2021 Allan Award

This article is based on the address given by the author at the meeting of the American Society of Human Genetics (ASHG) held virtually October 18-22, 2021. The video of the original address can be found at the ASHG website.

An unsupervised learning approach to identify novel signatures of health and disease from multimodal data

BACKGROUND

Modern medicine is rapidly moving towards a data-driven paradigm based on comprehensive multimodal health assessments. Integrated analysis of data from different modalities has the potential of uncovering novel biomarkers and disease signatures.

METHODS

We collected 1385 data features from diverse modalities, including metabolome, microbiome, genetics, and advanced imaging, from 1253 individuals and from a longitudinal validation cohort of 1083 individuals. We utilized a combination of unsupervised machine learning methods to identify multimodal biomarker signatures of health and disease risk.

RESULTS

Our method identified a set of cardiometabolic biomarkers that goes beyond standard clinical biomarkers. Stratification of individuals based on the signatures of these biomarkers identified distinct subsets of individuals with similar health statuses. Subset membership was a better predictor for diabetes than established clinical biomarkers such as glucose, insulin resistance, and body mass index. The novel biomarkers in the diabetes signature included 1-stearoyl-2-dihomo-linolenoyl-GPC and 1-(1-enyl-palmitoyl)-2-oleoyl-GPC. Another metabolite, cinnamoylglycine, was identified as a potential biomarker for both gut microbiome health and lean mass percentage. We identified potential early signatures for hypertension and a poor metabolic health outcome. Additionally, we found novel associations between a uremic toxin, p-cresol sulfate, and the abundance of the microbiome genera Intestinimonas and an unclassified genus in the Erysipelotrichaceae family.

CONCLUSIONS

Our methodology and results demonstrate the potential of multimodal data integration, from the identification of novel biomarker signatures to a data-driven stratification of individuals into disease subtypes and stages-an essential step towards personalized, preventative health risk assessment.

Predispositional genome sequencing in healthy adults: design, participant characteristics, and early outcomes of the PeopleSeq Consortium

BACKGROUND

Increasing numbers of healthy individuals are undergoing predispositional personal genome sequencing. Here we describe the design and early outcomes of the PeopleSeq Consortium, a multi-cohort collaboration of predispositional genome sequencing projects, which is examining the medical, behavioral, and economic outcomes of returning genomic sequencing information to healthy individuals.

METHODS

Apparently healthy adults who participated in four of the sequencing projects in the Consortium were included. Web-based surveys were administered before and after genomic results disclosure, or in some cases only after results disclosure. Surveys inquired about sociodemographic characteristics, motivations and concerns, behavioral and medical responses to sequencing results, and perceived utility.

RESULTS

Among 1395 eligible individuals, 658 enrolled in the Consortium when contacted and 543 have completed a survey after receiving their genomic results thus far (mean age 53.0 years, 61.4% male, 91.7% white, 95.5% college graduates). Most participants (98.1%) were motivated to undergo sequencing because of curiosity about their genetic make-up. The most commonly reported concerns prior to pursuing sequencing included how well the results would predict future risk (59.2%) and the complexity of genetic variant interpretation (56.8%), while 47.8% of participants were concerned about the privacy of their genetic information. Half of participants reported discussing their genomic results with a healthcare provider during a median of 8.0 months after receiving the results; 13.5% reported making an additional appointment with a healthcare provider specifically because of their results. Few participants (< 10%) reported making changes to their diet, exercise habits, or insurance coverage because of their results. Many participants (39.5%) reported learning something new to improve their health that they did not know before. Reporting regret or harm from the decision to undergo sequencing was rare (< 3.0%).

CONCLUSIONS

Healthy individuals who underwent predispositional sequencing expressed some concern around privacy prior to pursuing sequencing, but were enthusiastic about their experience and not distressed by their results. While reporting value in their health-related results, few participants reported making medical or lifestyle changes.

Profound Perturbation of the Metabolome in Obesity Is Associated with Health Risk

Obesity is a heterogeneous phenotype that is crudely measured by body mass index (BMI). There is a need for a more precise yet portable method of phenotyping and categorizing risk in large numbers of people with obesity to advance clinical care and drug development. Here, we used non-targeted metabolomics and whole-genome sequencing to identify metabolic and genetic signatures of obesity. We find that obesity results in profound perturbation of the metabolome; nearly a third of the assayed metabolites associated with changes in BMI. A metabolome signature identifies the healthy obese and lean individuals with abnormal metabolomes-these groups differ in health outcomes and underlying genetic risk. Specifically, an abnormal metabolome associated with a 2- to 5-fold increase in cardiovascular events when comparing individuals who were matched for BMI but had opposing metabolome signatures. Because metabolome profiling identifies clinically meaningful heterogeneity in obesity, this approach could help select patients for clinical trials.

Identification of Misclassified ClinVar Variants via Disease Population Prevalence

There is a significant interest in the standardized classification of human genetic variants. We used whole-genome sequence data from 10,495 unrelated individuals to contrast population frequency of pathogenic variants to the expected population prevalence of the disease. Analyses included the ACMG-recommended 59 gene-condition sets for incidental findings and 463 genes associated with 265 OrphaNet conditions. A total of 25,505 variants were used to identify patterns of inflation (i.e., excess genetic risk and misclassification). Inflation increases as the level of evidence supporting the pathogenic nature of the variant decreases. We observed up to 11.5% of genetic disorders with inflation in pathogenic variant sets and up to 92.3% for the variant set with conflicting interpretations. This improved to 7.7% and 57.7%, respectively, after filtering for disease-specific allele frequency. The patterns of inflation were replicated using public data from more than 138,000 genomes. The burden of rare variants was a main contributing factor of the observed inflation, indicating collective misclassified rare variants. We also analyzed the dynamics of re-classification of variant pathogenicity in ClinVar over time, which indicates progressive improvement in variant classification. The study shows that databases include a significant proportion of wrongly ascertained variants; however, it underscores the critical role of ClinVar to contrast claims and foster validation across submitters.

