Management of High-Throughput DNA Sequencing Projects: Alpheus

High-throughput DNA sequencing has enabled systems biology to begin to address areas in health, agricultural and basic biological research. Concomitant with the opportunities is an absolute necessity to manage significant volumes of high-dimensional and inter-related data and analysis. Alpheus is an analysis pipeline, database and visualization software for use with massively parallel DNA sequencing technologies that feature multi-gigabase throughput characterized by relatively short reads, such as Illumina-Solexa (sequencing-by-synthesis), Roche-454 (pyrosequencing) and Applied Biosystem's SOLiD (sequencing-by-ligation). Alpheus enables alignment to reference sequence(s), detection of variants and enumeration of sequence abundance, including expression levels in transcriptome sequence. Alpheus is able to detect several types of variants, including non-synonymous and synonymous single nucleotide polymorphisms (SNPs), insertions/deletions (indels), premature stop codons, and splice isoforms. Variant detection is aided by the ability to filter variant calls based on consistency, expected allele frequency, sequence quality, coverage, and variant type in order to minimize false positives while maximizing the identification of true positives. Alpheus also enables comparisons of genes with variants between cases and controls or bulk segregant pools. Sequence-based differential expression comparisons can be developed, with data export to SAS JMP Genomics for statistical analysis.

Protein tyrosine phosphatases: strategies for distinguishing proteins in a family containing multiple drug targets and anti-targets

PTP1B, but also proteins that are essential to cell development and survival. The availability of sequences and representative structures for the PTP family allows better identification of anti-targets, closely related family members likely to cross-react with directed inhibitors. Eight PTP subfamilies, classified by active site information and overall PTP catalytic domain structure similarity, are reviewed here: 1) the tyrosine-specific PTPs, 2) the dual-specificity PTPs, 3) the cdc25 subclass; 4) the Pten subclass; 5) the myotubularins, 6) the PRL subclass, 7) the low molecular weight PTPs, and 8) the newly defined cdc14 subclass. PTP subfamily classification and structure information can be incorporated into design strategies aimed at identifying potent and selective small molecule inhibitors. The accumulating inhibition data for compounds screened against panels of PTPs is reviewed. The in vitro data can yield clues to specificity so that individual subfamilies can be matched with effective scaffolds to jumpstart lead design and reduce false starts.

Synergistic Computational and Experimental Proteomics Approaches for More Accurate Detection of Active Serine Hydrolases in Yeast

An analysis of the structurally and catalytically diverse serine hydrolase protein family in the Saccharomyces cerevisiae proteome was undertaken using two independent but complementary, large-scale approaches. The first approach is based on computational analysis of serine hydrolase active site structures; the second utilizes the chemical reactivity of the serine hydrolase active site in complex mixtures. These proteomics approaches share the ability to fractionate the complex proteome into functional subsets. Each method identified a significant number of sequences, but 15 proteins were identified by both methods. Eight of these were unannotated in the Saccharomyces Genome Database at the time of this study and are thus novel serine hydrolase identifications. Three of the previously uncharacterized proteins are members of a eukaryotic serine hydrolase family, designated as Fsh (family of serine hydrolase), identified here for the first time. OVCA2, a potential human tumor suppressor, and DYR-SCHPO, a dihydrofolate reductase from Schizosaccharomyces pombe, are members of this family. Comparing the combined results to results of other proteomic methods showed that only four of the 15 proteins were identified in a recent large-scale, "shotgun" proteomic analysis and eight were identified using a related, but similar, approach (neither identifies function). Only 10 of the 15 were annotated using alternate motif-based computational tools. The results demonstrate the precision derived from combining complementary, function-based approaches to extract biological information from complex proteomes. The chemical proteomics technology indicates that a functional protein is being expressed in the cell, while the computational proteomics technology adds details about the specific type of function and residue that is likely being labeled. The combination of synergistic methods facilitates analysis, enriches true positive results, and increases confidence in novel identifications. This work also highlights the risks inherent in annotation transfer and the use of scoring functions for determination of correct annotations.

