|Home | About | Journals | Submit | Contact Us | Français|
Alterman MA, Gogichayeva NV, Kornilayev BA. Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry-based amino acid analysis. Analytical Biochemistry 335;2004:184–191. [PubMed]
Quantitation of amino acids from protein acid hydrolysates is demonstrated using MALDI-TOF mass spectrometry with α-cyano-4-hydroxycinnamic acid as matrix. All the standard protein amino acids that are measured using conventional chromatographic separation procedures are included, except that leucine and isoleucine are not discriminated. No ion suppression effects are observed with the methods described. Methyltyrosine is used as internal standard. Linear responses are documented between 20 and 300 μM concentration with correlation coefficients between 0.983 and 0.999. Limits of quantitation are between 0.03 μM and 3.7 μM. The main advantages of the method are that it requires no derivatization or chromatographic separation of the amino acids. Data acquisition is therefore exceedingly fast.
Samyn B, Sergeant K, Castanheira P, Faro C, van Beeumen J. A new method for C-terminal sequence analysis in the proteomic era. Nature Methods 2;2005:193–200. [PubMed]
Proteins in solution, in gels, or on polyvinyl difluoride blots are digested with cyanogen bromide under conditions that maximize the abundance of homoserine lactone relative to homoserine at the resulting peptide C-termini. The unfractionated peptide mixture is then digested with carboxypeptideases Y and P. Peptides with C-terminal homoserine lactone are refractory to digestion. The peptide derived from the C-terminus of the intact protein, however, is cleaved to give a ladder of signals revealed by MALDI-TOF mass spectrometry from which the C-terminal sequence can be deduced. The technique is convenient in circumstances where mass measurements of carboxypeptidase digests of the intact protein are impractical to perform with sufficient sensitivity and accuracy.
Matsuzaki H, et al. Genotyping over 100,000 SNPs on a pair of oligonucleotide arrays. Nature Methods 1;2004:109–111. [PubMed]
Genotyping of 116,204 human single nucleotide polymorphisms (SNPs) using Affymetrix oligonucleotide arrays is documented. Genomic DNA is digested using XbaI and HindIII separately, in parallel. Restriction fragments are ligated to adaptors and amplified with Platinum Pfx polymerase from Invitrogen to produce fragments of 250–2000 bp. Each SNP is represented on the arrays by 40 probes—10 groups of 4 representing perfect match and mismatch pairs for both alleles. Genotyping is accomplished by an algorithm that calculates the log-likelihood of the possible genotypes (homozygotes AA and BB, heterozygote AB, and null) based on the observed hybridization intensities to the groups of 4. Suitable confidence limits for acceptance are imposed. The average distance between markers is 23.6 kb, and 92% of the genome is estimated to be within 100 kb of a marker. The methodology provides a broadly applicable vehicle for genome-wide association studies.
Murray SS, Oliphant A, Shen R, McBride C, Steeke RJ, Shannon SG, Rubano T, Kermani G, Fan J-B, Chee MS, Hansen MST. A highly informative SNP linkage panel for human genetic studies. Nature Methods 1;2004:113–117. [PubMed]
A panel of 4763 SNP markers is presented for genetic linkage studies within families. Genotyping is performed with arrays of beads using the Illumina BeadArray technology. Each bead type has a specific oligonucleotide probe sequence attached. The arrays are randomly assembled collections of beads, such that each bead type is represented an average of 30 times per array. Each bead is located in a microwell at the end of an optical fiber. The fibers are bundled, and bundles are arranged in a 96-well format with a different array in each well for convenience in processing multiple samples in parallel. This system is being used to generate half of the genotyping data in the International HapMap Project. The SNP panel is used in the present work to type 518 individuals in 28 large families and hence to construct a genetic map of the markers. The mean genetic map distance between markers is 1.5 cM. The system provides a broadly applicable, high-throughput platform for linkage studies of disease markers.