Aberrant GlyRS-HDAC6 interaction linked to axonal transport deficits in Charcot-Marie-Tooth neuropathy

Dominant mutations in glycyl-tRNA synthetase (GlyRS) cause a subtype of Charcot-Marie-Tooth neuropathy (CMT2D). Although previous studies have shown that GlyRS mutants aberrantly interact with Nrp1, giving insight into the disease’s specific effects on motor neurons, these cannot explain length-dependent axonal degeneration. Here, we report that GlyRS mutants interact aberrantly with HDAC6 and stimulate its deacetylase activity on α-tubulin. A decrease in α-tubulin acetylation and deficits in axonal transport are observed in mice peripheral nerves prior to disease onset. An HDAC6 inhibitor used to restore α-tubulin acetylation rescues axonal transport deficits and improves motor functions of CMT2D mice. These results link the aberrant GlyRS-HDAC6 interaction to CMT2D pathology and suggest HDAC6 as an effective therapeutic target. Moreover, the HDAC6 interaction differs from Nrp1 interaction among GlyRS mutants and correlates with divergent clinical presentations, indicating the existence of multiple and different mechanisms in CMT2D.

Mutations in glycyl-tRNA synthetase (GlyRS) cause Charcot-Marie-Tooth disease, a neuromuscular disorder characterized by axonal degeneration. Here the authors show that mutant GlyRS interacts with histone deacetylase 6, resulting in increased deacetylation of α-tubulin and axonal transport deficits.

Profiling of Short-Tandem-Repeat Disease Alleles in 12,632 Human Whole Genomes

Short tandem repeats (STRs) are hyper-mutable sequences in the human genome. They are often used in forensics and population genetics and are also the underlying cause of many genetic diseases. There are challenges associated with accurately determining the length polymorphism of STR loci in the genome by next-generation sequencing (NGS). In particular, accurate detection of pathological STR expansion is limited by the sequence read length during whole-genome analysis. We developed TREDPARSE, a software package that incorporates various cues from read alignment and paired-end distance distribution, as well as a sequence stutter model, in a probabilistic framework to infer repeat sizes for genetic loci, and we used this software to infer repeat sizes for 30 known disease loci. Using simulated data, we show that TREDPARSE outperforms other available software. We sampled the full genome sequences of 12,632 individuals to an average read depth of approximately 30× to 40× with Illumina HiSeq X. We identified 138 individuals with risk alleles at 15 STR disease loci. We validated a representative subset of the samples (n = 19) by Sanger and by Oxford Nanopore sequencing. Additionally, we validated the STR calls against known allele sizes in a set of GeT-RM reference cell-line materials (n = 6). Several STR loci that are entirely guanine or cytosines (G or C) have insufficient read evidence for inference and therefore could not be assayed precisely by TREDPARSE. TREDPARSE extends the limit of STR size detection beyond the physical sequence read length. This extension is critical because many of the disease risk cutoffs are close to or beyond the short sequence read length of 100 to 150 bases.

An international effort towards developing standards for best practices in analysis, interpretation and reporting of clinical genome sequencing results in the CLARITY Challenge

Background

There is tremendous potential for genome sequencing to improve clinical diagnosis and care once it becomes routinely accessible, but this will require formalizing research methods into clinical best practices in the areas of sequence data generation, analysis, interpretation and reporting. The CLARITY Challenge was designed to spur convergence in methods for diagnosing genetic disease starting from clinical case history and genome sequencing data. DNA samples were obtained from three families with heritable genetic disorders and genomic sequence data were donated by sequencing platform vendors. The challenge was to analyze and interpret these data with the goals of identifying disease-causing variants and reporting the findings in a clinically useful format. Participating contestant groups were solicited broadly, and an independent panel of judges evaluated their performance.

Results

A total of 30 international groups were engaged. The entries reveal a general convergence of practices on most elements of the analysis and interpretation process. However, even given this commonality of approach, only two groups identified the consensus candidate variants in all disease cases, demonstrating a need for consistent fine-tuning of the generally accepted methods. There was greater diversity of the final clinical report content and in the patient consenting process, demonstrating that these areas require additional exploration and standardization.

Conclusions

The CLARITY Challenge provides a comprehensive assessment of current practices for using genome sequencing to diagnose and report genetic diseases. There is remarkable convergence in bioinformatic techniques, but medical interpretation and reporting are areas that require further development by many groups.

Adult genetic risk screening

Recent advances in genetic analysis especially DNA sequencing technology open a new strategy for adult disease prevention by genetic screening. Physicians presently treat disease pathology with less emphasis on disease risk prevention/reduction. Genetic screening has reduced the incidence of untreatable childhood genetic diseases and improved the care of newborns. The opportunity exists to expand screening programs and reduce the incidence of adult onset diseases via genetic risk identification and disease intervention. This article outlines the approach, challenges, and benefits of such screening for adult genetic disease risks.

Truncating mutations of MAGEL2 cause Prader-Willi phenotypes and autism

Prader-Willi syndrome (PWS) is caused by the absence of paternally expressed, maternally silenced genes at 15q11-q13. We report four individuals with truncating mutations on the paternal allele of MAGEL2, a gene within the PWS domain. The first subject was ascertained by whole-genome sequencing analysis for PWS features. Three additional subjects were identified by reviewing the results of exome sequencing of 1,248 cases in a clinical laboratory. All four subjects had autism spectrum disorder (ASD), intellectual disability and a varying degree of clinical and behavioral features of PWS. These findings suggest that MAGEL2 is a new gene causing complex ASD and that MAGEL2 loss of function can contribute to several aspects of the PWS phenotype.

P5-12-04: Genetic Linkage between Acquired and Primary Lymphedema Evaluated through Whole Exome Sequencing and NIR Fluorescence Lymphatic Imaging

Acquired lymphedema is thought to arise from the damage of the lymphatic vasculature that transports excess fluid and macromolecules away from tissues for return to the blood vasculature. The onset of the cancer acquired disease can occur months to years after lymph node dissection and manifests itself as an accumulation of fluid and macromolecules in tissues that leads to edema and irresolvable swelling. The rare disease of primary lymphedema is identical to cancer acquired lymphedema, with the exception that there is no trauma or cancer treatment that can be attributed as its cause. Primary lymphedema has been attributed to genetic causes since the late nineteenth century. Although there are five known genetic causes of hereditary or primary lymphedema, the majority of patients with lymphedema do not possess mutations in these genes. More recently, it has been proposed that a genetic link between cancer acquired and primary lymphedema exists. If a genetic susceptibility for cancer acquired lymphedema could be found, then we could predict which survivors will encounter the disease and could develop new therapies which are more effective than the current treatments that have remained unchanged for the past 80 years.

In an FDA approved investigational study, we used near-infrared (NIR) fluorescence imaging to phenotype the lymphatic architecture of subjects with both acquired and primary lymphedema, as well as their unaffected family members. We collected blood for DNA analyses. NIR fluorescence provided the phenotype of abnormal lymphatic function while whole exome sequencing provided the genotype. Bioinformatics analyses were then used to identify causative genes using cosegregation of familial genotypes using the phentotypes found through NIR fluorescence imaging.

