Selected reaction monitoring for quantitative proteomics: a tutorial

Systems biology relies on data sets in which the same group of proteins is consistently identified and precisely quantified across multiple samples, a requirement that is only partially achieved by current proteomics approaches. Selected reaction monitoring (SRM)—also called multiple reaction monitoring—is emerging as a technology that ideally complements the discovery capabilities of shotgun strategies by its unique potential for reliable quantification of analytes of low abundance in complex mixtures. In an SRM experiment, a predefined precursor ion and one of its fragments are selected by the two mass filters of a triple quadrupole instrument and monitored over time for precise quantification. A series of transitions (precursor/fragment ion pairs) in combination with the retention time of the targeted peptide can constitute a definitive assay. Typically, a large number of peptides are quantified during a single LC‐MS experiment. This tutorial explains the application of SRM for quantitative proteomics, including the selection of proteotypic peptides and the optimization and validation of transitions. Furthermore, normalization and various factors affecting sensitivity and accuracy are discussed.     

Aligning LC peaks by converting gradient retention times to retention index of peptides in proteomic experiments

MOTIVATION: Liquid chromatography-tandem mass spectrometry (LC-MS/MS) is a powerful tool in proteomics studies, but when peptide retention information is used for identification purposes, it remains challenging to compare multiple LC-MS/MS runs or to match observed and predicted retention times, because small changes of LC conditions unavoidably lead to variability in retention times. In addition, non-contiguous retention data obtained with different LC-MS instruments or in different laboratories must be aligned to confirm and utilize rapidly accumulating published proteomics data. RESULTS: We have developed a new alignment method for peptide retention times based on linear solvent strength (LSS) theory. We found that log k(0) (logarithm of retention factor for a given organic solvent) in the LSS theory can be utilized as a 'universal' retention index of peptides (RIP) that is independent of LC gradients, and depends solely on the constituents of the mobile phase and the stationary phases. We introduced a machine learning-based scheme to optimize the conversion function of gradient retention times (t(g)) to log k(0). Using the optimized function, t(g) values obtained with different LC-MS systems can be directly compared with each other on the RIP scale. In an examination of Arabidopsis proteomic data, the vast majority of retention time variability was removed, and five datasets obtained with various LC-MS systems were successfully aligned on the RIP scale.     

Matrix effects break the LC behavior rule for analytes in LC-MS/MS analysis of biological samples

High-performance liquid chromatography (HPLC) and liquid chromatography-tandem mass spectrometry (LC-MS/MS) are generally accepted as the preferred techniques for detecting and quantitating analytes of interest in biological matrices on the basis of the rule that one chemical compound yields one LC-peak with reliable retention time (Rt.). However, in the current study, we have found that under the same LC-MS conditions, the Rt. and shape of LC-peaks of bile acids in urine samples from animals fed dissimilar diets differed significantly among each other. To verify this matrix effect, 17 authentic bile acid standards were dissolved in pure methanol or in methanol containing extracts of urine from pigs consuming either breast milk or infant formula and analyzed by LC-MS/MS. The matrix components in urine from piglets fed formula significantly reduced the LC-peak Rt. and areas of bile acids. This is the first characterization of this matrix effect on Rt. in the literature. Moreover, the matrix effect resulted in an unexpected LC behavior: one single compound yielded two LC-peaks, which broke the rule of one LC-peak for one compound. The three bile acid standards which exhibited this unconventional LC behavior were chenodeoxycholic acid, deoxycholic acid, and glycocholic acid. One possible explanation for this effect is that some matrix components may have loosely bonded to analytes, which changed the time analytes were retained on a chromatography column and interfered with the ionization of analytes in the MS ion source to alter the peak area. This study indicates that a comprehensive understanding of matrix effects is needed towards improving the use of HPLC and LC-MS/MS techniques for qualitative and quantitative analyses of analytes in pharmacokinetics, proteomics/metabolomics, drug development, and sports drug testing, especially when LC-MS/MS data are analyzed by automation software where identification of an analyte is based on its exact molecular weight and Rt.     