Larsen MR, Hojrup P, Roepstorff P. Characterization of gel-separated glycoproteins using two-step proteolytic digestion combined with sequential microcolumns and mass spectrometry. Molecular and Cellular Proteomics 4;2005:107–119. [PubMed]
Gel bands of N-linked glycoprotein are subjected to a two-stage digestion procedure. First, trypsin is employed and an aliquot of the digest used for mass spectrometric protein identification. The remainder of the digest is further treated with the nonspecific proteinase K to yield small peptides. Unglycosylated peptides are removed upon passage through a microcolumn consisting of a GELoader tip (Eppendorf) packed with Poros 2. Glycosylated peptides are not retained, but are trapped on a second GELoader tip packed with graphite powder, washed to remove low molecular weight contaminants, and eluted with acetonitrile. Tandem mass spectrometry then provides amino acid sequence and partial glycan structure. Using this strategy on 8 pmol of ovalbumin applied to a gel, all 13 of the previously known glycan chains were identified, plus three additional ones.
Zhang H, Yi EC, Li X-j, Mallick P, Kelly-Spratt KS, Masselon CD, Camp DG II, Smith RD, Kemp CJ, Aebersold R. High throughput quantitative analysis of serum proteins using glycopeptide capture and liquid chromatography mass spectrometry. Molecular and Cellular Proteomics 4;2005:144–155. [PubMed]
Quantitative serum proteomics based on LC/MS analysis of tryptic peptides from serum proteins is compromised by the extreme complexity of the peptide mixtures produced by trypsin digestion. The present work offers a method of simplifying the mixture. Glycoproteins are immobilized by periodate oxidation of hydroxyl groups on their sugar side chains to aldehydes, then covalently binding them to hydrazide beads. The immobilized proteins are then digested with trypsin, and unbound peptides are washed away. The glycopeptides remaining immobilized are then released by digestion with peptide-N-glycosidase F, and are employed for quantitative LC/MS analysis. By restricting attention to those peptides that were previously glycosylated, the complexity of the peptide mixture is reduced, and the sensitivity and throughput of analyses are increased.
Lausted C, Dahl T, Warren C, King K, Smith K, Johnson M, Saleem R, Aitchison J, Hood L, Lasky SR. POSaM: A fast, flexible, open-source, inkjet oligoncleotide synthesizer and microarrayer. Genome Biology 5;2004:R58. [PubMed]
Noting the spur given to the microarray field by the early release of the Stanford design for a pin-spotting arrayer, and drawing attention to the advantages of synthesizing oligonucleotides in situ on chips to create custom arrays, the authors seek to remedy the lack of ready access to instrumentation for performing in situ, custom oligonucleotide synthesis in academic laboratories. A piezoelectric oligonucleotide synthesizer and arrayer is described. It uses a low-cost print head, high-quality motion controllers, and standard phosphoramidite chemistry, and rapidly produces arrays of 9800 features. The construction can be undertaken by most well-equipped molecular biology laboratories with modest organic chemistry and engineering expertise.
Cline DJ, Thorpe C, Schneider JP. General method for facile intramolecular disulfide formation in synthetic peptides. Analytical Biochemistry 335;2004:168–170. [PubMed]
4,4′-Dithiodipyridine is shown greatly to facilitate oxidation of peptides to form intramolecular disulfide bonds. The reaction takes place under the acidic conditions and in the presence of high concentrations of organic solvents that typify the conditions under which peptides are commonly purified by reverse-phase chromatography. This method for disulfide bond formation can therefore rapidly be performed following purification with a minimum of intervening manipulation.
Jackson SN, Wang H-YJ, Woods AS, Ugarov M, Egan T, Schultz JA. Direct tissue analysis of phospholipids in rat brain using MALDI-TOFMS and MALDI-ion mobility-TOFMS. Journal of the American Society of Mass Spectrometry 16;2005:133–138.
Frozen brain tissue is sectioned on a cryostat. Sections are placed on a target plate and spots of matrix solution—optimally, 6-aza-2-thiothymine or 2,6-dihydroxyacetophenone—are applied. Ions of three lipid classes, phosphatidylcholines, phosphatidylethanolamines, and sphingomyelin, are recorded in the resulting MALDI mass spectra. PC 32:0, PC 34:1, and SM 18:0 predominate. The identities of the phospholipid ions are confirmed in ion mobility studies. The method is amenable to use in tissue imaging, and to the detection of lipophilic drugs.
