gene ZNF313), rs27524 (gene ERAP1), rs702873 (gene REL), rs3213094 (gene IL-12B), rs12720356 (gene TYK2), and rs8016947 (gene NFKBI). Results. A set of data on the genotype structure and incidence in respective loci of 13 genes associated with predisposition to develop psoriasis was obtained for the first time based on a Russian sample of patients. At this stage of the study, we failed to reveal any reliable differences concerning the incidence of individual genotypes between psoriatic patients and healthy people because of the small size of the sample. Gene TYK2 (rs1272035) served as an exclusion. The homozygous T/T genotype was revealed more often (p < 0.01) in psoriatic patients (95.6%) vs. healthy people (77.8%), which makes it possible to consider the carriers of this genotype as people predisposed to the development of psoriasis.

DNA fingerprints of dogs and cats

Human minisatellite probes consisting of tandem repeats of the 'core' sequence, a putative recombination signal in human DNA, cross-hybridize to multiple polymorphic fragments in dog and cat DNA to produce individual-specific DNA 'fingerprints'. Pedigree analysis shows that most of the DNA fragments detected in an individual are heterozygous, and that these fragments are derived from multiple dispersed autosomal loci. DNA fingerprints of cats and dogs should prove suitable for individual identification and for establishing family relationships. They are also suitable for rapid marker generation in large pedigrees and could be applied to linkage analysis in these animals.

Anchor polymerase chain reaction display: a high-throughput method to resolve, score, and isolate dimorphic genetic markers based on interspersed repetitive DNA elements

Genes which confer a disease when mutated, or for which population variability contributes to a quantitative trait such as longevity or disease susceptibility, can be localized in the genetic map by use of an appropriately dense set of polymorphic DNA markers. Here we describe an anchor PCR method for high-throughput genotyping, which can be used to amplify the DNA segments flanking an interspersed repetitive sequence such as a transposon, and to limit the number of product bands per reaction to facilitate marker resolution. We used this method to amplify and display DNA fragments flanking the Tc1 transposable elements from different strains of the nematode Caenorhabditis elegans, varying widely in insert number, and to analyze marker segregation in recombinant inbred lines generated from an interstrain cross. Since essentially all eukaryotic genomes contain abundant interspersed repeat families, many of which are dimorphic (for presence or absence of specific elements) among populations, this method can be used for rapid genotyping and fine-scale chromosomal mapping in many species, including those for which extensive mapping and sequencing data do not yet exist.

The X chromosome and the rate of deleterious mutations in humans

Monosomy for the X chromosome in humans creates a genetic Achilles' heel for nature to deal with. We report that the human X chromosome appears to have one-third the density of the coding sequence of the autosomes and, because of partial shielding from the high mutation rate of the male sex, that it should also have a lower mutation rate than the autosomes (i.e.,.73). Hence, the X chromosome should contribute one quarter (.33x.73=.24) of the deleterious mutations expected from its DNA content. In this way, selection has possibly moderated risks from mutation in X-linked genes that are thought to have been fixed in their syntenic state since the onset of the mammalian lineage. The unexpected difference in the density of coding sequences indicates that our recent, hemophilia B-based estimate of the rate of deleterious mutations per zygote should be increased from 1.3 to 4 (1.3x3).

Relationship of human pancreatic cholesterol esterase gene structure with lipid phenotypes