Structure-based active site profiles for genome analysis and functional family subclassification

In previous work, structure-based functional site descriptors, fuzzy functional forms (FFFs), were developed to recognize structurally conserved active sites in proteins. These descriptors identify members of protein families according to active-site structural similarity, rather than overall sequence or structure similarity. FFFs are defined by a minimal number of highly conserved residues and their three-dimensional arrangement. This approach is advantageous for function assignment across broad families, but is limited when applied to detailed subclassification within these families. In the work described here, we developed a method of three-dimensional, or structure-based, active-site profiling that utilizes FFFs to identify residues located in the spatial environment around the active site. Three-dimensional active-site profiling reveals similarities and differences among active sites across protein families. Using this approach, active-site profiles were constructed from known structures for 193 functional families, and these profiles were verified as distinct and characteristic. To achieve this result, a scoring function was developed that discriminates between true functional sites and those that are geometrically most similar, but do not perform the same function. In a large-scale retrospective analysis of human genome sequences, this profile score was shown to identify specific functional families correctly. The method is effective at recognizing the likely subtype of structurally uncharacterized members of the diverse family of protein kinases, categorizing sequences correctly that were misclassified by global sequence alignment methods. Subfamily information provided by this three-dimensional active-site profiling method yields key information for specific and selective inhibitor design for use in the pharmaceutical industry.

Prediction of deleterious functional effects of amino acid mutations using a library of structure-based function descriptors

An automated, active site-focused, computational method is described for use in predicting the effects of engineered amino acid mutations on enzyme catalytic activity. The method uses structure-based function descriptors (Fuzzy Functional Forms trade mark or FFFs trade mark ) to automatically identify enzyme functional sites in proteins. Three-dimensional sequence profiles are created from the surrounding active site structure. The computationally derived active site profile is used to analyze the effect of each amino acid change by defining three key features: proximity of the change to the active site, degree of amino acid conservation at the position in related proteins, and compatibility of the change with residues observed at that position in similar proteins. The features were analyzed using a data set of individual amino acid mutations occurring at 128 residue positions in 14 different enzymes. The results show that changes at key active site residues and at highly conserved positions are likely to have deleterious effects on the catalytic activity, and that non-conservative mutations at highly conserved residues are even more likely to be deleterious. Interestingly, the study revealed that amino acid substitutions at residues in close contact with the key active site residues are not more likely to have deleterious effects than mutations more distant from the active site. Utilization of the FFF-derived structural information yields a prediction method that is accurate in 79-83% of the test cases. The success of this method across all six EC classes suggests that it can be used generally to predict the effects of mutations and nsSNPs for enzymes. Future applications of the approach include automated, large-scale identification of deleterious nsSNPs in clinical populations and in large sets of disease-associated nsSNPs, and identification of deleterious nsSNPs in drug targets and drug metabolizing enzymes.

Function first: a powerful approach to post-genomic drug discovery

In the post-genomic era, pharmaceutical researchers must evaluate vast numbers of protein sequences and formulate novel, intelligent strategies for identifying valid targets and discovering leads against them. The identification of small molecules that selectively target proteins or protein families will be aided by knowing the function and/or the structure of the target(s). By identifying protein function first, efficiencies are gained that allow subsequent focus of resources on particular protein families of interest. This article reviews current proteomic-scale approaches to identifying function as a way of accelerating lead discovery.

Engineered improvements in DNA-binding function of the MATa1 homeodomain reveal structural changes involved in combinatorial control

We have engineered enhanced DNA-binding function into the a1 homeodomain by making changes in a loop distant from the DNA-binding surface. Comparison of the free and bound a1 structures suggested a mechanism linking van der Waals stacking changes in this loop to the ordering of a final turn in the DNA-binding helix of a1. Inspection of the protein sequence revealed striking differences in amino acid identity at positions 24 and 25 compared to related homeodomain proteins. These positions lie in the loop connecting helix-1 and helix-2, which is involved in heterodimerization with the alpha 2 protein. A series of single and double amino acid substitutions (a1-Q24R, a1-S25Y, a1-S25F and a1-Q24R/S25Y) were engineered, expressed and purified for biochemical and biophysical study. Calorimetric measurements and HSQC NMR spectra confirm that the engineered variants are folded and are equally or more stable than the wild-type a1 homeodomain. NMR analysis of a1-Q24R/S25Y demonstrates that the DNA recognition helix (helix-3) is extended by at least one turn as a result of the changes in the loop connecting helix-1 and helix-2. As shown by EMSA, the engineered variants bind DNA with enhanced affinity (16-fold) in the absence of the alpha 2 cofactor and the variant alpha 2/a1 heterodimers bind cognate DNA with specificity and affinity reflective of the enhanced a1 binding affinity. Importantly, in vivo assays demonstrate that the a1-Q24R/S25Y protein binds with fivefold greater affinity than wild-type a1 and is able to partially suppress defects in repression by alpha 2 mutants. As a result of these studies, we show how subtle differences in residues at a surface distant from the functional site code for a conformational switch that allows the a1 homeodomain to become active in DNA binding in association with its cofactor alpha 2.

