Abebaw B. Jemere

Multifunctional protein processing chip with integrated digestion, solid-phase extraction, separation and electrospray

We describe a microfluidic device in which integrated tryptic digestion, SPE, CE separation and electrospray ionization for MS are performed. The chip comprised of 10×30 μm channels for CE, and two serially connected 150 μm deep, 800 μm wide channels packed with 40 to 60 μm diameter beads, loaded with either immobilized trypsin, reversed‐phase packing or both. On‐chip digestion of cytochrome c using the trypsin bed showed complete consumption of the protein in 3 min, in contrast to the 2 h required for conventional solution phase tryptic digestion. SPE of 0.25 μg/mL solutions of the peptides leu‐enkephalin, angiotensin II and LHRH gave concentration enhancements in the range of 4.4–12, for a ten times nominal volume ratio. A 100 nM cytochrome c sample concentrated 13.3 times on‐chip gave a sequence coverage of 85.6%, with recovery values ranging from 41.2 to 106%. The same sample run without SPE showed only five fragment peaks and a sequence coverage of 41.3%. When both on‐chip digestion and SPE (13.3 volume ratio concentration enhancement) were performed on 200 nM cytochrome c samples, a sequence coverage of 76.0% and recovery values of 21–105% were observed. Performing on‐chip digestion alone on the same sample gave only one significant fragment peak. The above digestion/peptide concentration step was compared to on‐chip protein concentration by SPE followed by on‐chip digestion with solution phase trypsin. Both procedures gave similar recovery results; however, much larger trypsin autodigestion interference in the latter approach was apparent.

On-chip solid phase extraction and enzyme digestion using cationic PolyE-323 coatings and porous polymer monoliths coupled to electrospray mass spectrometry.

We evaluate the compatibility and performance of polymer monolith solid phase extraction beds that incorporate cationic charge, with a polycationic surface coating, PolyE-323, fabricated within microfluidic glass chips. The PolyE-323 is used to reduce protein and peptide adsorption on capillary walls during electrophoresis, and to create anodal flow for electrokinetically driven nano-electrospray ionization mass spectrometry. A hydrophobic butyl methacrylate-based monolithic porous polymer was copolymerized with an ionizable monomer, [2-(methacryloyloxy)ethyl] trimethylammonium chloride to form a polymer monolith for solid phase extraction that also sustains anodal electroosmotic flow. Exposure of the PolyE-323 coating to the monolith forming mixture affected the performance of the chip by a minor amount; electrokinetic migration times increased by ∼5%, and plate numbers were reduced by an average of 5% for proteins and peptides. 1-mm long on-chip monolithic solid phase extraction columns showed reproducible, linear calibration curves (R(2)=0.9978) between 0.1 and 5 nM BODIPY at fixed preconcentration times, with a capacity of 2.4 pmol or 0.92 mmol/L of monolithic column for cytochrome c. Solution phase on-bed trypsin digestion was conducted by capturing model protein samples onto the monolithic polymer bed. Complete digestion of the proteins was recorded for a 30 min stop flow digestion, with high sequence coverage (88% for cytochrome c and 56% for BSA) and minimal trypsin autodigestion product. The polycationic coating and the polymer monolith materials proved to be compatible with each other, providing a high quality solid phase extraction bed and a robust coating to reduce protein adsorption and generate anodal flow, which is advantageous for electrospray.

A regenerating ultrasensitive electrochemical impedance immunosensor for the detection of adenovirus

We report on the development of a regenerable sensitive immunosensor based on electrochemical impedance spectroscopy for the detection of type 5 adenovirus. The multi-layered immunosensor fabrication involved successive modification steps on gold electrodes: (i) modification with self-assembled layer of 1,6-hexanedithiol to which gold nanoparticles were attached via the distal thiol groups, (ii) formation of self-assembled monolayer of 11-mercaptoundecanoic acid onto the gold nanoparticles, (iii) covalent immobilization of monoclonal anti-adenovirus 5 antibody, with EDC/NHS coupling reaction on the nanoparticles, completing the immunosensor. The immunosensor displayed a very good detection limit of 30 virus particles/ml and a wide linear dynamic range of 10(5). An electrochemical reductive desorption technique was employed to completely desorb the components of the immunosensor surface, then re-assemble the sensing layer and reuse the sensor. On a single electrode, the multi-layered immunosensor could be assembled and disassembled at least 30 times with 87% of the original signal intact. The changes of electrode behavior after each assembly and desorption processes were investigated by cyclic voltammetry, electrochemical impedance spectroscopy and X-ray photoelectron spectroscopy techniques.