The first family analyzed had members with primary and acquired lymphedema in which mutations encoding for proteins that participate in the HGF/c-MET and PI3K pathways could potentially explain the inheritance of lymphedema in this family. The father and affected daughters were heterozygous for a de novo SNP HGF in the kringle binding domain that interacts with tyrosine kinase receptor c-MET. The father had a normal lymphatic phenotype. On the other hand, the mother and daughters were heterozygous for the de novo mutation of INPPL1 (SHIP-2), adjacent to the SH2 domain of the protein that is known to bind to the multifunctional docking site of c-MET and associates with proteins in the Rho pathway for cytoskeletal reorganization. The daughters possessed both HGF and INPPL1 mutations and were diagnosed with primary lymphedema while the mother, who possessed the INPPL1 mutation, was diagnosed at the time of NIR imaging with acquired lymphedema. Analyses of remaining families as well as breast cancer related lymphedema patients are underway to confirm whether INPPL1 may be a candidate susceptibility gene for acquired lymphedema. Supported in parts by R01 HL092923 and CA128919, The Texas Star Award, and the Cullen Foundation.

Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P5-12-04.

Deficiencies of glucosamine-6-sulfate or galactosamine-6-sulfate sulfatases are responsible for different mucopolysaccharidoses

[1-3H]Galactitol-6-sulfate, N- [1-3H]acetylgalactosaminitol-6-sulfate, N-[1-3H]acetylglucosaminitol-6-sulfate, N-acetylglucosamine-6-sulfate, and 6-sulfated tetrasaccharides from chondroitin-6-sulfate have been used for the measurement of 6-sulfatase activity of extracts of normal skin fibroblasts and of fibroblasts cultured from patients with genetic mucopolysaccharidoses. With these substrates, extracts of fibroblasts derived from Morquio patients lack or have greatly reduced activities for galactitol-6-sulfate, N-acetylgalactosaminitol-6-sulfate, and 6-sulfated tetrasaccharides but have normal activity for N-acetylglucosamine-6-sulfate and its alditol; those derived from a patient with a newly discovered mucopolysaccharidosis have greatly reduced activity for N-acetylglucosamine-6-sulfate and its alditol but normal activity for galactitol-6-sulfate, N-acetylgalactosaminitol-6-sulfate, and the 6-sulfated tetrasaccharides. These findings demonstrate the existence of two different hexosamine-6-sulfate sulfatases, specific for the glucose or galactose configuration of their substrates. Their respective deficiencies, causing inability to degrade keratan sulfate and heparan sulfate in one case and keratan sulfate and chondroitin-6-sulfate in the other, are responsible for different clinical phenotypes.

The drug development crisis: efficiency and safety

Despite advancements in genetics, chemistry, and protein engineering, recent years have seen fewer approvals of new drugs, increases in development costs, and high-profile drug withdrawals. This article focuses on technologic methods for improving drug development efficiency. These technologies include high-content cell screening, expression profiling, mass spectroscopy, mouse models of disease, and a post-launch screening program that enables investigations of adverse drug effects. Implementation of these new technologies promises to improve performance in drug development and safety.

Correction of chromosomal mutation and random integration in embryonic stem cells with helper-dependent adenoviral vectors

For gene therapy of inherited diseases, targeted integration/gene repair through homologous recombination (HR) between exogenous and chromosomal DNA would be an ideal strategy to avoid potentially serious problems of random integration such as cellular transformation and gene silencing. Efficient sequence-specific modification of chromosomes by HR would also advance both biological studies and therapeutic applications of a variety of stem cells. Toward these goals, we developed an improved strategy of adenoviral vector (AdV)-mediated HR and examined its ability to correct an insertional mutation in the hypoxanthine phosphoribosyl transferase (Hprt) locus in male mouse ES cells. The efficiency of HR was compared between four types of AdVs that contained various lengths of homologies at the Hprt locus and with various multiplicities of infections. The frequency of HR with helper-dependent AdVs (HD AdVs) with an 18.6-kb homology reached 0.2% per transduced cell at a multiplicity of infection of 10 genomes per cell. Detection of random integration at DNA levels by PCR revealed extremely high efficiency of 5% per cell. We also isolated and characterized chromosomal sites where HD AdVs integrated in a random manner. In contrast to retroviral, lentiviral, and adeno-associated viral vectors, which tend to integrate into genes, the integration sites of AdV was distributed randomly inside and outside genes. These findings suggest that HR mediated by HD AdVs is efficient and relatively safe and might be a new viable option for ex vivo gene therapy as well as a tool for chromosomal manipulation of a variety of stem cells.

Monitoring expression of genes involved in drug metabolism and toxicology using DNA microarrays

Oligonucleotide DNA microarrays were investigated for utility in measuring global expression profiles of drug metabolism genes. This study was performed to investigate the feasibility of using microarray technology to minimize the long, expensive process of testing drug candidates for safety in animals. In an evaluation of hybridization specificity, microarray technology from Affymetrix distinguished genes up to a threshold of approximately 90% DNA identity. Oligonucleotides representing human cytochrome P-450 gene CYP3A5 showed heterologous hybridization to CYP3A4 and CYP3A7 RNAs. These genes could be clearly distinguished by selecting a subset of oligonucleotides that hybridized selectively to CYP3A5. Further validation of the technology was performed by measuring gene expression profiles in livers of rats treated with vehicle, 3-methylcholanthrene (3MC), phenobarbital, dexamethasone, or clofibrate and by confirming data for six genes using quantitative RT-PCR. Responses of drug metabolism genes, including CYPs, epoxide hydrolases (EHs), UDP-glucuronosyl transferases (UGTs), glutathione sulfotransferases (GSTs), sulfotransferases (STs), drug transporter genes, and peroxisomal genes, to these well-studied compounds agreed well with, and extended, published observations. Additional gene regulatory responses were noted that characterize metabolic effects or stress responses to these compounds. Thus microarray technology can provide a facile overview of gene expression responses relevant to drug metabolism and toxicology.

