Dukjin Kang

Proteomic Analysis of Exosomes from Human Neural Stem Cells by Flow Field-Flow Fractionation and Nanoflow Liquid Chromatography-Tandem Mass Spectrometry

Exosomes, small membrane vesicles secreted by a multitude of cell types, are involved in a wide range of physiological roles such as intercellular communication, membrane exchange between cells, and degradation as an alternative to lysosomes. Because of the small size of exosomes (30-100 nm) and the limitations of common separation procedures including ultracentrifugation and flow cytometry, size-based fractionation of exosomes has been challenging. In this study, we used flow field-flow fractionation (FlFFF) to fractionate exosomes according to differences in hydrodynamic diameter. The exosome fractions collected from FlFFF runs were examined by transmission electron microscopy (TEM) to morphologically confirm their identification as exosomes. Exosomal lysates of each fraction were digested and analyzed using nanoflow LC-ESI-MS-MS for protein identification. FIFFF, coupled with mass spectrometry, allows nanoscale size-based fractionation of exosomes and is more applicable to primary cells and stem cells since it requires much less starting material than conventional gel-based separation, in-gel digestion and the MS-MS method.

Separation of mitochondria by flow field-flow fractionation for proteomic analysis

Flow field-flow fractionation (FlFFF) has been utilized for size-based separation of rat liver mitochondria. Collected fractions of mitochondria of various sizes were examined by confocal microscopy, and mitochondria of each fraction were lysed and analyzed by two-dimensional polyacrylamide gel electrophoresis (2D-PAGE) for the comparison of protein patterns in differently sized mitochondria by densitometric measurements, and for protein characterization of some gel spots with nanoflow liquid chromatography-electrospray ionization-tandem mass spectrometry (nLC-ESI-MS-MS). FlFFF fractions of the mitochondria were also tryptically digested for shotgun proteomic characterization of mitochondrial proteins/peptides by nLC-ESI-MS-MS. Peak area (integrated ion counts) of some peptides extracted from LC-MS chromatograms were examined at different fractions for the quantitative comparison. Among 130 proteins, 105 unique proteins were found to be mitochodrial from the off-line combination of FlFFF and nLC-ESI-MS-MS analysis. It also showed that 23 proteins were found in all fractions but some proteins were found exclusively in certain fractions. Among 25 proteins listed from other subcellular species, seven proteins were known to exist in mitochondria as well as in other subcellular locations, which may support the possible translocation or multiple localizations of proteins among organelles. This study demonstrated effective use of FlFFF for the isolation and/or enrichment of intact mitochondria isolated from cells, as well as its potential use for the fractionation of other subcellular components in the framework of subcellular functional proteomics.

Field and flow programming in frit-inlet asymmetrical flow field-flow fractionation.

The separation of wide molecular mass (Mr) ranges of macromolecules using frit inlet asymmetrical flow field-flow fractionation (FI-AFlFFF) has been improved by implementing a combination of field and flow programming. In this first implementation, field strength (governed by the cross flow-rate through the membrane-covered accumulation wall) is decreased with time to obtain faster elution and improved detection of the more strongly retained (high Mr) materials. The channel outlet flow-rate is optionally held constant, increased, or decreased with time. With circulation of the flow exiting the accumulation wall to the inlet frit, the dual programming of cross flow and channel outlet flow could be implemented using just two pumps. With this flow configuration, the channel outlet flow-rate is always equal to the channel inlet flow-rate, and these may be programmed independently of the cross flow-rate through the membrane. FI-AFlFFF retains its operational advantage over conventional asymmetrical flow FFF (AFlFFF). Unlike conventional AFlFFF, FI-AFlFFF does not require time consuming, and experimentally inconvenient, sample focusing and relaxation steps involving valve switching and interruption of sample migration. The advantages of employing dual programming with FI-AFlFFF are demonstrated for sets of polystyrene sulfonate standards in the molecular mass range of 4 to 1000 kDa. It is shown that programmed FI-AFlFFF successfully expands the dynamic separation range of molecular mass.

Dual lectin-based size sorting strategy to enrich targeted N-glycopeptides by asymmetrical flow field-flow fractionation: profiling lung cancer biomarkers.

A dual lectin-based size sorting and simultaneous enrichment strategy for selectively isolating N-linked glycopeptides was developed using asymmetrical flow field-flow fractionation (AF4). AF4 is an elution-based method for separating biological macromolecules that has been utilized for the separation of lectin-glycopeptide complexes formed by mixing serum peptides with lectin cocktails according to the difference in diffusion coefficients. It has also been used for simultaneous depletion of nonglycosylated peptides. The dual lectin-based enrichment method was applied to proteolytic peptides from lung cancer serum samples with two lectins (WGA, GlcNAc-specific, and SNA, Sia-specific), and the whole mixture was separated by AF4. The lectin-glycopeptide complex fractions collected during AF4 separation were endoglycosidically digested with PNGase F. The resulting deamidated glycopeptides were analyzed by nanoflow liquid chromatography-electrospray ionization-tandem mass spectrometry (nLC-ESI-MS-MS) to semiquantitatively profile the N-linked glycopeptides from the sera of lung cancer patients and healthy controls. The AF4 enrichment strategy coupled with nLC-ESI-MS-MS identified 16/24 (up/down-regulated by at least 10-fold compared to normal sera) N-linked glycopeptides from a WGA complex fraction of lung cancer sera and 18/3 from a SNA fraction.

