High-throughput variation detection and genotyping using microarrays

The genetic dissection of complex traits may ultimately require a large number of SNPs to be genotyped in multiple individuals who exhibit phenotypic variation in a trait of interest. Microarray technology can enable rapid genotyping of variation specific to study samples. To facilitate their use, we have developed an automated statistical method (ABACUS) to analyze microarray hybridization data and applied this method to Affymetrix Variation Detection Arrays (VDAs). ABACUS provides a quality score to individual genotypes, allowing investigators to focus their attention on sites that give accurate information. We have applied ABACUS to an experiment encompassing 32 autosomal and eight X-linked genomic regions, each consisting of approximately 50 kb of unique sequence spanning a 100-kb region, in 40 humans. At sufficiently high-quality scores, we are able to read approximately 80% of all sites. To assess the accuracy of SNP detection, 108 of 108 SNPs have been experimentally confirmed; an additional 371 SNPs have been confirmed electronically. To access the accuracy of diploid genotypes at segregating autosomal sites, we confirmed 1515 of 1515 homozygous calls, and 420 of 423 (99.29%) heterozygotes. In replicate experiments, consisting of independent amplification of identical samples followed by hybridization to distinct microarrays of the same design, genotyping is highly repeatable. In an autosomal replicate experiment, 813,295 of 813,295 genotypes are called identically (including 351 heterozygotes); at an X-linked locus in males (haploid), 841,236 of 841,236 sites are called identically.

Comparison of two blood collection techniques for measles antibody detection in children

BACKGROUND & OBJECTIVES

Though several seroprevalence and seroconversion studies have been conducted for measles using the filter paper haemagglutination inhibition (FPHAI), very few studies have compared the conventional serum HAI with the FPHAI. The present study was aimed at the evaluation of whole blood samples on filter paper as an alternative to serum specimens for detection of antibodies to measles virus.

METHODS

Serum and whole blood samples soaked on filter paper were collected from 165 randomly selected healthy children in the age groups of 1-2 yr. HAI test was performed on both sets of samples and the results compared.

RESULTS

Ninety samples that gave a titre of 8 or more by serum HAI also had titres of > or = 8 by FPHAI showing 100 per cent agreement between the two assays. Seventeen samples that had a serum HAI titre of 2 and 4 were missed by FPHAI due to the starting dilution of 8 of the latter. However, FPHAI gave no false positive results compared with serum HAI.

INTERPRETATION & CONCLUSION

Collection of the FP samples by finger prick is more acceptable and requires less expertise than venepuncture. Thus, in spite of the small percentage of missed cases by the filter paper method, this technique of sampling was found to be a convenient and reliable alternative to venepuncture, for detection of measles virus antibody especially in large scale seroepidemiological studies.

Sequence variation within the fragile X locus

The human genome provides a reference sequence, which is a template for resequencing studies that aim to discover and interpret the record of common ancestry that exists in extant genomes. To understand the nature and pattern of variation and linkage disequilibrium comprising this history, we present a study of approximately 31 kb spanning an approximately 70 kb region of FMR1, sequenced in a sample of 20 humans (worldwide sample) and four great apes (chimp, bonobo, and gorilla). Twenty-five polymorphic sites and two insertion/deletions, distributed in 11 unique haplotypes, were identified among humans. Africans are the only geographic group that do not share any haplotypes with other groups. Parsimony analysis reveals two main clades and suggests that the four major human geographic groups are distributed throughout the phylogenetic tree and within each major clade. An African sample appears to be most closely related to the common ancestor shared with the three other geographic groups. Nucleotide diversity, pi, for this sample is 2.63 +/- 6.28 x 10(-4). The mutation rate, mu is 6.48 x 10(-10) per base pair per year, giving an ancestral population size of approximately 6200 and a time to the most recent common ancestor of approximately 320,000 +/- 72,000 per base pair per year. Linkage disequilibrium (LD) at the FMR1 locus, evaluated by conventional LD analysis and by the length of segment shared between any two chromosomes, is extensive across the region.

Prognostic and pathologic significance of quantitative protein expression profiling in human gliomas

PURPOSE

Analysis of tumor-derived genetic lesions has provided insights into molecular pathogenesis of human gliomas. Because these changes represent only one of several mechanisms that alter gene expression during tumorigenesis, it is likely that further information will be obtained from a careful analysis of important regulatory proteins present in these tumors.

