|Home | About | Journals | Submit | Contact Us | Français|
Bandeira N, Pham V, Arnott D, Lill JR. Automated de novo protein sequencing of monoclonal antibodies. Nature Biotechnology 26;2008:1336–1338.
Amino acid sequencing at the protein level is generally employed for detection of mutations and modifications of therapeutic monoclonal antibodies, as well as for characterization of antibodies from immunized hosts and commercial sources. Mass spectrometry methods are here described for rapid, complete sequencing of antibodies and other unknown proteins without recourse to time-consuming Edman degradation. They involve the use of optimized protease cocktails of Lys-C, Glu-C, Asp-N, chymortypsin, pepsin, and trypsin for generating information-rich sequence ladders, and adopt a computational strategy for assembling the data similar to comparative fragment assembly in DNA sequencing. Spectral alignments are generated from overlapping peptides to give spectral contigs, and then protein contigs, and because uninterpreted spectra are subject to alignment, difficulties of accurately interpreting all MS/MS spectra are avoided, and both modified and unmodified peptides are encompassed, including peptides with unexpected modifications. Data interpretation is highly automated, and the time to sequence a complete monoclonal antibody is reduced to 36 hours.
Eid J et al. Real-time DNA sequencing from single polymerase molecules. Science 323;2009:133–138.
Amid the many recent leaps forward in DNA sequencing methods, this article provides proof-of-principal for a new technology that promises further fundamental change in throughput and read-length. It succeeds in using DNA polymerase as a real-time sequencing engine. The temporal order of incorporation of distinct, fluorescently labeled deoxyribonucleotide triphosphates (dNTPs) is recorded for a single DNA polymerase molecule performing uninterrupted template-dependent synthesis. The reaction is monitored in a nanophotonic structure, called a zero-mode waveguide, that reduces the volume of observation by more than three orders of magnitude relative to confocal fluorescence microscopy. This enables single fluorofore detection with good signal-to-noise ratio in the presence of the high concentrations of dNTPs required to support adequate rates of nucleotide incorporation by the polymerase. Fluorescent reporter groups are conjugated to the terminal phosphate of the dNTPs. DNA sequence is determined by detecting the increase in color-coded fluorescence that occurs when a dNTP binds in correctly base-paired fashion within the active site of the polymerase. The signal lasts until the polyerase cleaves the dye-linker-phosphate group from the growing chain ready for a new reaction cycle. The polymerase operates at an average synthesis rate of 4.7 bases per second. The polymerase chosen for this work, 29 DNA polymerase, operates in a highly processive manner and is able to complete multiple laps of DNA strand displacement synthesis around the circular template, enabling many reads from the same template molecule. With just 15 molecules, a consensus sequence with 99.3% median accuracy can be produced with no detectable sequence context bias and a uniform error profile within reads, and this level of accuracy is proving to be amenable to further improvement. A zero-mode wave-guide is capable of producing 400 kb of sequence per day, and the authors estimate that with multiplexing of just 14,000 functioning zero-mode waveguides (attainable with existing instrumentation) a raw read throughput equivalent to 1-fold coverage of a diploid human genome per day is practical.
Another aspect of the work that will be of interest to enzymologists is the system’s ability to perform kinetic measurements on single polymerase molecules. DNA polymerization activity lasts for thousands of seconds. The pulse-width data provide a measure of the rate of nucleotide incorporation. These data reveal two characteristic polymerization rates of approximately 2 bases/sec and 4 bases/sec, suggesting that two different states of the polymerase exist that can interconvert. Furthermore, the polymerase pauses at sites corresponding to locations of DNA secondary structure.
Gnirke A, Melnikov A., Maguire J, Rogov P, LeProust EM, Brockman W, Fennell T, Giannoukos G, Fisher S, Russ C, Gabriel S, Jaffe DB, Lander ES, Nusbaum C. Solution hybrid selection with ultra-long oligonucleotides for massively parallel targeted sequencing. Nature Biotechnology 27;2009:182–189.