Identification of receptors for neuromedin U and its role in feeding

Neuromedin U (NMU) is a neuropeptide with potent activity on smooth muscle which was isolated first from porcine spinal cord and later from other species. It is widely distributed in the gut and central nervous system. Peripheral activities of NMU include stimulation of smooth muscle, increase of blood pressure, alteration of ion transport in the gut, control of local blood flow and regulation of adrenocortical function. An NMU receptor has not been molecularly identified. Here we show that the previously described orphan G-protein-coupled receptor FM-3 (ref. 15) and a newly discovered one (FM-4) are cognate receptors for NMU. FM-3, designated NMU1R, is abundantly expressed in peripheral tissues whereas FM-4, designated NMU2R, is expressed in specific regions of the brain. NMU is expressed in the ventromedial hypothalamus in the rat brain, and its level is significantly reduced following fasting. Intracerebroventricular administration of NMU markedly suppresses food intake in rats. These findings provide a molecular basis for the biochemical activities of NMU and may indicate that NMU is involved in the central control of feeding.

Optimization of the helper-dependent adenovirus system for production and potency in vivo

Helper-dependent (HD) adenoviral vectors devoid of all viral coding sequences provide for safe and highly efficient gene transfer with long-lasting transgene expression. High titer stocks of HD vectors can be generated by using the cre-recombinase system. However, we have encountered difficulties with this system, including rearranged HD vectors and variable efficiency of HD vector rescue. These problems represent a major hindrance, particularly with regard to large-scale production. To overcome these limitations, we have modified the system in two ways: We constructed a new helper virus with a modified packaging signal and enhanced growth characteristics. We also redesigned the vector backbones by including noncoding adenovirus sequences adjacent to the right inverted terminal repeat and by incorporated a number of different segments of noncoding DNA of human origin as "stuffer." Comparison of these vectors showed that the nature of the stuffer sequence affects replication of the HD vector. Optimization of the system resulted in a more robust and consistent production of HD vectors with low helper contamination and high in vivo potency.

Overexpression of M68/DcR3 in human gastrointestinal tract tumors independent of gene amplification and its location in a four-gene cluster

Fas-mediated apoptosis is an important regulator of cell survival, and abnormalities in this system have been shown to result in a number of human pathological conditions. A secreted member of the tumor necrosis factor receptor superfamily, DcR3, was recently reported to be amplified in human lung and colon cancers as a negative regulator of Fas-mediated apoptosis. We identified this gene, which we call M68. M68 genomic DNA, mRNA, and protein levels were examined in a series of human gastrointestinal tract tumors. Using M68 immunohistochemistry and a scoring system similar to that used for HER-2/neu, we found that M68 protein was overexpressed in 30 of 68 (44%) human adenocarcinomas of the esophagus, stomach, colon, and rectum. Tumors examined by Northern blot revealed M68 mRNA highly elevated in a similar fraction of primary tumors from the same gastrointestinal tract regions, as well as in the colon adenocarcinoma cell lines SW480 and SW1116. Further, we found M68 protein to be overexpressed in a substantial number of tumors in which gene amplification could not be detected by fluorescence in situ hybridization or quantitative genomic PCR, suggesting that overexpression of M68 may precede amplification in tumors. Finally, we find that M68 lies within a four-gene cluster that includes a novel helicase-like gene (NHL) related to RAD3/ERCC2, a plasma membrane Ras-related GTPase and a member of the stathmin family, amplification or overexpression of which may also contribute to cell growth and tumor progression.

Retina-specific nuclear receptor: A potential regulator of cellular retinaldehyde-binding protein expressed in retinal pigment epithelium and Müller glial cells

In an effort to identify nuclear receptors important in retinal disease, we screened a retina cDNA library for nuclear receptors. Here we describe the identification of a retina-specific nuclear receptor (RNR) from both human and mouse. Human RNR is a splice variant of the recently published photoreceptor cell-specific nuclear receptor [Kobayashi, M., Takezawa, S., Hara, K., Yu, R. T., Umesono, Y., Agata, K., Taniwaki, M., Yasuda, K. & Umesono, K. (1999) Proc. Natl. Acad. Sci. USA 96, 4814-4819] whereas the mouse RNR is a mouse ortholog. Northern blot and reverse transcription-PCR analyses of human mRNA samples demonstrate that RNR is expressed exclusively in the retina, with transcripts of approximately 7.5 kb, approximately 3.0 kb, and approximately 2.3 kb by Northern blot analysis. In situ hybridization with multiple probes on both primate and mouse eye sections demonstrates that RNR is expressed in the retinal pigment epithelium and in Müller glial cells. By using the Gal4 chimeric receptor/reporter cotransfection system, the ligand binding domain of RNR was found to repress transcriptional activity in the absence of exogenous ligand. Gel mobility shift assays revealed that RNR can interact with the promoter of the cellular retinaldehyde binding protein gene in the presence of retinoic acid receptor (RAR) and/or retinoid X receptor (RXR). These data raise the possibility that RNR acts to regulate the visual cycle through its interaction with cellular retinaldehyde binding protein and therefore may be a target for retinal diseases such as retinitis pigmentosa and age-related macular degeneration.

Identification of urotensin II as the endogenous ligand for the orphan G-protein-coupled receptor GPR14

Urotensin II (UII) is a neuropeptide with potent cardiovascular effects. Its sequence is strongly conserved among different species and has structural similarity to somatostatin. No receptor for UII has been molecularly identified from any species so far. GPR14 was cloned as an orphan G protein-coupled receptor with similarity to members of the somatostatin/opioid receptor family. We have now demonstrated that GPR14 is a high affinity receptor for UII and designate it UII-R1a. HEK293 cells and COS-7 cells transfected with rat GPR14 showed strong, dose-dependent calcium mobilization in response to fish, frog, and human UII. Radioligand binding analysis showed high affinity binding of UII to membrane preparations isolated from HEK293 cells stably expressing rat GPR14. In situ hybridization analysis showed that GPR14 was expressed in motor neurons of the spinal cord, smooth muscle cells of the bladder, and muscle cells of the heart. The identification of the first receptor for UII will allow better understanding of the physiological and pharmacological roles of UII.

Triplet repeat disorders: discussion of molecular mechanisms

Comparison of the growing number of disorders known to be associated with triplet repeat expansions reveals both common features and a diversity of molecular pathways. Despite significant progress towards the characterization of proteins coded by the mutant genes, the complex nature of these disorders requires identification of all molecular components of the triplet repeat pathways. In this brief review we will discuss recent progress in determining the molecular mechanisms of disorders with unstable trinucleotide mutations.