EXPERIMENTAL DESIGN

We have quantified the levels of key cell cycle/signaling proteins in 94 prospectively collected, meticulously preserved, "snap frozen" glioma specimens and have compared these levels with histopathological data and patient outcome.

RESULTS

The results of these experiments confirm that the levels of wild-type tumor suppressor proteins, such as p53, pRB, PTEN, p14(ARF), and p16(INK4), are lost or severely reduced in most gliomas, and that epidermal growth factor receptor, 2human telomerase reverse transcriptase, and cyclin-dependent kinase 4 are overexpressed frequently and with a few exceptions, almost exclusively, in glioblastomas. In addition, we report frequent underexpression of E2F-1 (in 55% of gliomas) and cyclin E overexpression (in 26% of gliomas), which have not yet been reported on the genomic level. Several of these markers significantly correlated with histopathological grade, and the levels of five proteins showed significant association with patient outcome. In particular, overexpression of epidermal growth factor receptor, human telomerase reverse transcriptase, cyclin-dependent kinase 4, and cyclin E was largely restricted to glioblastomas and was significantly associated with reduced patient survivals.

CONCLUSIONS

We conclude that the quantitation of cell cycle/signaling proteins from meticulously preserved glioma specimens provides further insights into the molecular pathogenesis of human gliomas and yields valuable prognostic information.

EDNRB/EDN3 and Hirschsprung disease type II

The study of vertebrate pigmentary anomalies has greatly improved our understanding of melanocyte biology. One such disorder, Waardenburg syndrome (WS), is a mendelian trait characterized by hypopigmentation and sensorineural deafness. It is commonly subdivided into four types (WS1-4), defined by the presence or absence of additional symptoms. WS type 4 (WS4), or Shah-Waardenburg syndrome, is also known as Hirschsprung disease Type II (HSCR II) and is characterized by an absence of epidermal melanocytes and enteric ganglia. Mutations in the genes encoding the endothelin type-B receptor (EDNRB) and its physiological ligand endothelin 3 (EDN3) are now known to account for the majority of HSCR II patients. Null mutations in the mouse genes Ednrb and Edn3 have identified a key role for this pathway in the normal development of melanocytes and other neural crest-derived lineages. The pleiotropic effects of genes in this pathway, on melanocyte and enteric neuron development, have been clarified by the embryologic identification of their common neural crest (NC) ancestry. EDNRB and EDN3 are transiently expressed in crest-derived melanoblast and neuroblast precursors, and in the surrounding mesenchymal cells, respectively. The influence of EDNRB-mediated signaling on the emigration, migration, proliferation, and differentiation of melanocyte and enteric neuron precursors, in vivo and in vitro has recently been the subject of great scrutiny. A major emergent theme is that EDN3-induced signaling prevents the premature differentiation of melanocyte and enteric nervous system precursors and is essential between 10 and 12.5 days post-coitum. We review the present understanding of pigment cell development in the context of EDNRB/EDN3--a receptor-mediated pathway with pleiotropic effects.

Testing for colon neoplasia susceptibility variants at the human COX2 locus

BACKGROUND

Siblings and other first-degree relatives of patients with "sporadic" (i.e., apparently nonfamilial) colorectal cancer or precursor adenomatous colon polyps have an increased risk of developing colon neoplasia. This observation suggests the presence of inherited genetic determinants for sporadic colon neoplasia. Mice homozygous for a null cyclooxygenase 2 (COX2) (also called PTGS2) allele have a dramatically reduced susceptibility to the development of intestinal adenomas. In humans, use of pharmacologic inhibitors of COX2 enzyme activity are associated with reduced risk of colon neoplasia. This study examined whether the human COX2 locus may be linked to colon neoplasia in humans.

METHODS

We used the affected sibling-pair method to test for linkage of the human COX2 locus to colon neoplasia.

RESULTS

We examined 74 concordantly affected sibling pairs from 46 sibships with colon neoplasia. One hundred five siblings from these sibships were diagnosed with either colorectal cancer or colon adenomatous polyps before age 65 years. No linkage between COX2 and colon neoplasia was found by use of a multipoint model-free linkage analysis (estimate of allele sharing was 0.44; standard error = +/-0.04; 95% confidence interval = 0.36 to 0.52). Moreover, even allowing for heterogeneity, the potential that a COX2 colon neoplasia susceptibility variant was present within a substantial subset of these sibships was strongly excluded under either a recessive or a dominant inheritance model (95% confidence to exclude a model in which 2.7% or more of the sibling pairs harbor a dominant susceptibility allele).