Pancreatic cholesterol esterase is one of the enzymes that plays a pivotal role in cholesterol absorption. Differences in the genotype of this enzyme could affect the susceptibility of individuals to dyslipidemia and/or cardiovascular disease. We undertook this study to investigate if any correlation exists between restriction fragment length polymorphism in the human pancreatic cholesterol esterase gene and serum lipid levels. DNA from 96 healthy adults was restricted with Stu I, Southern blotted, and probed with cDNA of human pancreatic cholesterol esterase. Results revealed six distinct patterns which were classified as A, B, C, D, E, and F which had a population frequency of 1%, 34.5%, 49%, 12.5%, 1% and 2% respectively. Correlation of the distribution of lipid and lipoprotein levels by pattern and sex revealed a significant interaction between pattern type and HDL (p=0.03) in the most common group (group C) for males. Male patients of pattern C tended to have a lower LDL cholesterol than non-pattern C males (p=0.07); in addition, 80% of all males in the study population with LDL cholesterol under 100 mg/dl were found in pattern C. Thus, the most common Stu I RFLP genotype is associated with a favorable lipid phenotype. This report shows an association between the human pancreatic cholesterol esterase genotype and serum lipid levels. Further analysis of a larger study group with Stu I and alternative polymorphic restriction enzymes is warranted, to confirm this biologically plausible result.

Genetic variability in white leghorns revealed by chicken liver expressed sequence tags

A total of 92 expressed sequence tags from chicken liver (CLEST) were searched for homology with known genes. Among the CLEST, 29% had no sequence similarities with known genes, 34% showed sequence similarity to rRNA, 9% to mitochondrial genes, 23% to known nuclear genes, and 5% to human expressed sequence tags. Among the nuclear CLEST (excluding rRNA), clones with sequence similarity to aldolase B were represented four times, whereas all the other clones represented unique genes. The presence of MspI and TaqI restriction fragment length polymorphisms (RFLP) associated with CLEST were analyzed by bulk Southern blotting in 16 strains of White Leghorn chickens derived from five different genetic bases. No RFLP were observed with rRNA CLEST and a single MspI RFLP was observed with mitochondrial CLEST. The nuclear CLEST with sequence similarity to known nuclear genes were grouped into two classes on the basis of their involvement in intermediary metabolism. Among the nine genes coding for metabolic enzymes, all but one were polymorphic at MspI and/or TaqI sites in at least one of the strains, whereas among the other genes six of nine were polymorphic. The average frequency of clones revealing RFLP per cDNA clone and restriction enzyme for the two classes were 0.7 and 0.3, respectively. The analysis indicated that in White Leghorns, RFLP markers in the vicinity of nuclear CLEST are relatively frequent. Further, RFLP in the vicinity of genes coding for metabolic enzymes were significantly more frequent than near genes coding for other proteins.

Nuclear DNA diversity in worldwide distributed human populations

Nucleotide variation was examined in an 8 kb intronic DNA bordering exon 44 of the human dystrophin gene on Xp21. Thirty-six polymorphisms (substitutions, small insertions/deletions and one (T)n microsatellite) were found using SSCP/heteroduplex analysis of DNA samples from mixed Europeans, Papua New Guineans as well as from six African, three Asian and two Amerindian populations. In this way the European bias in the nuclear polymorphism ascertainment has been avoided. In a maximum likelihood tree constructed from the frequency data, Africans clustered separately from the non-African populations. Fifteen polymorphisms were shared among most of the populations compared, whereas 13 sites were found to be endemic to Africans and four to non-Africans. The common sites contributed most to the average heterozygosity (Hn=0.101%+/-0.023), whereas the endemic ones, being rare, had little effect on this estimate. The F(ST) values were lower for Africans (0.072) than for non-Africans (0.158), suggesting a higher level of gene exchange within Africa, corroborating the observation of a greater number of segregating sites on this continent than elsewhere. The data suggest a recent common origin of the African and non-African populations, where a greater geographical isolation of the latter resulted in a smaller number of newly acquired polymorphisms.

Screening the monoamine oxidase B gene in 100 male patients with schizophrenia: a cluster of polymorphisms in African-Americans but lack of functionally significant sequence changes