Characterization of the human cysteinyl leukotriene CysLT1 receptor

The cysteinyl leukotrienes-leukotriene C4(LTC4), leukotriene D4(LTD4) and leukotriene E4(LTE4)-are important mediators of human bronchial asthma. Pharmacological studies have determined that cysteinyl leukotrienes activate at least two receptors, designated CysLT1 and CysLT2. The CysLT1-selective antagonists, such as montelukast (Singulair), zafirlukast (Accolate) and pranlukast (Onon), are important in the treatment of asthma. Previous biochemical characterization of CysLT1 antagonists and the CysLT1 receptor has been in membrane preparations from tissues enriched for this receptor. Here we report the molecular and pharmacological characterization of the cloned human CysLT1 receptor. We describe the functional activation (calcium mobilization) of this receptor by LTD4 and LTC4, and competition for radiolabelled LTD4 binding to this receptor by the cysteinyl leukotrienes and three structurally distinct classes of CysLT1-receptor antagonists. We detected CysLT1-receptor messenger RNA in spleen, peripheral blood leukocytes and lung. In normal human lung, expression of the CysLT1-receptor mRNA was confined to smooth muscle cells and tissue macrophages. Finally, we mapped the human CysLT1-receptor gene to the X chromosome.

Evaluation of the Best disease gene in patients with age-related macular degeneration and other maculopathies

Vitelliform macular dystrophy (VMD2, Best disease, MIM153700) is an early onset, autosomal, dominant macular degeneration characterized by the deposition of lipofuscin-like material within and below the retinal pigment epithelium (RPE); it is associated with degeneration of the RPE and overlying photoreceptors. Recently, we cloned the gene bestrophin, which is responsible for the disease, and identified a number of causative mutations in families with VMD2. Here, we report that the analysis of bestrophin in a collection of 259 age-related macular degeneration (AMD) patients provides evidence that mutations in the Best disease gene do not play a significant role in the predisposition of individuals to AMD. However, our results suggest that, in addition to Best disease, mutations within the bestrophin gene could be responsible for other forms of maculopathy with phenotypic characteristics similar to Best disease and for other diseases not included in the VMD category.

DNA chips: promising toys have become powerful tools

DNA chips are glass surfaces that represent thousands of DNA fragments arrayed at discrete sites. Hybridization of RNA or DNA-derived samples to DNA chips allows us to monitor expression of mRNAs or the occurrence of polymorphisms in genomic DNA. The technology holds great promise for identifying gene polymorphisms that predispose man to disease, gene regulation events involved in disease progression, and more-effective disease treatments.

Mapping of hKCa3 to chromosome 1q21 and investigation of linkage of CAG repeat polymorphism to schizophrenia

CAG trinucleotide polymorphisms in the neuronal small conductance calcium-activated potassium channel gene hKCa3 have been reported to be associated with schizophrenia. Attempts to confirm this finding have met with mixed results. We investigated hKCa3 CAG allele lengths in families from the National Institute of Mental Health (NIMH) Schizophrenia Genetics Initiative, by comparing transmission to discordant siblings and parental transmission to affected offspring. Overall, there was no convincing evidence that hKCa3 CAG lengths differ between schizophrenics and controls. We did, however, observe a trend (P = 0.063) toward over-representation of long (> or = 19) CAG repeats in the shorter of the two hKCa3 alleles in schizophrenics. There was no evidence of excessive parental transmission of long CAG repeat alleles to affected offspring. In addition, we re-mapped hKCa3 and found that it resides on chromosome 1q21, in a region which has been linked to familial hemiplegic migraine, but not to schizophrenia. These data provide no significant support for the association of hKCa3 with schizophrenia.

Identification of two hERR2-related novel nuclear receptors utilizing bioinformatics and inverse PCR

Identification of novel nuclear receptors based on the highly conserved DNA-binding domain (DBD) has previously depended mainly on low stringency hybridization of cDNA libraries and degenerate PCR. Establishment of the expressed sequence tag (EST) database in recent years has provided an alternative approach for the discovery of novel members of gene families. The rate-limiting step is the conversion of ESTs to full-length cDNA. This article describes the identification of two novel nuclear receptors (hERRbeta2 and hERRgamma2) related to human estrogen-receptor-related receptor 2 (hERR2) by mining the EST database and retrieving of full-length cDNA via inverse PCR on subdivided primary cDNA library pools. The deduced protein sequences of hERRbeta2 and hERRgamma2 contain 500 and 458 amino acid (aa) residues respectively. Sequence analysis revealed that hERRbeta2 and hERRgamma2 respectively share 95% and 77% overall aa sequence identity with hERR2. However, the extra C-terminal domain in hERRbeta2 and extra N-terminal domain in hERRgamma2 are not present in the closely related hERR2 or mouse ERR2 (mERR2). Extensive sequence verification revealed that hERR2 previously reported as a human gene is actually a rat gene, whereas hERRbeta2 is the true human ortholog of hERR2 and mERR2. Tissue distribution studies showed that hERRgamma2 was expressed in a broader panel of tissues at a higher level than hERRbeta2. hERRbeta2 was mapped to cytogenetic locus 14q24.3 approximately -14q31, a region containing multiple loci involved in genetic diseases, including Alzheimer and diabetes. hERRgamma2 was mapped to 1q32. Given the high sequence homology between hERRbeta2 and mERR2, the two receptors may have similar biological function in vivo.

Mice deficient in the urea-cycle enzyme, carbamoyl phosphate synthetase I, die during the early neonatal period from hyperammonemia

Ammonia liberated during amino acid catabolism in mammals is highly neurotoxic and is detoxified by the five enzymes of the urea cycle that are expressed within the liver. Inborn errors of each of the urea cycle enzymes occur in humans. Carbamoyl phosphate synthetase I (CPSase I; EC 6.3.4.16) is located within the inner mitochondrial matrix and catalyzes the initial rate-limiting step of the urea cycle. Unless treated, complete deficiency of CPSase I, a rare autosomal recessive disease, causes death in newborn infants. Survivors are often mentally retarded and suffer frequent hyperammonemic crises during intercurrent illness or other catabolic stresses. Biochemically, CPSase I deficiency is characterized by high levels of blood ammonia, glutamine, and alanine, with low or absent citrulline and arginine levels. As a first step toward the development of gene therapy directed to the hepatocyte, we have generated a CPSase I-deficient mouse by gene targeting. Mice with homozygous disruption of CPSase I (CPSase [-/-] mice) die within 36 hours of birth with overwhelming hyperammonemia, and without significant liver pathology. This animal is a good model of human CPSase I deficiency.