CONCLUSIONS

This study of concordantly affected sibling pairs thus demonstrates that variations in the COX2 gene are unlikely to be a source of individual susceptibility to colon neoplasia in humans.

The winged helix/forkhead transcription factor Foxq1 regulates differentiation of hair in satin mice

Satin (sa) homozygous mice have a silky coat with high sheen arising from structurally abnormal medulla cells and defects in differentiation of the hair shaft. We demonstrate that the winged helix/forkhead transcription factor, Foxq1 (Forkhead box, subclass q, member 1) is mutant in sa mice. An intragenic deletion was identified in the radiation-induced satin mutant of the SB/Le inbred strain; a second allele, identified by an N-ethyl-N-nitrosourea (ENU) mutagenesis screen, has a missense mutation in the conserved winged helix DNA-binding domain. Homozygous mutants of the two alleles are indistinguishable. We show that Foxq1 is expressed during embryogenesis and exhibits a tissue-restricted expression pattern in adult tissues. The hair defects appear to be restricted to the inner structures of the hair; consequently, Foxq1 has a unique and distinct function involved in differentiation and development of the hair shaft. Despite an otherwise healthy appearance, satin mice have been reported to exhibit suppressed NK-cell function and alloimmune cytotoxic T-cell function. We show instead that the immune defects are attributable to genetic background differences.

Linkage disequilibrium analysis of biallelic DNA markers, human quantitative trait loci, and threshold-defined case and control subjects

Linkage disequilibrium (LD) mapping has been applied to many simple, monogenic, overtly Mendelian human traits, with great success. However, extensions and applications of LD mapping approaches to more complex human quantitative traits have not been straightforward. In this article, we consider the analysis of biallelic DNA marker loci and human quantitative trait loci in settings that involve sampling individuals from opposite ends of the trait distribution. The purpose of this sampling strategy is to enrich samples for individuals likely to possess (and not possess) trait-influencing alleles. Simple statistical models for detecting LD between a trait-influencing allele and neighboring marker alleles are derived that make use of this sampling scheme. The power of the proposed method is investigated analytically for some hypothetical gene-effect scenarios. Our studies indicate that LD mapping of loci influencing human quantitative trait variation should be possible in certain settings. Finally, we consider possible extensions of the proposed methods, as well as areas for further consideration and improvement.

Patterns of meiotic recombination on the long arm of human chromosome 21

In this study we quantify the features of meiotic recombination on the long arm of human chromosome 21. We constructed a 67. 3-centimorgan (cM) high-resolution, comprehensive, and accurate genetic linkage map of chromosome 21q using 187 highly polymorphic markers covering almost the entire long arm; 46 loci, consisting of mutually recombining marker sets, were ordered with greater than 1000:1 odds and with average interlocus distance of 1.46 cM. These markers were used to accurately identify all exchanges in 186 female and 160 male meioses and to show (1) significant excess of recombination in female versus male meioses, (2) an overall decline in female:male recombination between the centromere and the telomere, (3) greater positive chiasma interference in male than in female meioses, and (4) lack of correlation between exchange frequency and parental age. By comparing the genetic map with the 21q sequence map, we show a general trend of increasing male, but near-constant female, recombination versus physical distance across 21q, explaining the gender-specific recombination effect. The recombination rate varies considerably between genders across 21q but is the greatest (eightfold) in the pericentromeric region, with a rate of approximately 250 kb/cM in females and approximately 2125 kb/cM in males. We used information on the locations of all exchanges to construct an empirical map function that confirms the statistical findings of positive interference. These analyses reveal that occurrence of recombination on 21q is not only gender-specific but also region-specific and that recombination suppression at the centromere is not universal. We also find evidence that male exchange location is highly correlated with gene density.

Lack of association of the angiotensinogen-6 polymorphism with blood pressure levels in the comprehensive NHLBI Family Blood Pressure Program. National Heart, Lung and Blood Institute

OBJECTIVE

To evaluate the association between the angiotensinogen-6 polymorphism (AGT-6) and blood pressure levels.