The monoamine oxidase B (MAO-B) gene was examined in 100 alleles derived from 80 Caucasian, 10 African-American, 5 Asian, and 5 Native American male patients with schizophrenia to identify sequence changes that might be associated with the disease. Approximately 235 kb of genomic sequence, primarily in coding regions, were screened by dideoxy fingerprinting, a modification of single-strand conformational polymorphism (SSCP) analysis that detects virtually 100% of sequence changes [Sarkar et al. (1992): Genomics 13:441-443; Liu and Sommer (1994): PCR Methods Appl 4:97-108]. No sequence changes of likely functional significance were identified, suggesting that mutations affecting the structure of the MAO-B protein are uncommon in the general population and are unlikely to contribute significantly to the genetic predisposition to schizophrenia. Eight polymorphisms were identified in African-Americans and Native Americans, but none were identified among Caucasians. Of the eight observed polymorphisms, a set of five transitions and one microdeletion was identified within approximately 17 kb of genomic sequence in the same 3 African-American individuals, while the remaining 7 African-Americans had a sequence identical to that in Caucasians. The presence of two such haplotypes, without intermediates, is compatible with the hypothesis that germline mutations can occur in clusters, as also suggested by other recent findings.

Geographic clustering of human Y-chromosome haplotypes

Five polymorphic markers on the Y-chromosome (mostly microsatellites) were typed in 121 individuals from 13 populations around the world. With these markers 78 different haplotypes were detected. Haplotypes present more than once tend to be shared by individuals from the same population or continent. A reconstruction of haplotype phylogeny also indicates significant geographic structure in the data. Based on the similarity of the haplotypes, population relationships were examined and found to be largely concordant with those obtained with other markers. Even though the sample size and the number of markers are small, there is very signficant clustering of the haplotypes by continent of origin.

Use of RAPD analyses to estimate population genetic parameters in the alfalfa leaf-cutting bee,Megachile rotundata

RAPD analyses were performed on five geographically isolated populations of Megachile rotundata. We used haploid males of the alfalfa leaf-cutting bee, M. rotundata, to overcome the limitation of the dominance of RAPD markers in the determination of population genetic parameters. Sixteen primers gave rise to 130 polymorphic and 31 monomorphic bands. The unbiased estimators calculated in this study include within- and between-population heterozygosity, nucleotide divergence, and genetic distance. The genetic diversity (H = 0.32–0.35) was found to be about 10 times that of previous estimates (H = 0.033) based on allozyme data. Contrary to the data obtained at the protein level, our results suggest that Hymenoptera do not have a lower level of genetic variability at the DNA level compared with other insect species. Regardless of the different assumptions underlying the calculation of heterozygosity, divergence, and genetic distance, all five populations showed a parallel interrelationship for the three parameters. We conclude that RAPD markers are a convenient tool to estimate population genetic variation in haploid M. rotundata and that with an adequate sample size the technique is applicable to the evaluation of divergence in diploid populations. Key words : Megachile rotundata, RAPD, heterozygosity, genetic distance, nucleotide divergence.

DNA polymorphisms of lipogenesis genes and analysis of linkage with fatness in turkeys

DNA variability of lipogenesis genes was evaluated by Restriction Fragment Length Polymorphism (RFLP) on four selected lines of turkeys. Thirteen polymorphisms were obtained, four on the acetyl-coenzyme A carboxylase gene, one on the apoprotein AI gene, two on the malic enzyme gene, four on the fatty acid synthase gene, and two on the stearoyl-coenzyme A mesaturase gene. Six of these RFLP were studied in more detail on 64 female turkeys sampled in two lines. Large differences in terms of number of DNA profiles and haplotype frequencies were observed between the two genes, in accordance with the variability of recorded fatness performances. An additive model including the effects of RFLP profiles was used to assess linkage between polymorphisms on lipogenesis genes and fatness measured either in vivo by an ultrasonic instrument or, after slaughter, by the ratio of skin plus subcutaneous fat weight to the whole leg weight. In each line, two polymorphisms suggested a statistical association with fat deposit, one on the malic enzyme gene detected by DNA digestion with HindIII and one on the fatty acid synthase gene detected by MspI. A multiple comparison of means associated with each RFLP profile enabled the identification of two lean profiles in one strain, and a fat profile in the other.

