Proteomics researchers have long deplored the lack of readily available, sensitive, and specific antibodies or similar capture reagents — crucial tools for the characterization of novel proteins. But now HUPO and others are preparing to address this need, though many details still need to be resolved.
HUPO announced its antibody initiative last week at a workshop hosted by the NCI and FDA at the National Institutes of Health. The long-term aim is to establish a resource of antibodies against every human protein. “It should be freely accessible to researchers and as free as possible to companies,” said Matthias Mann of the University of Southern Denmark, who presented the HUPO initiative. Several projects across the world with funding from different agencies could work towards this goal, coordinated by HUPO, which would also specify standards for the reagents. Meanwhile, in a separate effort, researchers in Europe are getting ready to apply for funding for a multi-center European project with a similar mission.
Although thousands of antibodies are already available from commercial and academic resources, probably more than half of human proteins have never been expressed or purified before, so no capture agents exist for them. In addition to academic scientists, pharma and biotech companies are also interested in a public resource, because they are reluctant to dedicate the lab resources and manpower required for such a wide-ranging project. And many agree that public efforts to raise antibodies will complement, rather than compete with, the numerous commercial efforts to develop antibody technologies. Reagent companies, for example, are unlikely to raise antibodies to proteins that are of little commercial interest. “It doesn’t make economic sense for them to do this because most of the antibodies will not be ordered very often,” said Mann.
One of HUPO’s first tasks is to determine which technology would be most suitable for creating a uniform resource of high-quality antibodies that would make experiments more comparable. “We don’t favor anything off-hand, but it has to be mature and it should be free of patent considerations up front,” said Mann. This might rule out technologies that are either patent-protected or still in development, such as phage-display, scaffold technologies, or aptamers, unless licensing fees were waived or their suitability was proven in a pilot project. But depending on which projects will be able to attract public funding, HUPO might be left with several technologies after all. At the moment, HUPO is looking for technology proposals, both from academic institutions and from companies. To date, these include proposals from San Diego-based GenWay Biotech, which makes chicken polyclonal antibodies, a group at Pacific Northwest National Laboratory, which generates single-chain human antibodies in yeast, and the German Society for Proteome Research, which is planning to create a resource of monoclonal mouse antibodies.
At a workshop in about three months, HUPO wants to decide on criteria to judge these proposals, and it hopes by then to have raised funding from pharmaceutical companies to finance one or two pilot projects, chosen from the proposals, Mann said.
However, Mike Taussig, group leader at the Babraham Institute in Cambridge, UK, and chairman of the “Integrated Approaches for Functional Genomics” program at the European Science Foundation, does not think a specific technology should be chosen. He is planning to coordinate a consortium of antibody-producing centers across Europe that would apply for funding from the 6th European framework program. “I think it would be wrong to put all your eggs in one basket,” he said. “The different centers would do these things in different ways, and then the user could pick and choose what he wanted.” This consortium would likely concentrate on antibodies to novel proteins, and maybe standardize existing reagents, he said.
In fact, the European Union is a funding source both initiatives are considering. The monoclonal antibody project proposed by the German Society for Proteome Research and Taussig’s planned consortium want to apply to the 6th European framework program, which has put out calls for “expressions of interest” by June and is expected to call for full-scale applications later this year. The German initiative is looking to raise 40 million euros ($36 million) from the EU, on top of 40 million euros from pharmaceutical companies. The project already has a commitment from Roche for 5 million euros, said Mann. Other funding sources for HUPO-endorsed projects could include Asian governments, such as China, or the NIH, he added. “We heard from NIH that they are quite interested that more antibodies are made.” However, NIH applications would require a more disease-focused approach, he said.
Once reagents are available, they could be distributed to researchers in different ways. HUPO might entrust a company with this, said Mann, because funding agencies might not permit antibodies to be sent overseas using government funds. Taussig said he hopes to establish an administrative center that would distribute the reagents produced by his proposed consortium, “where [the antibodies] could be collected, the database could be established, [and] people could then obtain them.”
But it will be several years before any resource is available, largely for technical reasons. One of them is expressing the proteins — or even parts of them — to raise or select antibodies against them. “There are many different systems that people use, and none of them work well for all proteins,” cautioned Leigh Anderson, CSO of Large Scale Biology. And making antibodies in high-throughput mode is no easy task, even with established methods. “Few people have ever tried to systematize or automate the process of making monoclonals so far,” Anderson said. “In the course of scaling up these processes, we are likely to have many lessons to be learned.”
The antibody initiatives might also run into the problem of gene patents that cover antibodies to the gene product. “We are trying to get some clarification on this,” said Mann. “If people do say that you can’t put this particular antibody on your protein chip because I have a patent, then we are in a very bad situation.” But unless an antibody is of high commercial value, for example as a therapeutic, patent holders might not enforce their claims.
Finally, will several large-scale rivaling efforts to provide an antibody resource benefit researchers? “If everybody goes off and does their own thing, there will be no coordination,” said Taussig. He and Mann were unaware of each other’s plans at the time of this interview. On the other hand, Large Scale Biology’s Anderson thinks overlapping initiatives might not necessarily tackle the same proteins, and if they did, “it wouldn’t be bad because they would be generating different antibodies,” he said, providing a richer variety for researchers to access.
— JK