DESIGN

Data were analysed from the first 4,322 subjects of the NHLBI Family Blood Pressure Program (FBPP), consisting of four networks (GenNet, GENOA, HyperGEN and SAPPHIRe), each conducting a multicentre observational family study to identify and characterize the genetic determinants of hypertension and blood pressure. The four studies use different designs (concordant sibpairs, discordant pairs, sibships, extended pedigrees), target different ethnic groups (Caucasian, African-American, Japanese, Chinese), and have different inclusion/exclusion criteria. However, the protocols and definitions were standardized across networks before data collection to allow maximum poolability.

METHODS

Each network/racial group was analysed separately, using generalized linear models that accounted for the non-independence of family members and/or the confounding of anti-hypertensive medications as needed. The results were also pooled using a pre-planned meta-analysis technique.

RESULTS

AGT-6 was not significantly associated with blood pressure in any network/racial group. In the meta-analysis, the pooled effect of AGT-6 was small [hazard ratio = 1.10, 95% confidence interval (CI) = 0.99-1.22, P= 0.0647 for systolic; hazard ratio = 1.04, 95% CI = 0.89-1.21, P= 0.6383 for diastolic]. A post-hoc analysis restricting to subjects meeting JNC VI criteria for Stage I hypertension (blood pressure > 140/90 mmHg or medicated) showed a stronger statistically significant relationship for systolic blood pressure (hazard ratio = 1.44, 95% CI = 1.04-2.00, P= 0.0283).

CONCLUSIONS

AGT-6 has minimal to no effect on the inter-individual variation of blood pressure levels, and is at best a 'minor gene' for blood pressure in the population as a whole.

Evidence for altered Fragile-X mental retardation protein expression in response to behavioral stimulation

The Fragile-X mental retardation protein, the protein absent in Fragile-X syndrome, is synthesized near synapses upon neurotransmitter activation. Humans and mice lacking this protein exhibit abnormal dendritic spine lengths and numbers. Here we investigated Fragile-X protein levels in animals exposed to behavioral paradigms that induce neuronal morphological change. Fragile-X protein immunoreactivity was examined in visual cortices of rats reared in a complex environment for 10 or 20 days, motor cortices of rats trained on motor-skill tasks for 3 or 7 days, and either visual or motor cortices of inactive controls. Rats exposed to a complex environment for 20 days or trained for 7 days on motor-skill tasks exhibited increased Fragile-X protein immunoreactivity in visual or motor cortices, respectively. These results provide the first evidence for a behaviorally induced alteration of Fragile-X protein expression and are compatible with previous findings suggesting synaptic regulation of its expression. These results also strengthen the association of Fragile-X mental retardation protein expression with the alteration of synaptic structure.

Molecular evidence for a relationship between LINE-1 elements and X chromosome inactivation: the Lyon repeat hypothesis

X inactivation is a chromosome-specific form of genetic regulation in which thousands of genes on one homologue become silenced early in female embryogenesis. Although many aspects of X inactivation are now understood, the spread of the X inactivation signal along the entire length of the chromosome remains enigmatic. Extending the Gartler-Riggs model [Gartler, S. M. & Riggs, A. D. (1983) Annu. Rev. Genet. 17, 155-190], Lyon recently proposed [Lyon, M. F. (1998) Cytogenet. Cell Genet. 80, 133-137] that a nonrandom organization of long interspersed element (LINE) repetitive sequences on the X chromosome might be responsible for its facultative heterochromatization. In this paper, we present data indicating that the LINE-1 (L1) composition of the human X chromosome is fundamentally distinct from that of human autosomes. The X chromosome is enriched 2-fold for L1 repetitive elements, with the greatest enrichment observed for a restricted subset of LINE-1 elements that were active <100 million years ago. Regional analysis of the X chromosome reveals that the most significant clustering of these elements is in Xq13-Xq21 (the center of X inactivation). Genomic segments harboring genes that escape inactivation are significantly reduced in L1 content compared with X chromosome segments containing genes subject to X inactivation, providing further support for the association between X inactivation and L1 content. These nonrandom properties of L1 distribution on the X chromosome provide strong evidence that L1 elements may serve as DNA signals to propagate X inactivation along the chromosome.

