Yichu Shan

Application of probabilistic neural network in the clinical diagnosis of cancers based on clinical chemistry data

Abstract As a recently developed and powerful classification tool, probabilistic neural network was used to distinguish cancer patients from healthy persons according to the levels of nucleosides in human urine. Two datasets (containing 32 and 50 patterns, respectively) were investigated and the total consistency rate obtained was 100% for dataset 1 and 94% for dataset 2. To evaluate the performance of probabilistic neural network, linear discriminant analysis and learning vector quantization network were also applied to the classification problem. The results showed that the predictive ability of the probabilistic neural network is stronger than the others in this study. Moreover, the recognition rate for dataset 2 can achieve to 100% if combining these three methods together, which indicated the promising potential of clinical diagnosis by combining different methods.

Organic-Inorganic Hybrid Silica Monolith Based Immobilized Titanium Ion Affinity Chromatography Column for Analysis of Mitochondrial Phosphoproteome

A novel kind of immobilized metal affinity chromatography (IMAC) column based on organic-inorganic hybrid silica monolith has been developed. The monolithic support was prepared in a 250 microm i.d. capillary by the sol-gel method with tetraethoxysilane (TEOS) and 3-aminopropyltriethoxysilane (APTES) as precursors. Subsequently, amine groups were functionalized by glutaraldehyde, and then activated with (aminomethyl) phosphonic acid, followed by Ti(4+) chelation. By such a hybrid silica monolithic Ti(4+)-IMAC column, 15 phosphopeptides were effectively isolated from the digest mixture of alpha-casein and BSA with the molar ratio as low as 1:200, illustrating its superior selectivity. With a synthetic phosphorylated peptide, YKVPQLEIVPNSpAEER, as the sample, the loading capacity and recovery of the Ti(4+)-IMAC monolithic column were measured to be 1.4 micromol/mL and 69%, respectively. Such an IMAC monolithic column was further applied to enrich phosphopeptides from rat liver mitochondria. In total, 224 unique phosphopeptides, corresponding to 148 phosphoprotein groups, were identified by duplicate nanoRPLC-LTQ MS/MS/MS runs with a false-positive rate of less than 1% at the peptide level. These results demonstrate that the hybrid silica monolith based Ti(4+)-IMAC column might provide a promising tool for large-scale phosphopeptide enrichment, facilitating the in-depth understanding of the biological functions of phosphoproteomes.

Integrated Device for Online Sample Buffer Exchange, Protein Enrichment, and Digestion

An integrated sample treatment device, composed of a membrane interface and a monolithic hybrid silica based immobilized enzymatic reactor (IMER), was developed for the simultaneous sample buffer exchange, protein enrichment, and online digestion, by which for the sample buffer, the acetonitrile content was reduced to approximately 1/10 of the initial one, and the pH value was adjusted from approximately 3.0 to approximately 8.0, compatible for online trypsin digestion. Furthermore, the signal intensity of myoglobin digests was improved by over 10 times. Such an integrated device was successfully applied to the online treatment of three protein eluates obtained by reverse-phase liquid chromatography (RPLC) separation, followed by further protein digest analysis with microreverse-phase liquid chromatography-electrospray ionization-tandem mass spectrometry (microRPLC-ESI-MS/MS). The experimental results showed that the performance of such an integrated sample treatment device was comparable to that of the traditional offline sample treatment method, including lyophilization and in-solution digestion. However, the consumed time was reduced to 1/192. All these results demonstrate that such an integrated sample treatment device could be further online coupled with protein separation, peptide separation, and identification, to achieve high-throughput proteome analysis.

Coupling Formic Acid Assisted Solubilization and Online Immobilized Pepsin Digestion with Strong Cation Exchange and Microflow Reversed-Phase Liquid Chromatography with Electrospray Ionization Tandem Mass Spectrometry for Integral Membrane Proteome Analysis

In this study, a facile system for membrane proteome profiling was established, in which membrane proteins were solubilized by formic acid, online digested by a pepsin-based immobilized enzyme reactor (pepsin-IMER), and analyzed by strong cation exchange and microflow reversed-phase liquid chromatography with electrospray ionization tandem mass spectrometry (SCX-μRPLC-ESI-MS/MS). Under optimized conditions, such a system showed excellent compatibility between all crucial steps and was successfully applied for analyzing integral membrane proteins extracted from rat liver microsomes. Out of the 235 unique proteins positively identified, 39% (91/235) were annotated as membrane proteins with one or more transmembrane domains (TMDs). It is anticipated that the efficient sample treatment and the relevant online analytical system might provide a promising tool for automated and comprehensive profiling of membrane proteomes.

Integration of capillary isoelectric focusing with monolithic immobilized pH gradient, immobilized trypsin microreactor and capillary zone electrophoresis for on‐line protein analysis

An integrated platform consisting of protein separation by CIEF with monolithic immobilized pH gradient (M-IPG), on-line digestion by trypsin-based immobilized enzyme microreactor (trypsin-IMER), and peptide separation by CZE was established. In such a platform, a tee unit was used not only to connect M-IPG CIEF column and trypsin-IMER, but also to supply adjustment buffer to improve the compatibility of protein separation and digestion. Another interface was made by a Teflon tube with a nick to couple IMER and CZE via a short capillary, which was immerged in a centrifuge tube filled with 20  mmol/L glutamic acid, to exchange protein digests buffer and keep electric contact for peptide separation. By such a platform, under the optimal conditions, a mixture of ribonuclease A, myoglobin and BSA was separated into 12 fractions by M-IPG CIEF, followed by on-line digestion by trypsin-IMER and peptide separation by CZE. Many peaks of tryptic peptides, corresponding to different proteins, were observed with high UV signals, indicating the excellent performance of such an integrated system. We hope that the CE-based on-line platform developed herein would provide another powerful alternative for an integrated analysis of proteins.