Two methods have been utilized to target large numbers of specific DNA regions for massively parallel sequencing in an effort to minimize resequencing costs. In the first, microarray capture, templates are hybridized to a DNA microarray. In the second, multiplex amplification, olignucleotides are first synthesized on a microarray and then cleaved off and amplified in solution by PCR to circularize and copy templates rather than capture them. The present paper employs elements of both methods to improve on the performance of either one. Oligonucleotides of 200-mer length are synthesized in array format for economy and cleaved from the array. Each oligonucleotide consists of a target-specific 170-mer sequence with universal primers on each side to permit PCR amplification. After this PCR amplification, a T7 promoter is added in a second round of PCR, and single-stranded RNA bait is then made by in vitro transcription in the presence of UTP-biotin. The product is a single-stranded RNA that can be used as bait and then extracted with strepavidin-coated magnetic beads. This method is tested by targeting >15,000 exons and four large DNA regions and sequencing with the Illumina system. Coverage is found to be relatively even, with about 60% of bases targeted in the exon selection, and about 80% in the regional selection having at least half the mean coverage. The concordance of sequence-based genotype calls with known HapMap genotypes is high (99.4%), with little allelic bias or drop-out at SNP loci.
Tagliavia M, Gianguzza F. Complete decontamination and regeneration of DNA purification silica columns. Analytical Biochemistry 385;2009:182–183.
Silica columns are used extensively for purification of DNA, but suffer from retention of substantial amounts of DNA that must be removed before the column can be used for a new sample. The present paper prescribes a new method for column clean-up that is rapid and effective. The method consists of sequential alkaline and acid treatments. Alkali is used to denature DNA and acid to depurinate. These treatments are performed in the presence of Triton X-100, which is shown to enhance removal, possibly by assisting the acid solution to permeate the silica matrix into regions that might otherwise harbor DNA that resists acid depurination. The method allows column regeneration in less than one hour, regardless of the size of the DNA molecules previously purified, with no detectable carryover contamination.
Gibson DG, Benders GA, Axelrod KC, Zaveri J, Algire MA, Moodie M, Montague MG, Venter JC, Smith HO, Hutchison CA, II. One-step assembly in yeast of 25 overlapping DNA fragments to form a complete synthetic Mycoplasma genitalium genome. Proceedings of the National Academy of Sciences, U.S.A. 105;2008:20404–20409.
The authors previously demonstrated the assembly of the complete 592-kb circular genome of Mycoplasma genitalium in yeast cells using a combination of in vitro enzymic recombination for the initial stages of the synthesis and in vivo recombination in yeast for the final stage. Here, they show that yeast can assemble DNA fragments even from the earlier stages, reducing the need for in vitro manipulations. Specifically, yeast cells are shown to be capable of assembling 25 fragments of approximately 24-kb length in one step. This finding promises accelerated progress in synthetic biology.
Katritzky AR, Haase DN, Johnson JV, Chung A. Benzotriazole-assisted solid-phase assembly of Leu-enkephalin, amyloid β segment 34–42, and other “difficult” peptide sequences. Journal of Organic Chemistry 74;2009:2028–2032.
During solid-phase peptide synthesis, formation of peptide links between amino acids with β-branched side-chains (Val, Ile, Thr) may result in partial racemization, or formation of deletion sequences, or aspartamides. 1-Hydroxybenzotriazole hydrate (HOBt) is a widely used coupling additive that suppresses racemization when used in combination with carbodiimides such as dicyclohexylcarbodiimide. Here, N-(Fmoc-α-aminoacyl)benzotriazoles are used as building blocks for carbodiimide-free synthesis of six “difficult” peptides using microwave acceleration. Chirally homogeneous products are obtained. In comparison to carbodiimide-based methods, the coupling reactions require no base, the conditions are relatively mild, and reactions are more rapid. Crude yields of peptides of 34–86% are demonstrated.
Ejsing CS, Sampaio JL, Surendranath V, Duchoslav E, Ekroos K, Klemm RW, Simons K, Shevchenko A. Global analysis of the yeast lipidome by quantitative shotgun mass spectrometry. Proceedings of the National Academy of Sciences, USA 106;2009:2136–2141.
