Matthew B. Kerby

A fluorogenic assay using pressure-driven flow on a microchip.

A fluorogenic assay for human T‐cell phosphatase (TCPTP) was conducted on an etched glass microchip using pressure driven flow. The TCPTP enzyme catalyzes the removal of a phosphate group from 6,8‐difluoro‐4‐methylumbelliferyl/phosphate (DiFMUP) to produce the fluorogenic product 6,8‐difluoro‐4‐methylumbelliferone (DiFMU). Enzyme assays with real‐time on‐chip dilution were performed in both low‐viscosity (1 cP) buffer and an enzyme solution containing 50% glycerol (6 cP). Single side channels connect a series of reagent wells to a main channel where the fluorescent product of the enzyme reaction passes the detector region. Flow regulation of mixed viscosity fluids requires a pressure control on each arm of the chip contributing to the overall flow. An 8‐channel pressure controller was built to regulate the air pressure above all wells feeding channels of the chip, thereby controlling the dilution ratios of buffer, substrate and enzyme. Well pressures maintained a constant concentration of enzyme in the detector channel while adjusting the flow contribution of substrate and buffer. The substrate concentration was stepped over two orders of magnitude while verifying fluid dilutions using marker dyes. The kinetic parameters, Km, Vmax, and Kcat, showed good agreement with the values determined using a standard well plate and fluorometer.

Electrophoretic migration of proteins in semidilute polymer solutions

We present a systematic study of the electrophoretic migration of 10–200 kDa protein fragments in dilute‐polymer solutions using microfluidic chips. The electrophoretic mobility and dispersion of protein samples were measured in a series of monodisperse polydimethylacrylamide (PDMA) polymers of different molecular masses (243, 443, and 764 kDa, polydispersivity index <2) of varying concentration. The polymer solutions were characterized using rheometry. Prior to loading onto the microchip, the polymer solution was mixed with known concentrations of SDS (SDS) surfactant and a staining dye. SDS‐denatured protein samples were electrokinetically injected, separated, and detected in the microchip using electric fields ranging from 100 to 300 V/cm. Our results show that the electrophoretic mobility of protein fragments decreases exponentially with the concentration c of the polymer solution. The mobility was found to decrease logarithmically with the molecular weight of the protein fragment. In addition, the mobility was found to be independent of the electric field in the separation channel. The dispersion is relatively independent of polymer concentration and it first increases with protein size and then decreases with a maximum at about 45 kDa. The resolution power of the device decreases with concentration of the PDMA solution but it is always better than 10% of the protein size. The protein migration does not seem to correspond to the Ogston or the reptation models. A semiempirical expression for mobility given by van Winkle fits the data very well.

Kinetic Measurements of Protein Conformation in a Microchip

This paper presents a microchip‐based system for collecting kinetic time‐based information on protein refolding and unfolding. Dynamic protein conformational change pathways were studied in microchannel flow using a microfluidic device. We present a protein‐conserving approach for quantifying refolding by dynamically varying the concentration of the chemical denaturants, guanidine hydrochloride and urea. Short diffusion distances in the microchannel result in rapid equilibrium between protein and titrating solutions. Dilutions on the chip were tightly regulated using pressure controls rather than syringe‐based flow, as verified with extensive on‐chip tracer dye controls. To validate this protein assay method, folding transition experiments were performed using two well‐characterized proteins, human serum albumin (HSA) and bovine carbonic anhydrase (BCA). Transition events were monitored through fluorescence intensity shifts of the protein dye 8‐anilino‐1‐naphthalenesulfonic acid (ANS) during dilutions of protein from urea or guanidine hydrochloride solutions. The enzymatic activity of refolded BCA was measured by UV absorption through the conversion of p‐nitrophenyl acetate (p‐NPA). The microchip protein refolding transitions using ANS were well‐correlated with conventional plate‐based experiments. The microfluidic platform enables refolding studies to identify rapidly the optimal folding strategy for a protein using small quantities of material.

Free-solution electrophoretic separations of DNA-drag-tag conjugates on glass microchips with no polymer network and no loss of resolution at increased electric field strength.