Mass Defect-Based Pseudo-Isobaric Dimethyl Labeling for Proteome Quantification

Discovering differentially expressed proteins in various biological samples requires proteome quantification methods with accuracy, precision, and wide dynamic range. This study describes a mass defect-based pseudo-isobaric dimethyl labeling (pIDL) method based on the subtle mass defect differences between (12)C/(13)C and (1)H/(2)H. Lys-C protein digests were labeled with CD2O/(13)CD2O and reduced with NaCNBD3/NaCNBH3 as heavy and light isotopologues, respectively. The fragment ion pairs with mass differences of 5.84 mDa were resolved by high-resolution tandem mass spectrometry (MS/MS) and used for quantification. The pIDL method described here resulted in highly accurate and precise quantification results with approximately 100-fold dynamic range. Furthermore, the pIDL method was extended to 4-plex proteome quantification and applied to the quantitative analysis of proteomes from Hca-P and Hca-F, two mouse hepatocarcinoma ascites syngeneic cell lines with low and high lymph node metastasis rates.

Recent advances in stable isotope labeling based techniques for proteome relative quantification

The large scale relative quantification of all proteins expressed in biological samples under different states is of great importance for discovering proteins with important biological functions, as well as screening disease related biomarkers and drug targets. Therefore, the accurate quantification of proteins at proteome level has become one of the key issues in protein science. Herein, the recent advances in stable isotope labeling based techniques for proteome relative quantification were reviewed, from the aspects of metabolic labeling, chemical labeling and enzyme-catalyzed labeling. Furthermore, the future research direction in this field was prospected.

Protein separation using free‐flow electrophoresis microchip etched in a single step

A one-step etching method was developed to fabricate glass free-flow electrophoresis microchips with a rectangle separation microchamber (42 mm-long, 23 mm-wide and 28 microm-deep), in which two glass bridges (0.5 mm-wide) were made simultaneously to prevent bubbles formed by electrolysis near the Pt electrode from entering the separation chamber. By microchip free-flow zone electrophoresis, with 200 V voltage applied, the baseline separation of three FITC labeled proteins, ribonuclease B, myoglobin and beta-lactoglobulin, was achieved, with resolution over 1.78. Furthermore, with 2.5 mM Na(2)SO(4) added into the electrode buffer to form higher electrical field strength across separation microchamber than electrode compartments, similar resolution of samples was achieved with the applied voltage decreased to 75 V, which could obviously decrease Joule heat during continuous separation. All these results demonstrate that the free-flow electrophoresis microchip fabricated by one-step etching method is suitable for the continuous separation of proteins, which might become an effective pre-fractionation method for proteome study.

Monoliths with immobilized zirconium ions for selective enrichment of phosphopeptides

To meet the demands of protein phosphorylation study, immobilized zirconium ion affinity chromatography (Zr(4+)-IMAC) monolith was prepared by combining UV-initiated polymerization of monolithic support and subsequent photografting in both capillary columns and microchannels. Hydrophilic poly(2-hydroxyethyl methacrylate (HEMA)-co-ethylene dimethacrylate (EDMA)) monolithic support was prepared under UV irradiation at the wavelength of 365 nm with monomer HEMA, crosslinker EDMA and 2,2-dimethoxy-2-phenylacetophenone as photoinitiator in 1-decanol solution, which provides good biocompatibility and permeability for biomolecule analysis. To introduce chelating ligands, such as phosphate groups, on the pore surface of monolith for metal ion immobilization, photografting of ethylene glycol methacrylate phosphate with benzophenone as the photoinitiator was performed at 254 nm for 300 s. The grafting process and metal ion immobilization can be monitored by measuring the electroosmotic flow produced by the modified monolith, providing a quantitative evaluation of post-modification. This new method for the preparation of Zr(4+)-IMAC monolith simplifies the optimization of monolith preparation and avoids the time-consuming chemical modification process. Additionally, advantages include facile preparation in microdevices, easy regenerability and good reproducibility. After optimization, the microchip-based Zr(4+)-IMAC monolith was used for phosphopeptide analysis and showed good selectivity in phosphopeptide enrichment with matrix-assisted laser desorption ionization mass spectrometry detection.

Development of a highly efficient 2-D system with a serially coupled long column and its application in identification of rat brain integral membrane proteins with ionic liquids-assisted solubilization and digestion.

Two dimensional high performance liquid chromatography-electrospray ionization-tandem mass spectrometry (2D-HPLC-ESI-MS/MS) is one of the most powerful techniques for high resolution, efficiency, and throughput separation and identification of proteomes. For a bottom-up strategy-based proteome analysis, usually multistep salt elution was needed in the first dimension separation by SCX, to simplify the peptides for the further second dimensional separation by RPLC. Here, by using a 30 cm-long serially coupled long column (SCLC) in the second dimension, we reduced the salt steps of SCX from 13 to 5 to shorten the total analysis time. Compared to the commonly applied 2D-HPLC with over 10-step salt elution in SCX and microRPLC with a short column (SC), named as SC-2D, the peak capacity of 2D-HPLC with a SCLC column, named as SCLC-2D, was increased 3.3-folds while the analysis time was increased by only 1.17-folds. Therefore, the time-based protein identification efficiency was ∼55 protein groups/h, nearly 2-fold of that for SC-2D (∼28 protein groups/h). With the further combination of assisted solubilization by ionic liquids and SCLC-2D, 608 integral membrane proteins (IMPs) (27.66% of the total 2198 proteins, FDR < 1%) were identified from rat brain, more than those obtained by the traditional urea method (252 unique IMPs, occupying 17.03% of total 1480 proteins). All of these results demonstrate the promise of the developed technique for large-scale proteome analysis.