A group of Yale researchers has developed 5,800-spot yeast proteome chips to find numerous new yeast proteins.
The protein chip experiments, which Heng Zhu and colleagues reported in a July 27 Science article, “Global Analysis of Protein Activities Using Proteome Chips,” were not only an exploration into the yeast proteome, but a validation effort for novel protein chip technologies.
“We realized that DNA chips are powerful for a variety of uses,” said Michael Snyder, one of Zhu’s co-authors from Yale’s department of molecular, cellular, and developmental biology. “Hence proteins, which is where the action is at in terms of directly analyzing activity, should be orders of magnitude more powerful.
“This technology should be extremely powerful for analyzing not only protein function and regulation, but also for drug discovery,” Snyder said.
To build their yeast proteome, the researchers first developed a collection of 5,800 open reading frames (ORFs) from the yeast genome, which comprised 93.5 percent of the organism’s total ORFs, and used recombination cloning to clone them into a high-copy yeast expression vector. Then they fused the cloned proteins to the GST-HisX6 molecule, which served as binding molecules to the surface of a nickel-coated slide.
The researchers previously tried binding the proteins directly to an aldehyde-treated slide, but settled on the GST-HisX6 and nickel-coated slides because it offered superior results.
The GST-HisX6 works better, said Snyder, “because it attaches the protein through its tag and not the protein itself, therefore keeping the rest of the protein accessible for binding its interacting partners and for enzymatic activity. It [also] gives an extra round of purification when proteins are first purified through GST and then bound to the substrate using the HisX6 tag.”
After purification and monitoring steps, the researchers used a Virtek Pro DNA microarrayer to spot down the proteins on the slides. On each slide, they spotted down 6,566 protein preparations in duplicate.
The researchers tested the chips by probing them with anti-GST antibodies, and found that over 93.5 percent of the probes gave signals significantly above the background signal.
“Our results … demonstrate that it is feasible to spot 13,000 protein samples in one half the area of a standard microscope slide with excellent resolution,” Zhu and his colleagues wrote.
After constructing the arrays, the researchers used them to detect protein-protein and protein-lipid interactions by applying samples of known proteins as well as liposome samples.
The researchers not only detected numerous previously known interactions, they also identified a previously unknown post-translational modification of a protein. In the lipid-protein experiments, they found 52 uncharacterized lipid-binding proteins, and 45 membrane -associated lipid binding proteins.
Previously, a noted limit of protein chips had been the inability to detect membrane proteins, but the researchers’ methods of purification surmounted this obstacle.
“Because the proteins are tagged at the N terminus, they probably do not normally enter the secretory pathway properly and thus remain in the cytoplasm where we purify them,” Snyder said.
However, the researchers did acknowledge that the N terminus tagging method would be likely to under-represent extracellular domains and secreted proteins, since it would prevent these proteins from being folded or modified properly and delivered to the secretory pathway.
“Regardless, the use of proteome chips has significant advantages over existing approaches,” the authors wrote. “Using similar procedures it is clearly possible to prepare protein arrays of 10-100,000 proteins for global proteome analysis in humans and other eukaryotes.”
Zhu’s lab has abundant plans for further proteome chip experiments, including probing for enzymatic activities. The lab is currently making human protein chips, said Snyder.