Expanded-capacity adenoviral vectors--the helper-dependent vectors

Significant advances have recently been made in the development of vectors and gene-delivery systems for gene therapy. Experiments performed over the past decade have revealed how vectors will have to be modified to make them a clinically viable treatment option. In the case of adenovirus (Ad) vectors, which have been particularly useful as gene delivery vehicles, the main drawback associated with their use is vector-mediated immunogenicity. Recent modifications of the Ad backbone have led to the development of helper-dependent (HD) Ad vectors, which are completely devoid of all viral protein-coding sequences. These modifications have significantly reduced the immunogenicity of Ad vectors and have enhanced their safety. It is expected that HD vectors will become important tools for future clinical gene therapy.

Inducible control of gene expression: prospects for gene therapy

A number of drug-related gene expression systems are available for controlling target gene transcription through the use of small-molecule inducing compounds. While the utility of such systems has been demonstrated in vitro and in transgenic mice, recent improvements are likely to make these systems more amenable for use in a therapeutic context, such as gene therapy. These improvements include further optimization of the antiprogestin-regulated gene switch, rendering it more sensitive to RU486, and the synthesis of nonimmunosuppressive rapamycin analogs for use in dimerization-based strategies of gene regulation.

Isolation and characterization of LRP6, a novel member of the low density lipoprotein receptor gene family

A novel member of the low density lipoprotein receptor (LDLR) gene family has been identified and characterized. This gene, termed LDL receptor-related protein 6 (LRP6), encodes a transmembrane protein which has 71% identity and is structurally similar to the protein encoded by LRP5, a proposed candidate gene for type 1 diabetes located on human chromosome 11q13. LRP6 maps to human chromosome 12p11-p13. Mouse Lrp6 encodes a protein that has 98% identity to human LRP6 and maps to chromosome 6. Unlike other members of the LDLR family, LRP6 and LRP5 display a unique pattern of four epidermal growth factor (EGF) and three LDLR repeats in the extracellular domain. The cytoplasmic domain of LRP6 is not similar to other members of the LDLR family, while comparison with LRP5 reveals proline-rich motifs that may mediate protein-protein interactions. Thus, it is likely that LRP6 and LRP5 comprise a new class of the LDLR family.

An adenoviral vector deleted for all viral coding sequences results in enhanced safety and extended expression of a leptin transgene

Adenoviral (Ad)-mediated in vivo gene transfer and expression are limited in part by cellular immune responses to viral-encoded proteins and/or transgene immunogenicity. In an attempt to diminish the former responses, we have previously developed and described helper-dependent (HD) Ad vectors in which the viral protein coding sequences are completely eliminated. These HD vectors have up to 37 kb insert capacity, are easily propagated in a Cre recombinase-based system, and can be produced to high concentration and purity (>99.9% helper-free vector). In this study, we compared safety and efficacy of leptin gene delivery mediated by an HD vector (HD-leptin) and a first-generation E1-deleted Ad vector (Ad-leptin) in normal lean and ob/ob (leptin-deficient) mice. In contrast to evidence of liver toxicity, inflammation, and cellular infiltration observed with Ad-leptin delivery in mice, HD-leptin delivery was associated with a significant improvement in associated safety/toxicity and resulted in efficient gene delivery, prolonged elevation of serum leptin levels, and associated weight loss. The greater safety, efficient gene delivery, and increased insert capacity of HD vectors are significant improvements over current Ad vectors and represent favorable features especially for clinical gene therapy applications.

Identification of the gene responsible for Best macular dystrophy

Best macular dystrophy (BMD), also known as vitelliform macular dystrophy (VMD2; OMIM 153700), is an autosomal dominant form of macular degeneration characterized by an abnormal accumulation of lipofuscin within and beneath the retinal pigment epithelium cells. In pursuit of the disease gene, we limited the minimum genetic region by recombination breakpoint analysis and mapped to this region a novel retina-specific gene (VMD2). Genetic mapping data, identification of five independent disease-specific mutations and expression studies provide evidence that mutations within the candidate gene are a cause of BMD. The 3' UTR of the candidate gene contains a region of antisense complementarity to the 3' UTR of the ferritin heavy-chain gene (FTH1), indicating the possibility of antisense interaction between VMD2 and FTH1 transcripts.

Identification and cloning of an orphan G protein-coupled receptor of the glycoprotein hormone receptor subfamily

Mining of the EST database identified a human EST that was predicted to encode a novel member of the glycoprotein hormone receptor subfamily. Based on the sequence information, the full-length coding region of this gene was isolated and sequenced. This gene, designated HG38, is predicted to encode a polypeptide of 907 amino acid residues with a putative signal peptide sequence at its very N-terminus. HG38 is most closely related to members of the glycoprotein hormone receptor subfamily with approximately 35% overall identity at the protein sequence level. As with the glycoprotein hormone receptors, HG38 contains a long extracellular domain with a total of 16 leucine-rich repeats. Northern blot analysis showed that HG38 was expressed in skeletal muscle, placenta, spinal cord, and various regions of the brain. Radiation hybrid mapping placed HG38 into human chromosome 12q22-23. HG38 is most likely to be a receptor for a novel class of glycoprotein ligands.

Uncoupling protein-3: a muscle-specific gene upregulated by leptin in ob/ob mice

We identified and partially characterized another member of the uncoupling protein termed UCP3. Human and mouse UCP3 protein sequences are 86% identical to each other, and 73% and 59% identical to UCP2 and UCP1, respectively. Expression of human UCP3 in yeast resulted in a drastic decrease of mitochondria membrane potential. Northern analysis showed that UCP3 was highly expressed in skeletal muscle in human, rat, and mouse. Mapping of UCP3 placed it to the same chromosomal region of UCP2 in both human and mouse, a region that is linked to obesity and hyperinsulinemia. Furthermore, adenovirus-mediated leptin expression in obese ob/ob mice led to increased expression of UCP3 in skeletal muscle. The data indicate that UCP3 encodes a muscle-specific uncoupling protein that may play an important role in the regulation of energy expenditure and development of obesity.

Leptin gene therapy and daily protein administration: a comparative study in the ob/ob mouse

We have compared the efficacy of daily injection of recombinant leptin protein (rh-leptin) with adenovirus-mediated delivery of the murine or human leptin gene (Ad-leptin) for treatment of obesity in the obese (ob/ob) mouse model. We demonstrate an improved correction profile for obesity and associated surrogate markers using the adenovirus delivery method. Rate of weight loss and percentage satiety were significantly greater in the mice treated with Adleptin. These findings were associated with lower peak serum leptin levels with Ad-leptin (22.9 +/- 2.6 ng/ml for the human gene, and 48.9 +/- 11.5 ng/ml for the murine gene) compared to rh-leptin (385.2 +/- 36.0 ng/ml). (Values are given as mean +/- standard error of the mean.) Importantly rh-leptin and ex vivo-expressed Ad-leptin were equivalently active in a functional cell-based assay. The primary difference in the two therapeutic approaches is the continuous chronic secretion of leptin mediated by gene delivery, versus the intermittent bolus delivery and rapid clearance of the daily injection of rh-leptin protein. Thus, in vivo findings suggest that leptin effects are better achieved at lower steady-state levels, a pharmacological feature attained here by gene therapy. These findings may have implications for the potential use of leptin in the treatment of obesity.

