If a cell typically contains up to 10,000 to 100,000 proteins or subcomponent peptides, then just finding one, or discerning distinct identities of a few, can seem like looking for a proverbial needle in a haystack.
The task’s all the harder if your starting material is hardly more than a few cells. Then, provided the target has been sorted out, how do you manipulate that bit to analyze it further? Standard screening currently requires cumbersome and lengthy processing steps using equipment the size of kitchen appliances.
Anup Singh (8130) and colleagues are developing microfabricated devices for protein and peptide analysis (dubbed a µProLab) through the Molecular Integrated Microsystems (MIMS) grand challenge Laboratory Directed Research and Development project.
They’ve found, Anup says, that "by miniaturizing, we can actually do better." Using microchannels a few centimeters long, and in some cases just a few millimeters long, on glass chips, they’ve demonstrated separation of six proteins and peptides in 45 seconds — one-tenth the time it would take if performed in longer capillaries, and with 1/1,000th the starting sample needed for laboratory-bench-top-scale separations using porous-matrix-filled tubes called chromatography columns.
Chromatography works by selectively delaying different groups of proteins for different lengths of time in the matrix as the sample, applied to the top like pouring a liquid into a funnel, is rinsed through the material with a buffered solution and collected in a row of vials at the bottom. Different types of proteins drip out at different times, forming isolated "peaks" in the collection vials.
Separations can be tailored to proteins’ different physical properties, based on the material used for the spongelike sieving matrix and the liquid used to rinse it.
MIMS aims to integrate steps needed to sort and identify small amounts of proteins or peptides by "addressing" smart materials on chip assemblies to "do certain things at certain times in a certain place," Anup says. In addition to running chromatography and other separations at microscale, the chips will include components such as valves to control and manipulate movement of fluids and concentrators to permit pre- and postanalysis concentration of dilute samples.
Anup hit upon his patent-applied preconcentrator invention by serendipity. He was working determinedly to get ready for a conference presentation. A minuscule, pico-liter-sized protein sample he’d injected onto a microchannel that had been carefully packed with porous beads should have emerged, based on theory, after an electric field was applied. Anup suspected the initial sample injection didn’t work. He used a hand-held syringe to push the fluid out of the channel. The detector happened to still be on, and to his surprise it registered a huge peak of concentrated protein.
"If not for that conference, I might not have discovered it," he says. "I was just working day and night."
He and collaborators termed the technique electrokinetic trapping. Sharp, concentrated peaks form by using an electric field to focus charged analytes into a small spot in the separation channel. The preconcentration technique is addressable and reversible; proteins can be trapped and concentrated at specific locations by turning the voltage on and released by turning the voltage off.
The investigators, including Tim Shepodd (8722), have created a new method of creating in place sieving gels by using ultraviolet light to polymerize a porous matrix whose composition can be fine-tuned for various separations. Select locations can be polymerized by using a mask.
For controlling flow through a branching array of intersecting channels, Brian Kirby (8358) has used a moving plug of polymerized material to shuttle flow through a bypass, thus creating a sort of nonreturn, or check, valve.
The team intends to combine separation techniques to "fingerprint" proteins, as is currently done in bench-top processes, separating by both charge and size dimensions. Already, Anup and coworkers, including Jongyoon Han and Dan Throckmorton (8130), have seen separation efficiencies in a single dimension two to three times greater than the larger techniques allow.
"We hope to cut the time, overall, 10- to 100-fold," he says, "and work with a small sample — possibly a few cells." The ultimate goal is that, once sorted by charge and size, a spot of protein can be microfluidically transported to a mass spectrometer for analysis of its constituent elements — ideally through a completely automatic transfer when the device integration is complete.