Broad-scale analysis of lipids has typically been characterized by the need for complex extraction schemes to separate lipid classes, and consumption of large amounts of material. The present paper uses a simple two-stage extraction method to replace the conventional Bligh and Dyer protocol: first chloroform/methanol, 17/1 (v/v) to extract apolar classes, then chloroform/methanol, 2/1 (v/v) to extract the remaining aqueous phase to rescue polar classes. The resulting pools are analyzed by infusion into the mass spectrometer, but 0.2 mM methylamine replaces the standard ammonium acetate as the infusion buffer to enhance the ionization of anionic lipids. Mass spectrometers providing high resolution and mass accuracy are employed for the analysis, and 21 class-specific internal standards are spiked in to provide absolute quantification using dedicated software. The resulting methodology is estimated to cover about 95% of the yeast lipidome, and to allow quantification of 250 molecular species belonging to 21 major lipid classes. (The lipidome of yeast grown on lipid-free minimal medium is considerably simpler then the mammalian lipidome.) Sensitivity is improved up to 125 x relative to established methods, allowing a full analysis with only 2 million cells.
Yang W-C, Sedlak M, Regnier FE, Mosier N, Ho N, Adamec J. Simultaneous quantification of metabolites involved in central carbon and energy metabolism using reversedphase liquid chromatography-mass spectrometry and in vitro 13C labeling. Analytical Chemistry 80;2008:9508–9516.
A general method is described for quantifying the intermediates in carbon and energy metabolism (glycolysis, the pentose phosphate pathway, and the tricarboxylic acid cycle), including nucleotides and cofactors. With the exception of pyruvate, the method permits quantification of all the salient metabolites in a single 30-min reversephase chromatography run with on-line mass spectrometry. Carbonyl, phosphoryl and carboxyl groups on metabolites are derivatized by reaction with aniline to augment chromatographic retention. 13C6 aniline is used to derivatize a standard metabolite mixture for use as an internal standard in quantification. Metabolites that are not distinguishable by mass spectrometry alone are adequately discriminated by chromatography.
Sellick CA, Hansen R, Maqsood AR, Dunn WB, Stephens GM, Goodacre R, Dickson AJ. Effective quenching process for physiologically valid metabolite profiling of suspension cultured mammalian cells. Analytical Chemistry 81;2009:174–183.
Many metabolites are subject to very rapid turnover in cell culture, and the degree of stability varies widely from metabolite to metabolite. To ensure that metabolomic measurements reflect physiological concentrations, therefore, all enzymatic activity must be rapidly terminated upon sampling the cells. The present paper examines the efficiency of alternative methods for quenching as applied to mammalian cells in suspension culture. Based on the recovery of the especially labile metabolites, ATP, ADP, AMP, NAD+, NADH, and glucose-6-phosphate, 60% methanol supplemented with 0.85% ammonium bicarbonate is judged to be optimal.
Szájli E, Fehér T, Medzihradszky KF. Investigating the quantitative nature of MALDI-TOF MS. Molecular and Cellular Proteomics 7;2008:2410–2418.
The precision and accuracy of quantitative measurements made with MALDI mass spectrometry are addressed in this article. MALDI is difficult to use for quantification because of variability in signal strength due to “hot spots,” signal suppression, variation between analytes in ionization efficiency, and variation in laser power. The expected inaccuracies are documented in measurements of the relationship between relative signal intensities and concentration ratios. Signal strength ratios are strongly influenced by the introduction of E. coli cells as a change in the sample milieu. Use of internal standards is strongly recommended, and is shown to improve precision greatly. Confidence limits are improved by increasing the number of measurements that are averaged and the target area over which they are acquired. However, linearity of signal strength ratio is maintained over a very small concentration range, namely 5 x. Preliminary measurements to assess the amounts of standards that need to be added are therefore required.
Brenner N, Haapala M, Vuorensola K, Kostiainen R. Simple coupling of gas chromatography to electrospray ionization mass spectrometry. Analytical Chemistry 80;2008:8334–8339.
A gas chromatograph (GC) is interfaced with standard electrospray sources merely by placing the GC outlet capillary between the electrospray emitter and the mass spectrometer inlet. Neutral compounds are dissolved in the charged elecrospray droplets and, upon ion evaporation, undergo proton transfer reactions with the protonated solvent. Selectivity can be adjusted by choice of electrospray solvent.
Palumbo AM, Reid GE. Evaluation of gas-phase rearrangment and competing fragmentation reactions on protein phosphorylation site assignment using collision induced dissociation-MS/MS and MS3. Analytical Chemistry 80;2008:9735–9747.