Sequence- and structure-based protein function prediction from genomic information

Existing functional annotation transfer is fraught with inaccuracies that may hinder forward interpretation and mining of genomic data. Hand-curation of the annotation placed into databases is not practical. In lieu of experimental evidence, computational biological approaches offer high-throughput tools to predict function accurately; however, these methods are still notably deficient in defining and describing the complexity of protein function. Enriching genomic sequences obtained from sequencing efforts and expression array methods with protein function information and classification will be an efficient first step for incorporating genomic data into drug discovery programs.

Cooperative ordering in homeodomain-DNA recognition: solution structure and dynamics of the MATa1 homeodomain

The mating type homeodomain proteins, MATa1 and MATalpha2, combine to form a heterodimer to bind DNA in diploid yeast cells. The a1-alpha2 heterodimer tightly and specifically binds haploid-specific gene operators to repress transcription. On its own, however, the a1 homeodomain does not bind DNA in a sequence-specific manner. To help understand this interaction, we describe the solution structure and backbone dynamics of the free a1 homeodomain. Free a1 in solution is an ensemble of structures having flexible hinges at the two turns in the small protein fold. Conformational changes in the a1 homeodomain upon ternary complex formation are located in the loop between helix 1 and helix 2, where the C-terminal tail of alpha2 binds to form the heterodimer, and at the C-terminus of helix 3, the DNA recognition helix. The observed differences, comparing the free and bound a1 structures, suggest a mechanism linking van der Waals stacking changes to the ordering of a final turn in the DNA-binding helix of a1. The tail of alpha2 induces changes in loop 1 of a1 that push it toward a properly folded DNA binding conformation.

Identification of the Archaeoglobus fulgidus endonuclease III DNA interaction surface using heteronuclear NMR methods

BACKGROUND

Endonuclease III is the prototype for a family of DNA-repair enzymes that recognize and remove damaged and mismatched bases from DNA via cleavage of the N-glycosidic bond. Crystal structures for endonuclease III, which removes damaged pyrimidines, and MutY, which removes mismatched adenines, show a highly conserved structure. Although there are several models for DNA binding by this family of enzymes, no experimental structures with bound DNA exist for any member of the family.

RESULTS

Nuclear magnetic resonance (NMR) spectroscopy chemical-shift perturbation of backbone nuclei (1H, 15N, 13CO) has been used to map the DNA-binding site on Archaeoglobus fulgidus endonuclease III. The experimentally determined interaction surface includes five structural elements: the helix-hairpin-helix (HhH) motif, the iron-sulfur cluster loop (FCL) motif, the pseudo helix-hairpin-helix motif, the helix B-helix C loop, and helix H. The elements form a continuous surface that spans the active site of the enzyme.

CONCLUSIONS

The enzyme-DNA interaction surface for endonuclease III contains five elements of the protein structure and suggests that DNA damage recognition may require several specific interactions between the enzyme and the DNA substrate. Because the target DNA used in this study contained a generic apurinic/apyrimidinic (AP) site, the binding interactions we observed for A. fulgidus endonuclease III should apply to all members of the endonuclease III family and several interactions could apply to the endonuclease III/AlkA (3-methyladenine DNA glycosylase) superfamily.

Configuration of the catalytic GIY-YIG domain of intron endonuclease I-TevI: coincidence of computational and molecular findings

I-TevI is a member of the GIY-YIG family of homing endonucleases. It is folded into two structural and functional domains, an N-terminal catalytic domain and a C-terminal DNA-binding domain, separated by a flexible linker. In this study we have used genetic analyses, computational sequence analysis andNMR spectroscopy to define the configuration of theN-terminal domain and its relationship to the flexible linker. The catalytic domain is an alpha/beta structure contained within the first 92 amino acids of the 245-amino acid protein followed by an unstructured linker. Remarkably, this structured domain corresponds precisely to the GIY-YIG module defined by sequence comparisons of 57 proteins including more than 30 newly reported members of the family. Although much of the unstructured linker is not essential for activity, residues 93-116 are required, raising the possibility that this region may adopt an alternate conformation upon DNA binding. Two invariant residues of the GIY-YIG module, Arg27 and Glu75, located in alpha-helices, have properties of catalytic residues. Furthermore, the GIY-YIG sequence elements for which the module is named form part of a three-stranded antiparallel beta-sheet that is important for I-TevI structure and function.

