|Home | About | Journals | Submit | Contact Us | Français|
This column highlights recently published articles that are of interest to the readership of this publication. We encourage ABRF members to forward information on articles they feel are important and useful to Clive Slaughter, Hartwell Center, St. Jude Children’s Research Hospital, 332 North Lauderdale St., Memphis TN 38105-2794. Tel; (901) 495-4844: Fax; (901) 495-2945: Email; gro.edujts@rethgualS.evilC or to any member of the editorial board. Article summaries reflect the reviewer’s opinions and not necessarily those of the Association.
Taouatas N, Drugan MM, Heck AJR, Mohammed S. Straightforward ladder sequencing of peptides using a Lys-N metallopeptidase. Nature Methods 5;2008:405–407.
Lysyl-endopeptidase-N cleaves peptides and proteins with high efficiency and specificity on the N-terminal side of lysine residues, producing peptides with a single N-terminal lysine. The peptide products of Lys-N digestion are here shown to produce electron transfer dissociation (ETD) spectra containing predominantly c-type ions. Such spectra therefore represent simple sequence ladders that are readily interpreted de novo. Collision-induced dissociation (CID) also favors the formation of N-terminal product ions (b-ions), but the preference is less strong than with ETD. The enzyme works well on proteins in bands from SDS-polyacrylamide gels as well as on untreated proteins in solution.
Kuyama H, Sonomura K, Nishimura O, tsunasawa S. A method for N-terminal de novo sequence analysis of proteins by matrix-assisted laser desorption/ionization mass spectrometry. Analytical Biochemistry 380;2008:291–296.
This paper incorporates the use of Lys-N into a scheme for isolation and mass spectrometric sequencing the N-terminal peptide from a protein. All the peptides yielded by the enzyme are expected to contain an N-terminal lysine residue except for the peptide from the substrate protein’s N-terminus. The peptide mixture is then derivatized with succinimidyloxycarbonylmethyltris (2,4,6-trimethoxyphenyl)phosphonium bromide (TMPP-Ac-OSu) to produce TMPP-Ac-tagged peptides. Adjusting the pH of the derivatization reaction to 8.2 renders it specific for Na-amino groups. All the derivatized peptides will then have a single amino group on the side-chain of their N-terminal lysine residue, except for the N-terminal peptide, which will lack any amino groups. The peptides with a free amino group are finally allowed to react with a p-phenylenediisothiocyanate (DITC) resin, leaving only the N-terminal peptide in solution and available for mass-spectrometric sequencing. If the N-terminus of the protein is blocked, the N-terminal Lys-N peptide will have no amino group anyway, so it too will be incapable of reacting with the DITC-resin.
Snovida Si, Rak-Banville JM, Perreault H. On the use of DHB/aniline and DHB/N,N-dimethylalanine matrices for improved detection of carbohydrates: Automated identification of oligosaccharides and quantitative analysis of sialylated glycans by MALDI-TOF mass spectrometry. Journal of the American Society for Mass Spectrometry 19;2008:1138–1146.
The addition of aniline or dimethylaniline to 2,5-dihy-droxybenzoic acid matrix for MALDI has been shown to improve detection sensitivity through the formation of matrix crystals of more favorable morphology. Here, aniline is shown to form a Schiff’s base with the reducing end of oligosaccharides on-target, although the reaction does not go to completion. However, dimethylaniline does not form Schiff’s bases at all. Comparison of spectra acquired in the two matrices therefore assists in spectral interpretation. The matrices are also employed successfully for analysis of sialylated peptides in the positive-ion mode using laser energies low enough to avoid loss of sialic acid.
Toyoda M, Ito H, Matsuno Y, Narimatsu H, Kameyama A. Quantitative derivatization of sialic acids for the detection of sialoglycans by MALDI MS. Analytical Chemistry 80;2008:5211–5218.
This study shows that sialoglycans can be quantitatively amidated by acetohydrazide and 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) at pH 2.5 to yield derivatives with stable glycosidic bonds between sialic acids and the remainder of the oligosaccharide. This overcomes the problem of lability of such bonds in MALDI mass-spectral analysis of sialoglycans. The reaction proceeds to completion with both α2,3- and α2,6-linked sialic acids.