Multiplex genotype determination at a large number of gene loci

To facilitate large-scale genotype analysis, an efficient PCR-based multiplex approach has been developed. For simultaneously amplifying the target sequences at a large number of genetic loci, locus-specific primers containing 5' universal tails are used. Attaching the universal tails to the target sequences in the initial PCR steps allows replacement of all specific primers with a pair of primers identical to the universal tails and converts the multiplex amplification into "uniplex." Simultaneous amplification of 26 genetic loci with this approach is described. The multiplex amplification can be coupled with genotype determination. By incorporating a single-base mismatch between a primer and the template into the target sequences, a polymorphic site can be converted into a desirable restriction fragment length polymorphism when it is necessary. In this way, the allelic PCR products for the polymorphic loci can be discriminated by gel electrophoresis after restriction enzyme digestion. In this study, 32 loci were typed in such a multiplex way.

Multiplex genotype determination at a DNA sequence polymorphism cluster in the human immunoglobulin heavy-chain region

We have developed a method for multilocus genotype determination. The method involves using restriction fragment length polymorphisms (RFLPs) for allele discrimination. If a polymorphism is not an RFLP, it is converted into an RFLP during the polymerase chain reaction (PCR). After amplification and restriction enzyme digestion, samples are analyzed by sequential gel loading during electrophoresis. The efficiency of this method was demonstrated by determining the genotypes of 108 semen samples at seven DNA sequence polymorphic sites identified in the human immunoglobulin heavy-chain variable region. It was shown that more than 1000 PCR products could be easily analyzed per day per investigator. To show the reliability of this method, some of the typing results were confirmed by DNA sequence analysis. By computer simulation, most (98%) polymorphisms were shown to be natural or convertible (by changing 1 bp close to or next to each polymorphic site) RFLPs for the commercially available 4-base cutters.

Screening the dystrophin gene suggests a high rate of polymorphism in general but no exonic deletions in schizophrenics

The dystrophin gene, located at chromosome Xp21, was evaluated as a candidate gene in chronic schizophrenia in response to the report of a large family in which schizophrenia cosegregated with Becker muscular dystrophy [Zatz et al., 1991: Am J Hum Genet 49: A364; 1992: J Med Genet 30(2):131-134]. Genomic DNA from 94 men with chronic schizophrenia was evaluated by Southern blot analysis using cDNA probes that span exons 1-59. No exonic deletions were identified. An unexpectedly high rate of polymorphism was calculated in this study and two novel polymorphisms were found, demonstrating the usefulness of the candidate gene approach even when results of the original study are negative.

Recurrent missense mutations at the first and second base of codon Arg243 in human lipoprotein lipase in patients of different ancestries

Mutations in the lipoprotein lipase (LPL) gene are the most common cause of familial chylomicronemia. Here we define the molecular basis of LPL deficiency in four patients of German, French, Dutch, and Chinese descent. We show that two of the probands of Dutch and Chinese origin have a previously described Arg243His mutation while the patients of German and French descent have a novel Arg243Cys substitution in their LPL gene. Haplotype analysis is in favour of two separate origins for the Arg243Cys substitution which together with the Arg243His mutation would implicate three recurrent mutations involving the first and second nucleotides of the codon encoding Arg243 of the LPL gene. The recurrent mutations affecting the first and second nucleotide of CGC coding for the normal Arg residue are support for the high mutability of CpG dinucleotides within the LPL gene.

The study of variation in the human genome

Regions of the genome showing high evolutionary stability are often conserved as a result of functional constraints. Conversely, more variable regions are likely to represent DNA with no functional or structural importance. However, as in the case of immunologically important regions, sequence divergence does not always indicate lack of functional importance. There is thus a wealth of information from both a functional and an evolutionary point of view that comes from studies of DNA sequence variation, a neglected aspect of the genome endeavor. Naturally, one cannot sequence hundreds of individuals in full, but a useful compromise is to use less expensive methods and to limit the more expensive types of analysis to an appropriately chosen sample of loci. The sample could be determined after careful consideration of categories of DNA segments with respect to individual variation. The study of such categories of DNA variation patterns can help in the understanding of the role of each gene and vice versa. One other important application requiring a study of DNA variation in different human populations is forensic DNA typing. This study requires a knowledge of allele frequencies in different human populations. Evidence of a match between two DNA samples is meaningless if the approximate population frequency of the DNA pattern is not known. It has been suggested (E. Lander) that one use the highest frequency for the most common allele as a baseline frequency estimate. Obviously, systems in which this is employed require an extensive analysis of population-specific allele frequencies. In general, the best way of studying interindividual variation when detecting or describing new polymorphisms is to include interethnic variation.(ABSTRACT TRUNCATED AT 250 WORDS)