蛋白质组学中色谱保留时间对齐算法的研究进展北大核心CSCD

色谱是目前蛋白质组学流程中的一个基本环节,而色谱的保留时间对齐是有效提高鉴定和定量准确性的重要步骤之一。经过多年的发展,目前已经产生了一系列保留时间对齐算法。文中主要从可用性角度对蛋白质组学分析中色谱保留时间对齐算法及工具进行了系统总结,并对其发展趋势及应用方向进行了展望。     

Determination of dextropropoxyphene and nordextropropoxyphene in urine by liquid chromatography-electrospray ionization mass spectrometry

Dextropropoxyphene and nordextropropoxyphene were extracted from urine samples with mixed mode solid-phase extraction cartridges. After elution and evaporation to dryness, the eluate was dissolved in mobile phase and each sample was injected in a LC-ESI-MS system. Quantification was carried out in the selected ion monitoring mode. This article shows the possibility to analyse drugs of abuse substances in urine with a single quadrupole mass spectrometer if only a thorough work-up procedure and a sufficient chromatographic separation is accomplished. In order to enhance the fragmentation of the analytes, in-source fragmentation was carried out. One fragment and the pseudomolecular ion per analyte together with chromatographic retention times were sufficient to verify that the sought compound was found in the samples. In- and between day variation was lower than 10% and the recovery was well above 90%. The analytes were quantified in the range 100-10000 ng/ml urine.     

Data filtration for more robust alignment of chromatograms of complex peptide mixtures

In carrying out proteomic researches using mass-spectrometry there often arises a need to compare experimental data with each other (e.g. control of pathology, the labeled to unlabelled samples). If for peptide identification in different experiments one uses only their exact mass measurements and the retention time in the chromatographic column, difficulties with the identification of chromatographic peaks belonging to the same substances in different chromatograms come up (retention time normalization). Due to inevitable discrepancies in chromatographic conditions of experiments (replacement of chromatographic columns, small changes in mobile phase flow rate or solvent concentration) retention times of the same peptides will diverge from experiment to experiment. In this paper we offer a reliable method for selecting peaks from mass-chromatograms corresponding to the same peptides, which can later be used for retention time normalization (either linear or any other monotone function).     

Informatics for peptide retention properties in proteomic LC-MS

Retention times in HPLC yield valuable information for the identification of various analytes and the prediction of peptide retention is useful for the identification of peptides/proteins in LC-MS-based proteomics. Informatics methods such as artificial neural networks and support vector machines capable of solving nonlinear problems made possible the accurate modeling of quantitative structure-retention relationships of peptides (including large polymers) up to 5 kDa to which classical linear models cannot be applied, as well as the proteome-wide prediction of peptide retention. Proteome-wide retention prediction and accurate mass-information facilitate the identification of peptides in complex proteomic samples. In this review, we address recent developments in solid informatics methods and their application to peptide-retention properties in 'bottom-up' shotgun proteomics. We also describe future prospects for the standardization and application of retention times.     

Systematic and comprehensive strategy for reducing matrix effects in LC/MS/MS analyses

A systematic, comprehensive strategy that optimizes sample preparation and chromatography to minimize matrix effects in bioanalytical LC/MS/MS assays was developed. Comparisons were made among several sample preparation methods, including protein precipitation (PPT), liquid-liquid extraction (LLE), pure cation exchange solid-phase extraction (SPE), reversed-phase SPE and mixed-mode SPE. The influence of mobile phase pH and gradient duration on the selectivity and sensitivity for both matrix components and basic analytes was investigated. Matrix effects and overall sensitivity and resolution between UPLC technology and HPLC were compared. The amount of specific matrix components, or class of matrix components, was measured in the sample preparation extracts by LC/MS/MS with electrospray ionization (ESI) using both precursor ion scanning mode and multiple reaction monitoring (MRM). PPT is the least effective sample preparation technique, often resulting in significant matrix effects due to the presence of many residual matrix components. Reversed-phase and pure cation exchange SPE methods resulted in cleaner extracts and reduced matrix effects compared to PPT. The cleanest extracts, however, were produced with polymeric mixed-mode SPE (both reversed-phase and ion exchange retention mechanisms). These mixed-mode sorbents dramatically reduced the levels of residual matrix components from biological samples, leading to significant reduction in matrix effects. LLE also provided clean final extracts. However, analyte recovery, particularly for polar analytes, was very low. Mobile phase pH was manipulated to alter the retention of basic compounds relative to phospholipids, whose retention tends to be relatively independent of pH. In addition to the expected resolution, speed and sensitivity benefits of UPLC technology, a paired t-test demonstrated a statistically significant improvement with respect to matrix effects when this technology was chosen over traditional HPLC. The combination of polymeric mixed-mode SPE, the appropriate mobile phase pH and UPLC technology provides significant advantages for reducing matrix effects resulting from plasma matrix components and in improving the ruggedness and sensitivity of bioanalytical methods.     

