Collection, focusing, and metering of DNA in microchannels using addressable electrode arrays for portable low-power bioanalysis

Abstract

Although advances in microfluidic technology have enabled increasingly sophisticated biosensing and bioassay operations to be performed at the microscale, many of these applications employ such small amounts of charged biomolecules (DNA, proteins, and peptides) that they must first be preconcentrated to a detectable level. Efficient strategies for precisely handling minute quantities of biomolecules in microchannel geometries are critically needed; however, it has proven challenging to achieve simultaneous concentration, focusing, and metering capabilities with current-generation sample-injection technology. By using microfluidic chips incorporating arrays of individually addressable microfabricated electrodes, we demonstrate that DNA can be sequentially concentrated, focused into a narrow zone, metered, and injected into an analysis channel. This technique transports charged biomolecules between active electrodes upon application of a small potential difference (1 V) and is capable of achieving orders of magnitude concentration increases within a small device footprint. The collected samples are highly focused, with sample zone size and shape defined solely by electrode geometry.

Fig. 1.

The influence of focusing and concentration of an injected sample on the ability to detect and resolve two distinct components by electrophoretic separation. (A) Injection of a nonconcentrated and unfocused sample zone requires a long separation distance to distinguish each component; however, the corresponding signal from each species falls below the detectable range as the zones spread by diffusion and dispersion during the time elapsed while traveling this distance. (B) Injection of a concentrated but unfocused sample zone allows each component to be detected, but a long separation distance is still required. (C) Injection of a concentrated and focused sample zone allows each component to be detected within a considerably shorter separation distance.

Fig. 2.

Simultaneous concentration and focusing of DNA using an array of individually addressable microelectrodes. (A) Illustration of microdevice construction and assembly. PC, printed circuit. (B) Image of the electrode array within the device. Channel dimensions are 275 μm wide by 45 μm tall, and electrodes are 50 μm wide with 225-μm edge-to-edge spacing. A 100-bp double-stranded DNA ladder [12 μg/ml in 1× TBE buffer with 10% (vol/vol) BME] fluorescently labeled with YOYO-1 intercalating dye is loaded inside the microchannel. (C) A potential of 1 V is applied to the first two electrodes in the array, resulting in migration toward the anode (electrode 2) of DNA initially between the two electrodes, after which it becomes captured at the anode (migration is from left to right). (D) The anodic potential is switched to electrode 3, resulting in release of the captured DNA. (E) The cathodic potential is switched to electode 2, and migration continues toward electrode 3 until all DNA initially between electrodes 1 and 3 becomes captured at electrode 3. This process is repeated until a desired amount of enrichment is achieved. (F) Electrophoretic separation of a 100-bp double-stranded DNA ladder sample initially at 6 μg/ml [1× TBE buffer with 10% (vol/vol) BME] after concentration and injection into a gel matrix photoploymerized inside the microchannel immediately downstream of the final electrode in the capture array (5% T-crosslinked polyacrylamide gel, E = 23 V/cm). (G) All fragments in the DNA ladder are resolved in a separation length of 3 mm. (H) Without focusing and concentration before injection into the gel, fragments are unresolvable at the same separation length. The total area under the peaks in G is approximately double that in H, as expected because the quantity of DNA injected in the focused/concentrated separation of G was nearly double that in the unfocused separation of H (based on the initial sample concentration and volume contained between electrodes in each case).

Fig. 3.

Characterization of the electrode capture process. (A) The achievable concentration level is robust and essentially independent of buffer composition, amount of BME, and electrode material. (B) Influence of buffer composition, concentration, and amount of BME on the kinetics of the electrode capture process. Kinetics are significantly accelerated in the case of histidine buffer. (C) Comparison of simulated capture time constants with corresponding experimental results for transient enrichment in DNA concentration at the anode. Simulations were performed for the case of DNA initially at 12 μg/ml in TBE buffers of various concentrations under a 1-V applied potential (simulation details are provided in Supporting Materials and Methods, which is published as supporting information on the PNAS web site).

Fig. 4.

Advanced applications of the electrode capture process. (A) Illustration of dual focusing achieved by applying a negative potential on both sides of a capture electrode to further confine the sample to the anode surface [DNA sample initially at 12 μg/ml in 1× TBE buffer with 10% (vol/vol) BME]. (B) A 50-fold increase in normalized fluorescence intensity represents the concentration enhancement achieved by applying a sequential capture–release process across 15 electrodes (occupying a downstream distance of 3.9 mm). The additional concentration increase at the final two electrodes upon application of the dual focusing scheme reflects tighter confinement of DNA at the electrode surface where the detection window for intensity measurements is located. (C) Dual focusing is simultaneously applied to alternating electrodes in the array to collect precise quantities of DNA that can be metered for subsequent analysis. (D–F) Illustration of a buffer exchange process performed by superimposing a bulk hydrodynamic flow over a captured DNA sample. (D) A DNA sample initially at 12 μg/ml in 1× TBE buffer with 10% (vol/vol) BME is captured at the central anode by dual focusing. (E) A hydrodynamic flow is introduced by placing a drop of 50 mM histidine (no BME added) labeled with carboxylated polystyrene microsphere tracers at the microchannel inlet, generating velocities ranging from 11 to 18 μm/s (flow direction is from left to right). Depending on the flow velocity, the captured DNA is partially swept downstream from the electrode surface but remains confined between cathodes (2.4-V potential). (F) Captured DNA returns to the anode when the flow stops. The sample can then be released and resuspended in the new buffer environment when the potential is switched off.