Mcalister GC, Berggren WT, Griep-Raming J, Horning S, Marakov A, Phanstiel D, Stafford G, Swaney DL, Syka JEP, Zabrouskov V, coon JJ. A proteomics grade electron transfer dissociation–enabled hybrid linear ion trap–orbitrap mass spectrometer. Journal of Proteome Research 7;2008:3127–3136.
A hybrid linear ion-trap–orbitrap mass spectrometer is modified with a second, chemical ionization source to produce fluoranthene radical ions for electron transfer dissociation (ETD). The reagent anions are transported to the hybrid mass spectrometer’s c-trap via an octapole. From the c-trap they are transported to the linear ion-trap for reaction with analyte ions. The ETD product ions may be analyzed either in the low-resolution linear ion trap for speed and sensitivity or in the orbitrap for high mass accuracy and high resolution. It takes only 4–8 msec to inject sufficient reagent anions into the linear trap. With this instrument system, the authors demonstrate complex mixture analysis, localization of post-translational modifications, quantification using isotope-labeled peptides, and characterization of large peptides and intact proteins.
Ferguson PL, Konermann L. Nonuniform isotope patterns produced by collision-induced dissociation of homogeneously labeled ubiquitin: Implications for spatially resolved hydrogen/deuterium exchange ESI-MS studies. Analytical Chemistry 80;2008:4078–4086.
Extensive use has been made of solution-phase exchange of amide hydrogen/deuterium atoms followed by intact protein collision-induced dissociation (CID) to explore solution-phase folding, structure, dynamics, and molecular interaction of the protein. Backbone cleavage during CID involves rapid intramolecular migration of protons or deuterons, and this is expected to scramble the isotope exchange pattern acquired by the protein during solution-phase labeling to some extent. It is often assumed that complete scrambling will result in identical deutera-tion levels of CID fragments and that this level will match that of the intact protein. Deviations from uniformity are generally believed to result from incomplete scrambling interpretable in terms of the solution-phase behavior of the protein. Contrary to these assumptions, using ubiq-uitin ions generated by electrospray ionization for low-energy fragmentation either in-source or in a collision cell, the present report demonstrates that b-ions exhibit deu-teration levels significantly below that of the intact protein and that these levels do not correlate with the spatial isotope distribution in solution.
Bache N, Rand KD, Roepstorff P, Jørgensen TJD. Gas-phase fragmentation of peptides by MALDI in-source decay with limited amide hydrogen (1H/2H) scrambling. Analytical Chemistry 80;2008:6431–6435.
This paper examines scrambling of amide hydrogen/ deuterium atoms during MALDI in-source decay with 2,5-dihydroxybenzoic acid as the matrix. Using model peptides with known regioselective labeling in solution, this paper is able to conclude that the level of scrambling under these conditions is low, and that c-ions and sodi-ated y-ions in particular may be used to obtain site-specific information about solution-phase deuteration levels.
Barrera NP, Di Bartolo N, Booth PJ, Robinson CV. Micelles protect membrane complexes from solution to vacuum. Science 321;2008:243–246.
Interactions between subunits of integral membrane proteins have hitherto been impossible to study by elec-trospray mass spectrometry because hydrophobic contacts are readily disrupted during transition to the gas phase and because the mass spectra are usually dominated by detergents needed to keep membrane proteins soluble. This report, however, shows that by using a mass spectrometry–compatible detergent well above the critical micelle concentration, interactions between cytoplasmic and membrane subunits of a well-characterized complex are preserved. The protein studied is BtuC2D2, a vitamin B12 importer from E. coli, which consists of two trans-membrane subunits with 20 transmembrane helices per subunit, plus two cytoplasmic subunits that bind ATP. The non-ionic detergent n-dodecyl-β-D-maltoside is used to solubilize the complex. With maximum accelerating voltages in the source and across the collision cell, a diffuse peak consisting of the protein plus heterogeneous micellar species containing 100 or more detergent molecules is observed. As the pressure in the collision cell is increased, detergent molecules dissociate from the assembly, revealing a series of peaks that can be assigned to the protein tetramer. In the presence of ATP/Mg2+, a mass increase consistent with the binding of two molecules of ATP together with their Mg2+ adducts is observed.