Simultaneous determination of dibucaine and naphazoline in human serum by monolithic silica spin column extraction and liquid chromatography-mass spectrometry

A simple, sensitive, and specific liquid chromatography-mass spectrometry (LC-MS) method for simultaneous determination of dibucaine and naphazoline from serum was developed and validated. The extraction procedure was performed using a monolithic silica spin column. Chromatographic separation of dibucaine and naphazoline was achieved on a C(18) reverse phase column with a mobile phase gradient (mobile phase A: 10mM ammonium formate and mobile phase B: acetonitrile) at a flow rate of 0.2mL/min. LC-MS was operated under the selective ion monitoring mode using the electrospray ionization technique in the positive mode. The retention times for naphazoline, dibucaine, and the internal standard (IS) were 6.7, 7.8, and 8.0min, respectively. A linear graph was obtained for dibucaine and naphazoline with correlation coefficients >0.998 for all analytes by this method. The limit of quantification of dibucaine and naphazoline was 10 and 25ng/mL, respectively. The mean recoveries were greater than 70%. Both compounds were stable under conditions of short-term storage, long-term storage as well as after freeze-thaw cycles. Monolithic spin column extraction and LC-MS analysis enabled the separation of dibucaine and naphazoline within 20min.     

Thermodynamics of interactions between amino acid side chains: experimental differentiation of aromatic-aromatic, aromatic-aliphatic, and aliphatic-aliphatic side-chain interactions in water

A stationary phase for high-pressure liquid chromatography has been prepared by derivatizing microparticulate silica gel with functionality mimicking the side chain of isoleucine. The chromatographic retentions of a series of hydrophobic and amphiphilic amino acid analytes on this stationary phase (Ile MSP) using an aqueous mobile phase were measured as a function of temperature from 273 K to 323 K. Observed temperature dependencies are consistent with a constant change in heat capacity, DeltaC degrees P, upon binding of the analyte to the stationary phase. The curvatures of plots of retention data versus temperature (related to the magnitude of DeltaC degrees P) are distinctly different for retention of aromatic and aliphatic analytes, with retention of aliphatic analytes Val, Ile, and Leu exhibiting the characteristic signature of the hydrophobic effect, i.e., a large negative DeltaC degrees P upon desolvation from water and a maximum of retention around room temperature. Retention of aromatic analytes (Trp, Phe, and Tyr) involves smaller heat capacity changes and pronounced negative enthalpies of interaction with the stationary phase. Estimates of DeltaC degrees P for the interactions of analyte side chains with the Ile side chain were obtained by fitting the temperature dependence of retention to an expression derived from thermodynamic considerations and chromatographic theory. Similar estimates were made for interactions with the Phe side chain, using previously published data for a phenylalanine mimic stationary phase (Phe MSP) (. Protein Sci. 1:786-795). As with the Ile MSP, the retentions of aliphatic analytes show temperature dependencies markedly different from those of aromatic analytes. Data from both phases indicate that a realistic differentiation can be made between the interactions of various types of amino acid side chains tested (i.e., aliphatic/aliphatic, aliphatic/aromatic, and aromatic/aromatic) by comparison of the corresponding thermodynamic functions for pairwise interactions. The retention of leucine on the Phe MSP and that of phenylalanine on the Ile MSP showed similar DeltaC degrees P values, suggesting that the aromatic-aliphatic interaction is reasonably independent of the residue attached to the stationary phase. This result is consistent with a one-to-one interaction and suggests a simple way to estimate the column-dependent phase factor, making it possible to compare entropies and free energies of interaction obtained using different MSPs. The possibilities for using MSP-derived interaction potentials in folding simulations are discussed.     

