HIGHLIGHTS

---

Research Highlights

Continuous flow DNA and RNA amplification on the microscale

Recent progress in the Human Genome Project (and other related genome projects) has generated a vast amount of sequence data and thus a need for a new generation of high-throughput DNA/RNA analysis techniques. Not surprisingly, the development of microfluidic systems for nucleic acid analysis has proceeded at a rapid rate to meet these needs and has afforded significant gains in efficiency, throughput and functionality. Of special importance in modern molecular biology is the enzymatic amplification of specific DNA or RNA sequences. In the case of RNA this is accomplished by synthesis of complementary DNA using reverse transcriptase followed by the polymerase chain reaction (PCR).

Theodore Christopoulos and co-workers have recently described the fabrication and operation of a microfabricated device for continuous flow DNA and RNA amplification. The device incorporates both reverse transcription and PCR on a monolithic glass device (Fig. 1), and importantly allows for dynamic selection of cycle number (between 25 and 40 cycles) within the amplification step. Specifically, the glass chip incorporates a single microchannel network that is sectioned into four temperature zones (one for reverse transcription and three for PCR). Using the device, the authors demonstrate successful amplification of a single-copy gene from genomic DNA and also amplification of three DNA samples flowing simultaneously through the chip (separated by 2 μL water plugs). In addition, high-throughput reverse transcription followed by PCR was performed on four RNA samples serially flowing through the device. Importantly, the use of ‘washing’ plugs between sample ‘plugs’ inhibits contamination or carryover between samples.

Fig. 1 Schematic diagram of a microfabricated device for continuous-flow PCR and RT-PCR with cycle number selection. There are two inlets for PCR, two inlets for RT, and five outlets for product collection at 20, 25, 30, 35, and 40 cycles. The chip is placed over four heating blocks, a–d, with appropriate temperatures for denaturation, extension, and annealing, respectively. I, intersection between the RT and the PCR channels. (Adapted with permission. Copyright 2003, The American Chemical Society.)

Interestingly, the authors also demonstrate successful PCR amplification using manual injection and transport of sample in continuous flow. The high level of functional integration within the device combined with the obvious operational flexibility should yield wide application in both field and laboratory based environments.

(Analytical Chemistry, 2003, 75, 288)

Chemical temperature control in microchannels

Heating and cooling of small sample volumes can be performed in extremely short times due to low thermal masses. In addition, the high surface-to-volume ratios typically encountered within microfluidic systems provide an efficient route to improved heat dissipation and thus thermal control. The ability to control temperature within a chemical system is crucial in many applications including chemical synthesis, DNA amplification and capillary electrophoresis, and consequently many microfluidic systems have been shown to exhibit superior performance when compared to macroscale instruments. Typically, integrated heating of samples within microchannels is achieved through the use of thin-film resistive heaters deposited on the interior or exterior surface of the chip substrate. Similarly, cooling of microfluidic systems is normally performed using external components (such as Peltier elements) or by convection. In almost all of these cases these additional components increase the instrumental footprint, cost and complexity. In this issue of Lab on a Chip, Elisabeth Verpoorte and colleagues at the University of Neuchâtel and Delft University of Technology address this limitation by using chemical and physical processes to regulate the temperature within microchannel circuits. Specifically, the authors exploit exothermic and endothermic reactions in microchannels to heat or cool a sample in an adjacent microchannel. Cooling is achieved by drawing air and acetone through a microchannel by vacuum. As the components are transported along the channel they mix, causing the acetone to evaporate which induces a decrease in the local temperature. In a similar way, heating is effected by dissolution of a concentrated sulfuric acid flow in a water flow, with temperature in the adjacent (sample) channel controlled by variation of flow rate ratios in the temperature control channels. Initial studies demonstrate minimum and maximum temperatures in the sample channel of −3 °C and 76 °C respectively, and thermal ramps of approximately 1 °C s⁻¹. Since the integration of the heating or cooling system involves no additional fabrication steps, the approach should prove valuable in a wide range of analytical applications where control of local temperatures is required.

(Lab on a Chip, 2003, 3, 1)

Microfluidic knots and spirals

The evolution of increasingly complex microfluidic systems has been successful in extending their applicability to a variety of chemical, biological and mathematical problems. Unfortunately, the creation of three-dimensional microstructures, although feasible, often involves the use of convoluted fabrication protocols. Recently, the advent of soft lithography has been shown to be an effective alternative to standard lithographic methods for fabricating complex, multilayer structures. The most popular way to make three-dimensional microfluidic structures involves the fabrication of conventional two-dimensional structures, which are subsequently aligned, stacked and connected to produce a multi-level fluidic network. Although successful, stacking methods are often ill-defined and time-consuming. To this end, George Whitesides and co-workers at Harvard University have described a simple method for fabricating complex, three-dimensional microfluidic networks that relies on the inherent flexibility of elastomeric polymers. ‘Pseudo-3D’ channels are created by bending planar channels out of plane into three-dimensional shapes. ‘True 3D’ channels are formed by decomposing network patterns into connectable substructures which are either planar or pseudo-3D. Using this general approach the authors demonstrate the creation of a variety of fluidic structures including a knot (Fig. 2), a spiral channel and a ‘basketweave’ of channels. In addition, more complex three-dimensional structures such as braided channel networks and 3D grids are reported.

