关键词： 功能性蛋白芯片
生物谷报道：蛋白质芯片是近年来研究的热点，而功能性调节蛋白芯片包含了许多纯化的人类调节蛋白，为揭示这些调节蛋白的神秘功能将起到重要作用。
They contain complete and purified proteins, rather than peptide fragments or tagged proteins used in existing products, says Hui Ge, vice president and chief scientific officer at Maryland-based firm ProteinOne. Three such "active" arrays are to go on sale in mid-February.
"Everything's ready," said Ge, who is speaking at the Cambridge Health Institute's 3rd annual "PepTalk" in San Diego in January. ProteinOne's new products will allow simultaneous screening of interactions with several regulatory human proteins in a single experiment, he says.
The most common microarrays are glass slides onto which chunks of DNA are fixed in a grid formation. RNA extracted from a cell is usually used as a template to generate cDNA, which is used to probe the array. The spots at which hybridization occurs reveal which genes were being expressed in the cell. As yet however, protein microarrays are still in their infancy. This is mainly because it is difficult to isolate and tether proteins to a slide without disrupting their structure and hence binding properties.
But, says Ge, ProteinOne's new arrays get round the problem by using nitrocellulose as the supporting material. Proteins form natural hydrophobic bonds with this substrate, he says. It's an interaction that retains the protein's conformation and activity as it would exist in an aqueous solution, he adds.
Ge's "transcription array" contains 28 different transcription factors, such as TFIIB and p300; his "receptor array" features 18 nuclear receptors, including the PPARs; and a "cancer array" offers 23 tumor suppressor and onco-proteins. These are the first pure protein arrays to go onto the market, claims Ge, who is keen to distinguish his product from other protein arrays that are already commercially available.
Since late 2003, for example, Connecticut-based firm Protometrix has been selling its yeast proteome chip - a modified microscope slide spotted with some 5000 different yeast proteins. This is nearly the entire yeast proteome, according to Michael Snyder, professor and Chairman of Molecular, Cellular, and Developmental Biology at Yale, and the scientist behind this impressive array.
But Protometrix has used GST tags to purify the proteins for their patented 'Yeast ProtoArray', says purist Ge. "If you have such a big tag connected to the protein, sometimes it's OK," he said. "But sometimes - maybe 20-40% of chances - you change the protein structure, you change the protein conformation, and you change the protein property," he said. This can generate a lot of false-negative and -positive signals, he fears.
Ge's team has used bacterial and baculoviral systems to over-express human recombinant proteins. These then go through a regime of extensive purification, he says. "Right now we pretty much have 150 proteins," he said. "Within the next two or three years we should reach between 700 and maybe even 1000 proteins."
But protein purity aside, both Ge and Snyder agree that the real advantage of the protein array is that it can be probed with just about anything. They can inform not only on protein-protein interactions, but also on protein-DNA, protein-RNA, protein-ligand, and protein-small molecule interactions, depending on the probe used.
Working out binary interactions such as these is a useful way to gain insight on poorly understood internal complex organization, says Bertrand Séraphin, director of the Center for Molecular Genetics in Gif sur Yvette, France. But engineering the perfect protein array is still tremendously difficult, he warns.
Proteins, unlike nucleic acids, are structurally heterogeneous, and all behave differently to each other, says Séraphin. "It is thus difficult to find conditions that will be applicable to a large set of proteins to give meaningful large-scale results," he said.
The investment that ProteinOne has made in purifying proteins for its new arrays is certainly reflected in the price. A kit containing two arrays is expected to retail at between $300 and $400, says Ge. Spot for spot, DNA microarrays work out cheaper by an order of magnitude.
