Full list of NHGRI presentations at the ASHG meeting.
Abstracts of NHGRI Contributions at the 2011 ASHG Meeting
October 11-15, 2011
Identification of a mosaic activating mutation as the molecular basis of Proteus syndrome using massively parallel sequencing of affected tissues
L. G. Biesecker, et al
Includes from NHGRI: L. G. Biesecker, M. J. Lindhurst, J. C. Sapp, J. K. Teer, J. J. Johnston, K. Peters, J. Turner, P. L. Schwartzberg, J. C. Mullikin
Proteus syndrome manifests mosaic overgrowth in skin, connective tissue, brain, and other tissues. It was hypothesized to be caused by a mosaic mutation, lethal in the non-mosaic state. We tested this hypothesis using massively parallel sequencing of DNA from Proteus syndrome tissues, comparing affected to unaffected tissues. Seventeen samples from 11 patients were sequenced using targeted exome capture. These samples included four affected-normal sample pairs, two affected samples from a fifth patient, one pair of samples from discordant monozygotic twins, and five unaffected parents. We generated ~1.6 Gb of sequence data, with >87% coverage of exons with a high quality genotype. A novel variant was initially identified in a single affected sample using an affected-unaffected filter. Upon manual examination, we found this variant in three additional affecteds but not in the parents of affecteds or the unaffected identical twin, 1000 genomes (n=634), or ClinSeq (n=401). The sequence variant data were confirmed and extended by a custom restriction enzyme assay of >150 samples from 29 patients. Most affected samples were specimens removed from clinically abnormal areas at surgery. Twenty-seven of 29 patients with Proteus syndrome were found to have the identical mutation in this gene, but it was not present in >20 cell lines and tissues from persons with unrelated disorders. Tissues and cell lines from patients with Proteus syndrome harbored admixtures of mutant alleles that varied from 1% to ~50%. Patient-derived cell lines with the mutation showed evidence of activation of this gene product using anti-phosphoprotein antibodies and western blot analyses. We also show that a pair of single cell clones established from the same starting culture and differing only by their mutation status had differential activation of this protein. We conclude that a somatic mosaic activating mutation in this gene causes Proteus syndrome, validating the Happle mosaicism hypothesis. That this mutation is the same in all patients, mosaic, and not in a CpG dinucleotide explains the rarity of this disorder. These data show the power of massively parallel sequencing to identify causative genes in disorders that are not amenable to positional cloning. Further, these results provide a therapeutic target for this severe and progressive disorder. (The author confirms that the gene and mutation will be disclosed at the ICHG meeting, should the abstract be selected).
NBEAL2 is mutated in Gray Platelet Syndrome and is required for biogenesis of platelet alpha-granules.
T. Vilboux, et al
Includes from NHGRI: C. F. Boerkoel, Y. Huang, D. Maynard, H. Dorward, K. Berger, J. C. Mullikin, M. Huizing, W. A. Gahl, M. Gunay- Aygun
Gray Platelet Syndrome (GPS) is a rare autosomal recessive disorder characterized by bleeding tendency, myelofibrosis, thrombocytopenia, and large platelets that lack ±-granules. The causative gene has been sought for decades. We mapped the locus for GPS to a 9.4Mb interval on 3p21.1-22.1 that included 197 protein-coding genes. We sequenced these genes using a combination of next generation and Sanger sequencing in 15 independent GPS families. We identified 15 different mutations in NBEAL2 (neurobeachin-like 2); 5 missense, 4 frameshift, 3 nonsense and 3 splice site mutations. The protein encoded by NBEAL2 has no known function, yet; however, it belongs to a family of proteins that contains 3 domains (BEACH, ARM and WD40) that are crucial for protein-protein interactions, membrane dynamics and vesicle trafficking. Another protein from this family, LYST, a lysosomal trafficking regulator protein, is defective in Chédiak-Higashi syndrome, a disorder associated with platelet dense granule abnormalities in addition to giant secretory granules in leukocytes and other cell types. RNA analysis showed that at least 7 NBEAL2 mRNA transcripts are expressed in hematopoietic cells, including megakaryocytes and platelets. Mass spectrometry of discontinuous sucrose gradient platelet fractions localized NBEAL2 protein to the platelet dense tubular system (endoplasmic reticulum). Microarray data in GPS fibroblasts showed overexpression of fibronectin, essential for proplatelet formation in cultured megakaryocytes and critical for megakaryocyte-matrix interactions. This could explain the myelofibrosis seen in GPS patients and therefore can be explored as a therapeutic target. Understanding NBEAL2 function will likely lead to the discovery of novel pathways of organelle formation and maturation.