Fig. 2 Microchannels produced by bending PDMS microchannels into pseudo-3D geometries. (A) Left: scheme of a figure-eight knot. Right: optical micrographs of the channel having the shape of a figure-eight knot. (B) Left: scheme of a 2D serpentine fluidic channel rolled into a pseudo-3D structure. Right: optical micrograph of the final structure. The channels were visualized by filling with an aqueous solution of fluorescein and illuminated with UV light. (Adapted with permission. Copyright 2003, The American Chemical Society.)

(Journal of the American Chemical Society, 2003, 125, 554)

Biomembrane microfluidics

Lipid bilayers are the fundamental structural feature of cellular membranes, constituting a barrier to the passage of polar molecules and ions. Substrate-supported lipid bilayers provide a useful model system with which to study the function of membrane biomolecules, and are generally formed by associating the membrane with a hydrophilic support material such as glass. Biomolecules in these supported systems can diffuse freely through the membrane, mimicking a property of cellular membranes that is essential for many cellular functions. Furthermore, the ability to influence and control supported lipid bilayers allows the creation of a new range of cellular experiments. Recently, Steven Boxer and Lance Kam of Stanford University have described a methodology to selectively remove, collect and reconstitute lipid bilayers from discrete regions of a planar surface (Fig. 3). The successful implementation of the method lies in the distinctive stability and formation properties of supported lipid bilayers. Specifically, targeted regions of a preformed bilayer can be selectively removed by guiding a flow of detergent over the substrate. Since flow occurs at low Reynolds numbers, detergent fluid streams can be easily confined with a precision of a few microns. A new surface-associated lipid bilayer (containing new biomolecular species) can then be formed by directing lipid vesicles over the exposed surface. Using this patterning approach combined with electrophoretic manipulation, the authors demonstrate a first-generation, membrane-based separation/purification strategy.

Fig. 3 Micropatterning, electrophoresis, and stripping of lipid bilayers. (A) Schematic of a supported lipid bilayer confined by microfabricated barriers. Mobile species, illustrated by lipids with red and green headgroups, freely diffuse and mix, approaching a uniform concentration across the extent of the lipid bilayer. (B) Application of an electric field induces migration of charged membrane components (the red lipids); the neutral lipids (green) do not respond to this applied field and remain homogeneous in the corral. (C) A stream of stripping solution is flowed over part of the surface under laminar flow conditions, leaving an open region of the substrate. (D) A new lipid bilayer, indicated by the lipid with blue headgroups, is introduced. (Adapted with permission. Copyright 2003, The American Chemical Society.)

Additionally, the authors also discuss novel experimental systems based on these concepts, including membrane-based separations, determination of protein–ligand association and dynamics, and the analysis and assembly of membrane biomolecule complexes. They also foresee the creation of highly integrated microfluidic systems for the preparation and analysis of biological membranes.

(Langmuir, 2003, 19, in press)

High temperature microfluidic lithography

Soft lithography describes the molding of elastomeric polymers such as poly(dimethylsiloxane) using master templates and has become highly popular over recent times as a facile, flexible, high-resolution and low-cost route to the creation of micro-sized features on planar substrates. A soft lithographic method of much interest is MIMIC (MIcro Moulding In Capillaries). In MIMIC, a fluid capable of subsequent solidification (for example, a liquid pre-polymer) wicks spontaneously by capillary action into a network of channels formed by a conformal contact between a mould (normally an elastomeric material) and a substrate. After an organic or inorganic material in the fluid crystallizes, cures, polymerizes, adsorbs, adheres, or reacts with the surface of a substrate, the elastomeric component is removed, and a pattern of the material remains on the substrate. The pattern in the master is thus replicated in a structure supported on the surface of the substrate. Although MIMIC has proven extremely successful (due the simplicity of the fabrication process) its use in large-scale applications is limited due to the slow rate of pattern formation, which is governed by the pressure difference driving the capillary motion of the fluid filling the mould.

To this end, Dario Pisignagno and co-workers at the Instituto Nazionale di Fisica della Materia have recently described a simple microfluidic technique to significantly improve the speed of soft-lithographic patterning. Their approach is based on the reduction of polymer viscosity by controlled variation of system temperatures. Since filling times are inversely proportional to fluid viscosity the method provides a direct route to increasing filling rates, without having to utilize vacuum-assisted techniques that act to increase the pressure difference driving the capillary motion of the fluids. Importantly this means that the technique is easily applied to the fabrication of high-resolution features (<5 μm). In initial studies, the authors demonstrate improvements in the filling speed of polyurethane within microfluidic channels of up to a factor of 60 at 80 °C. Due to the simplicity and flexibility of the approach its application is not limited by resolution or choice of inorganic substrate, thus making it suitable for industrial mass production.

(Advanced Materials, 2002, 14, 1565)

Andrew J. de Mello

---