Pursuant to recent reports of gas-phase rearrangement reactions of peptide ions under standard conditions for collision-induced dissociation (CID), this paper examines a series of 33 synthetic phosphopeptides to determine the conditions under which gas-phase transfer reactions involving phosphate groups may occur in ion-trap mass spectrometers (in which ion activation times are relatively long), thereby leading to mis-assignment of phosphorylation sites. All the chosen peptides contained multiple potential sites of phosphorylation (i.e., Ser, Thr and Tyr residues). As many as 15 of the 33 peptides (45%) yielded ions that resulted from transfer of phosphate from a phosphoamino acid residue to a previously unphosphorylated hydroxyl side chain. The incidence of such reactions is correlated with restricted proton mobility in the gas-phase ion, being commonest in singly charged ions with more than one basic amino acid. Furthermore, the common practice of assigning phosphorylation sites on the basis of MS3 fragmentation of ions that have undergone neutral loss of 98 Da (ostensibly loss of H3PO4) is shown to be unreliable. The assumption is that loss of 98 Da is due to loss of H3PO4 from a phosphorylated residue, and that the site of phosphorylation can be identified on the basis of the dehydro side-chain left behind. However, the loss of 98 Da may instead occur via the combined loss of HPO3 from one residue and H20 from another. In this case, the dehydro amino acid remaining will be mistaken for the site of phosphorylation. For these reasons, phosphorylation sites assigned to specific residues by CID using ion-trap mass spectrometers should be treated with great caution. On a less gloomy note, gas-phase rearrangments of these kinds are not observed during electron transfer dissociation.
Price WN II, Chen Y, Handelman SK, Neely H, Manor P, Karlin R, Nair R, Liu J, Baran M, Everett J, Tong SN, Forouhar F, Swaminathan SS, Acton T, Xiao R, Luft JR, Lauricella A, DeTitta GT, Rost B, Montelione GT, Hunt JF. Understanding the physical properties that control protein crystallization by analysis of large-scale experimental data. Nature Biotechnology 27;2009:51–57.
Overall, only about 12% of purified proteins form crystals of high enough quality for structure determination. The importance of understanding the physical properties of proteins that limit the propensity to form crystals lies in the expectation that this information will lead to formulae for modifying proteins to increase the likelihood of successful crystallization. Here, the body of data accumulated by the Northeast Structural Genomics Consortium is analyzed to gain insight. As long suspected, the most important single factor is the prevalence of well-ordered (low entropy) surface features. Sequence correlates previously reported appear to be surrogates for low entropy surfaces. However, frequency of glycine, alanine and phenylalanine are independent correlates, presumably because they permit backbone hydrogen bonding in crystal packing interactions (in the case of Gly and Ala), or hydrophobic interactions (in the case of Phe). Tendency to aggregate in dilute aqueous solution is detrimental, even though Lys and Glu, which promote solubility, also tend to diminish crystallization. The best surface features are ones that confer strong stereospecific interactions under the conditions pertaining in crystallization solutions.
Wu S-L, Jiang H, Lu Q, Dai S, Hancock WS, Karger BL. Mass spectrometric determination of disulfide linkages in recombinant therapeutic proteins using online LC -MS with electron-transfer dissociation. Analytical Chemistry 81;2009:112–122.
A mass spectrometric strategy is described for determining the disulfide linkages for therapeutic proteins without the need for two separate experiments, one with and the other without chemical reduction. The products of three types of product ion scan are compared. First, CID-MS2 produces peptide bond cleavage yielding b and y ions. Second, ETD produces c and z ions from peptide bond cleavage, but additionally produces disulfide bond cleavage. H• serves as a donor to the disulfide bond, Cys-S-S-Cys, yielding a free protonated product, Cys-SH, and an odd-electron species, Cys-S•, on the complementary cleavage product. Each disulfide-linked peptide therefore dissociates to produce two populations, the proton transfer form, Cys-SH, and the electron transfer (charge reduced) form, Cys-S•. Third, an MS3 step is used to produce an ETD cleavage pattern for the charge-reduced species and a CID cleavage pattern for the non-charge-reduced species to assign the disulfide linkages. This strategy is used to characterize recombinant human growth hormone (a simple case), a therapeutic monoclonal antibody (a case of moderate complexity), and tissue plasminogen activator (a complex case).
Na Y-R, Park C. Investigating protein unfolding kinetics by pulse proteolysis. Protein Science 18;2009:268–276.