Tunable thick polymer coatings for on-chip electrophoretic protein and peptide separation.

We report a variety of procedures for fabricating confinement-induced polymer coatings, used to eliminate non-specific protein adsorption and to control electroosmotic flow for microchip capillary electrophoresis. The coating strategy generates relatively thick polymer wall coatings (100-700 nm) and can easily be tuned by adjusting the monomer concentration. 2-hydroxyethyl methacrylate (HEMA) polymer coating, photopatterned in microfluidic channels, effectively reduced protein non-specific adsorption and rendered high efficiency (N/m=∼3 × 10⁶) for protein separation. The coating strategy provides rapid and effective means to create robust wall coatings, with the ability to photograft various surface chemistries onto the coating. [2-(methacryloyloxy) ethyl] trimethylammonium chloride grafted HEMA coated channels showed high durability and reproducibility for generating EOF (RSD=2.6%, n=64) over a period of 15 days. Sulfobetaine methacrylate grafted HEMA coated channels allowed separation of BSA digest, 15 peaks resolved in 25s, with an average N/m of 4 × 10⁵.

Capillary electrochromatography with packed bead beds in microfluidic devices

Microchip‐based bead‐packed columns for electrochromatography are described for several types of stationary phases. Chromatography columns 2 mm in length were used for the separation of proteins and peptides by size‐ and ion‐exchange modes of separation, respectively. In size‐exclusion electrochromatograpgy, FITC‐IgG and FITC‐insulin were baseline resolved in less than 10 s, with efficiencies of up to 139,000 plates/m for FITC‐insulin. In strong cation‐exchange electrochromatography, a mixture of three fluorescently labeled peptides was baseline resolved in less than 40 s, with efficiencies up to 400,000 plates/m. The RSD for the analytes retention times were<3% in both size‐exclusion and ion‐exchange modes of separations. The use of a 1‐mm‐long reverse‐phase column for the semiquantitative evaluation of pharmaceutical formulations in drug solubility tests illustrates the use of this microfluidic chip‐based electrochromatographic approach to drug development.

Matrix‐free laser desorption/ionization mass spectrometry using silicon glancing angle deposition (GLAD) films

Glancing angle deposition (GLAD) was used to fabricate nanostructured silicon (Si) thin films with highly controlled morphology for use in laser desorption/ionization mass spectrometry (DIOS-MS). Peptides, drugs and metabolites in the mass range of 150-2500 Da were readily analyzed. The best performance was obtained with 500 nm thick films deposited at a deposition angle of 85 degrees . Low background mass spectra and attomole detection limits were observed with DIOS-MS for various peptides. Films used after three months of dry storage in ambient conditions produced mass spectra with negligible low-mass noise following a 15 min UV-ozone treatment. The performance of the Si GLAD films was as good as or better than that reported for electrochemically etched porous silicon and related materials, and was superior to matrix-assisted laser desorption/ionization (MALDI)-MS for analysis of mixtures of small molecules between 150-2500 Da in terms of background chemical noise, detection limits and spot-to-spot reproducibility. The spot-to-spot reproducibility of signal intensities (100 shots/spectrum) from 21 different Si GLAD film targets was +/-13% relative standard deviation (RSD). The single shot-to-shot reproducibility of signals on a single target was +/-19% RSD (n = 7), with no indication of sweet spots or mute spots.

