September 1, 2018 (Vol. 38, No. 15)
Data Sharing and Consensus Building Guide Abcam’s Antibody and Assay Validation Efforts
“Take nobody’s word for it.” This motto has inspired scientists ever since it was adopted by the Royal Society in 1660. It emphasizes the centrality of experiments and facts—verifiable facts. But what if you try to verify facts and fail? What if you cannot even take your own “word for it”? That’s the predicament faced by many life sciences researchers who rely on antibodies, which have been known to vary from lab to lab, from supplier to supplier, and even from batch to batch. To get around the problem of antibody variability, antibody tools supplier Abcam suggests that researchers consider another approach: take everybody’s word for it, provided “everybody” means a networked community of antibody-savvy scientists.
The Abcam approach begins with data sharing, which involves scientists making their data available to other scientists. In fact, Abcam itself started as an exercise in data sharing. As the company website states, Abcam was founded in 1998 to provide “the best available antibodies accompanied by comprehensive and open data.”
Abcam co-founder Jonathan Milner, Ph.D., persuaded the scientists who used the company’s antibody products to share data about their results. Then Dr. Milner put that data in a standard format and kept it in one location, making it easy for scientists to choose the best antibodies and assays for their specific applications.
Abcam instituted the Abreviews program, which asks scientists to submit information on product performance. “Review our products for other scientists,” the Abcam website suggests, “[and] tell your peers what works and what doesn’t.” Essentially, Abreviews is designed to bring the power of social networking to antibody characterization. By consulting Abreviews, scientists may benefit from the collective wisdom of their peers. To incentivize scientists to contribute to Abreviews, Abcam offers Abpoints, which can be exchanged for rewards.
Sharing the Validation Load
When Abcam was founded, commercial assays were quite new, and scientists still routinely developed their own lab-grown assays. Validating assays—even for commercial offerings—required combing through the literature or validating each antibody oneself. Since then, the tables have turned. Commercial assays have become the norm.
Commercial assays are often more reliable than laboratory assays. Nonetheless, commercial assays are still products, unique to their producers, rather than interchangeable commodities. Consequently, commercial assays must be carefully scrutinized, insists Michael Weiner, Ph.D., vice president, molecular sciences, Abcam.
“Typically, scientists require two antibodies that work together well and bind to the antigen,” he says. “They often have to validate them to ensure they work together at the desired sensitivities. At Abcam, we’ve already done that for our customers.”
Dr. Weiner asserts that Abcam’s approach to validation sets it apart from other assay developers. “We’re vertically integrated,” he points out. “We have one team that makes the antigen, one that produces the antibody using either a hybridoma approach or phage display, and one that performs quality control. Having a separate quality control group is important, to ensure results are reproducible.”
Abcam relies on a control group that is separate from its development team to ensure reproducible results for each of its assays.
Planning for Innovation
Recognize a need, then fill it. That’s what Abcam did when it instituted its data sharing and e-commerce programs. But Abcam isn’t finished. The company’s mission, says Dr. Weiner, is to get ahead of the needs that will become pressing in three to five years.
“We look at the literature to see what tools are missing and then find a way to supply them,” he says. “In the short term, this involves improving our pipeline to be more efficient in terms of how we make hybridomas, or how we manage manufacturing and the supply chain.”
An example of how efficiency can be increased concerns hybridoma technology. Making a hybridoma typically takes several months, but it can be done, Dr. Weiner has shown, in just 19 days. Dr. Weiner’s method is an in vitro phage display system pioneered at AxioMx. (This company, which Dr. Weiner founded in 2012, was acquired by Abcam in 2015.) The AxioMx system is currently being refined at Abcam.
“At AxioMx, we were part of an NIH program to make recombinant antibodies, in response to the challenge of getting reproducible agents,” Dr. Weiner says. “But antigens are a challenge to make, also. We’re targeting the class[es] that are difficult, but doable. These are antigens a grad student might spend a year or more making.” Following the adage (attributed to mathematician Norbert Wiener) that “the best model of a cat is a cat…preferably the same cat,” Dr. Weiner says, “we like to use that antigen in its natural state.”