Wrammert J, Smith K, Miller J, Langley WA, Kokko K, Larsen C, Zheng N-Y, Mays I, Garman L, Helms C, James J, Air GM, Capra JD, Ahmed R, Wilson PC. Rapid cloning of high-affinity human monoclonal antibodies against influenza virus. Nature 453;2008:667–672.
Rapid production of high-affinity, fully human mono-clonal antibodies against an important pathogen, the influenza virus, is reported in this publication. Annual vaccinations with currently circulating strains are administered to maintain protective, antibody-mediated immunity. Specific IgG+ antibody-secreting plasma cells are here reported to peak at 7 d following a booster vaccination. These cells reach 6% of peripheral B cells, and are readily distinguishable from memory B cells by cell sorting; up to 80% of antibody-secreting plasma cells are influenza specific. Moreover, the response is derived from a highly restricted clonal repertoire of antibodies, although these show a very high rate of somatic mutation (>20 somatic mutations from germ-line sequence per clone). Individual cells from this pool are used to produce over 50 human monoclonal antibodies. The process yields specific, high-affinity antibodies within 30 d of booster vaccination. The method is applicable to the production of therapeutic reagents to prevent the spread of viruses causing a variety of acute infections, hastening recovery from infection.
Junutula JR, Raab H, Clark S, Bhakta S, Leipold DD, Weir S, Chen Y, Simpson M, Tsai SP, Dennis MS, Lu Y, Meng YG, Ng C, Yang J, Lee CC, Duenas E, Gorrell J, Katta V, Kim A, McDorman K, Flagella K, Venook R, Ross S, Spencer SD, Lee Wong W, Lowman HB, Vandlen R, Sliwkowski MX, Scheller RH, Polakis P, Mallet W. Site-specific conjugation of a cytotoxic drug to an antibody improves the therapeutic index. Nature Biotechnology 26;2008:925–932.
Antibodies for targeted therapy of tumors often lack therapeutic activity on their own, so are used as delivery vehicles for cytotoxic drugs. Antibody-drug conjugates show lower systemic toxicity than traditional small-molecule chemotherapeutic agents. Conjugation is normally either through lysine side-chains, or through cysteine suly-dryl groups created by reducing interchain disulfide bonds. Unfortunately, both these approaches yield heterogeneous products with different pharmacologic characteristics. In this work, antibodies are engineered with cysteine residues at positions on the light or heavy chains that do not perturb folding or assembly of the immunoglobulin but are available for drug binding. The reagents are called THIOMABS. One such antibody against an ovarian cancer antigen is conjugated to the microtubule depolymerizing agent monomethyl auristatin E, and is shown to be as efficacious as conventional antibody-drug conjugates and to be better tolerated by experimental animals owing to its lower systemic cytotoxicity.
Petrak J, Ivanek R, Toman O, Cmejla R, Cmejlova J, Vyoral D, Zivny J, Vulpe CD. Déjà vu in proteomics. A hit parade of repeatedly identified differentially expressed proteins. Proteomics 8;2008:1744–1749.
This article probes the general impression that when reviewing the results of proteomic studies based on 2-D gel electrophoresis, certain proteins are repeatedly found to be expressed differentially in a wide variety of different biological contexts. The present authors compile a dataset containing 4700 protein identifications from 169 articles describing 186 2-D gel experiments, and calculate the frequency of occurrence of the most often identified differentially expressed proteins and protein families in human and rodent tissues. There is much commonality: for example, the most frequently identified protein, enolase 1, is differentially expressed in 30% of all 2-D electrophoresis experiments in human and rodent tissues. The question arises as to whether such proteins are genuinely changing expression level or whether the observations are simply artifactual. Analysis of mRNA expression data indicates that the changes are genuine. These observations suggest that the commonly observed protein expression changes may be of fundamental biological interest, and that one’s own results of proteomic studies should be considered in light of the other’s findings.
