Researchers from Harvard University have created a new protein microarray they claim can enable the large-scale expression of proteins for the first time on such a platform.
The technology, described in the online edition of the current Nature Methods, is the next generation of a technology called nucleic acid programmable protein array, or NAPPA, that uses new chemistry to enable the production of proteins en masse — a capability that so far has not been attainable with protein microarrays, Niroshan Ramachandran, first author on the paper and formerly a scientist in Joshua LaBaer’s lab at the Harvard Institute of Proteomics at the Harvard Medical School, told ProteoMonitor this week.
NAPPA, developed by LaBaer’s group, consists of cDNA vectors coupled with a capture antibody and allows for functional proteins to be synthesized in situ directly from printed cDNAs at the time of the assay [See PM 01/14/05]. The original concept behind the technology, described in a 2004 paper, was to develop an easier and less expensive way to produce functional protein arrays, an area dominated by Invitrogen’s ProtoArray product line.
Proteins are translated using a T7-coupled rabbit reticulocyte lysate in vitro transcription-translation system. The expressed proteins are then captured locally with an antibody to a C-terminal glutathione S-transferase tag on each protein. Because the proteins are produced just in time for the assay, they eliminate the need for high-throughput protein isolation, therefore reducing variability in the amount of proteins that can be produced and variability while increasing the stability of the proteins.
At about a one-tenth to one-hundredth of the cost of other functional arrays, NAPPA also represents a significantly more cost-effective method of producing such arrays, according to the researchers.
But while their 2004 research showed that the method could work, “we had no idea how robust or reliable the technology would be,” Ramachandran said. “It was at best a proof of concept with a handful of proteins. What the first paper said was that this is a way to make stable arrays, and a faster, cheaper, [and] easier way to make arrays.”
When they went to scale with their method, however, they found it was “very difficult to do en masse to thousands of genes. And we realized that we couldn’t do that reproducibly: every day we made an array that looked very different from the other,” because everything was done on a gene-by-gene basis and not in a parallel process, Ramachandran said.
Similarly other strategies for making protein microarrays that have been developed since the original NAPPA technology also have limitations with scale. They include the multiple spotting technique, in which an E. coli-based in vitro transcription-translation system extract is printed directly on a printed PCR template. Another approach, DNA array to protein array, has also been developed. Here, proteins are translated to a cDNA array and then diffused across a cell-free extract-infused membrane to a protein-capture surface.
But these methods, the researchers write in Nature Methods, have been tested using only small amounts of proteins compared to printing purified proteins and “have yet to demonstrate the robust ability to produce the high content needed to justify protein microarrays as a routine proteomics tool.”
In their article, Ramachandran and his co-authors describe their next-generation NAPPA, which they say retains the ease of handling and production and stability of the original array, while now allowing for high content and reproducibility.
In the iteration of the technology described in the article, the researchers demonstrated NAPPA’s utility for 1,000 genes. In upcoming articles, its performance will be scaled up to hundreds of thousands of genes from human genes to pathogens, Ramachandran said.
Chemistry Makes the Difference
The key to the next-generation platform was a “home-brewed DNA-purification system” that uses a resin derivatized with diamine chemistry. The original NAPPA system used commercial chemistries for DNA purification, but each one that they tested — according to Ramachandran, the researchers tested just about every one that is available — had insufficient yield and purity.
By adopting diamine chemistry from German firm Macherey-Nagel to a 96-well format from its original bead format, they were able to capture all the DNA that they produced with “very little loss,” Ramachandran said.
“We had no idea how robust or reliable the technology would be.”
“DNA bound the positively charged diamines at low pH and eluted when the diamines became neutrally charged under alkaline conditions,” the authors said. Using their method, a researcher can process 5,000 samples per week and yield 18 micrograms of supercoiled DNA with l milliliter of culture.
On the purity, the DNA they produce is good enough to be used to transfect mammalian cells, considered the gold standard in DNA purity. In fact, the DNA is being used for transfection in other research being done in LaBaer’s lab, Ramachandran said.
The group also developed a new printing chemistry that relies on the ability of bovine serum albumin to “dramatically improve DNA-binding efficiency. The BSA and capture antibody are coupled to the amine-coated glass surface via an activated ester-terminated homo-bifunctional cross linker,” they said. Using fluorescently labeled DNA, an estimated 64 percent of the DNA is captured onto the surface.
In testing their method, they found that for about 1,000 genes selected for which plasmids were isolated from single colonies, DNA signal was detected for 99 percent of the samples. In addition 93 percent of the membrane proteins tested showed good signal. Ramachandran said they have not yet figured out the reason, or how well folded they are, or whether they interact with the proteins they’re supposed to, “but the fact that they’re expressing [is] half the battle,” Ramachandran said. “Most times you can’t even express them to study them.”
One important area that NAPPA is not particularly well suited for is post-translational modifications, and in the article, the authors say their system lacks “most post-translation modifications. However, because it is an open system, it is possible to add modifying enzymes or extracts, such as kinases or canine pancreatic microsomal membranes, to test the effect of post-translational modifications.”
David Lubman, a professor of surgical immunology at the University of Michigan Medical Center, who also is developing his own protein arrays, said that if the next-generation NAPPA technology can express proteins in high-content fashion, it could represent an important breakthrough in protein microarray development.
“Obviously, the more proteins you can study, the better off you are. In diseases such as cancer there are so many signaling pathways involved,” he said. “If you have 10,000 proteins expressed, many of them may not be important, but many of them are, and you need to express a large number of the total protein content of the cell to find the ones that are important.”
But the inability of NAPPA to express post-translational modifications as expressed in the disease state could be a drawback, Lubman said.
“They claim they can use kinases and other agents to make these post-translational modifications, but the real problem is that if you are studying cells, especially disease cells, they are post-translationally modified in ways related to disease, including glycosylations, that are difficult to reproduce,” he said.
Regardless, Ramachandran and his colleagues are in the commercialization phase with the platform. Last year, they formed a company called Auguron Biosciences to bring NAPPA to market and inked an option with Plexera providing it exclusive access to NAPPA for use as a content source on Plexera’s own label-free array platform.
In late March, however, Plexera was shut down when its parent company, Lumera, merged with GigOptix [See PM 04/03/08]. As Lumera looks for a potential buyer for Plexera, the option with Auguron will be part of the Plexera’s assets being shopped.
In an e-mail, a spokeswoman for Lumera said the option “could be part of a potential buyer’s interest in the assets.”
In the meantime, Auguron has been seeking funding. According to Jim Richey, CEO of the company, the venture capital community has not been receptive to the NAPPA platform.
“The VC community is not all that keen on platform companies right now,” Richey said.
As a result, Auguron has shifted away from its original plan to base its business on a product model, in which it would compile a proprietary library of cDNA constructed to enable NAPPA and then ship sample formats to clients for custom library development and printing. Its business model is now based on strategic partnerships in which the arrays would be used as antigen panels for the immuno-profiling of disease.