Kong XL, Huang LCL, Hsu C-M, Chen W-H, Han C-C, Chang H-C. High-affinity capture of proteins by diamond nanoparticles for mass spectroscopic analysis. Analytical Chemistry 77;2005:259–265. [PubMed]
Proteins are concentrated from dilute solution by virtue of their ability to interact with diamond nanoparticles. Diamond powder with nominal particle size of 100 nm is treated with strong oxidative acid to yield a particle surface that is carboxylated and oxidized. The particles are hydrophilic and stable to storage in aqueous suspension. Protein complexed to diamond is sedimented by centrifugation, and is amenable to analysis by MALDI without prior removal of the particles.
Cargile BJ, Bundy JL, Stephenson JL. Jr. Potential for false positive identifications from large databases through tandem mass spectrometry. Journal of Proteome Research 3;2004:1082–1085. [PubMed]
There is a general and growing concern about the assumption that protein identifications made in shotgun proteomics are unquestionably correct. The authors here create a protein sequence database of a mythical creature (Medusa) consisting of 40,000 randomly generated protein sequences. The sizes of the proteins and the frequencies of the amino acids are made to emulate the human proteome. This database is used for “identification” of proteins from rat testis based on searches of MS/MS spectral data derived from tryptic peptides. Using scoring cut-offs of moderate stringency, 1400 proteins are “identified” with SEQUEST and 500 with MASCOT. Considering only those proteins “identified” on the basis of two or more peptides, 30 proteins are identified with SEQUEST and 2 proteins with MASCOT. Not all the matches can be eliminated by manual spectral validation. Such spurious results are to be anticipated because large databases, such as those available for eukaryotes, contain enough peptides for random matches to be likely to occur with a high degree of significance. Of equal concern is the danger that correct identifications are missed through attempts to eliminate false identifications by elevating scoring cut-offs. The authors advocate introducing additional matching criteria, such as peptide isoelectric point, to eliminate false-positives, and hence to allow scoring cut-offs to be relaxed without risking a large increase in false identifications. Peptides may be selected by isoelectric focusing, and the measured pI values compared with those computed for sequences matched on the basis of mass spectrometry.
Leichert LI, Jakob U. Protein thiol modifications visualized in vivo. PloS Biology 2;2004:e333. [PubMed]
Sethuraman M, McComb ME, Huang H, Huang S, Heibeck T, Costello CE, Cohen RA. Isotope-coded affinity tag (ICAT) approach to redox proteomics: Identification and quantitation of oxidant-sensitive cysteine thiols in complex protein mixtures. Journal of Proteome Research 3;2004:1228–1233. [PubMed]
Both these papers present methodology for identifying proteins whose activity is regulated by redox potential through reversible modification of reactive cysteine residues. Disulfide bond formation, nitrosylation, glutathionylation, or sulfenic acid formation cause conformational changes that lead to protein activation or inactivation as part of the cellular response to oxidative stress. Leichert and Jakob lyse cells in trichloroacetic acid to quench thiol exchange. They then alkylate cysteines present in reduced state using iodoacetamide. Cysteines present in an oxidized state are then reduced with dithiothreitol and reacted with 14C-labeled iodoacetamide. The radioactivity of proteins separated by 2-dimensional gel electrophoresis then provides an indication of their oxidation status. Comparison between control cells and cells subjected to oxidative stress identifies proteins putatively subject to redox regulation. Sethuraman et al. measure susceptibility of protein thiols to oxidation in vitro. Cysteines are labeled with the Applied Biosystems’ cleavable ICAT reagent before and after treatment of tissue homogenates with hydrogen peroxide. Only peptides containing reduced cysteine react with the reagent, so cysteines susceptible to oxidation are recognized by depletion of the corresponding ICAT-labeled peptides.