Hydrogen exchange behavior of [U-15N]-labeled oxidized and reduced iso-1-cytochrome c

Heteronuclear NMR spectroscopy was used to measure the hydrogen-deuterium exchange rates of backbone amide hydrogens in both oxidized and reduced [U-15N]iso-1-cytochrome c from the yeast Saccharomyces cerevisiae. The exchange data confirm previously reported data [Marmorino et al. (1993) Protein Sci. 2, 1966-1974], resolve several inconsistencies, and provide more thorough coverage of exchange rates throughout the cytochrome c protein in both oxidation states. Combining the data previously collected on unlabeled C102T with the current data collected on [U-15N]C102T, exchange rates for 53 protons in the oxidized state and 52 protons in the reduced state can now be reported. Most significantly, hydrogen exchange measurements on [U-15N]iso-1-cytochrome c allowed the observation of exchange behavior of the secondary structures, such as large loops, that are not extensively hydrogen-bonded. For the helices, the most slowly exchanging protons are found in the middle of the helix, with more rapidly exchanging protons at the helix ends. The observation for the Omega-loops in cytochrome c is just the opposite. In the loops, the ends contain the most slowly exchanging protons and the loop middles allow more rapid exchange. This is found to be true in cytochrome c loops, even though the loop ends are not attached to any regular secondary structures. Some of the exchange data are strikingly inconsistent with data collected on the C102S variant at a different pH, which suggests pH-dependent dynamic differences in the protein structure. This new hydrogen exchange data for loop residues could have implications for the substructure model of eukaryotic cytochrome c folding. Isotopic labeling of variant forms of cytochrome c can now be used to answer many questions about the structure and folding of this model protein.

Assignment of 15N chemical shifts and 15N relaxation measurements for oxidized and reduced iso-1-cytochrome c

A protocol for complete isotopic labeling of iso-1-cytochrome c from the eukaryote Saccharomyces cerevisiae is reported. Assignments are reported for the vast majority of the 15N amide resonances in both oxidized and reduced states. 15N heteronuclear relaxation experiments were collected to study the picosecond-nanosecond backbone dynamics of this protein. Relaxation rates were computed and fit to spectral density functions by a model-free analysis. Backbone amides in the overlapping loop B/C region are the most flexible on the picosecond-nanosecond time scale in both forms of the protein. The results show that, on average, the protein backbone is slightly more dynamic in the oxidized than the reduced state, though not significantly so. Exchange terms, which suggest significant motion on a time scale at least an order of magnitude slower than the overall correlation time of 5.2 ns, were required for only two residues in the reduced state and 27 residues in the oxidized state. When analyzed on a per-residue basis, the lower order parameters found in the oxidized state were scattered throughout the protein, with a few continuous segments found in loop C and the C-terminal helix, suggesting greater flexibility of these regions in the oxidized state. The results provide dynamic interpretations for previously presented structural and functional data, including redox-dependent changes that occur in the protein. The way is now paved for extensive dynamic analysis of variant cytochromes c.

High level, context dependent misincorporation of lysine for arginine in Saccharomyces cerevisiae a1 homeodomain expressed in Escherichia coli

The Saccharomyces cerevisiae a1 homeodomain is expressed as a soluble protein in Escherichia coli when cultured in minimal medium. Nuclear magnetic resonance (NMR) spectra of previously prepared a1 homeodomain samples contained a subset of doubled and broadened resonances. Mass spectroscopic and NMR analysis demonstrates that the heterogeneity is largely due to a lysine misincorporation at the arginine (Arg) 115 site. Arg 115 is coded by the 5'-AGA-3' sequence, which is quite rare in E. coli genes. Lower level mistranslation at three other rare arginine codons also occurs. The percentage of lysine for arginine misincorporation in a1 homeodomain production is dependent on media composition. The dnaY gene, which encodes the rare 5'-AGA-3' tRNA(ARG), was co-expressed in E. coli with the a1-encoding plasmid to produce a homogeneous recombinant a1 homeodomain. Co-expression of the dnaY gene completely blocks mistranslation of arginine to lysine during a1 overexpression in minimal media, and homogeneous protein is produced.