Clinical transcriptome sequencing and analysis of a patient with spiny follicular hyperkeratoses of unknown etiology.
K. V. Fuentes Fajardo, et al
Includes from NHGRI: M. Huizing, M. Nehrebecky, F. Gill, C. F. Boerkoel, A. R. Cullinane, W. A. Gahl
Advances in Next Generation Sequencing technology (NGS) have made transcriptome sequencing a practical reality for clinical research applications. Exome sequencing as a tool for finding genetic causes of diseases has become prominent in the literature, and is chosen as the major NGS technology when following up cases in the NIH Undiagnosed Diseases Program (UDP) Protocol. To date, no diagnosis has been arrived at through leads from transcriptome research. The UDP's first attempt at transcriptome sequencing in a clinical setting involved a 50 year-old Caucasian woman with a unique phenotype that included widespread spiny follicular hyperkeratosis resulting in scarring alopecia, follicular plugging and skin abscesses. Her initial milder lesions were exacerbated by a combination of UV-A light treatment and oral retinoids. Thorough dermatologic and medical evaluations to rule out infectious, endocrinologic and paraneoplastic etiologies were unhelpful in delineating an etiology and in clearing the skin problem. Histology and candidate gene sequencing were not diagnostic. Here we present the analysis of transcriptome sequence variations found in RNA from keratinocytes and fibroblasts from involved skin contrasting and comparing them to those found in RNA from fibroblasts and keratinocytes from unaffected skin, blood, and the reference human genome sequence. We identified 18 mutations unique to the keratinocytes and fibroblasts from affected areas, including two genes that indicate a "second-hit," causing additional damage to a gene already harboring a missense mutation. We also observed that deleterious germline mutations were enriched for genes in cell adhesion, cell transport and immune response (p=0.00002, Bonferroni). These data demonstrate the ability to use RNA-Seq to find genes that carry deleterious mutations and demonstrate dysregulation of expression. These findings can then guide follow-up targets for diagnostic and treatment decisions.
Whole exome sequencing identifies a novel mitochondrial enzyme as the gene responsible for Combined Malonic and Methylmalonic Aciduria.
J. L. Sloan, et al
Included from NHGRI: J. J. Johnston, I. Manoli, R. J. Chandler, C. Krause, N. Carrillo-Carrasco, S. D. Chandrasekaran, J. R. Sysol, K. O'Brien, N. S. Hauser, J. C. Sapp, H. M. Dorward, M. Huizing, N. I. H. Intramural Sequencing Center, L. G. Biesecker, C. P. Venditti
Combined methylmalonic and malonic aciduria (CM a.m.MA) is characterized by increased urinary methylmalonic acid (MMA) and malonic acid (MA), with MMA>MA, and normal malonyl-CoA decarboxylase activity. CM a.m.MA was first reported in a child with failure to thrive, seizures and immunodeficiency and a dog with neurodegeneration but the molecular etiology was unknown. We performed exome sequencing of a patient with CM a.m.MA and her parents and filtered variants to identify 12 candidate genes. One of these genes had a putative mitochondrial leader. We sequenced this gene in 9 patients and found two mutations in this gene in 8 of 9 patients. DNA from the dog also had a homozygous mutation in this gene. Mutations included 9 missense, 1 in-frame deletion and 1 nonsense mutation. Eight missense mutations and the in-frame deletion were in functional motifs conserved among members of this enzyme family. One patient with 2 mutations was identified in an exome cohort of subjects not ascertained for metabolic disease and had the distinct biochemical features of CM a.m.MA. The age of diagnosis and symptoms in the nine subjects with CM a.m.MA were highly variable. MMA and MA were elevated in plasma and urine using a new GC-MS assay developed to measure MA. Fibroblasts from 4 subjects had a cellular metabolic defect; increased production of MMA (2.4- to 6-fold) compared to controls, after loading with 5 mM propionate. Viral overexpression of the candidate gene, but not GFP, corrected the metabolic defect. Immunostaining of fibroblasts overexpressing a C-terminal GFP fusion protein or the native enzyme showed a mitochondrial distribution and co-localized with a mitochondrial antibody. These data establish the causative gene for CM a.m.MA and describe the first disease association with a member of this enzyme family. Mutant alleles occur with a minor allele frequency (MAF) of 0.0058 in ~1,000 control individuals predicting a CM a.m.MA population incidence of ~1:30,000. This predicts that CM a.m.MA is one of the most common forms of MMAemia, and perhaps, one of the more common inborn errors of metabolism. The spectrum of symptoms and natural history of this disorder are highly variable and require further delineation. The identification of an affected using exome sequencing highlights an interesting and alternative diagnostic approach because CM a.m.MA is not identified through routine newborn screening by elevated propionylcarnitine.