Hauser NJ, Han H, Mcluckey SA, Basile F. Electron transfer dissociation of peptides generated by microwave D-cleavage digestion of proteins. Journal of Proteome Research 7;2008:1867–1872.
When the temperature of a protein solution at pH < 2.1 is raised rapidly above 108°C by microwave irradiation, polypeptide cleavage specifically at aspartic acid residues becomes frequent. The resulting peptides are generally large and contain one or more basic residues in their sequence. For these reasons, they tend to yield ions with high charge states during electrospray ionization. The fragmentation of such ions by CID is difficult to interpret, but their characteristics would be expected to be well suited to ETD or ECD analysis. This paper demonstrates that peptides generated by microwave D-cleavage are indeed readily amenable to ETD and indicates that this cleavage method provides a rapid and specific alternative to enzymatic cleavage for bottom-up proteomics.
Nielsen ML, Vermeulen M, Bonaldi T, Cox J, Moroder L, Mann M. Iodoacetamide-induced artifact mimics ubiquitination in mass spectrometry. Nature Methods 5;2008:459–460.
Sites of ubiquitination are commonly assigned on the basis of detecting the diglycine adduct (114 Da) from the C-terminus of ubiquitin that remains on target lysine residues after tryptic digestion of the complex. Proteins are frequently prepared for digestion by alkylating cysteine residues with iodoacetamide. Unfortunately, iodoacet-amide treatment may, under some conditions, form covalent adducts with lysine side-chains. These adducts are expected to consist of one acetamide molecule (57 Da). However, the authors of the present article reasoned that acetamidoacetamide adducts (114 Da) might also be formed. The atomic composition of such adducts would be indistinguishable from diglycine. By treating histone proteins from cultured cells with either deuterium-labeled or unlabeled iodoacetamide, they confirmed that such artifacts occur. How widespread the resulting misidentifica-tion of ubiquitination sites might be is unknown. The use of chloracetamide in place of iodoacetamide is suggested because of its greater selectivity for cysteine residues.
Thingholm TE, Larsen MR, Ingrell CR, Kassem M, Jensen ON. TiO2-based phosphoproteomic analysis of the plasma membrane and the effects of phosphatase inhibitor treatment. Journal of Proteome Research 7;2008:3304–3313.
Phosphorylation of plasma membrane proteins initiates many signal transduction pathways. However, the problematic physical characteristics of membrane proteins and the low stoichiometry of protein phosphoryla-tion make such phosphorylation events difficult to analyze. A strategy for handling such analyses is described here. Membrane proteins from the plasma membrane, endoplasmic reticulum, and Golgi apparatus of cultured human mesenchymal stem cells are enriched first by differential centrifugation of membranes in sucrose, and then by extraction with sodium carbonate. After prote-olytic digestion, phosphopeptides are purified by titanium dioxide chromatography and analyzed by LC-MS. Some 703 phosphorylation sites are identified in 376 phospho-proteins. Of particular interest is the effect of various protein phosphatase inhibitors. Cells were incubated with phosphatase inhibitors for 30 min prior to harvesting to maximize phosphorylation levels. Sodium pervanadate, Calyculin A, and the two phosphatase inhibitor cocktails from Sigma were tested. A total gain of 40% in the number of identified phosphoproteins was realized by using these inhibitors. Calcynurin A was particularly effective in increasing the number of serine and threonine phospho-rylation sites, and sodium pervanadate, a tyrosine phos-phatase inhibitor, enabled as many as 60 phosphotyrosine peptides to be identified, the highest number recovered from a cell lysate without enrichment of phosphoproteins with anti-phosphotyrosine antibodies.
Kim S, Gupta N, Pevzner PA. Spectral probabilities and generating functions of tandem mass spectra: A strike against decoy databases. Journal of Proteome Research 7;2008:3345–3363.
