Antibody proteins have been the fastest-growing class of approved drugs over the last two decades, but current antibody drug discovery methods are expensive, time-consuming and labour-intensive.
A new technology created by researchers at the National Institutes of Health’s Vaccine Center, the University of Texas and the University of Kansas, which was described in the January 2018 issue of Nature Biotechnology, aims to overcome all of these issues and has so far yielded impressive results. Not only is the screening process much faster than with existing technologies, it has also led to the discovery of more potent antibodies.
Instead of working at the single-cell level, the new technique processes millions of cells at a time. Additionally, instead of only using live cells, which means researchers only have one shot at discovery from each sample, it ‘recreates’ antibody proteins from those cells so screening can be carried out repeatedly, giving a greater chance of identifying valuable antibodies.
The technology is already being used at several research institutions and one of its creators, Brandon DeKosky, assistant professor of chemical and petroleum engineering and pharmaceutical chemistry at the University of Kansas, has high hopes that its widespread application could accelerate our understanding of how to best leverage our immune systems to prevent and cure human diseases.
Elly Earls catches up with DeKosky to find out more about how the technique works and its potential to identify new antibodies and help us design more effective vaccine strategies.
Elly Earls (EE): Why are antibodies such an important source for drug development?
Brandon DeKosky (BD): Antibodies attach to their protein target with high specificity and signal it for destruction by the body’s immune system. The high specificity allows antibodies to provide a potent therapeutic benefit, often without major side effects, making antibodies the fastest growing class of therapeutics over the last 15-20 years.
The top-selling antibody drug is Humira, which is used to treat auto immune diseases like rheumatoid arthritis. Others include Herceptin for breast cancer and Palivizumab, which is used in the prevention of respiratory syncytial virus (RSV) infections.
EE: What are the limitations of existing technologies for antibody drug discovery?
BD: Previous technologies have involved working at the single-cell level in 96- or 384-well plates to isolate the antibody genes from single cells. These technologies are slow, expensive, and require a lot of labour. They also only give a small glimpse of the total antibodies that might be made by a single individual. Using traditional technologies, you can identify around 200 different antibodies in a couple of months.
EE: How does the new technology overcome these limitations?
BD: We process millions of cells at a time, and we even take it a step further to ‘recreate’ those antibody proteins on yeast inside a laboratory. The yeast each display one unique antibody molecule on their surface, and we can use these artificial yeast populations (we call them libraries) to discover new antibodies. For example, we were able to determine new antibodies that targeted Ebola virus by screening the yeast libraries.
This is a major advantage over traditional technologies because if you use live cells, you only have one shot at discovery from each sample and if you’re dealing with samples from, say, a clinical vaccine trial or a longitudinal HIV infected patient cohort, there are just not that many vials around. Even if you know there are valuable antibodies in them, you might not find those high value antibodies on the first try.
With our technology, we’ve be able to screen many more antibodies than you can with current technologies – orders of magnitude more – and the antibodies we’ve found are extremely potent, in some cases more potent than therapeutics that are currently in clinical trials.
EE: Can you explain in layman’s terms how the technology works?
BD: We isolate single cells into individual droplets. This collection of droplets is similar to when you shake a bottle of salad dressing – there are many droplets of water suspended in oil. We use each of those millions of droplets as an individual compartment to perform single-cell reactions, letting us process millions of cells in the span of an afternoon. Then once the antibody genes are collected, we transfer those antibody genes into yeast so that the yeast cells can make antibody proteins on their surface. From there we see which of the yeast bind to proteins of interest (e.g. an Ebola virus envelope protein), and we isolate those yeast and sequence the antibody genes that they encode.
EE: Talk me through the development of the technology. What were the biggest challenges and how did you overcome them?
BD: We had a significant challenge for getting the single-cell reactions to work inside droplets, which we were able to solve by coming up with a new way of capturing and purifying the genetic material using magnetic beads. We also had a major challenge in getting the broad range of human antibodies to express on the yeast surface, which we were able to solve by introducing new helper proteins and gene sequences. Last, we needed to set up the systems for screening yeast and then recovering the antibody sequences from those yeast libraries, and we tried several things before settling on a workflow that we were happy with.
EE: Now that the technology is ready, what are the next steps?
BD: I’m very excited about applying these technologies for antibody discovery against rapidly evolving viruses, especially HIV-1 and Epstein-Barr virus (which causes infectious mononucleosis). I think it will also give us a broader overview of how we achieve immune protection against viral infections and help us design more effective vaccine strategies.
We are also improving the system both experimentally and with respect to the computation and data analysis to make things faster and more streamlined. Right now, the computation is a little bit of a bottleneck and we’re trying to address that.
EE: How much could this technology speed up the drug discovery process?
BD: I’m hopeful that we will be able to discover antibodies that we couldn’t find before, that maybe will show us new ways to target these viruses that would lead to better therapeutics or vaccines. I think these technologies will really accelerate our understanding of how to best leverage the power of our immune systems to prevent and cure human diseases.
Dr. Brandon DeKosky is an assistant professor at the University of Kansas’ Departments of Chemical Engineering and Pharmaceutical Chemistry, where his laboratory leverages recent advances in next-generation DNA sequencing technologies to achieve a more comprehensive understanding of immune function and to accelerate the development of new vaccines and therapeutics.