Assessing the significance of conserved genomic aberrations using high resolution genomic microarrays

Genomic aberrations recurrent in a particular cancer type can be important prognostic markers for tumor progression. Typically in early tumorigenesis, cells incur a breakdown of the DNA replication machinery that results in an accumulation of genomic aberrations in the form of duplications, deletions, translocations, and other genomic alterations. Microarray methods allow for finer mapping of these aberrations than has previously been possible; however, data processing and analysis methods have not taken full advantage of this higher resolution. Attention has primarily been given to analysis on the single sample level, where multiple adjacent probes are necessarily used as replicates for the local region containing their target sequences. However, regions of concordant aberration can be short enough to be detected by only one, or very few, array elements. We describe a method called Multiple Sample Analysis for assessing the significance of concordant genomic aberrations across multiple experiments that does not require a-priori definition of aberration calls for each sample. If there are multiple samples, representing a class, then by exploiting the replication across samples our method can detect concordant aberrations at much higher resolution than can be derived from current single sample approaches. Additionally, this method provides a meaningful approach to addressing population-based questions such as determining important regions for a cancer subtype of interest or determining regions of copy number variation in a population. Multiple Sample Analysis also provides single sample aberration calls in the locations of significant concordance, producing high resolution calls per sample, in concordant regions. The approach is demonstrated on a dataset representing a challenging but important resource: breast tumors that have been formalin-fixed, paraffin-embedded, archived, and subsequently UV-laser capture microdissected and hybridized to two-channel BAC arrays using an amplification protocol. We demonstrate the accurate detection on simulated data, and on real datasets involving known regions of aberration within subtypes of breast cancer at a resolution consistent with that of the array. Similarly, we apply our method to previously published datasets, including a 250K SNP array, and verify known results as well as detect novel regions of concordant aberration. The algorithm has been fully implemented and tested and is freely available as a Java application at http://www.cbil.upenn.edu/MSA.     

Alignment and merging of electron microscope images of frozen hydrated crystals of the T4 DNA helix destabilizing protein gp32*I.

Low dose cryoelectron microscopy has been used to record images and electron diffraction patterns of frozen hydrated crystals of the single-stranded DNA binding protein gp32*I. Fourier transforms from 13 image areas, corresponding to approximately 40,000 unit cells, were aligned by a minimal phase residual search and merged by vector addition in reciprocal space. Phases from the resulting composite transform were combined with amplitudes from electron diffraction patterns to reconstruct the projected mass density of the gp32*I crystal at 8.4 A resolution.     

Structural conservation of the pores of calcium-activated and voltage-gated potassium channels determined by a sea anemone toxin.

The structurally defined sea anemone peptide toxins ShK and BgK potently block the intermediate conductance, Ca(2+)-activated potassium channel IKCa1, a well recognized therapeutic target present in erythrocytes, human T-lymphocytes, and the colon. The well characterized voltage-gated Kv1.3 channel in human T-lymphocytes is also blocked by both peptides, although ShK has a approximately 1,000-fold greater affinity for Kv1.3 than IKCa1. To gain insight into the architecture of the toxin receptor in IKCa1, we used alanine-scanning in combination with mutant cycle analyses to map the ShK-IKCa1 interface, and compared it with the ShK-Kv1.3 interaction surface. ShK uses the same five core residues, all clustered around the critical Lys(22), to interact with IKCa1 and Kv1.3, although it relies on a larger number of contacts to stabilize its weaker interactions with IKCa1 than with Kv1.3. The toxin binds to IKCa1 in a region corresponding to the external vestibule of Kv1.3, and the turret and outer pore of the structurally defined bacterial potassium channel, KcsA. Based on the NMR structure of ShK, we deduce the toxin receptor in IKCa1 to have x-y dimensions of approximately 22 A, a diameter of approximately 31 A, and a depth of approximately 8 A; we estimate that the ion selectivity lies approximately 13 A below the outer lip of the toxin receptor. These dimensions are in good agreement with those of the KcsA channel determined from its crystal structure, and the inferred structure of Kv1.3 based on mapping with scorpion toxins. Thus, these distantly related channels exhibit architectural similarities in the outer pore region. This information could facilitate development of specific and potent modulators of the therapeutically important IKCa1 channel.     

Myeloperoxidase-catalyzed redox-cycling of phenol promotes lipid peroxidation and thiol oxidation in HL-60 cells

Various types of cancer occur in peroxidase-rich target tissues of animals exposed to aryl alcohols and amines. Unlike biotransformation by cytochrome P450 enzymes, peroxidases activate most substrates by one-electron oxidation via radical intermediates. This work analyzed the peroxidase-dependent formation of phenoxyl radicals in HL-60 cells and its contribution to cytotoxicity and genotoxicity. The results showed that myeloperoxidase-catalyzed redox cycling of phenol in HL-60 cells led to intracellular formation of glutathionyl radicals detected as GS-DMPO nitrone. Formation of thiyl radicals was accompanied by rapid oxidation of glutathione and protein-thiols. Analysis of protein sulfhydryls by SDS-PAGE revealed a significant oxidation of protein SH-groups in HL-60 cells incubated in the presence of phenol/H2O2 that was inhibited by cyanide and azide. Additionally, cyanide- and azide-sensitive generation of EPR-detectable ascorbate radicals was observed during incubation of HL-60 cell homogenates in the presence of ascorbate and H2O2. Oxidation of thiols required addition of H2O2 and was inhibited by pretreatment of cells with the inhibitor of heme synthesis, succinylacetone. Radical-driven oxidation of thiols was accompanied by a trend toward increased content of 8-oxo-7,8-dihydro-2'-deoxyguanosine in the DNA of HL-60 cells. Membrane phospholipids were also sensitive to radical-driven oxidation as evidenced by a sensitive fluorescence HPLC-assay based on metabolic labeling of phospholipids with oxidation-sensitive cis-parinaric acid. Phenol enhanced H2O2-dependent oxidation of all classes of phospholipids including cardiolipin, but did not oxidize parinaric acid-labeled lipids without addition of H2O2. Induction of a significant hypodiploid cell population, an indication of apoptosis, was detected after exposure to H2O2 and was slightly but consistently and significantly higher after exposure to H2O2/phenol. The clonogenicity of HL-60 cells decreased to the same extent after exposure to H2O2 or H2O2/phenol. Treatment of HL-60 cells with either H2O2 or H2O2/phenol at concentrations adequate for lipid peroxidation did not cause a detectable increase in chromosomal breaks. Detection of thiyl radicals as well as rapid oxidation of thiols and phospholipids in viable HL-60 cells provide strong evidence for redox cycling of phenol in this bone marrow-derived cell line.     