This paper distinguishes “false discovery rate,” the proportion of incorrect identifications among all identifications judged correct, and “false positive rate,” the probability that a random peptide matches a particular spectrum with a score equal to or larger than the value given by the scoring function in use. False-discovery rate is a property of multiple spectra that may be measured by searching a decoy database, whereas false-positive rate is a property of an individual spectrum. The authors argue that assigning the same false-positive rate to all identifications with identical scores (the assumption implicit in the decoy database strategy) is a dangerous oversimplification because the scoring functions of existing search tools are not based on rigorous probabilistic models and are often inaccurate, thus failing to provide accurate probabilities for individual spectral matches (most importantly, “one-hit wonders”). The present paper provides the means for rigorous computation of values for the statistical significance of individual spectral identifications, defined as the number of peptides among all possible peptide sequences that match a given spectrum with the score assigned by the search engine. It improves the trade-off between sensitivity and specificity, addresses the problem of one-hit wonders and eliminates the need for decoy database searches. The software is available as open source from http://www.cs.ucsd.edu/users/ppevzner/software.html.
Junqueira M, Spirin V, Balbuena S, Waridel P, Surendranath V, Kryukov G, Adzhubei I, Thomas H, Sunyaev S, Shevchenko A. Separating the wheat from the chaff: Unbiased filtering of background tandem mass spectra improves protein identification. Journal of Proteome Research 7;2008:3382–3393.
A very large proportion of the spectra acquired in LC-MS experiments fail to match to target peptides in database searches. Among these are background spectra that originate from keratins, autolysis products of prote-olytic reagent enzymes, components of culture media and expression host, antibodies, non-peptide species such as plasticizers and detergents, and, in the case of protein interaction experiments, from proteins expressed at high levels but interacting with low affinity. Searching such spectra is time consuming and amplifies the rate of false-positive assignment, especially in homology-based searches. The present paper describes methodology for removing background spectra. It works by comparison with spectra from background libraries compiled from control or blank LC-MS runs. The algorithm computes statistical estimates of dissimilarity between query spectra and library spectra, and does not rely upon any mass spectrometer platform, fragmentation model, or protein sequence resource. It also does not rely upon matching the intensities of peaks. Filtering enhances the speed, sensitivity, and statistical confidence of protein identifications. In sequence similarity searches, it reduces the number of assignments not related to the target proteins or to background contaminant proteins and requiring manual validation, by an average of 30-fold. It decreases the false-positive rate in stringent database searches. The software can be accessed at http://genetics.bhw.harvard.edu/cgi-bin/msfilter/eagleeye.cgi, and a stand-alone version is available upon request.
Lange V, Malmström JA, Didion J, King NL, Johansson BP, Schäfer J, Rameseder J, wong C-H, Deutsch EW, Brusniak M-Y, Bühlmann P, Björck L, Domon B, Aebersold R. Targeted quantitative analysis of Streptococcus pyrogenes virulence factors by multiple reaction monitoring. Molecular and Cellular Proteomics 7;2008:1489–1500.
This paper contains a systematic description of a targeted mass-spectrometric approach to the quantification of a predetermined list of proteins in multiple samples. The proteome is first studied by multidimensional fractionation and tandem mass spectrometry. From the resulting data, peptides are selected that uniquely identify each of the proteins of interest. Multiple reaction monitoring transitions are chosen for the peptide set, and for a corresponding set of heavy isotope–labeled peptides to be used as internal quantification standards. The transitions are validated, and then used for measurements on test samples. The methodology is applied to the quantification of virulence factors of Streptococcus pyrogenes under varying growth conditions. Using a 4000 QTRAP mass spectrometer from Applied Biosystems, six transitions per peptide are monitored for a total of 48 peptides derived from 9 virulence proteins and 12 housekeeping proteins included for normalization purposes.
Hannah MA, Redestig H, Leisse A, Willmitzer L. Global mRNA changes in microarray experiments. Nature Bio-technology 26;2008:741–742.