Ethical issues in international genomics research collaborations: Perspectives from Sub-Saharan Africa.
C. Rotimi NHGRI/NIH
Large-scale genomics research projects — especially those that involve plans for the broad release of samples and data — raise many ethical issues. These issues are heightened in the context of international collaborations (for example, the International HapMap Project, the 1000 Genomes Project, and The Cancer Genome Atlas, and the International Human Microbiome Consortium), where different sets of cultural values and norms, and different legal and regulatory requirements, may have to be reconciled. For example, approaches to recruitment, informed consent, and individual autonomy and privacy vary considerably in different parts of the world. Attitudes about the role (if any) for community consultation or engagement and the appropriateness of particular research governance mechanisms can also differ. In addition, people in different parts of the world (and often even in the same part of the world) may have very different views about whether, or how, research results or incidental findings should be returned to participants in genomics research studies (where the samples and data have not been anonymized). The scope of the right to withdraw samples or data from repositories or databases is also an area of potential difference-especially where full realization of a project's end goals depends on having the samples and data available through a central repository to multiple researchers over an extended period of time. This session will explore these and other ethical issues that arise in large-scale international genomics research collaborations. Perspectives will be presented from several continents and countries.
A liver-specific transgenic mouse model identifies new disease-associated biomarkers and establishes antioxidants as an ameliorative treatment for the renal disease of methylmalonic acidemia (MMA).
E. Manoli, et al
Includes from NHGRI: J. R. Sysol, J. L. Sloan, R. J. Chandler, C. P. Venditti
We generated mice that express methylmalonyl-CoA mutase (Mut) cDNA under the control of a liver-specific promoter on a knockout background (Mut-/-;TgINS-Alb-Mut) to model extrahepatic manifestations, study pathophysiology and examine therapeutic interventions. Low-level hepatic Mut expression conferred complete rescue from the neonatal lethality displayed by Mut-/- mice and allowed disease-associated renal pathology to be induced with a high-protein diet. Ingestion of a high-protein chow for 2 months resulted in elevated plasma methylmalonic acid levels (¼M) in the Mut-/-;TgINS-Alb-Mut mice (1500 620) compared to similarly-treated Mut+/- littermates (7.4 ±0.6), growth failure and increased mortality. Mut-/-;TgINS-Alb-Mut mice developed tubulointerstitial nephritis associated with a decreased glomerular filtration rate (GFR) [37.6 3.9% of Mut+/- GFR, p<0.0001] and elevated creatinine levels. Mitochondria of proximal tubular epithelial cells were enlarged and had shortened cristae; kidney immunohistochemistry showed increased succinate dehydrogenase and decreased cytochrome c oxidase staining. Expression analysis using whole kidney RNA from the protein-challenged Mut-/-;TgINS-Alb-Mut mice compared to age, sex and diet matched littermates revealed differentially expressed mRNAs from several pathways including immune response, lipid metabolism, ketone biosynthesis and cell survival. One significantly up-regulated gene encoding a secreted glycoprotein was also increased in the plasma and urine of Mut-/-;TgINS-Alb-Mut mice and the concentration correlated inversely with GFR (r=-0.45; p<0.01). This protein was further studied in the plasma of 46 mut MMA patients (NCT00078078); levels were strongly associated with renal indices (patient plasma creatinine and cystatin-C values; p<0.01 for both). The inclusion of ubiquinol and vitamin E in the high protein diet for 2 months ameliorated the loss of GFR in the Mut-/-;TgINS-Alb-Mut mice (37.6 ±3.9% [pre-] vs. 60 ±4.8% of Mut+/- GFR [post-treatment], p<0.01) and normalized urinary and plasma disease-associated biomarkers despite persisting metabolite elevations induced by the dietary challenge. This novel mouse model has allowed identification of new biomarkers using genomic approaches and provided evidence for a therapeutic effect of antioxidants on the renal disease of MMA. These results should have broad extension to other metabolic disorders manifesting mitochondrial dysfunction.
Indexing and deep sequencing of a point mutation in mosaic samples from Proteus syndrome: Alternative detection strategies.