Pulse proteolysis refers to the cleavage of unfolded or partially folded proteins by high concentrations of protease (e.g., 0.2 mg/mL) for very short times (e.g., 1 min) before quenching the reaction. Cleavage occurs at disordered regions due to exposure to proteolytic attack. In the present paper, the technique is shown to be an effective way of measuring unfolding kinetic constants. The disappearance of the band corresponding to the intact protein is quantitatively monitored by SDS-PAGE. A distinctive feature of the approach is that it doesn’t require pure protein. Accuracy of the method is established in studies of the unfolding of E. coli maltose binding protein and E. coli ribonuclease H and comparing the measured values for kinetic constants with those measured by circular dichroism. The presence of the maltose binding protein ligand, maltose, is shown to stabilize the protein from unfolding.
Stocks BB, Konermann L. Structural characterization of short-lived protein unfolding intermediates by laser-induced oxidative labeling and mass spectrometry. Analytical Chemistry 81;2009:20–27.
Pan Y, Stocks BB, Brown L, Konermann L. Structural characterization of an integral membrane protein in its natural lipid environment by oxidative methionine labeling and mass spectrometry. Analytical Chemistry 81;2009:28–35.
Charvátova O, Foley BL, Bern MW, Sharp JS, Orlando R, Woods RJ. Quantifying protein interference footprinting by hydroxyl radical oxidation and molecular dynamics simulation: Application to galectin-1. Journal of the American Society for Mass Spectrometry 19;2008:1692–705.
These papers describe the use of hydroxyl radicals (•OH) generated by pulsed laser dissociation of hydrogen peroxide (H2O2) as a structural probe for proteins, monitoring the location of adduct formation by mass spectrometry following proteolytic digestion. Stocks et al. study the rapid acid-induced unfolding of holo-myoglobin in a continuous-flow, rapid mixing apparatus in which a single 18-ns laser pulse is delivered after the protein has been subjected to acid for a variable period on the ms time-scale. The labeling duration is restricted to about 1 μs by adding 15 mM glutamine as a radical scavenger. To avoid the need to consider the large variety of products formed during labeling, these authors monitor only the disappearance of unmodified peptides by diminution in their mass spectral intensities. They document the order in which the myoglobin helices unfold, and are able to ascribe the special stability of the slowly unfolding helices to persistence of heme-protein contacts in the initial phases of the unfolding process. Pan et al. study the structure of bacteriorhodopsin in its native membrane (without the need for detergent solubilization) using similar methods, and indicate that, unexpectedly, under the experimental conditions pertaining, methionine residues are the only ones oxidized, and that labeling of Met residues in solvent-exposed loops predominates; Met residues deeply buried show much less labeling. Charvátova et al. characterize the dimer interface in the protein, galectin-1 by its protection from labeling. These authors use a longer laser pulse and no quenching, and observe complicated labeling patterns. To cope with the complexity, they describe customized software to identify peptides, many of which are multiply labeled, and to quantify the extent of labeling at each individual residue.
Banik SP, Pal S, Ghorai S, Chowdhury S, Khowala S. Interference of sugars in the Coomassie Blue G dye binding assay of proteins. Analytical Biochemistry 286;2009:113–115.
Assay of protein concentration by binding of Coomassie Blue G dye (the Bradford assay) is shown to suffer significant interference in the presence of sugars. Polysaccharides and disaccharides cause signal enhancement: millibram levels mimick microgram levels of protein. Monosaccharides do not themselves produce signal, but sequester the dye so that signal is lost due to diminution of binding to protein. It is recommended that carbohydrate contaminants be removed by a suitable method such as protein precipitation using TCA or acetone prior to protein assay by this method.
Rockberg J, Löfblom J, Hjelm B, Uhlén M, Ståhl S. Epitope mapping of antibodies using bacterial surface display. Nature Methods 5;2008:1039–1045.
Combinatorial methodology for epitope mapping that is capable of analysis on the whole-proteome scale is presented. A gene encoding the antigen is fragmented by sonication, ligated into a display vector, and the fragment displayed on the surface of the gram-positive bacterium, Staphylococcus carnosus. Cells capable of binding the antibody under test are then isolated by flow cytometry. Inserts from these cells are then sequenced by pyrosequencing. Epitopes are identified by alignment with the antigen sequence. Surface expression is quantified by measuring the binding of labeled human serum albumin to albumin binding protein expressed as a fusion with the antigen fragment. This allows for normalization of binding signal against expression level, and enhances definition of linear epitope boundaries. The method has the advantage over phage display that it is suitable for fluorescence-activated cell sorting. It also allows detection of cross-reactivities to other proteins.