Optimum activity of the phosphofructokinase from Ascaris suum requires more than one metal ion

Phosphofructokinase (PFK) catalyzes the phosphorylation of fructose 6-phosphate (F6P) to give fructose 1,6-bisphosphate (FBP) using MgATP as the phosphoryl donor. As the concentration of Mg(2+) increases above the concentration needed to generate the MgATP chelate complex, a 15-fold increase in the initial rate was observed at low MgATP. The effect of Mg(2+) is limited to V/K(MgATP), and initial rate studies indicate an equilibrium-ordered addition of Mg(2+) before MgATP. Isotope partitioning of the dPFK:MgATP complex indicates a random addition of MgATP and F6P at low Mg(2+), with the rate of release of MgATP from the central E:MgATP:F6P complex 4-fold faster than the net rate constant for catalysis. This can be contrasted with the ordered addition of MgATP prior to F6P at high Mg(2+). The addition of fructose 2,6-bisphosphate (F26P(2)) has no effect on the mechanism at low Mg(2+), with the exception of a 4-fold increase in the affinity of the enzyme for F6P. At high Mg(2+), F26P(2) causes the kinetic mechanism to become random with respect to MgATP and F6P and with MgATP released from the central complex half as fast as the net rate constant for catalysis. The latter is in agreement with previous studies [Gibson, G. E., Harris, B. G., and Cook, P. F. (1996) Biochemistry 35, 5451-5457]. The overall effect of Mg(2+) is a decrease in the rate of release of MgATP from the E:MgATP:F6P complex, independent of the concentration of F26P(2).     

Excess of de novo deleterious mutations in genes associated with glutamatergic systems in nonsyndromic intellectual disability

Little is known about the genetics of nonsyndromic intellectual disability (NSID). We hypothesized that de novo mutations (DNMs) in synaptic genes explain an important fraction of sporadic NSID cases. In order to investigate this possibility, we sequenced 197 genes encoding glutamate receptors and a large subset of their known interacting proteins in 95 sporadic cases of NSID. We found 11 DNMs, including ten potentially deleterious mutations (three nonsense, two splicing, one frameshift, four missense) and one neutral mutation (silent) in eight different genes. Calculation of point-substitution DNM rates per functional and neutral site showed significant excess of functional DNMs compared to neutral ones. De novo truncating and/or splicing mutations in SYNGAP1, STXBP1, and SHANK3 were found in six patients and are likely to be pathogenic. De novo missense mutations were found in KIF1A, GRIN1, CACNG2, and EPB41L1. Functional studies showed that all these missense mutations affect protein function in cell culture systems, suggesting that they may be pathogenic. Sequencing these four genes in 50 additional sporadic cases of NSID identified a second DNM in GRIN1 (c.1679_1681dup/p.Ser560dup). This mutation also affects protein function, consistent with structural predictions. None of these mutations or any other DNMs were identified in these genes in 285 healthy controls. This study highlights the importance of the glutamate receptor complexes in NSID and further supports the role of DNMs in this disorder.     

Consistent probabilistic outputs for protein function prediction

In predicting hierarchical protein function annotations, such as terms in the Gene Ontology (GO), the simplest approach makes predictions for each term independently. However, this approach has the unfortunate consequence that the predictor may assign to a single protein a set of terms that are inconsistent with one another; for example, the predictor may assign a specific GO term to a given protein ('purine nucleotide binding') but not assign the parent term ('nucleotide binding'). Such predictions are difficult to interpret. In this work, we focus on methods for calibrating and combining independent predictions to obtain a set of probabilistic predictions that are consistent with the topology of the ontology. We call this procedure 'reconciliation'. We begin with a baseline method for predicting GO terms from a collection of data types using an ensemble of discriminative classifiers. We apply the method to a previously described benchmark data set, and we demonstrate that the resulting predictions are frequently inconsistent with the topology of the GO. We then consider 11 distinct reconciliation methods: three heuristic methods; four variants of a Bayesian network; an extension of logistic regression to the structured case; and three novel projection methods - isotonic regression and two variants of a Kullback-Leibler projection method. We evaluate each method in three different modes - per term, per protein and joint - corresponding to three types of prediction tasks. Although the principal goal of reconciliation is interpretability, it is important to assess whether interpretability comes at a cost in terms of precision and recall. Indeed, we find that many apparently reasonable reconciliation methods yield reconciled probabilities with significantly lower precision than the original, unreconciled estimates. On the other hand, we find that isotonic regression usually performs better than the underlying, unreconciled method, and almost never performs worse; isotonic regression appears to be able to use the constraints from the GO network to its advantage. An exception to this rule is the high precision regime for joint evaluation, where Kullback-Leibler projection yields the best performance.