Here, we demonstrate the potential for high‐resolution electrophoretic separations of ssDNA–protein conjugates in borosilicate glass microfluidic chips, with no sieving media and excellent repeatability. Using polynucleotides of two different lengths conjugated to moderately cationic protein polymer drag‐tags, we measured separation efficiency as a function of applied electric field. In excellent agreement with prior theoretical predictions of Slater et al., resolution is found to remain constant as applied field is increased up to 700 V/cm, the highest field we were able to apply. This remarkable result illustrates the fundamentally different physical limitations of free‐solution conjugate electrophoresis (FSCE)‐based DNA separations relative to matrix‐based DNA electrophoresis. ssDNA separations in “gels” have always shown rapidly declining resolution as the field strength is increased; this is especially true for ssDNA > 400 bases in length. FSCEs ability to decouple DNA peak resolution from applied electric field suggests the future possibility of ultra‐rapid FSCE sequencing on chips. We investigated sources of peak broadening for FSCE separations on borosilicate glass microchips, using six different protein polymer drag‐tags. For drag‐tags with four or more positive charges, electrostatic and adsorptive interactions with poly(N‐hydroxyethylacrylamide)‐coated microchannel walls led to appreciable band‐broadening, while much sharper peaks were seen for bioconjugates with nearly charge‐neutral protein drag‐tags.

Nanoneedle method for high-sensitivity low-background monitoring of protein activity.

We describe a new method for measuring the activity of protein in miniscule quantities using a carbon nanotube nanoneedle. The unique features of this new method are (a) the immobilization of a few molecules; (b) subsequent translocation and isolation of them near the tip of a position-actuated nanoneedle; and (c) a fixed, defined, and unhindered molecular position to allow rapid real-time sensing and monitoring. The kinetic bioactivity of immobilized alkaline phosphatase (AP) molecules was measured as test model. Results show no decrease in enzymatic activity compared to that of the solution-phase enzyme reaction, suggesting that the immobilization provided unhindered access for ligand binding and minimal conformational modulation caused by undesired surface interactions.

Direct Sequence Detection of Structured H5 Influenza Viral RNA

We describe the development of sequence-specific molecular beacons (dual-labeled DNA probes) for identification of the H5 influenza subtype, cleavage motif, and receptor specificity when hybridized directly with in vitro transcribed viral RNA (vRNA). The cloned hemagglutinin segment from a highly pathogenic H5N1 strain, A/Hanoi/30408/2005(H5N1), isolated from humans was used as template for in vitro transcription of sense-strand vRNA. The hybridization behavior of vRNA and a conserved subtype probe was characterized experimentally by varying conditions of time, temperature, and Mg2+ to optimize detection. Comparison of the hybridization rates of probe to DNA and RNA targets indicates that conformational switching of influenza RNA structure is a rate-limiting step and that the secondary structure of vRNA dominates the binding kinetics. The sensitivity and specificity of probe recognition of other H5 strains was calculated from sequence matches to the National Center for Biotechnology Information influenza database. The hybridization specificity of the subtype probes was experimentally verified with point mutations within the probe loop at five locations corresponding to the other human H5 strains. The abundance frequencies of the hemagglutinin cleavage motif and sialic acid recognition sequences were experimentally tested for H5 in all host viral species. Although the detection assay must be coupled with isothermal amplification on the chip, the new probes form the basis of a portable point-of-care diagnostic device for influenza subtyping.

Increase of separation resolution through field enhancement in microchips

An enhanced ability to separate charged species from neutral compounds in a microfluidic chip is demonstrated using a chip design with low‐resistance electrode channels operating with a multiport pressure/voltage controller. A factor of 2.7 improvement in resolution was obtained from chips made using identical mask designs but different etch depth protocols. Greater separation power allows one to cover a wider dynamic range for compounds with different electrophoretic mobilities.