Hansen RE, Roth D, Winther JR. Quantifying the global cellular thiol-disulfide status. Proceedings of the National Academy of Sciences, U.S.A. 106;2009:422–427.
The methodology described here for quantifying cellular thiols and disulfides incorporates several precautions to ensure accurate, specific results. Proteins are initially precipitated by trichloroacetic acid (TCA), which not only separates them from glutathione for separate quantification but also ensures cessation of oxidation and thiol-disulfide exchange. Protein sulfhydryls are quantified by deriviatizing with 4-dithiodipyridine, which forms the chromogenic compound, 4-thiopyridone in a thiol-specific manner at pH values low enough to prevent thiol-disulfide reactions. For quantification of protein disulfides, free sulfhydryls are alkylated with N-ethylmaleimide (NEM) and disulfides are reduced with sodium borohydride for measurement. Because borohydride inactivates NEM, alkylation of newly reduced thiols is avoided. And excess borohydride is readily removed by acid precipitation of proteins, so reactivity of the chromogen with borohydride is avoided. To measure mixed disulfides between proteins and glutathione, disulfides are reduced with tris(hydroxypropyl)phosphine and sulfhydryls are labeled with 7-fluorobenzo-2-oxa-1,3-diazol-4-sulfonate and measured by HPLC with fluorescence detection to ensure selectivity for glutathione. Phosphines don’t react with this particular fluorophore, so the reducing agent may be present during the derivatization to avoid thiol oxidation during the reaction. Finally, protein content of TCA pellets is determined by amino acid analysis to ensure accuracy. These conditions will be of interest to all with an interest in measurement of thiols and disulfides and in control of their reactivity.
Svensson M, Borén M, Sköld K, Fälth M, Sjögren B, Andersson M, Svenningsson P, Andrén PE. Heat stabilization of the tissue proteome: a new technology for improved proteomics. Journal of Proteome Research 8;2009:974–981.
Ex vivo degradation of tissue components due to endogenous proteases, phosphatases and other enzymes may be extremely rapid. As a means of controlling such degradation, tissue samples are here heated automatically to 95°C in a controlled manner between metal heating blocks that exert a pre-determined pressure on the tissue, within an evacuated container. The method reduces enzymatic levels to background, and eliminates degradation in measured instances, without compromising protein extractability with SDS buffer. The procedure is found to work more efficiently than addition of enzymatic inhibitors during extraction, and more efficiently than microwave irradiation, possibly because of difficulties eliminating cold spots due to inhomogenous heating during the latter treatment.
Bakalarski CE, Elias JE, Villén J, Haas W, Gerber SA, Everley PA, Gygi SP. The impact of peptide abundance and dynamic range on stable-isotope-based quantitative proteomic analyses. Journal of Proteome Research 7;2008:4756–4765.
This article approaches the issue of accuracy in the quantification of relative peptide abundance by the SILAC method. Ion-trapping instruments capable of very high mass accuracy and resolution are employed for the study in hope of reducing the incidence of interference from peptide and non-peptide species of closely similar m/z to peptides identified for quantification. As anticipated, accuracy depends strongly on signal/noise ratio (s/n) of precursor ion signals. There is a high proportion (52%) of identified peptides with poor s/n (≤10) at the apex of the extracted ion current chromatogram peak, even for 1:1 mixture, and the proportion becomes progressively greater with widening divergence from 1:1. The overall standard deviation in ratios for a 1:1 mixture is 0.22 for peptides with s/n > 10, and is 0.83 for peptides with s/n < 10. Attempts to increase s/n by accumulating more ions in the ion-trap are circumscribed by the attendant degradation of mass accuracy. Software for discriminating reliable from unreliable ratio measurements is offered.
Liu K, Zhang J, Wang J, Zhao L, Peng X, Jia W, Ying W, Zhu Y, Xie H, He F, Qian X. Relationship between sample loading amount and peptide identification and its effects on quantitative proteomics. Analytical Chemistry 81;2009:1307–1314.
