One of my parents favourite movies was “Kind Hearts and Coronets” which describes the personal history of a nobleman who was I think 11th in line to a title. All of the characters shown in the poster were actually played by Alec Guinness. He successfully executes plots to eliminate all of the individuals standing between him and the title. He is undone at the end by his writing the story of what he had done. I will attempt the same here, while I hope avoiding undoing what we have accomplished by providing a personal history of how we commercialized the world’s first aptamer based diagnostic kit.
This will be part I: dealing with our first discovery of aptamers for a target molecule.
We started to try to develop aptamers for ochratoxin A in 2006. Ochratoxin is a mycotoxin produced by fungi that infect grain and grapes and several other food products. Our first attempts at aptamer selection carefully following protocols published in the literature resulted in failure. All of our top sequences exhibited the same level of binding to our immobilized support resin as they did to the target molecule.
We decided that we needed to test whether this wild idea, that single stranded oligonucleotides could actually mimic antibodies in their ability to bind to target molecules, was actually valid. Our first several tries with published aptamers did not work. In one case we contacted the authors to try to help us out. They told us that the sequence published was wrong and provided us with a different sequence. The new sequence did not work any better. In all cases it was not that the aptamers did not work as well as advertised, they did not work at all. Some bound to resin, some seemed to nothing at all. We faithfully and carefully reproduced every step of the binding assay, including the use of the exact buffer used.
We were starting to think that maybe starting an aptamer company was not the best idea we had ever had. In the end though, we realized that we could actually order cocaine from Sigma and that the cocaine aptamer did work as promised. Armed with this courage, we went back to selection for aptamers for ochratoxin A. We pushed this to 17 rounds of selection, cloned the library into E. coli and characterized individual clones in binding assays (this was before we pioneered the application of next generation sequencing in aptamer selection. I am amazed now that we could actually succeed by cloning 50 sequences). By pushing the selection this hard we found aptamers that worked extremely well. The aptamer that we published for ochratoxin has been widely cited globally. This is still probably one of the best aptamers for a small molecule ever discovered. We learned a lot from this first selection, but probably the most important things that we learned is that one; yes, aptamers are real, this is not a myth, and two; the selection strategy for an aptamer is the most important part of the process
One issue that we ran into with ochratoxin A was that it is sparingly soluble in aqueous solutions. We did our selections with the aptamer in the presence of methanol and in the end discovered aptamers that could tolerate up to 20% methanol, 10% ethanol, and 5% DMSO without a loss in performance.
Just like in the movie “Kind Hearts and Coronets” the methods used required refinement, and Alec Guinness improved the execution of his plots (pardon the pun) better as he went along. At NeoVentures we still devise a custom selection strategy for each new target we face, and we are still learning from our mathematical models on how to improve the selection process.