Heterodimerization of the yeast homeodomain transcriptional regulators alpha 2 and a1: secondary structure determination of the a1 homeodomain and changes produced by alpha 2 interactions

The homeodomain proteins, a1 and alpha 2, act cooperatively to regulate cell-type specific genes in yeast. The basis of this cooperativity is an interaction between the two proteins, forming a heterodimer that binds DNA tightly and specifically. A fragment containing the homeodomain of a1, a1(66-126), has been studied by NMR spectroscopy to gain secondary structure information and to characterize the changes in a1 upon heterodimerization with alpha 2. Heteronuclear (1H-15N) NMR methods were used to assign backbone resonances of the 61 amino acid fragment. The a1(66-126) secondary structure was determined using NOE connectivities, 3JHN alpha coupling constants and hydrogen exchange kinetic data. NMR data identify three helical segments separated by a loop and a tight turn that are the characteristic structural elements of homeodomain proteins. The a1 fragment was titrated with alpha 2(128-210), the homeodomain-containing fragment of alpha 2, to study changes in a1(66-126) spectra produced by alpha 2 binding. The a1(66-126) protein was labeled with 15N and selectively observed using isotope-edited NMR experiments. NMR spectra of bound a1(66-126) indicate that residues in helix 1, helix 2, and the loop connecting them are directly involved in the binding of the alpha 2 fragment. Relatively minor effects on the resonances from residues in helix 3, the putative DNA-binding helix, were noted upon alpha 2 binding. We have thus located a region of the a1 homeodomain important for specific protein recognition.

Conformational properties of DNA hairpins with TTT and TTTT loops

The DNA hairpins d[CGATCG-Tn-CGATCG] (n = 3, 4) have been studied by NMR in order to gain information on hairpin conformation and flexibility. Resonance assignments were made using a combination of DQF-COSY, DQF-COSY[31P], NOESY, and 1H-31P-COSY. These data also provide approximate coupling constant information which points out exceptionally flexible regions of the phosphate backbone. The data for both hairpins reveal substantial flexibility within the loop segments. For n = 4, NOESY data alone are insufficient to distinguish between two loop-folding motifs, although coupling constant data favor a conformation in which Tb is folded toward the minor groove and is highly exposed to solvent. This is in agreement with chemical shift data and susceptibility to modification by KMnO4. The phosphate backbone between Tc and Td is exceptionally flexible, undergoing a facile exchange between (beta t,gamma+) and (beta+,gamma t) conformers. A similar flexible phosphate is observed between Tc and C7 when n = 3. Differences in stem conformation and dynamics in both hairpins are restricted to the two base pairs adjacent to the stem-loop junction. The C7pG8 stem phosphate appears to flip easily between (zeta-,alpha-) and (zeta-,alpha t) conformers when n = 4 but not when n = 3. Hairpin loop size thus affects the conformational flexibility of the adjacent stem segment.

Sequence-dependent oligonucleotide-target duplex stabilities: rules from empirical studies with a set of twenty-mers

We were interested in developing a better method to predict the thermal stability of specific oligonucleotide-target duplexes. Recognizing that the base sequence can have important effects, we investigated the use of a simple parameter based on nearest-neighbor stacking interactions, the mean stacking temperature. We took values for doublet stabilities from the literature and used a computer program to calculate mean stacking temperatures for all oligonucleotides of specified length and G + C content in the M13 phage genome. As expected, the program predicted a fairly broad range of stabilities for different sequences of equal G + C content. We selected 20-mer sequences representing the highest and lowest mean stacking temperatures at 25%, 50% and 75% G + C and synthesized them for use as probes against M13 DNA immobilized on filters. By hybridizing and washing at different temperatures, we demonstrated that mean stacking temperatures correlate well with observed stabilities. Relative stabilities of the six oligos were predicted correctly in every case. We used conditions appropriate to oligonucleotide probing and polymerase chain reaction and we were able to derive simple linear equations relating the empirical data and mean stacking temperature for both. These observations should be useful in planning experiments with oligonucleotides.