Altered phosphorylation and intracellular distribution of a (CUG)n triplet repeat RNA-binding protein in patients with myotonic dystrophy and in myotonin protein kinase knockout mice

Myotonic dystrophy (DM) is associated with expansion of CTG repeats in the 3'-untranslated region of the myotonin protein kinase (DMPK) gene. The molecular mechanism whereby expansion of the (CUG)n repeats in the 3'-untranslated region of DMPK gene induces DM is unknown. We previously isolated a protein with specific binding to CUG repeat sequences (CUG-BP/hNab50) that possibly plays a role in mRNA processing and/or transport. Here we present evidence that the phosphorylation status and intracellular distribution of the RNA CUG-binding protein, identical to hNab50 protein (CUG-BP/hNab50), are altered in homozygous DM patient and that CUG-BP/hNab50 is a substrate for DMPK both in vivo and in vitro. Data from two biological systems with reduced levels of DMPK, homozygous DM patient and DMPK knockout mice, show that DMPK regulates both phosphorylation and intracellular localization of the CUG-BP/hNab50 protein. Decreased levels of DMPK observed in DM patients and DMPK knockout mice led to the elevation of the hypophosphorylated form of CUG-BP/hNab50. Nuclear concentration of the hypophosphorylated CUG-BP/hNab50 isoform is increased in DMPK knockout mice and in homozygous DM patient. DMPK also interacts with and phosphorylates CUG-BP/hNab50 protein in vitro. DMPK-mediated phosphorylation of CUG-BP/hNab50 results in dramatic reduction of the CUG-BP2, hypophosphorylated isoform, accumulation of which was observed in the nuclei of DMPK knockout mice. These data suggest a feedback mechanism whereby decreased levels of DMPK could alter phosphorylation status of CUG-BP/hNab50, thus facilitating nuclear localization of CUG-BP/hNab50. Our results suggest that DM pathophysiology could be, in part, a result of sequestration of CUG-BP/hNab50 and, in part, of lowered DMPK levels, which, in turn, affect processing and transport of specific subclass of mRNAs.

DMD-specific FISH probes are diagnostically useful in the detection of female carriers of DMD gene deletions

The challenge of Duchenne muscular dystrophy (DMD) carrier identification resides in the ability to identify the presence of a mutant gene over the background contributed by the normal allele. Current diagnosis of carrier status when a deletion has been identified in a proband is based on an analysis of a gene dosage. We present a diagnostic strategy that uses fluorescence in situ hybridization (FISH) to detect female carriers with major deletions in the dystrophin gene. We screened a human X-chromosome-derived genomic library with a full-length dystrophin cDNA and isolated 15 dystrophin-specific cosmids that contain DMD gene exons. Six cosmids were further tested as FISH probes in control individuals and subsequently applied on chromosomes from eight males with DMD and known deletions and on samples from three female carriers. As expected, X chromosomes in normal females displayed four signals, two for the DMD-specific probe and two for the X-chromosome centromeric probe. Hybridization on chromosomal spreads from carriers of deletions revealed only one signal from the DMD-specific probe and two from the control centromeric probe. Males carrying deletions showed no DMD-specific signal for the deleted exons tested. Our data indicate that FISH could represent an alternative method for the detection of female carriers with DMD gene deletions.

Hypermutable myotonic dystrophy CTG repeats in transgenic mice

Myotonic dystrophy (DM) is one of a growing number of inherited human disorders associated with the expansion of triplet repeat DNA sequences. Expanded alleles are highly unstable in both the germline and soma, accounting in large part for the unusual genetics of this disorder, its phenotypic variability and probably, the progressive nature of the symptoms. However, the molecular mechanisms and the genetic factors modulating repeat stability in DM and the other human disorders associated with expanded repeats are not well understood. To provide a model system in which the turnover of triplet repeats could be studied throughout mammalian development, we have generated five transgenic mouse lines incorporating expanded CTG/CAG arrays derived from the human DM locus. Transgene analysis has revealed germline hypermutability, including expansions, deletions and parent-of-origin effects, somatic and early embryonic instability and segregation distortion. Mutational differences between lines and sexes demonstrate that stability, as in humans, is modulated by as yet unidentified cis and trans acting genetic elements.

Molecular detection of new mutations, resolution of ambiguous results and complex genetic counseling issues in Huntington disease

Huntington disease (HD) is an autosomal dominant neurodegenerative disorder caused by expansion of a variable length (CAG)n repeat in the 5' coding region of a novel gene on chromosome 4p16.3. We provide comprehensive molecular analysis of a sporadic case of HD in which a paternally derived normal length allele expanded to an affected length allele. Linkage analysis and paternity testing confirm the paternal origin of the expansion and demonstrate that unequal crossing over during meiosis is an unlikely mechanism for de novo expansion in HD. This case identifies a complex genetic counseling issue for the families of sporadic cases since calculations of recurrence risk are not possible at this time. In addition, we describe utilization of a combination of polymerase chain reaction (PCR) based assays for examination of both the CAG repeat and an adjacent variable length CCG repeat in the huntingtin gene. The combination of these assays can increase the accuracy of molecular diagnosis for HD and may clarify any ambiguous results obtained during molecular testing of HD families.

Trinucleotide repeat disorders in humans: discussions of mechanisms and medical issues

Several human disorders are now known to be caused by expansion of unstable trinucleotide repeat sequences, including fragile X syndrome (FRAX), myotonic dystrophy (DM), spinal and bulbar muscular atrophy (SBMA, also known as Kennedy disease), Huntington disease (HD), dentatorubral-pallidoluysian atrophy (DRPLA), spinocerebellar ataxia type 1 (SCA1), Machado-Joseph disease (MJD), and Friedreich ataxia. As these diseases are studied in more detail, important differences have emerged in the nature of the unstable repeats and the mechanism by which the repeat expansions cause disease symptoms. There are already animal models of some of these disorders, and these are important resources for studying pathology and therapeutic strategies. Diagnostic procedures for these disorders are only beginning to be standardized, and effective therapy will have to wait for further information on disease mechanisms. Much has been learned since discovery of the fragile X syndrome gene in 1991, but much remains to be done.