Flow field-flow fractionation: a pre-analytical method for proteomics

Proteome analysis requires a comprehensive approach including high-performance separation methods, mass spectrometric analysis, and bioinformatics. While recent advances in mass spectrometry (MS) have led to remarkable improvements in the ability to characterize complex mixtures of biomolecules in proteomics, a proper pre-MS separation step of proteins/peptides is still required. The need of high-performance separation and/or isolation/purification techniques of proteins is increasing, due to the importance of proteins expressed at extremely low levels in proteome samples. In this review, flow field-flow fractionation (F4) is introduced as a complementary pre-analytical separation method for protein separation/isolation, which can be effectively utilized for proteomic research. F4 is a set of elution-based techniques that are capable of separating macromolecules by differences in diffusion coefficient and, therefore, in hydrodynamic size. F4 provides protein separation without surface interaction of the analyte with packing or gel media. Separation is carried out in an open channel structure by a flow stream of a mobile phase of any composition, and it is solely based on the interaction of the analytes with a perpendicularly-applied, secondary flow of the fluid. Therefore, biological analytes such as proteins can be kept under a bio-friendly environment without losing their original structural configuration. Moreover, proteins fractionated on a size/shape basis can be readily collected for further characterization or proteomic analysis by MS using, for instance, either on-line or off-line methods based on electrospray ionization (ESI) or matrix-assisted laser desorption-ionization (MALDI). This review focuses on the advantages of F4 compared to most-assessed separation/isolation techniques for proteomics, and on selected applications based on size-dependent proteome separation. New method developments based on the hyphenation of F4 with on-line or off-line MS, and with other separation methods such as capillary isoelectric focusing (CIEF) are also described.     

Identification of amygdalin and its major metabolites in rat urine by LC-MS/MS

Amygdalin and its metabolites in rat urine were identified using liquid chromatography-electrospray ionization (ESI) tandem ion-trap mass spectrometry. The purified rat urine sample was separated using a reversed-phase C18 column with 10 mM sodium phosphate buffer (pH 3.1) containing 30% methanol as the mobile phase, amygdalin and its metabolites were detected by on-line mass detector in selected ion monitoring (SIM) mode. The identification of the metabolites and elucidation of their structure were performed by comparing the changes in molecular masses (DeltaM), retention times and MS(2) spectral patterns of metabolites with those of parent drug. At least seven metabolites and the parent drug were found in rat urine after i.v. injection of 100 mg/kg doses of amygdalin. Among them, six metabolites were reported for the first time.     

Determination of picropodophyllin and its isomer podophyllotoxin in human serum samples with electrospray ionization of hexylamine adducts by liquid chromatography-tandem mass spectrometry

A liquid chromatography-tandem mass spectrometry (LC-MS/MS) method for determination of the new anticancer agent picropodophyllin (AXL1717) and its isomer podophyllotoxin levels in human serum has been developed. Monitoring of hexylamine adducts rather than proton adducts was used to optimize sensitivity. The chromatography system was an Acquity BEH C18, 2.1 mm × 50 mm 1.7 μm column with gradient elution (mobile phase A: 2.5 mM hexylamine and 5 mM formic acid in Milli-Q water and mobile phase B: methanol). The retention times were 1.4 min for picropodophyllin, 1.5 min for podophyllotoxin and 1.9 min for internal standard deoxypodophyllotoxin. The isomers were base-line separated. The analytes were detected after electrospray ionization in positive mode with selected reaction monitoring (SRM) with ion transitions m/z 516→102 for picropodophyllin and podophyllotoxin and m/z 500→102 for internal standard. The sample preparation was protein precipitation with acetonitrile (1:3) containing internal standard followed by dilution of the supernatant with mobile phase A (1:1). The limit of quantification (LOQ) was 0.01 μmol/L for picropodophyllin and podophyllotoxin. The limit of detection (LOD) at 3 times the signal to noise (S/N) was estimated below 0.001 μmol/L for picropodophyllin and podophyllotoxin. The quantification range of the method was between 0.01 μmol/L and 5 μmol/L for both isomers. The accuracy was within ±15% of the theoretical value for both picropodophyllin and podophyllotoxin and inter-assay precision did not exceed ±15%, except for the 0.016 μmol/L level of podophyllotoxin, which was 18%. The selectivity of the method was verified by analysis of two different product ions for each analyte and by analysis for interference of seven different batches of blank human serum. The combined recovery and matrix effects were about 83% for picropodophyllin and podophyllotoxin. The new LC-MS/MS method showed sufficient sensitivity and selectivity for determination of picropodophyllin and its isomer podophyllotoxin levels in human serum from subjects receiving therapeutic doses of AXL1717.     