Ferguson RE, Carroll HP, Harris A, Maher ER, Selby PJ, Banks RE. Housekeeping proteins: A preliminary study illustrating some limitations as useful references in protein expression studies. Proteomics 5;2005:566–571. [PubMed]
Variation in the expression levels of proteins commonly used as internal controls in protein expression studies is examined. Glyceraldehyde-3-phosphate dehydrogenase, β-actin, β-tubulin, and class I β-tubulin are quantitated by Western blotting as a function of total protein level in a series of renal cancer cell lines, in matched pairs of renal tumors and normal kidney cells, and in nine different human tissues. Each of the markers tested varies significantly in at least one biological context, often for reasons concerned with the various cellular functions they fulfill. It is therefore recommended that the choice of housekeeping proteins for internal standards is made carefully in relation to the cell and tissue types, experimental conditions, and disease states under consideration.
Dunkley TPJ, Watson R, Griffin JL, Dupree P, Lilley KS. Localization of organelle proteins by isotope tagging (LOPIT). Molecular and Cellular Proteomics 3;2004: 1128–1134. [PubMed]
Unambiguous assignment of the subcellular localization of proteins has long been problematic because pure preparations of organelles are exceedingly difficult to make. This paper presents a method for assigning subcellular localization that is suitable for use with organelles only partially separated by centrifugation through self-generating density gradients. Proteins sharing a common subcellular distribution are expected to exhibit a similar distribution in the gradient. Protein distributions are measured in pairwise comparisons of gradient fractions using the Applied Biosystems’ cleavable ICAT reagent with mass spectrometric quantitation. Multivariate data analysis is then used to match these distributions to the distributions of known organelle-specific markers to assign subcellular localization.
Rush J, Moritz A, Lee KA, Guo A, Goss VL, Spek EJ, Zhang H, Zha X-M, Polakiewicz RD, Comb MJ. Immunoaffinity profiling of tyrosine phosphorylation in cancer cells. Nature Biotechnology 23;2005:94–101.
Gembitsky DS, Lawlor K, Jacovina A, Yaneva M, Tempst P. A prototype antibody microarray platform to monitor changes in protein tyrosine phosphorylation. Molecular and Cellular Proteomics 3;2004:1102–1118. [PubMed]
These papers present different strategies for approaching the same problem—the global profiling of proteins that are subject to tyrosine phosphorylation. Rush et al. digest cell extracts with proteases such as trypsin, then isolate peptides containing phosphotyrosine by immunoaffinity chromatography using a pTyr-specific monoclonal antibody. The peptides are then identified by mass spectrometry. In this way, the pool of species containing a modification of relatively low abundance is enriched, allowing identification of many previously unknown modification targets. Gembitsky et al. detect changes in phosphorylation state of known tyrosine-phosphorylated proteins using an antibody array method. Antibodies against specified phosphoproteins are arrayed as capture antibodies. The array is then incubated with a whole-cell or tissue extract, and probed with a fluorescently labeled phosphotyrosine-specific monoclonal antibody. This method is capable of high sensitivity and throughput, and is amenable to multiplexing.
Zappacosta F, Annan RS. N-terminal isotope tagging strategy for quantitative proteomics: Results-driven analysis of protein abundance changes. Analytical Chemistry 76;2004:6618–6627. [PubMed]
An isotope tagging method is described that labels every peptide from a protein in a sequence-independent manner, and is also suitable for peptides containing posttranslational modifications. Peptides in a protein digest are subjected to acylation with either d0- or d5-propionic anhydride. Lysine side chains are blocked prior to the reaction, so labeling is restricted to the peptide N-terminus. Protein abundance differences are detected by mass spectrometric analysis of mixtures of isotopically distinguished peptides, and may be identified by data-dependent acquisition of MS/MS spectra.
Choe SE, Boutros M, Michelson AM, Church GM, Halfon MS. Preferred analysis methods for Affymetrix GeneChips revealed by a wholly defined control dataset. Genome Biology 6;2005:R16. [PubMed]
Choices between alternative methods for analysis of gene expression data to optimize sensitivity while minimizing false discovery rates can be made only by investigating differences in the performance of analysis methods in cases where the RNA abundance levels are already known. The present paper evaluates analysis options with wholly defined RNA mixtures. A mixture of 2551 defined RNA species provides a constant background, while 100–200 RNAs are spiked in at levels representing fold-changes from 1.2 to 4.0. Accurate estimates of false-positive and false-negative rates can thus be made at each fold-change level, and nonspecific signal strength can be measured using probe sets corresponding to RNAs that are genuinely absent from the sample. All normalization methods perform similarly. However, substantial improvements to false-negative and false-positive discovery rates are obtained by subtracting nonspecific signal from the perfect match probe intensities, performing an intensity-dependent normalization at the probe set level, and incorporating an intensity-dependent standard deviation in the test statistic.