It's the genes! EST access to human genome content

ESTs or 'expressed sequence tags' are DNA sequences read from both ends of expressed gene fragments. The Merck-WashU EST Project and several other public EST projects are being performed to rapidly discover the complement of human genes, and make them easily accessible. These ESTs are widely used to discover novel members of gene families, to map genes to chromosomes as 'sequence-tagged sites' (STSs), and to identify mutations leading to heritable diseases. Informatic strategies for querying the EST databases are discussed, as well as the strengths and weaknesses of the EST data. There is a compelling need to build on the informatic synthesis of human gene data, and to devise facile methods for determining gene functions.

Identification of a (CUG)n triplet repeat RNA-binding protein and its expression in myotonic dystrophy

Myotonic dystrophy (DM) is an autosomal dominant neuromuscular disease that is associated with a (CTG)n repeat expansion in the 3'-untranslated region of the myotonin protein kinase (Mt-PK) gene. This study reports the isolation and characterization of a (CUG)n triplet repeat pre-mRNA/mRNA binding protein that may play an important role in DM pathogenesis. Two HeLa cell proteins, CUG-BP1 and CUG-BP2, have been purified based upon their ability to bind specifically to (CUG)8 oligonucleotides in vitro. While CUG-BP1 is the major (CUG)8-binding activity in normal cells, nuclear CUG-BP2 binding activity increases in DM cells. Both CUG-BP1 and CUG-BP2 have been identified as isoforms of a novel heterogeneous nuclear ribonucleoprotein (hnRNP), hNab50. The CUG-BP/hNab50 protein is localized predominantly in the nucleus and is associated with polyadenylated RNAs in vivo. In vitro RNA-binding/photocrosslinking studies demonstrate that CUG-BP/hNab50 binds to RNAs containing the Mt-PK 3'-UTR. We propose that the (CUG)n repeat region in Mt-PK mRNA is a binding site for CUG-BP/hNab50 in vivo, and triplet repeat expansion leads to sequestration of this hnRNP on mutant Mt-PK transcripts.

In vivo muscle gene transfer of full-length dystrophin with an adenoviral vector that lacks all viral genes

Duchenne muscular dystrophy (DMD) is an important target for gene transfer because of the disease's high frequency and devastating course. To date, adenoviral vector-mediated gene transfer for DMD has been unavailable because (1) adenoviral vectors were unable to accommodate the full-length dystrophin cDNA (14 kb); and (2) adenoviral vectors induced inflammatory reactions in the gene transfer recipient. We addressed both problems with a novel adenoviral vector that contains no viral genes and encodes 28.2 kb of foreign DNA including both the full-length dystrophin cDNA with the muscle creatine kinase promoter for transcriptional control and a lacZ marker gene. This report presents the in vivo expression of dystrophin and beta-galactosidase from this vector in skeletal muscle of the mdx mouse, a mutant mouse that lacks dystrophin. Somatic delivery of the vector by intramuscular injection in 6-day-old mice resulted in the expression of full-length, recombinant dystrophin at the muscle membrane. Dystrophin-associated proteins were restored in muscle fibers expressing recombinant dystrophin. Mdx muscle injected with our vector showed a decrease in the proportion of fibers with nuclei located centrally; centrally placed nuclei in muscle fibers are characteristic of cycles of degeneration and regeneration suffered by dystrophin-deficient muscle tissue. These results are strong evidence that adenoviral vector-mediated full-length dystrophin delivery provides substantial somatic function.

Molecular dissection of a cosmid from a gene-rich region in 17q21 and characterization of a candidate gene for alpha-N-acetylglucosaminidase with two cDNA isoforms

A cosmid mapped to human Chromosome (Chr) 17q21, c140c10, was found to contain a CpG island. We completed the sequence analysis of c140c10 because of two considerations: the cosmid contained an STS from the 17-beta-hydroxysteroid dehydrogenase gene (17-HSD), which was believed to be a neighbor of the breast cancer susceptibility gene, BRCA1; CpG islands are usually associated downstream and/or upstream of human genes. Computer-based exon trapping of the cosmid sequence revealed putative additional exons. With two of those exons used as a probe to screen human placental cDNA libraries, two cDNA isoforms for a novel gene, designated as ufHSD, were isolated. The amino acid sequence of the open reading frames of the cDNA showed no significant homology to any protein in the data base. However, it is possible that our cDNAs are from the gene for alpha-acetylglucosaminidase, which has recently been localized to the same region. Northern analyses show that the major isoform is expressed in all tissues tested, with the highest expression in blood leukocytes and lowest in brain. Finally, our study has shown that the 46.7-kb cosmid c140c10 encompasses loci for five genes and pseudo-genes: PsiPTP4A, ufHSD, 17-HSDI, 17-HSDII, and 22A1.

Instability of the expanded (CTG)n repeats in the myotonin protein kinase gene in cultured lymphoblastoid cell lines from patients with myotonic dystrophy

The mutation associated with myotonic dystrophy (DM) is the expansion of an unstable trinucleotide repeat, (CTG)n, in the 3'-untranslated region of the myotonin protein kinase gene. Although expanded repeats show both germline and somatic instability, the mechanisms of the instability are poorly understood. To establish a model system in which somatic instability of the DM repeat could be studied in more detail, we established lymphoblastoid cell lines (LBCL) from DM patients. Analysis of the DNA from DM LBCL using Southern blotting showed that the (CTG)n repeats were apparently stable up to 29 passages in culture. To study infrequent repeat size mutations that are undetectable due to the size heterogeneity, we established LBCL of single-cell origins by cloning using multiple steps of limiting dilution. After expansion to approximately 10(6) cells (equivalent to approximately 20 cell cycles), the DNAs of these cell lines were analyzed by the small pool PCR technique using primers flanking the (CTG)n repeat region. Two types of mutations of the expanded (CTG)n repeat alleles were detected: (1) frequent mutations that show small changes of the (CTG)n repeat size, resulting in alleles in a normal distribution around the progenitor allele, and (2) relatively rare mutations with large changes of the (CTG)n repeat size, with a bias toward contraction. The former may represent the mechanism responsible for the somatic heterogeneity of the (CTG)n repeat size observed in blood cells of DM patients. This in vitro experimental system will be useful for further studies on mechanisms involved in the regulation of the somatic stability of the (CTG)n repeats in DM.