“In the long term,” Dr. Weiner continues, Abcam’s mission “involves identifying new opportunities and capabilities.” Dr. Weiner says the company is focused on designing antibodies against difficult targets—those that are not immunogenic or that are hard to quantify. Another involves designing the antibodies without access to the antigen, including antibodies that work against denatured or fully folded proteins.
“Antigens that work against the easily made targets already have been made,” he notes. Consequently, Dr. Weiner orients his group toward more challenging targets such as membrane-spanning proteins.
Abcam, says Dr. Weiner, is also looking to enhance its multiplexing prowess. Back in 2015, the company acquired Firefly BioWorks, which specialized in the development of multiplexed assays that could be used on standard laboratory instrumentation. Abcam has since built upon Firefly’s multiplexing platform to develop FirePlex assays, which assess up to 65 microRNAs or 75 proteins simultaneously, and the 384-well-plate FirePlex-HT Immunoassays.
In coming years, Dr. Weiner anticipates, researchers will want to look at the proteome the way they looked at the genome in the 1990s, which means they will want to begin conducting comprehensive surveys. For example, they may probe entire enzyme classes to isolate interesting changes.
“Assays could be designed to analyze all post-translational modifications at the same time,” Dr. Weiner suggests. “Researchers will want to trace the signal all the way to the nucleus of a cell, to investigate complete pathways. To do that, they need affinity reagents. That’s where we come in.”
Reviewing Meaningful Successes
“Getting users to contribute data was a huge challenge—and a huge success,” Dr. Weiner recalls. Scientists were concerned about protecting proprietary information, and the concept of sharing information about their assays was new. Abcam explained how sharing that data could advance science without threatening their intellectual property positions.
Another big challenge was persuading researchers to accept antibodies from commercial entities rather than from networks of colleagues. Eventually, researchers saw that commercial antibodies were standardized and, consequently, capable of generating more reproducible results than were lab-generated assays.
The company also was the first in the life sciences to make its catalog available electronically, on the Internet. By the time the company had more than 100,000 products, Dr. Weiner says, “a printed catalog couldn’t keep up with the additions,” at least, not without distributing a lot more paper and risking an unacceptable ecological impact.
More recently, Abcam has knockout validated nearly 1500 antibodies in its library. (These antibodies, which tend to be the ones most commonly used, bear a “knockout validated” seal.) Plans call for applying this validation method across the company’s antibody catalog.
Looking to the Future
Abcam has come a long way from its early days, when it delivered antibodies in an ice bucket to local clients. Now the company’s state-of-the-art temperature-controlled shipping system delivers antibodies to clients in more than 100 countries. Abcam has 11 locations and conducts research in the United States, Europe, and China.
Abcam plans to consolidate its position while expanding its geographical reach and development partnerships. To that end, Abcam recently signed a memorandum of understanding with Shuwen Biotech to develop and commercialize companion diagnostics in China.
The company also plans to continue innovating. For example, it intends to help researchers reach their ultimate goal, which is having one antibody that works for every application, such as flow cytometry and immunohistochemistry. “Researchers buy antibodies and try them in ways and applications for which they haven’t been tested. Not having the ability to do all the things a researcher might possibly do is something for us to work on,” Dr. Weiner says.
That goal is one for the distant future. In the meantime, Abcam does have antibodies that work well in a variety of different applications. Currently, Abcam is developing and validating additional antibodies and commercializing more kits.
“I’d like to see post-translational modification kits, and products that use antibodies in different formats,” says Dr. Weiner. “ChIP-Seq, which uses antibodies to measure gene expression levels in cells, is a very interesting area, along with immuno-oncology—a field that didn’t even exist when I started my career many years ago.”
Abcam