Mukherjee S, Berger MF, Jona G, Wang XS, Muzzey D, Snyder M, Young RA, Bulyk ML. Rapid analysis of the DNA-binding specificities of transcription factors with DNA microarrays. Nature Genetics 36;2004:1331–1339. [PubMed]
Liu X, Noll DM, Lieb JD, Clarke ND. DIP-chip: Rapid and accurate determination of DNA-binding specificity. Genome Research 15;2005:421–427. [PubMed]
These papers introduce the use of DNA microarrays for high throughput identification of transcription factor binding sites. The method of Mukherjee et al. measures the direct binding of transcription factors to DNA microarrays. Epitope-tagged transcription factors are applied to a spotted whole-genome intergenic array, and then detected using fluorescently labeled, tag-specific antibodies. The data are normalized relative to the amount of double-stranded DNA in each microarray feature by measuring binding of the dye SybrGreen I to a duplicate array. Three yeast transcription factors are studied, and, in addition to the detection of known binding sequences, new targets are identified in each case. Many are upstream of previously uncharacterized open reading frames, and many are evolutionarily conserved. Liu et al. isolate protein-DNA complexes by immunoprecipitation from an in vitro mixture of genomic DNA and the pure transcription factor. The DNA fragments are then identified using a whole-genome microarray. The method is validated using a yeast transcription factor of known specificity.
Cleary MD, Meiering CD, Jan E, Guymon R, Boothroyd JC. Biosynthetic labeling of RNA with uracil phosphoribosyltransferase allows cell-specific microarray analysis of mRNA synthesis and decay. Nature Biotechnology 23;2005: 232–237.
Steady-state levels of macromolecules are governed by their combined rates of synthesis and degradation. Furthermore, the speed with which a cell can respond to a change in circumstances, determined, for example, by the rate at which it can switch between different steady-state levels of a regulatory macromolecule, also depends on both synthesis and degradation rates of that macromolecule. The present paper describes a method for measuring the rates of synthesis and degradation of mRNAs that is compatible with standard microarray-based methods for determining mRNA abundance. The method utilizes an enzyme in the protozoan parasite, Toxoplasma gondii, UPRT, to add a phosphoribosyl group to thiouridine to form thio-UMP, which is then available for incorporation into RNA. UPRT is shown to be amenable to expression and function in human cells. Thiouracil is rapidly taken up by cells, and is also rapidly chased out by uracil, with no discernable effect on gene expression levels. Thio-RNA is isolated by labeling with biotin and then performing avidin affinity chromatography. Rates of synthesis are measured by applying the thio-RNA made during thiouracil pulse to a standard expression microarray, and rates of degradation are measured from mRNA made during the subsequent uracil chase.
Bertone P, Stolc V, Royce TE, Rozowsky JS, Urban AE, Zhu X, Rinn JL, Tongprasit W, Samanta M, Weissman S, Gerstein M, Snyder M. Global identification of human transcribed sequences with genome tiling arrays. Science 306;2004:2242–2246. [PubMed]
This paper provides a draft expression map for the entire human genome. Using maskless photolithography, 134 high-density oligonucleotide arrays are made representing 1.5 Gb of nonrepetitive genomic DNA. The probes are 36-mers and are synthesized with a feature density of 390,000 probes per array, and are positioned on average every 46 residues. The total number of probes is 51,874,388. Transcribed sequences are located by hybridizing the arrays to fluorescently labeled cDNA reverse-transcribed from poly(A+) RNA from pooled liver tissue. A total of 10,595 novel transcripts are identified, many of which are believed to be functional because of their homology with known mouse proteins. Many are located in regions distal to known genes. Some encode proteins of 300 amino acids or more. The remaining transcripts presumably encode small proteins, untranslated exons, or RNA species with presently unknown function.