Most normalization routines for mRNA expression microarray experiments work on the assumption that no changes occur in total mRNA levels. Yet this is frequently not the case. For example, in a test case of cold adaptation in Arabidopsis, a 15% increase occurs in total mRNA. As a consequence, standard expression estimates result in 772 up-regulated genes being missed and 664 genes that did not change being incorrectly classified as down-regulated. For the design of suitable spike-in controls for normalization, the present report exploits the observation that some probe sets on Affymetrix GenChips never seem to record positive hybridization signals. Such probe sets are here stringently identified from the results of about 1000 arrays, and checked to ensure firstly that the sequence spanning the probe set can be amplified from genomic DNA to avoid the need for spike-in cDNA clones, and secondly that the sequences are unlikely to cross-hybridize. Seventy-five out of 94 such sequences from Arabidopsis produced high-quality synthetic RNA controls. The same amount of sample is spiked with varying quantities of the external controls from near the detection limit to near signal saturation so as to improve accuracy by producing standard curves spanning the entire dynamic range. Such controls are simple to generate and are suitable for use with the current generation of arrays that lacks them.
Marioni JC, Mason CE, Mane SM, Stephens M, Gilad Y. RNA-seq: An assessment of technical reproducibility and comparison with gene expression arrays. Genome Research 18;2008:1509–1517.
New ultra-high-throughput DNA sequencing technologies (454 from 454 Life Sciences, Solexa from Illumina, and SOLiD from Applied Biosystems) that enable thousands of megabases to be sequenced in a few days have been employed to study genetic variation, transcription-factor binding sites, and DNA methylation. The technologies are now being applied to the quantitative problem of mRNA expression analysis. The present paper provides a detailed technological assessment of this application; specifically, an estimation of technical variance among replicate samples given by the Solexa system and a comparison of its ability to identify differentially expressed genes with that of conventional array techniques. This is achieved by analyzing gene expression differences between liver and kidney RNA samples. The sequencing data are shown to be highly reproducible, with very little difference between replicate samples in different lanes. Sequencing detects differential expression of 81% of genes with significant expression differences in Affymetrix arrays, and detects 30% more differentially expressed genes than the Affyme-trix arrays at the same false-discovery rate. Evidence is provided suggesting that the expression differences for a large portion of the latter group of genes are genuine. A single lane in the Solexaa system provides information otherwise comparable to a single Affymetrix GeneChip. However, sequencing supports the additional capabilities of detecting genes expressed at very low levels, analyzing alternatively spliced variants, and studying novel transcripts.
Wilhelm BT, Marguerat S, Watt S, Schubert F, Wood V, Goodhead I, Penkett CJ, Rogers J, Bähler J. Dynamic repertoire of a eukaryotic transcriptome surveyed at single-nucleotide resolution. Nature 453;2008:1239–1245.
A number of reports of mRNA expression analysis employing the new ultra-high-throughput DNA sequencing technologies have appeared during the last quarter. (See also commentary by Graveley BR, Nature 453;2008:1197–1198.) Among these, the present paper on the fission yeast, Saccharomyces pombe, describes the broadest dataset, which includes 122 million 39-base-pair sequences, corresponding to 250 genome equivalents. Coverage of known transcripts rises to 99.3% when 5 million bases are sequenced; 453 new transcripts are identified, of which 427 are apparently non-coding. Of particular interest, the ability of the system to sequence through exon-exon or exon-intron junctions provides information indicating that splicing efficiency is regulated during cell proliferation and differentiation. Splicing of 254 genes is found to be more efficient during meiosis. Furthermore, splicing efficiency appears to be correlated with transcription level, so that higher transcription levels are associated with more efficient splicing.
Seeley EH, Oppenheimer SR, Mi D, Chaurand P, Caprioli RM. Enhancement of protein sensitivity for MALDI imaging mass spectrometry after chemical treatment of tissue sections. Journal of the American Society for Mass Spectrometry 19;2008:1069–1077.
Various protocols for washing tissue sections are evaluated for the purpose of detecting proteins by MALDI mass spectrometry. Solvents are compared with respect to the efficiency of lipid removal, maintenance of tissue architecture, total signal strength from proteins, and numbers of protein signals detected. Alcohol-based washes are found to be the most effective, and isopropanol to be optimal among these. Protein detection is shown to be correlated with the efficiency of lipid removal. Alcohol-based washes are also found to preserve tissue sections for protein analysis for several days, sufficient time to allow shipment between laboratories.