Dysbiosis-induced IL-33 contributes to impaired antiviral immunity in the genital mucosa

Significance Protective mechanisms of commensal bacteria against viral infection are limited to how immune inductive signals are provided by commensal bacteria for enhancing immunity. Whether, or how, commensal bacteria might influence the effector arm of immune responses remains unknown. Here, we demonstrate that dysbiosis within the vaginal microbiota results in severe impairment of antiviral protection against herpes simplex virus type 2 infection. IL-33 released into the vaginal tract after antibiotic treatment blocks the ability of effector T cells to migrate into the vaginal tissue and secrete the antiviral cytokine, IFN-γ. Thus, our findings suggest a previously unstudied role of commensal bacteria in the effector phase of the antiviral immune response against genital herpes. Commensal microbiota are well known to play an important role in antiviral immunity by providing immune inductive signals; however, the consequence of dysbiosis on antiviral immunity remains unclear. We demonstrate that dysbiosis caused by oral antibiotic treatment directly impairs antiviral immunity following viral infection of the vaginal mucosa. Antibiotic-treated mice succumbed to mucosal herpes simplex virus type 2 infection more rapidly than water-fed mice, and also showed delayed viral clearance at the site of infection. However, innate immune responses, including type I IFN and proinflammatory cytokine production at infection sites, as well as induction of virus-specific CD4 and CD8 T-cell responses in draining lymph nodes, were not impaired in antibiotic-treated mice. By screening the factors controlling antiviral immunity, we found that IL-33, an alarmin released in response to tissue damage, was secreted from vaginal epithelium after the depletion of commensal microbiota. This cytokine suppresses local antiviral immunity by blocking the migration of effector T cells to the vaginal tissue, thereby inhibiting the production of IFN-γ, a critical cytokine for antiviral defense, at local infection sites. These findings provide insight into the mechanisms of homeostasis maintained by commensal bacteria, and reveal a deleterious consequence of dysbiosis in antiviral immune defense.

Development of an Online Microbore Hollow Fiber Enzyme Reactor Coupled with Nanoflow Liquid Chromatography-Tandem Mass Spectrometry for Global Proteomics

In this study, we report the development of a microbore hollow fiber enzyme reactor (mHFER) coupled to nanoflow liquid chromatography-tandem mass spectrometry (nLC-ESI-MS/MS) for the online digestion or selective enrichment of glycopeptides and analysis of proteins. With mHFER, enzymatic digestion of protein could be achieved by continuous flow within a very small volume (~10 μL) of mHF inserted in a PEEK tube. Digested peptides exited through the pores of the hollow fiber membrane wall to external single or multiplexed trap columns for nLC-ESI-MS/MS analysis. Evaluation of online mHFER-nLC-ESI-MS/MS system was made with bovine serum albumin (BSA) by varying the temperature of digestion and the amount of protein injected. We evaluated the ability of the mHFER system to enrich glycopeptides by injecting a mixture of lectin (concanavalin A) and digested peptides from α-1-acid glycoprotein (AGP) into the mHFER, followed by delivery of PNGase F for endoglycosidic digestion. Nonglycosylated peptides unbound to lectins eluted at the first breakthrough run while N-linked glycopeptides eluted after the endoglycosidic digestion. The developed method was applied to urine samples from patients with prostate cancer and controls; 67 N-linked glycopeptides were identified and relative differences in glycopeptide content between patient and control samples were determined.

Lectin-based enrichment method for glycoproteomics using hollow fiber flow field-flow fractionation: application to Streptococcus pyogenes.

This paper presents a new application of hollow fiber flow field-flow fractionation (HF5) as a preparative method to preconcentrate high mannose type N-linked glycoproteins from Streptococcus pyogenes by means of the mannose-specific binding affinity between concanavalian A (ConA) and N-linked glycosylated proteins. Prior to fractionation of N-linked glycoproteins from bacterial lysates, it was examined that ConA formed several types of multimers depending on the pH values (4, 6, and 8) of the carrier solution and it was confirmed that the molecular weight (MW) of ConA, spiked with alpha-1 acid glycoprotein (AGP) as a standard glycoprotein, increased due to binding with the mannose moiety of AGP. After adding ConA to bacterial lysates, mannose type N-linked glycoproteins were found to be enriched when the ConA fraction was isolated from whole bacterial lysates through HF5 run. For the identification of glycoproteins, the ConA fraction of HF5 was tryptically digested and followed by two-dimensional nanoflow strong cation exchange-reversed phase liquid chromatography-electrospray ionization-tandem mass spectrometry (2D SCX-RPLC-ESI-MS-MS) analysis to identify the N-linked glycoprotein species. From two-dimensional shotgun analyses, 45 proteins that exist on the Asn-Xaa-Ser/Thr sequence were identified as high mannose type N-linked glycoprotein. As a result, it was first demonstrated that HF5 is an alternative tool to enrich high mannose type N-linked glycoproteins using ConA-specific binding affinity.