Early In Vitro Transcription Termination in Human H5 Influenza Viral RNA Synthesis

Rapid diagnostic identification of the human H5 influenza virus is a strategic cornerstone for outbreak prevention. We recently reported a method for direct detection of viral RNA from a highly pathogenic human H5 influenza strain (A/Hanoi/30408/2005(H5N1)), which necessarily was transcribed in vitro from non-viral sources. This article provides an in-depth analysis of the reaction conditions for in vitro transcription (IVT) of full-length influenza H5 RNA, which is needed for diagnostic RNA production, for the T7 and SP6 phage promoter systems. Gel analysis of RNA transcribed from plasmids containing the H5 sequence between a 5′ SP6 promoter and 3′ restriction site (BsmBI) showed that three sequence-verified bands at 1,776, 784, and 591 bases were consistently produced, whereas only one 1,776-base band was expected. These fragments were not observed in H1 or H3 influenza RNA transcribed under similar conditions. A reverse complement of the sequence produced only a single band at 1,776 bases, which suggested either self-cleavage or early termination. Aliquots of the IVT reaction were quenched with EDTA to track the generation of the bands over time, which maintained a constant concentration ratio. The H5 sequence was cloned with T7 and SP6 RNA polymerase promoters to allow transcription in either direction with either polymerase. The T7 transcription product from purified, restricted plasmids in the vRNA direction only produced the 1,776-base full-length sequence and the 784-base fragment, instead of the three bands generated by the SP6 system, suggesting an early termination mechanism. Additionally, the T7 system produced a higher fraction of full-length vRNA transcripts than the SP6 system did under similar reaction conditions. By sequencing we identified a type II RNA hairpin loop terminator, which forms in a transcription direction-dependent fashion. Variation of the magnesium concentration produced the greatest impact on termination profiles, where some reaction mixtures were unable to produce full-length transcripts. Optimized conditions are presented for the T7 and SP6 phage polymerase systems to minimize these early termination events during in vitro transcription of H5 influenza vRNA.

Circulating IgSF Proteins Inhibit Adhesion of Antibody Targeted Microspheres to Endothelial Inflammatory Ligands

Proposed methods for detecting circulatory system disease include targeting ultrasound contrast agents to inflammatory markers on vascular endothelial cells. For antibody-based therapies, soluble forms of the targeted adhesion proteins of the immunoglobulin superfamily (IgSF) reduce adhesion yet were left unaccounted in prior reports. Microspheres labeled simply with a maximum level of antibodies can reduce the diagnostic sensitivity by adhering to proteins expressed normally at a low level, while sparsely coated particles may be rendered ineffective by circulating soluble forms of the targeted proteins. A new microdevice technique is applied to simultaneously measure the adhesion profile to a series of IgSF-protein-coated surfaces. In this investigation, we quantify the in vitro binding characteristics of 5-μm microspheres to oriented intercellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1) protein-coated surfaces in the presence of human serum at physiological concentrations. Defined regions of a slide were coated with recombinant chimeric Fc-human ICAM-1 and VCAM-1 in variable ratios but constant total concentration. Monoclonal human anti-ICAM-1 or anti-VCAM-1 antibodies in competition with non-binding mouse anti-rabbit antibodies coat the microsphere surface at a constant surface density with variable yet controlled surface activities. Using multiple slide surface IgSF protein and microsphere antibody concentrations, an adhesion profile was developed for the microspheres with and without IgSF proteins from human serum, which demonstrated that exposure to serum reduced microsphere binding, on average, more than 50% compared to the no-serum condition.. The serum effects were limited to antibodies on the microsphere, since binding inhibition was reversed after rinsing serum from the system and fresh antibody-coated microspheres were introduced. This analysis quantifies the binding effects of soluble IgSF proteins from human serum on antibody-based targeted ultrasound detection and drug delivery methods.

Selective Ion Extraction for High Throughput Screening with a Microfluidic Enzyme Assay

This proof-of-concept paper describes the application of selective ion extraction to an assay of protein kinase A on a microfluidic chip platform. Selective ion extraction is a flux balance technique, where a combination of independent pressure control and voltage are used to selectively extract one ion from a mixture. The assay product is completely separated and diverted into a separate channel from the waste stream containing the unconverted substrate and enzyme. By detecting only product, background noise generated by the substrate is removed which increases the signal to noise ratio and assay sensitivity. This technique is intended for adapting kinase or protease assays with low conversion rates to an on-chip reaction format for HTS screening.