Ionizing radiation and genetic risks. I. Epidemiological, population genetic, biochemical and molecular aspects of Mendelian diseases

This paper reviews the currently available information on naturally occurring Mendelian diseases in man; it is aimed at providing a background and framework for discussion of experimental data on radiation-induced mutations (papers II and III) and for the estimation of the risk of Mendelian disease in human populations exposed to ionizing radiation (paper IV). Current consensus estimates indicate that a total of about 125 per 10(4) livebirths are directly affected by one or another naturally occurring Mendelian disease (autosomal dominants, 95/10(4); X-linked ones, 5/10(4); and autosomal recessives, 25/10(4). These estimates are conservative and take into account conditions which are very rare and for which prevalence estimates are unavailable. Most, although not all, of the recognized "common" dominants have onset in adult ages while most sex-linked and autosomal recessives have onset at birth or in childhood. Autosomal dominant and X-linked diseases (i.e., the responsible mutant alleles) presumed to be maintained in the population due to a balance between mutation and selection are the ones which may be expected to increase in frequency as a result of radiation exposures. Viewed from this standpoint, the above assumption seems safe only for a small proportion of such diseases; for the remainder, there is no easy way to discriminate between different mechanisms that may be responsible or to rigorously exclude some in favor of some others. Mutations in genes that code for enzymic proteins are more often recessive in contrast to those that code for non-enzymic proteins, which are more often dominant. At the molecular level, with recessives, a wide variety of changes is possible and these include specific types of point mutations, small and large intragenic deletions, multilocus deletions and rearrangements. In the case of dominants, however, the kinds of recoverable point mutations and deletion-type changes are less extensive because of functional constraints. The mutational potential of genes varies, depending on the gene, its size, sequence content and arrangement, location and its normal functions, and can be grouped into three groups: those in which only point mutations have been found to occur, those in which only deletions or other gross changes have been recovered and those in which both kinds of changes are known. Molecular data are available for about 75 Mendelian conditions and these suggest that in approximately 50% of them, the changes categorized to date are point mutations and in the remainder, intragenic deletions or other gross changes; there does not seem to be any fundamental difference between dominants and recessives with respect to the underlying molecular defect.(ABSTRACT TRUNCATED AT 400 WORDS)

Determining consanguinity by oligonucleotide fingerprinting with (GTG)5/(CAC)5

Simple tandemly organized (GTG)n/(CAC)n sequences are spread throughout the human chromosomes. The most informative DNA fingerprints for the testing of pedigrees and/or paternity were obtained with the simple triplet repeat probe (GTG)5 or its complement (CAC)5. These hypervariable simple-repeat fragments are stably inherited in a Mendelian fashion. Using these highly discriminating probes, all human individuals could, theoretically, be differentiated, except for genetically identical monozygotic twins. Examples from actual case work are reported and pertinent advantages of this methodology are discussed.