A soft preparative method for membrane proteome analysis using frit inlet asymmetrical flow field-flow fractionation: application in a prostatic cancer cell line.

Membrane proteins participate in a number of important biological functions such as signal transduction, molecular transport, and cell-cell interactions. However, due to the nature of membrane proteins, the development of a preparative method that produces a sufficient yield of purified membrane proteins from the cell remains a challenge. In the present study, frit inlet asymmetrical flow field-flow fractionation (FI-AFlFFF) was employed to fractionate membrane fragments containing membrane proteins from free cytoplasmic proteins of prostatic cancer cell (DU145 cell) lysates. The isolated membrane proteins were then digested and analyzed by nanoflow liquid chromatography/tandem mass spectrometry (nLC-ESI-MS-MS). Since fractionation of the cell lysate mixtures containing membrane fragments and cytoplasmic proteins could be achieved based on the differences of their sizes in FI-AFlFFF, membrane fragments were partially isolated from the cytoplasmic proteins and collected. The performance of FI-AFlFFF for prefractionation of the membrane proteome was examined by comparing the number of membrane proteins that were identified with the number identified using an ultracentrifugation method. The application of FI-AFlFFF to membrane proteomics produced an increased yield of purified membrane proteins with fewer cytoplasmic proteins compared to a conventional ultracentrifugation method.

Molecular mass sorting of proteome using hollow fiber flow field-flow fractionation for proteomics.

Hollow fiber flow field-flow fractionation (HF FlFFF) has been demonstrated as a tool for pre-fractionating proteomes by differences in molecular mass (Mr), where the resulting protein fractions are subsequently digested and analyzed by shotgun proteomics using two-dimensional liquid chromatography-electrospray ionization-tandem mass spectrometry (2D-LC-ESI-MS/MS). HF FlFFF is a separation device capable of fractionating proteins or cells by hydrodynamic radius, and protein fraction can be readily collected as intact conditions in aqueous buffer solutions. In this study, HF FlFFF was applied to fractionate the proteome of Corynebacterium glutamicum, a well known soil bacterium that has been widely used in bioindustry due to its remarkable ability to secrete high amounts of glutamic acid. The collected HF FlFFF fractions of different MW intervals were enzymatically digested for protein identification by 2D-LC-ESI-MS/MS. Experiments showed improvements in protein identification when HF FlFFF pre-fractionation was applied, due to decreases in the ionization suppression effect and the MS exclusion effect by spectral congestion. Pre-fractionation of C. glutamicum proteome allowed us to find 90 additional proteins by 2D-LC-ESI-MS/MS that were not found by a direct shotgun analysis without pre-fractionation. A total of 415 proteins were found overall with 203 proteins commonly found from experiments with and without pre-fractionation.

Large-Scale Identification by Shotgun Proteomics of Proteins Expressed in Porcine Liver and Salivary Gland

Abstract Protein catalogs containing a large number of proteins expressed in a variety of organs can be powerful tools for stem-cell research, because this requires accurate knowledge about how cells differentiate. Salivary gland progenitor (SGP) cells are somatic stem cells isolated from the salivary gland that can differentiate into hepatic or pancreatic cell lineages. Their differentiation state has been assessed by the expression of major protein markers, but to use these cells in regenerative medicine, it will be necessary to establish additional means of quality assessment. We examined the use of shotgun proteomics for porcine salivary gland (a source of SGP cells) and liver (a destination of differentiated SGP cells) for determining the state of SGP cell differentiation. Protein complexes from each organ were digested into peptides and separated by two-dimensional liquid chromatography involving strong cation-exchange chromatography followed by reversed-phase liquid chromatography. The separated peptides were analyzed by on-line electrospray ionization tandem mass spectrometry using a quadrupole-time of flight mass spectrometer (ESI Q-TOF MS/MS), and the spectra obtained were processed to search peptides against a mammalian database for protein identification. Using this method, we identified 117 proteins in porcine salivary gland and 154 proteins in porcine liver. Of these, 72 and 109 were specific to salivary gland and liver, respectively, and some of these were previously shown to be organ specific. The current study can be utilized in the future as a basis to study the pattern of differentiation in protein expression by stem cells.