Varying amounts of a digest of yeast whole-cell lysate are analyzed by LC-MS/MS and numbers of identified peptides are recorded. The results indicate that there is an optimal quantity of sample, above which numbers of identified peptides declines. When peak areas for selected peptides are plotted against quantity of sample loaded, some are observed to rise continually with amount of sample loaded, others saturate, and yet others rise to a maximun and then decline as more sample is loaded. When selecting peptides for use in monitoring protein quantity by label-independent methods, only those peptides giving a continuously increasing signal within the range to be tested should be chosen.
Fusaro VA, Mani DR, Mesirov JP, Carr SA. Predication of high-responding peptides for targeted protein assays by mass spectrometry. Nature Biotechnology 27;2009:190–198.
In order to quantify the abundance of proteins using multiple reaction monitoring (MRM) of tryptic peptides derived from them, readily detectable peptides that are unique to each protein of interest must be chosen for monitoring. This paper provides new computational tools for predicting the best peptides to choose. The method uses a battery of peptide physicochemical properties to predict which peptides will give the best mass spectrometer signal strength. Earlier algorithms aimed merely to predict which peptides are most likely to be detected. The present model uses the Random Forest algorithm to compute from the training set the most important variables for predictive purposes. The method yields a predictive model that is robust with respect to the choice of training datasets, and functions better than earlier algorithms, and significantly better than random choice of peptides. The availability of such a computational method is important because proteomic data may not be available to help in the choice of peptides, and may not include all potentially useful peptides even when such data are available. The method is also important for reducing expense associated with the synthesis of putative target peptides because it reduces the number that must be synthesized in order to find two or more acceptable peptides per protein.
Castle JC, Zhang C, Shah JK, Kulkarni AV, Kalsotra A, Cooper TA, Johnsom JM. Expression of 24,426 human alternative splicing events and predicted cis regulation in 48 tissues and cell lines. Nature Genetics 40;2008:1416–1425.
This first genome-wide survey of the incidence of alternative splicing in the human genome indicates a very high prevalence of the phenomenon. A whole-transcript microarray design targeting every exon and every junction is used for the study. The analysis is more complete than exon arrays and cassette exon splicing arrays. In 48 human tissues and cell lines, the authors measure each isoform produced by a total of 24,426 detected splicing events, and provide a catalog of annotated splicing events along with expression data for each probe, each gene exon, each splicing event and each splicing motif in each tissue. 143 motifs are associated with alternative splicing events that vary between tissues.
Ingolia NT, Ghaemmaghami S, Newman JRS, Weissman JS. Genome-wide analysis in vivo of tranlation with nucleotide resolution using ribosome profiling. Science 324;2009:218–223.
It is well known that mRNA abundance is a poor measure of protein production. Polysome profiling, in which mRNAs are isolated for microarray analysis based on their ribosome content, has proved useful but has limited resolution and accuracy. Here, a new technique based on protection of those portions of mRNA protected from nuclease digestion by ribosome binding is described. The ribosome footprint is ~30 nucleotides. These mRNA fragments are converted into a cDNA library in a form suitable for deep sequencing to measure the abundance of different footprint sequences. The technique is used to characterize translation in budding yeast. Along a given message, the position of the 5′ ends of the footprints are observed to start abruptly 12–13 bases upstream of the start codon, end 18 bases upstream of the stop codon, and show strong 3-base periodicity. These observations indicate that the data allow the region being translated to be precisely defined and even the reading frame to be identified. Ribosome density decreases along the coding sequence, suggesting differences in the functional states of ribosomes that affect the rate of elongation or processivity early and late in peptide elongation. Extensive translational control is demonstrated, and widespread, regulated initiation at non-AUG codons is observed. This technique is anticipated to make a substantial impact on characterizing tissue-specific and disease-related expression, defining the full proteome of complex organisms, and exploring how effects such as ribosome pausing modulate protein synthesis and folding.
Doherty MK, Hammond DE, Clague MJ, Gaskell SJ, Beynon RJ. Turnover of the human proteome: determination of protein intracellular stability by dynamic SILAC . Journal of Proteome Research 8;2009:104–112.