Polymorphisms in the human X-linked pyruvate dehydrogenase E1 alpha gene

Pyruvate dehydrogenase E1 alpha deficiency is an X-chromosome-linked disorder, often with fatal consequences. We have searched for genetically useful polymorphisms in or near this gene. No restriction fragment length polymorphisms were detected using a battery of 36 different restriction enzymes and probing with a full-length cDNA fragment, or two single-copy genomic fragments located within intron 8, and 15 kb 3' of the coding region, respectively. The chemical cleavage method was then applied to the detection of base changes in or near the gene. One polymorphism was found in exon 8 of the coding region. However, no base changes were detected in intron 3 or in the part of intron 8 covered by fragment gB2. Three blocks of microsatellite DNA containing variable numbers of CA-repeats were isolated from the 5' end of the gene and characterized. Length polymorphisms in these microsatellite DNAs were analysed using the polymerase chain reaction. Although the three loci are tightly linked, the polymorphisms appear not to be in disequilibrium, making them useful markers in linkage studies of the pyruvate dehydrogenase E1 alpha gene. Of 31 females analysed 12(39%) were heterozygous for at least one length polymorphism of the three (CA)n alleles.

Individual-specific DNA fingerprinting in man using the oligonucleotide probe (GTG)5/(CAC)5

Restriction fragment length polymorphisms (RFLPs) associated with interspersed simple repetitive DNA arise from DNA fragment lengths that contain variable numbers of the repeated motifs. Using restriction enzymes with different 4 base pair recognition sites and the simple triplet repeat hybridization probe, (GTG)5/(CAC)5, DNA multilocus fingerprints can be obtained in man. Only the DNAs of monozygous twins show indistinguishable banding patterns. Since the bands are inherited according to Mendelian laws, DNA fingerprints can be used for identification of individuals and paternity analysis. The discriminatory power in the DNA fingerprinting technique in forensic science is demonstrated and examples of paternity testing are given.

Genotyping of macerated stillborn fetuses

It is generally impossible to collect blood or to culture tissue from a macerated stillborn fetus. Accurate genotyping of such a fetus may, however, be critical for the diagnosis of genetic diseases and appropriate genetic counseling. In the East Flanders Prospective Twin Study, placental tissue of twin and triplet sets, in some of which one or both members were stillborn and macerated, has been stored at -20 degrees C. Of all these fetuses, sex and zygosity could be determined accurately on the placental deoxyribonucleic acid. We tested the possibility of nongenetic changes in deoxyribonucleic acid that result from maceration or tissue degradation over time in storage on placental samples from monochorionic twins in which only one member was stillborn and macerated. The deoxyribonucleic acid variants in these monozygotic twins were identical whether or not either cotwin was macerated. Thus deoxyribonucleic acid variants can be determined accurately on the placental tissue of macerated fetuses, even after prolonged freezing.

Application of single-locus hypervariable region DNA probes to deficiency cases in paternity testing

Seven DNA probes which recognize single-locus hypervariable region (HVR) were applied to a paternity test in which the putative father and his wife were deceased. Three legitimate children, an illegitimate child and her mother were available for analysis. The cumulative paternity index of the illegitimate child derived from 15 conventional blood group markers was 18.71 and from 7 DNA probes 92,572.08, that is, 4,948 times higher than the former. Thus the DNA analyses gave nearly conclusive evidence that the putative father was the biological father of the child. The application of highly discriminating polymorphisms of DNA which recognize single HVR loci is considered to be extremely informative in cases of disputed parentage.

Polymorphism of the human gamma chain fibrinogen gene

Nucleotide sequence comparisons of the published cDNAs and genomic sequences coding for the gamma-chain of human fibrinogen revealed several differences. We isolated two independent human cDNA clones, coding for part of this protein and compared their sequences, which are identical with the published relevant data. Our sequence allowed us to solve a conflict for aminoacid 88. All the remaining differences resulting from this comparison occurred in the third position of a codon and did not change codon properties or restriction sites. The level of polymorphism of this gene is discussed, taking into account also the nucleotide differences among all the published relevant data.

The human Y chromosome shows a low level of DNA polymorphism

Six new Y-specific probes have been isolated and are reported. Along with another six already described they have been used in a systemic search for male specific RFLPs. An overall number of 46515 nucleotides have been screened with 12 enzymes and no polymorphic pattern observed. Our data reveal a greatly reduced level of polymorphism compared with other chromosomes.

