Previously I noted that FBDD provides a means for academic groups (and start ups) to negate the advantage that Big Pharma’s massive screening collections give them. Fragment screening and structural characterisation of fragment binding broadens the scope of a structural biology group’s activities. I do believe that a package of fragment binding modes and affinities is something in which a pharmaceutical company would be interested. But what if an academic group wants to move the fragment hits (which I refuse to call ‘frits’ because I don’t work there any more) further along the optimisation trajectory? I’ll outline some of the issues that need to be addressed and this post is intended to stimulate discussion rather than being a last word on academic FBDD. Please feel free to comment if you’d like to challenge anything or flag up anything that’s been overlooked or oversimplified.
The post-screening phase of lead generation will generally require chemical synthesis. Academic synthetic chemists typically focus on synthesising complex natural products or developing new synthetic methodology so it can be difficult to interest them in the more mundane business of lead generation. To be fair the synthetic chemistry required for lead generation is unlikely to be of sufficient novelty or complexity to earn a graduate student the PhD in synthesis which will be his or her primary objective. There are medicinal chemists in academia but in some cases these happen to be synthetic chemists who think that medicinal chemistry is simply a branch of synthetic chemistry. Also synthetic chemists in academia tend not to be interested in molecular design. The net result is that it can be difficult for a protein structure and fragment screening group to find academic collaborators to take the project into the post-screening phase even when there may be synthetic chemists in the same institution.
As you move from screening to post-screening phases of lead generation the work becomes more multi-disciplinary. Unfortunately ‘multi-disciplinary’ isn’t something that usually gets done well in academic institutions although the problems are often less to do with skills than with organisation (and occasionally egos). In addition to the molecular design and synthesis, it will become necessary to run assays to demonstrate that affinity translates into inhibition of the target enzyme (it’s likely to be an enzyme if you’re doing FBDD). For some targets (e.g. antibacterial or kinases) it’ll be necessary to demonstrate some cellular activity. As you approach micromolar potency you may want to check that compounds in your lead series have sufficient aqueous solubility and don’t have particular affinity for anti-targets such as hERG and CYPs. This is much less of an issue if the main objective is publication but is something to be considered if you’re hoping to flog the results of the work to a pharmaceutical company.
That last comment gets me onto a tough issue for an academic lead generation group. How might you persuade a pharmaceutical company to buy your lead series? The main problem is what you’re trying to sell is information and you’ll need to show that you’ve got something good without giving it away. Life is easier if you own the relevant intellectual property but the synthetic chemistry that needs to get done to secure patent cover is not always going to get people PhDs in synthesis. The other point to remember is that Pharma people may not be willing to look at what you’ve got under a confidentiality agreement because of potential for compromising their own IP position. This can become a serious issue if you’re trying to stake a claim for future activity against related targets (e.g. all tyrosine kinases) or to negotiate exclusivity by preventing a company from using leads from competing programs.
So far this post has focussed on the difficulties (which in Pharma-speak would re-branded as personal development opportunities by the happy-smiley folk who inhabit the HR ether) of doing post-screening fragment-based work in an academic environment. If the primary objective is publication then you can write up at any point that is convenient which means that even a small amount of synthesis can have impact. In contrast, commercial lead discovery organisations need to create a secure IP position before they can publish. When publishing affinities and structures of protein-ligand complexes it’s always worth looking out for results that have relevance that goes beyond the specific project. Examples of synthetic elaboration of a fragment leading to a change in its binding mode are particular relevant and the prototypical (low molecular complexity) nature of fragments means that differences in affinity are more easily interpreted.
Getting pharmaceutical companies interested in the output of an academic fragment project is not trivial. A lot depends on the value of the target and the quality of the leads that have been generated. However, getting to leads requires organisation and realising the value of them requires commercial awareness. Organisation is about persuading people that they’re better off working together and can take time to put in place. Commercial awareness is more difficult to acquire and it’s probably best to try to keep things as simple as possible when starting out. Although this may all seem a bit daunting it’s worth remembering that synthesised compounds will typically be novel and that synthesis can be directed away from known ligands. Also one should not forget the ‘supporting data’ of crystal structures and measured affinities.
This is a good point to wrap up. I believe that FBDD provides an excellent framework in which to both train researchers and do high quality science. FBDD also extends the range of options available to academic researchers for collaborating with industrial partners. That's where I'm going to leave it so feel free to comment if anything that I've said (or not said) has annoyed you.