Identification of dihydroetorphine in biological fluids by gas chromatography-mass spectrometry

A method for the monitoring of dihydroetorphine hydrochloride, a powerful anaesthetic and analgesic drug, in biological fluids was developed, involving GC-MS with multiple selected-ion monitoring. Dihydroetorphine was extracted from human blood and urine with dichloromethane and then derivatized with N-heptafluorobutyrylimidazole after having been concentrated to dryness. A dihydroetorphine monoheptafluorobutyl derivative was formed, which showed good behaviour in GC-MS with electron impact ionization. Its molecular ion, m/z 609, and its main fragments, m/z 576, 534, 522 and 508, were selected as the ions for identification owing to their relative peak intensities and characteristics. The target drug was identified based on its retention time, its selected multiple ions and their relative intensities. This method was successfully used for the detection of dihydroetorphine in blood and urine from a dihydroetorphine addict and a poisoned patient, respectively.     

Quantitative analysis of cocaine and its metabolites in whole blood and urine by high-performance liquid chromatography coupled with tandem mass spectrometry

AIM: In forensic toxicology it is important to have specific and sensitive analysis for quantification of illicit drugs in biological matrices. This paper describes a quantitative method for determination of cocaine and its major metabolites (ecgonine methyl ester, benzoylecgonine, norcocaine and ethylene cocaine) in whole blood and urine by liquid chromatography coupled with tandem mass spectrometry LC/MS/MS. METHOD: The sample pre-treatment (0.20 g) consisted of acid precipitation, followed by centrifugation and solid phase extraction of supernatant using mixed mode sorbent columns (SPEC MP1 Ansys Diag. Inc.). Chromatographic separation was performed at 30 degrees C on a reverse phase Zorbax C18 column with a gradient system consisting of formic acid, water and acetonitrile. The analysis was performed by positive electrospray ionisation with a triple quadropole mass spectrometer operating in multiple reaction monitoring (MRM) mode. Two MRM transitions of each analyte were established and identification criteria were set up based on the retention time and the ion ratio. The quantification was performed using deuterated internal analytes of cocaine, benzoylecgonine and ecgonine methyl ester. The calibration curves of extracted standards were linear over a working range of 0.001-2.00 mg/kg whole blood for all analytes. The limit of quantification was 0.008 mg/kg; the interday precision (measured by relative standard deviation-%RSD) was less than 10% and the accuracy (BIAS) less than 12% for all analytes in whole blood. Urine samples were estimated semi-quantitatively at a cut-off level of 0.15 mg/kg with an interday precision of 15%. CONCLUSION: A liquid chromatography mass spectrometric (LC/MS/MS) method has been developed for confirmation and quantification of cocaine and its metabolites (ecgonine methyl ester, benzoylecgonine, norcocaine and ethylene cocaine) in whole blood and semi-quantitative in urine. The method is specific and sensitive and offers thereby an excellent alternative to other methods such as GC-MS that involves derivatisation.     

Mobile phase modifier effects in multimodal cation exchange chromatography

This study examines protein adsorption behavior and the effects of mobile phase modifiers in multimodal chromatographic systems. Chromatography results with a diverse protein library indicate that multimodal and ion exchange resins have markedly different protein binding behavior and selectivity. NMR results corroborate the stronger binding observed for the multimodal system and provide insight into the structural basis for the observed binding behavior. Protein-binding affinity and selectivity in multimodal and ion exchange systems are then examined using a variety of mobile phase modifiers. Arginine and guanidine are found to have dramatic effects on protein adsorption, yielding changes in selectivity in both chromatographic systems. While sodium caprylate leads to slightly weaker chromatographic retention for most proteins, certain proteins exhibit significant losses in retention in both systems. The presence of a competitive binding mechanism between the multimodal ligand and sodium caprylate for binding to ubiquitin is confirmed using STD NMR. Polyol mobile phase modifiers are shown to result in increased retention for weakly bound proteins and decreased retention for strongly bound proteins, indicating that the overall retention behavior is determined by a balance between changes in electrostatic and hydrophobic interactions. This work provides an improved understanding of protein adsorption and mobile phase modifier effects in multimodal chromatographic systems and sets the stage for future work to develop more selective protein separation systems.     