DNA sequence polymorphisms in Alu repeats

We have developed an efficient method for detection of sequence differences in genomic DNA based on a new principle (M. Orita et al., 1989, Genomics 5: 874-879). Using this method, we show here that approximately half the Alu repeats interspersed in the human genome are significantly polymorphic. Analysis of Alu repeat polymorphism should be useful in construction of a high-resolution map and also in identifying genotypes of individuals for clinical and other purposes because the repeats are ubiquitous and the technique for their detection is simple.

Differential diagnosis between complete mole and hydropic abortus by deoxyribonucleic acid fingerprints

We used a new method of deoxyribonucleic acid fingerprint analysis to obtain the differential diagnosis between complete mole and hydropic abortus. This method with a deoxyribonucleic acid minisatellite probe requires only a small amount of tissue sample and peripheral blood, and presents individual specific restriction fragment length polymorphisms (deoxyribonucleic acid "fingerprints") by simultaneous detection of many hypervariable regions (minisatellite regions) widely dispersed in the human genome. Southern blot hybridization showed that in cases of complete mole, all polymorphic fragments were exclusively inherited from the father. Some of the polymorphic bands of paternal deoxyribonucleic acid were not observed in molar deoxyribonucleic acid. However, in the hydropic abortus, the polymorphic fragments could be traced back to its parent. These results indicate that deoxyribonucleic acid fingerprints could distinguish the abnormal fertilization of complete mole (androgenesis) from the normal fertilization of hydropic abortus by identifying the difference in genetic variations between complete mole and hydropic abortus at the deoxyribonucleic acid level.

The mutational spectrum of single base-pair substitutions causing human genetic disease: patterns and predictions

Reports of single base-pair substitutions that cause human genetic disease and that have been located and characterized in an unbiased fashion were collated; 32% of point mutations were CG----TG or CG----CA transitions consistent with a chemical model of mutation via methylation-mediated deamination. This represents a 12-fold higher frequency than that predicted from random expectation, confirming that CG dinucleotides are indeed hotspots of mutation causing human genetic disease. However, since CG also appears hypermutable irrespective of methylation-mediated deamination, a second mechanism may also be involved in generating CG mutations. The spectrum of point mutations occurring outwith CG dinucleotides is also non-random, at both the mono- and dinucleotide, levels. An intrinsic bias in clinical detection was excluded since frequencies of specific amino acid substitutions did not correlate with the 'chemical difference' between the amino acids exchanged. Instead, a strong correlation was observed with the mutational spectrum predicted from the experimentally measured mispairing frequencies of vertebrate DNA polymerases alpha and beta in vitro. This correlation appears to be independent of any difference in the efficiency of enzymatic proofreading/mismatch-repair mechanisms but is consistent with a physical model of mutation through nucleotide misincorporation as a result of transient misalignment of bases at the replication fork. This model is further supported by an observed correlation between dinucleotide mutability and stability, possibly because transient misalignment must be stabilized long enough for misincorporation to occur. Since point mutations in human genes causing genetic disease neither arise by random error nor are independent of their local sequence environment, predictive models may be considered. We present a computer model (MUTPRED) based upon empirical data; it is designed to predict the location of point mutations within gene coding regions causing human genetic disease. The mutational spectrum predicted for the human factor IX gene was shown to resemble closely the observed spectrum of point mutations causing haemophilia B. Further, the model was able to predict successfully the rank order of disease prevalence and/or mutation rates associated with various human autosomal dominant and sex-linked recessive conditions. Although still imperfect, this model nevertheless represents an initial attempt to relate the variable prevalence of human genetic disease to the mutability inherent in the nucleotide sequences of the underlying genes.

Hypervariable polymorphic VNTR loci for parentage testing and individual identification

Three kinds of variable number of tandem repeat DNA probes (VNTR: pYNZ22, pYNH24, and pYNZ2) showing hypervariable polymorphisms were studied. Allelic frequencies and their confidence intervals among Japanese individuals were obtained. Co-dominant segregation of the polymorphism was confirmed in family studies. Two a priori probabilities were calculated for each VNTR locus: exclusion probabilities for an alleged father/mother/child trio and for an alleged parent/child duo, and probabilities of matching of genotyped two unrelated individuals or two siblings. Availability as well as highly discriminating polymorphic pattern of VNTR loci makes it potentially very useful for forensic and human genetic purposes.