A regenerating self-assembled gold nanoparticle-containing electrochemical impedance sensor

We report on the development of an electrochemical reductive desorption protocol for repeated regeneration of gold electrodes modified with multi-layers of self-assembled surfaces for use in electrochemical sensing. The gold electrodes were first modified with 1,6-hexanedithiol to which gold nanoparticles were attached in a subsequent modification step. Attachment of thiolated single-stranded nucleic acid oligomers to the gold nanoparticles completed the electrochemical sensor. The changes of electrode behavior after each assembly and desorption processes were investigated by cyclic voltammetry, electrochemical impedance spectroscopy and X-ray photoelectron spectroscopy techniques. The self-assembled sensor showed a wide dynamic range (0.1-100 nM), a low detection limit (20 pM) and high reproducibility (4.4% RSD) for complementary nucleic acid target molecules, along with reusability. On a single gold electrode, the complete sensor-target structure could be assembled and disassembled at least four times with 90% of its original signal intact.

Integrated electrokinetic sample fractionation and solid-phase extraction in microfluidic devices.

A microfluidic device that performs “in space” sample fractionation, collection, and preconcentration for proteomics is described. Effluents from a 2.75 mm long fractionation channel, focused via sheath flow, were sequentially delivered into an array of 36‐collection channels containing monolithic polymer beds for SPE. Optimum conditions for the device design, and simultaneous photolytic fabrication of 36 monolithic columns in the 36 channels, as well as for their proper performance in electrokinetic sample fractionation and collection are described. A hydrophobic butyl methacrylate‐based monolithic porous polymer was copolymerized with an ionizable monomer, acryloamido‐methyl‐propane sulfonate, to form a polymer monolith for SPE that also sustains cathodic electroosmotic flow. The SPE bed was made deep enough to greatly reduce the linear flow rate within the bed, in order to compensate for the lower electroosmotic mobility of the cationically charged SPE bed relative to the glass walled device. Under these conditions, electrokinetic fractionation of a protein sample resulted in tightly focused sample zones delivered into each of the 36‐channel polymer beds with no observed crosscontamination. Monolithic columns showed reproducible performance with preconcentration factor of 30 for 2 min loading time. The ability to fractionate, collect, and preconcentrate samples on a microfluidic platform will be especially useful for automated or continuous operation of these devices in proteomics research.

Microfluidic devices for electrokinetic sample fractionation.

We present three generations of microchip‐based “in‐space” sample fractionators and collectors for use in proteomics. The basic chip design consisted of a single channel for CE separation of analytes that then intersects a fractionation zone feed into multiple high aspect ratio microchannels for fractionation of separated components. Achievements of each generation are discussed in relation to important design criteria. CE‐separated samples were electrokinetically driven to multiple collection channels in sequence without cross‐contamination under the protection of sheath streams. A 36‐channel fractionator demonstrated the efficacy of a high‐throughput fractionator with no observed cross‐contamination. A mixture of IgG and BSA was used to test the efficiency of the fractionator and collector. CE of the fractionated samples was performed on the same device to verify their purity. Our demonstration proved to be efficient and reproducible in obtaining non‐contaminated samples over 15 sample injections. Experimental results were found to be in close agreement with PSpice simulation in terms of flow behavior, contamination control and device performance. The design presented here has a great potential to be integrated in proteomic platforms.

Evaluation of protein separation mechanism and pore size distribution in colloidal self-assembled nanoparticle sieves for on-chip protein sizing

The separation behavior of 6.5–66 kDa proteins in silica particle array‐based sieves formed by colloidal self‐assembly in microchips is reported across a pore size range of 22–103 nm. The protein separation and resolution improves markedly with decreasing pore size. The variation of electrophoretic mobility with molecular weight of SDS–protein complexes and with particle size was evaluated using the Ogston sieving equation for a random pore gel structure, and using the modified Giddings equation developed by Wirth for uniform pore structures. The Wirth/Giddings equation provides the best fit for estimation of molecular weight of proteins, and demonstrates that even though experimental evidence shows there is some dispersion in measured pore sizes, these structures can best be described as having a uniform pore size.