Turnover of proteins is monitored on a proteomewide basis by a method based on SILAC. Cells are cultured on medium containing 13C6-arginine and the label is then chased for varying periods with medium containing unlabeled arginine. Proteins from each culture are separated by SDS-PAGE and each lane is divided into 40 slices for digestion with trypsin. The relative incorporation of the light and heavy amino acid is then ascertained by LC-MS/MS analysis of the tryptic peptides. The method allows individual proteins to be tracked, and general questions regarding the determinants of turnover rate to be addressed.
Fletcher JS, Rabbani S, Henderson A, Blenkinsopp P, Thompson SP, Lockyer NP, Vickerman JC. A new dynamic in mass spectral imaging of single biological cells. Analytical Chemistry 80;2008:9058–9064.
Carado A, Passarelli MK, Kozole J, Wingate JE, Winograd N, Loboda AV. C60 secondary ion mass spectrometry with a hybrid-quadrupole orthogonal time-of-flight mass spectrometer. Analytical Chemistry 80;2008:7921–7929.
The Vickerman group in Manchester, UK, and the Winograd group at Penn State provide independent solutions to a long-standing limitation in imaging by secondary ion mass spectrometry (SIMS). The advantage of SIMS over MALDI as the method for desorption of ions from tissue samples is its very high spatial resolution (single nm range for SIMS; 10s of μm range for MALDI), enabled by the superior focus achievable with primary ion beams. C60 clusters are most commonly used for SIMS imaging. However, they require inordinately long image acquisition times when used with time-of-flight mass spectrometry because the yield of secondary ions from a pulsed primary ion source is poor. Both groups therefore construct SIMS-TOF instruments that employ continuous primary ion beams. The Vickerman group employs a linear “ion buncher” to store secondary ions and extracts them with a pulsed extraction voltage. Winograd’s group store secondary ions in the linear ion trap of a modified QqTOF instrument. With these instrument systems, images are acquired in a realistic time frame, bringing SIMS imaging closer to practical application.
McGuire AL, Colgrove J, Whitney SN, Diaz CM, Bustillos D, Versalovic J. Ethical, legal, and social considerations in conducting the human microbiome project. Genome Research 18;2008:1861–1864.
It has been estimated that there are 10x as many microbial cells than human cells in and on our bodies. The microbiota on some sites is essential for maintenance of health. The Human Microbiome Project is intended to ascertain whether humans have a “core” microbiome common to all individuals, whether there is stability within individuals over time, whether there is similarity in microbiomes within families, communities and environments, and whether changes in the microbiome can be correlated with changes in health. Like other screening efforts, this project faces a variety of ethical, legal and social issues. Samples must be collected and experiments planned, with informed consent of participants, but, given presently limited knowledge in the field, the potential risks and benefits of participation in the project are uncertain. There is lack of clarity about the extent to which subjects should be informed of research results related to individual risk of disease or the presence of infectious agents. The issue of privacy is of increasing concern as it becomes possible to identify individuals solely on the basis of genomic information in databases from which individual identity has otherwise been stripped. Sampling should be limited in its degree of invasiveness according to a risk-benefit assessment. Although preliminary sampling will be minimally invasive, later stages of the project might involve more invasive procedures (e.g., endoscopy), and will require ongoing assessment. Finally, the project will require equitable selection of subjects according to ethnicity/race. Careful consideration of these various issues will be essential to the success of the project.
Huang DW, Sherman BT, Lempicki RA. Bioinformatics enrichment tools: Paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids Research 37;2009:1–13.
High-throughput studies typically generate long lists of interesting genes that must be interpreted biologically with the help of bioinformatics tools such as Gene Ontology. Such tools are proliferating: the present paper surveys 68 in total. To assist users make selections among them, these tools are classified into three groups according to the algorithms employed. In the first, traditional strategy, the list of “interesting” genes is tested for enrichment of each biological class, one annotation term at a time, relative to random sampling. In the second, all genes from an experiment are considered in the enrichment calculation, both the genes in the “interesting” list and others genes not qualifying for inclusion in that list. In the third, improved sensitivity and specificity are sought by considering interrelationships of GO terms in the enrichment calculations. The paper considers several issues that remain unresolved in the field, including understanding the limitation of multiple testing correction on enrichment P-values, how to cross-compare analysis results derived from multiple gene lists, how to choose the best gene reference background, and how best to utilize backend databases that are alternatives to GO.