Lack of association between a biallelic polymorphism in the adducin gene and blood pressure in whites and African Americans

Population-based candidate gene association analyses are becoming increasingly popular as a result of a greater number of genes and gene polymorphisms having been identified for which some functional information is available. Because many biochemical and physiologic systems impact blood pressure regulation and hypertension susceptibility, many of these identified genes and polymorphisms are candidates for population-level association studies involving blood pressure levels or hypertension status. Recent studies have suggested that the alpha-adducin gene may harbor polymorphisms that influence blood pressure level. Therefore, we embarked on a study to test one such polymorphism in two large US samples: one from an urban African American population (Maywood, IL) and another from a rural white population (Tecumseh, MI). We used both family-based association tests and tests that consider the impact of additional measured factors beyond adducin gene variation on blood pressure levels. We found no evidence for a significant effect of the chosen adducin polymorphism on blood pressure variation in either sample. We also found no association between Adducin genotypes and antihypertensive use. These facts, together with similar findings in companion studies, suggest that the alpha-adducin gene polymorphism does not have a pronounced effect on blood pressure variation in the populations studied. This does not suggest, however, that the alpha-adducin gene does not have a role in blood pressure regulation and hypertension susceptibility.

Parallel genotyping of human SNPs using generic high-density oligonucleotide tag arrays

Large scale human genetic studies require technologies for generating millions of genotypes with relative ease but also at a reasonable cost and with high accuracy. We describe a highly parallel method for genotyping single nucleotide polymorphisms (SNPs), using generic high-density oligonucleotide arrays that contain thousands of preselected 20-mer oligonucleotide tags. First, marker-specific primers are used in PCR amplifications of genomic regions containing SNPs. Second, the amplification products are used as templates in single base extension (SBE) reactions using chimeric primers with 3' complementarity to the specific SNP loci and 5' complementarity to specific probes, or tags, synthesized on the array. The SBE primers, terminating one base before the polymorphic site, are extended in the presence of labeled dideoxy NTPs, using a different label for each of the two SNP alleles, and hybridized to the tag array. Third, genotypes are deduced from the fluorescence intensity ratio of the two colors. This approach takes advantage of multiplexed sample preparation, hybridization, and analysis at each stage. We illustrate and test this method by genotyping 44 individuals for 142 human SNPs identified previously in 62 candidate hypertension genes. Because the hybridization results are quantitative, this method can also be used for allele-frequency estimation in pooled DNA samples.

Parental origin and phenotype of triploidy in spontaneous abortions: predominance of diandry and association with the partial hydatidiform mole

The origin of human triploidy is controversial. Early cytogenetic studies found the majority of cases to be paternal in origin; however, recent molecular analyses have challenged these findings, suggesting that digynic triploidy is the most common source of triploidy. To resolve this dispute, we examined 91 cases of human triploid spontaneous abortions to (1) determine the mechanism of origin of the additional haploid set, and (2) assess the effect of origin on the phenotype of the conceptus. Our results indicate that the majority of cases were diandric in origin because of dispermy, whereas the maternally-derived cases mainly originated through errors in meiosis II. Furthermore, our results indicate a complex relationship between phenotype and parental origin: paternally-derived cases predominate among "typical" spontaneous abortions, whereas maternally-derived cases are associated with either early embryonic demise or with relatively late demise involving a well-formed fetus. As the cytogenetic studies relied on analyses of the former type of material and the molecular studies on the latter sources, the discrepancies between the data sets are explained by differences in ascertainment. In studies correlating the origin of the extra haploid set with histological phenotype, we observed an association between paternal-but not maternal-triploidy and the development of partial hydatidiform moles. However, only a proportion of paternally derived cases developed a partial molar phenotype, indicating that the mere presence of two paternal genomes is not sufficient for molar development.

Evaluation of three commercially available kits for serological diagnosis of dengue haemorrhagic fever

Seventy one acute phase serum samples collected during an epidemic of dengue hemorrhagic fever were tested by immunoblot, a rapid immunochromatographic assay and Dengue Duo ELISA for presence of anti dengue IgM and IgG antibodies. A concordance of 81.7% and 76.1% was seen between the three tests for the detection of anti-dengue IgM antibodies and IgG antibodies respectively. The rapid test takes only five minutes, can be easily carried out in most laboratories and compares well with the ELISA and the immunoblot.