A novel human DNA polymorphism resulting from transfer of DNA from chromosome 6 to chromosome 16

A cloned minisatellite, termed lambda MS29, that is unusual because it detects two variable loci in human DNA has been isolated. One locus, DNF21S1, located in the terminal region of the short arm of human chromosome 6, is also present in great apes. The second minisatellite locus, DNF21S2, is located interstitially on chromosome 16p11 and is absent both from non-human primates and from some humans. Physical mapping and sequencing show that the second locus has arisen recently in evolution by duplication of a large (greater than 15 kb) segment of chromosome 6 DNA containing a minisatellite and transposition onto chromosome 16 into a member of a novel low-copy-number repetitive DNA family. This unusual duplication/transposition event appears to represent the first example of a human DNA polymorphism arising through DNA-mediated, rather than RNA-mediated, transfer between autosomes.

Increased detectability of somatic changes in the DNA from human tumours after probing with "synthetic" and "genome-derived" hypervariable multilocus probes

DNA fingerprinting with two minisatellite (33.15, M13) and two simple repeat probes [(GACA)4, (CAC)5/(GTG)s] was performed to screen for somatic changes in the DNA from various solid human tumours in comparison with constitutional DNA from the same patient. Loss of bands or changes in band intensities were observed. Together the probes 33.15 and (CAC)5/(GTG)5 detected deviating fingerprint patterns in 63% of the colorectal carcinomas investigated. In mammary and stomach carcinomas, only 1/11 and 2/11 tumours, respectively, showed differences with either of the three probes, 33.15, (GACA)4 and (CAC)5/(GTG)5.

DNA fingerprinting with the oligonucleotide probe (CAC)5/(GTG)5: somatic stability and germline mutations

DNA fingerprints were generated from various human somatic tissues and from peripheral blood of 179 children and their 80 parents using (CAC)5/(GTG)5 oligonucleotide probes. Whereas somatic stability of the fingerprint patterns was demonstrated, the average rate for germline mutations was estimated to be approximately 0.001 per DNA locus and gamete, with the three different restriction enzymes used. Seven out of eight mutations observed appeared to be of paternal origin.

Population bottlenecks in Polynesia revealed by minisatellites

Tandem-repetitive highly variable loci in the human genome (minisatellites) have been used in gene mapping and as DNA "fingerprints", but they have not yet found much application in population genetics. We have investigate the capacity of six minisatellites to discriminate between four populations in Oceania. We find that in comparison to Melanesians, Polynesians have a significant loss of heterozygosity (or gene diversity), not noted using more traditional markers. We show also that the number of alleles, the allele distribution and the mutation rates at the Polynesian minisatellite loci do not deviate from those predicted by the neutral mutation/infinite allele model. The low gene diversity is therefore likely to be a result of the maintenance of small population sizes and bottleneck effects during the colonization of the Pacific.

A new approach using DNA fingerprinting for the determination of androgenesis as a cause of hydatidiform mole

A new method of DNA analysis has been used for the determination of androgenesis as a cause of complete hydatidiform mole. This method, using a minisatellite core probe, requires only a small amount of DNA and detects the restriction fragment length polymorphisms (RFLPs) due to allelic differences in the number of tandem repeats containing the core sequence. Southern blot hybridization showed an individual-specific DNA fingerprint, and each polymorphic band in molar tissues could be identified as being of paternal, but not maternal, origin. Some polymorphic bands of paternal DNA were not observed in molar tissues, indicating that endoreduplication of a normal haploid sperm or fertilization by dispermy to an anuclear oocyte with no effective genome could be the cause of complete hydatidiform mole. This method is sufficiently reliable and rapid that differential diagnosis could be made between complete hydatidiform mole, partial mole and hydropic change.