Kim D-H, Behlke MA, Rose SD, Chang M-S, Choi S, Rossi JJ. Synthetic dsRNA dicer substrates enhance RNAi potency and efficacy. Nature Biotechnology 23;2005: 222–226.
Siolas D, Lerner C, Burchard J, Ge W, Linsey PS, Paddison PJ, Hannon GJ, Cleary MA. Synthetic shRNAs as potent RNAi triggers. Nature Biotechnology 23;2005:227–231.
These two papers report improvements in the design of RNAs to provide major enhancements in effectiveness as agents for RNA interference (RNAi). Kim et al. show that 25–30-mer duplexes can be up to 10-fold more potent than the corresponding 21-mer duplexes normally used, and that some sites not susceptible to silencing by 21-mers are effectively targeted by 27-mers, with silencing lasting as long as 10 days. They did not observe induction of interferon response or activation of protein kinase R. Siolas et al. show that short hairpin RNAs with 29-bp stems and 2-bp 3′ overhangs are also more effective than conventional reagents. These features are believed to work by making the RNAs better substrates for cleavage by the processing enzyme, Dicer.
Wheeler DB, Bailey SN, Guertin DA, Carpenter AE, Higgins CO, Sabatini DM. RNAi living-cell microarrays for loss-of-function screens in Drosphila melanogaster cells. Nature Methods 1;2004:127–132. [PubMed]
Earlier array-based methods for screening living cells for RNAi-based loss of function have employed mammalian cells transfected with vectors expressing 21–23-mer short interfering RNAs or short hairpin RNAs to avoid sequence-independent activation of interferon reponse. In the present work, slide chemistries are modified to permit the growth of Drosophila cells, in which RNAi can, instead, be induced by long, double-stranded (ds) RNAs, with very high efficiency and specificity. Genome-wide collections of dsRNAs are available for Drosophila, and a prototype array of 384 features is used in the present study. In the case of Drosophila cells, two distinct dsRNAs can be used to silence two genes simultaneously, allowing the method to be used to identify interactions of genetic suppressors, enhancers, and synthetic lethals.
Shih JH, Michalowska AM, Dobbin K, Ye Y, Qui TH, Green JE. Effects of pooling mRNA in microarray class comparisons. Bioinformatics 20;2004:3318–3325. [PubMed]
Pooling RNA from different samples for microarray-based class comparisons is done either when insufficient RNA from each individual is available for testing on its own array, or when the number of arrays is reduced to minimize costs. This paper assesses the consequences of pooling for the power of an experiment to detect differential gene expression between classes of individuals, taking into account the magnitude of experimental variation compared with the variation between individuals. To offset the loss of degrees of freedom due to pooling, multiple pools from different individuals must be used to achieve the same statistical power. The smaller the number of independent pools employed, the larger the number of individual samples the pools must contain to achieve comparable power. The savings of costs achieved by pooling may be outweighed by the added numbers of individuals required, depending on the relative costs of samples and chips. Formulae relating these variables are supplied to assist in experimental design.
Chang J, Van Remmen H, Ward WF, Regnier FE, Richardson A, Cornell J. Processing of data generated by 2-dimensional gel electrophoresis for statistical analysis: Missing data, normalization, and statistics. Journal of Proteome Research 3;2004:1210–1218. [PubMed]
In protein expression studies performed by 2D gel electrophoresis, it is common to observe spots that cannot be matched across all gels in a series for the purpose of quantitative comparison of spot staining intensity. Unmatched spots may arise through decreased quantity of the affected proteins, change in migration due to post-translational modification, experimental variations such as insufficient resolution, or failure to detect a spot because it falls below the intensity detection threshold. The resultant “missing data” present the problem of how to include unmatched spots in quantitative assessments of protein expression change. The present work suggests using a method involving K-nearest-neighbors–based imputation, and describes conditions under which the method should be invoked. Alternative methods for normalization and for statistical testing are also evaluated.