M. J. Lindhurst, et al
Includes from NHGRI: J. K. Teer, J. J. Johnston, E. M. Finn, J. C. Sapp, J. C. Mullikin, L. G. Biesecker
Detection of mutations in mosaic samples is challenging due to the variability in the level of the mutant allele that can occur in various tissues. Recently, we have shown that Proteus syndrome is caused by a somatic mutation that results in constitutive activation of a key signaling pathway in affected tissues. The variant was initially identified using next gen sequencing of 17 samples. To confirm and extend these findings, we screened 158 samples by Sanger sequencing. Seventy-one of these samples had clear evidence of the mutant allele on the electropherograms, 17 had electropherograms that suggested a low level of the mutation, and 70 were apparently negative. However, it was challenging to distinguish low levels (<1%) of mosaicism from background signals or noise in the electropherogram traces. We next developed a PCR/restriction endonuclease assay (MboII) that could be separated on the ABI3130 and was designed such that MboII digested only the mutant amplicon. Areas under the curve for the mutant and wild-type peaks on the instrument were assessed quantitatively. A validation experiment using dilutions of cloned mutant and wild-type DNA showed a correlation of r2=0.9993 with a lower limit of 1% mutant allele sensitivity. Using this assay, the mutant allele was detected in all 71 of the positive samples, 15 of the above 17 samples that were suggestive and 15 samples of the 70 samples that were negative by Sanger. This increased the number of patients with at least one positive sample from 23 to 27. The level of mutation in these samples ranged from 1% to ~50%. We screened 75 samples from non-Proteus individuals; all were negative for the mutation. While the PCR/MboII assay increased sensitivity compared to sequencing, variation increased at low mutant allele levels (again, <1%). To address this, we PCR-amplified the mutation from 12 samples using indexing primers and used next-gen sequencing to generate ultra-deep coverage. These samples were pooled, and sequenced on a single GAIIx lane (76 bp paired-end). 28,568,780 total reads were generated, and 335,378-1,649,523 reads were assigned to each sample. Allele frequencies were calculated at the variant position, and agreed well with the MboII results. Additional optimizations will be performed and have the potential to further improve sensitivity. (The authors confirm that the gene and mutation will be disclosed at the ICHG meeting should the abstract be selected for presentation).
Tryptic Peptide Analysis of WBC to Diagnose Genetic disorders: Application to Primary Immunodeficiency Disorders and Nephropathic Cystinosis
S. Hahn, et al
Includes from NHGRI T. Vilboux, W. Gahl
PURPOSE: There are many genetic diseases that would benefit from early diagnosis but that have no population screening platform. Examples include primary immunodeficiency disorders such as X-linked agammaglobulinemia (XLA), Wiskott Aldrich Syndrome (WAS), and Severe Combined Immune Deficiency (SCID), as well as nephropathic cystinosis. BTK, WASP, CD3 µ and cystinosin are low abundance proteins deficient in these conditions. They are localized in cell membranes, cytoplasm, or on the cell surface and are not detectable in plasma. We propose that these proteins can be observed in proteolytically digested extracts of WBCs and will be absent or significantly reduced in affected cells, making diagnosis possible. METHODS: Candidate peptides were screened by in silico trypsin digestion modeling followed by a BLAST search to insure that the sequences are unique within the human genome. The final signature" peptides were selected by evaluating the MRM chromatogram for the isotopically labeled peptide and the WBC digest peptide in control (n=20) and patient samples. Five lymphocyte cell lines were used to establish the absence of signature peptides for primary immunodeficiencies. Three PBMC samples from XLA were blindly analyzed. WBCs from 10 cystinosis patients with known genotypes were also tested. The amount of each peptide in the WBC was determined by taking the ratio of the peak area for the signature peptide to that of the labeled peptide and reported as normalized to actin. RESULTS: Three blinded samples lacked only BTK peptides, confirming that these samples were from XLA patients. All five cell lines clearly showed the biochemical phenotype of each cell line. Cystinosin was non-detectable in 9 cystinosis patients. One patient showed a cystinosin peak but at a very low concentration compared to control. CONCLUSION: Our method quantified the proteolytic peptides for the target proteins, BTK, WASP, CD3 µ and cystinosin in various cell lines and patient samples. Targeted proteins from these conditions were either absent or significantly diminished. This approach can be potentially utilized as part of a multiplex analysis for many genetic conditions in which the protein of interest is significantly reduced. While the transition from white blood cells to dried blood spots will be challenging, we believe that with further enrichment and optimization, this method can be potentially applied to newborn screening.
For a full list of NHGRI presentations at ASHG, including abstracts of research, please go to Participation from the National Human Genome Research Institute (NHGRI)
Last Reviewed: May 24, 2012