Researchers at the Massachusetts Institute of Technology have developed a technology that they say may eventually allow researchers to more quickly and efficiently separate proteins and eliminate the use of gels for sample preparation.
The technology is a molecular sieve based on photolithography techniques commonly used in the semiconductor industry to make electronic chips used in cell phones and computers. Though it needs further development to be of practical use, the MIT researchers said it could help address the long-standing complaint that gels make sample preparation a painstakingly tedious chore.
However, some researchers expressed doubts to ProteoMonitor this week that the technology would have practical applications for proteomics work.
The technology involves a sieve that is placed in a silicon chip. A biological sample containing proteins is then put on the sieve for separation through nanopores.
Unlike gels in which pores sizes vary, the size of the nanopores in the sieve is uniform. Because of this, the MIT researchers were able to control the protein sieving process, allowing them to separate the proteins more efficiently, they said.
“There has been a lot of research going on regarding reliable ways of making uniform nanopores or even nanomembranes for various applications,” Jongyoon Han, an associate professor of electrical engineering and biological engineering at MIT, told ProteoMonitor this week. “The method that we are using now is actually quite simple. We are basically taking advantage of standard photolithography techniques.”
The sieving process is based on the Ogston sieving mechanism, a theoretical model in which deep and shallow regions serve as barriers to molecules moving through a sieve. Under the model, the rate at which proteins pass through the sieve would depend on their size — the smallest would pass through first, the largest would pass last.
In a paper published in the December online edition of Applied Physics Letters, Han and his colleagues said that three sodium dodecyl sulfate proteins “were quickly separated within 30 [seconds] and a 570 µm separation length.” Smaller protein complexes, they said, migrated faster than larger ones.
Further on, the researchers said, “Since the SDS-protein complexes and the dsDNA molecules separated are smaller than the 60-nanometer nanofilter gap size, [our study] clearly demonstrates the effectiveness of Ogston sieving in the nanofilter array and further is a direct experimental confirmation of Ogston sieving in a well-defined, regular nanopore system.”
The researchers see their technology eventually replacing the use of gels in sample preparation. Indeed, 2D gel electrophoresis in the US has already been overtaken by liquid chromatography in popularity, partly due to the time- and labor-intensive demands of gels. In Europe, however 2D gel electrophoresis is still the method of choice for protein separation.
“The reason people are using liquid chromatography is that gel is so painful because it takes so long and sample recovery is really bad,” said Jianping Fu, a PhD student in mechanical engineering at MIT who assisted Han in the research. “If you want to recover protein after separation from the gel, it is very painful and efficiency is very low.”
But whether Han’s technology is the solution to the shortcomings of gels is far from certain. Aside from the need to further develop the sieve, some in the proteomics community said that current limitations to other technologies and techniques render practical adoption of the sieve virtually impossible.
“The reason people are using liquid chromatography is that gel is so painful because it takes so long and sample recovery is really bad.”
“The way gels are constructed now, you move the proteins until they get stuck in the gel — their mobility decreases with increasing distance form the sample introduction. Then they can be stained from the top,” Aran Paulus, R&D manager for Bio-Rad’s laboratory separation division told ProteoMonitor in an e-mail. “In [the MIT researchers’] set-up, you will need to have a flow-through since there will be no easy way to pull off the top.
“Also, in order to do 2D you need to couple in isoelectric focusing. I see no predictable way to do this now,” he said.
Pier Giorgio Righetti, a professor of biochemistry at Polytechnic Institute of Milan, Italy, noted that the technique used by the MIT researchers is a one-dimension technique and “Single-dimension analysis won’t get you anywhere in proteomic analysis because the proteome is enormously complex,” he said.
He further said that systems based on sieving have historically suffered from poor resolution.
“As an example, isoelectric focusing in immobilized pH gradients can separate proteins that differ by only one one-thousandth of a pH unit in isoelectric points, a superb resolving power,” he said.
“On the contrary, even the best SDS-PAGE [sodium dodecyl sulphate polyacrylamide gel electrophoresis] system can achieve separation between two adjacent proteins of average size only when the mass difference is at least 2000 to 3000 daltons. This mass difference corresponds to at least [a] 15-to 20-amino-acid difference in chain length.”
Han acknowledged the work done so far by his group only confirms the Ogston model. To make the sieve of use in sample prep, the fractionation step will need to be integrated with a sensor.
“Also if you wanted to use this just as a replacement for gel electrophoresis, then you have to make this device amenable to the volume scale of typical biological research,” Han said.