Reproducibility of LC-MS-based protein identification

Traditional analysis of liquid chromatography-mass spectrometry (LC-MS) data, typically performed by reviewing chromatograms and the corresponding mass spectra, is both time-consuming and difficult. Detailed data analysis is therefore often omitted in proteomics applications. When analysing multiple proteomics samples, it is usually only the final list of identified proteins that is reviewed. This may lead to unnecessarily complex or even contradictory results because the content of the list of identified proteins depends heavily on the conditions for triggering the collection of tandem mass spectra. Small changes in the signal intensity of a peptide in different LC-MS experiments can lead to the collection of a tandem mass spectrum in one experiment but not in another. Also, the quality of the tandem mass spectrometry experiments can vary, leading to successful identification in some cases but not in others. Using a novel image analysis approach, it is possible to achieve repeat analysis with a very high reproducibility by matching peptides across different LC-MS experiments using the retention time and parent mass over charge (m/z). It is also easy to confirm the final result visually. This approach has been investigated by using tryptic digests of integral membrane proteins from organelle-enriched fractions from Arabidopsis thaliana and it has been demonstrated that very highly reproducible, consistent, and reliable LC-MS data interpretation can be made.     

Chromatographic study of magnesium and calcium binding to immobilized human serum albumin

The use of immobilized human serum albumin (HSA) as a stationary phase in affinity chromatography has been shown to be useful in resolving optical antipodes or to investigate interactions between drugs and protein. However, to our knowledge, no inorganic ion binding has been studied on this immobilized protein type. To do this, the human serum albumin stationary phase was assimilated to a weak cation-exchanger by working with a mobile phase pH equal to 6.5. A study of the eluent ionic strength effect on ion retention was carried out by varying the buffer concentrations and the column temperatures. The thermodynamic parameters for magnesium and calcium transfer from the mobile to the stationary phase were determined from linear van’t Hoff plots. An enthalpy–entropy compensation study revealed that the type of interaction was independent of the mobile phase composition. A simple model based on the Gouy–Chapman theory was considered in order to describe the retention behavior of the test cations with the mobile phase ionic strength. From this theoretical approach, the relative charge densities of the human serum albumin surface implied in the binding process were estimated at different column temperatures.     

Aqueous normal phase chromatography improves quantification and qualification of homocysteine, cysteine and methionine by liquid chromatography-tandem mass spectrometry

Elevation of plasma homocysteine concentration is recognized as an independent predictor of cardiovascular disease risk. Therefore, quantification of homocysteine and related sulphur amino acids cysteine and methionine from plasma samples is routinely performed in clinical laboratories. Due to the highly hydrophilic character of these amino acids, previously reported LC-MS methods often suffered from very short chromatographic retention resulting in inadequate separation from matrix background and possible co-eluents. In the present method, aqueous normal phase (ANP) chromatography was introduced to improve chromatographic separation for liquid chromatography-electrospray ionization tandem mass spectrometry. Selective qualification of analytes and internal standards was achieved by qualifier ion monitoring. Using this enhanced selectivity, spurious co-eluents were identified and separated from the analyte signal by optimization of chromatographic conditions. Method validation proved high precision and accuracy (intra-assay reproducibility 1.2-4.3% CV, inter-assay reproducibility 3.4-6.1% CV, accuracy 91.3-105.9%). Total cycle time of 7 min and low costs per sample allow high-throughput application in clinical diagnostics and research trials.     

Proteomics-based consensus prediction of protein retention in a bacterial membrane

The availability of complete bacterial genome sequences allows proteome-wide predictions of exported proteins that are potentially retained in the cytoplasmic membranes of the corresponding organisms. In practice, however, major problems are encountered with the computer-assisted distinction between (Sec-type) signal peptides that direct exported proteins into the growth medium and lipoprotein signal peptides or amino-terminal membrane anchors that cause protein retention in the membrane. In the present studies, which were aimed at improving methods to predict protein retention in the bacterial cytoplasmic membrane, we have compared sets of membrane-attached and extracellular proteins of Bacillus subtilis that were recently identified through proteomics approaches. The results showed that three classes of membrane-attached proteins can be distinguished. Two classes include 43 lipoproteins and 48 proteins with an amino-terminal transmembrane segment, respectively. Remarkably, a third class includes 31 proteins that remain membrane-retained despite the presence of typical Sec-type signal peptides with consensus signal peptidase recognition sites. This unprecedented finding indicates that unknown mechanisms are involved in membrane retention of this class of proteins. A further novelty is a consensus sequence indicative for release of certain lipoproteins from the membrane by proteolytic shaving. Finally, using non-overlapping sets of secreted and membrane-retained proteins, the accuracy of different signal peptide prediction algorithms was assessed. Accuracy for the prediction of protein retention in the membrane was increased to 82% using a majority-vote approach. Our findings provide important leads for future identification of surface proteins from pathogenic bacteria, which are attractive candidate infection markers and potential targets for drugs or vaccines.