HotSpot Therapeutics recently announced a strategic collaboration with AbbVie to develop IRF5 transcription factor inhibitors for autoimmune diseases. The following blog post provides some valuable real-world perspective on why this initiative is such an important breakthrough in immunology therapeutics.
Cocktail party conversations can become rather interesting when you tell people you work for a biotech developing medicines for undruggable targets. But the beauty of working in the world of scientific research and drug discovery is that the status quo rarely ever stays that way for long. Accordingly, what’s undruggable today has the potential to become a breakthrough biologic target tomorrow.
Nowhere is this phenomenon on clearer display than in the work currently taking place to target Transcription Factors. These hair clip-like molecules clamp onto DNA, thereby turning many genes on and off. Essentially, they behave like switches, determining the expression of a gene in the right cell at the right time to drive a specific function. That can be a good thing: for example, when determining inherited traits and regulating critical cell functions. Or a bad thing: for example, when increased expression of certain genes contributes to the formation of cancer.
In fact, one of the first breakthrough pieces of research examining the links between transcription factors and cancer was James E. Darnell’s 2002 paper Transcription Factors as Targets for Cancer Therapy, in which he wrote:
“A limited list of transcription factors are overactive in most human cancer cells, which makes them targets for the development of anticancer drugs. That they are the most direct and hopeful targets for treating cancer is proposed, and this is supported by the fact that there are many more human oncogenes in signaling pathways than there are oncogenic transcription factors. But how could specific transcription-factor activity be inhibited?”
Cracking the Code on Transcription Factors
Although Transcription Factors have long been a drug discovery target, they have always been elusive – or undruggable – targets because they do not have active sites. As Darnell observed, the prospect of being able to regulate Transcription Factor function holds the potential to reframe the way we treat disease but achieving that goal with traditional small molecule screening approaches is impossible. Put simply, because Transcription Factors lack a catalytic engine, the traditional playbook of using substrate site inhibitors to modulate cellular activity does not work. On top of that, Transcription Factors contain a complicated DNA/protein complex, which significantly complicates researchers’ efforts to develop robust chemical screening paradigms.
What Darnell understood at the time was that small molecule modulation of transcription factor activity had the potential to transform cancer treatment but targeting the necessary protein-to-DNA or protein-to-protein interactions that control transcription factors was next-to-impossible. Specifically, he recognized that protein interaction surfaces are typically flat and do not present the deep pockets found in enzyme active sites, which makes it hard to create an inhibitor that will bind properly.
Computational Sciences Fill Missing Link
Thus, just thinking about how one would drug a protein that didn’t have an active site was a conundrum for scientists and the industry until new technologies were developed. Now, thanks to recent advances in technology that allow us to understand molecular function at a deeper, more granular level, it is possible to identify the specific molecules that will bind to these transcription factors in a predictable, repeatable manner.
This last part – predictability and repeatability – is important. For all of their elusiveness, Transcription Factors do have some common characteristics within their underlying structures, and identifying those patterns is the first step to being able to drug what was once undruggable. The key to unlocking those patterns has been advances in computational sciences that make it possible to develop machine learning algorithms that identify the fingerprints of natural hotspots, or the functional pockets on proteins beyond the active site that control cell function. By processing petabytes of data on detailed cellular structures and previously unknown regulatory relationships, we’ve been able to identify 1,500 proteins containing natural hotspots that control protein function outside of traditional active site cellular function.
From there, it becomes possible to understand how specific pockets drive the behaviors of the protein and thereby design pharmacological modulators that impact transcription factor function.
Ultimately, by using machine learning technology in concert with a cutting-edge biology, chemistry and biophysics toolkit, we are able to winnow down the total universe of transcription factors to a handful of molecular mechanisms that can be targeted, analyzed and modulated in ways that were never before possible. We’re seeing this capability come to life right now in our work developing small molecule antagonist directed towards interferon regulatory factor 5 (IRF5), a Transcription Factor that functions as a master regulator of innate human immunity. IRF5 is a genetically validated Transcription Factor that is important in a variety of diseases and has always been considered undruggable by small molecules because it doesn’t have an active site. But it does contain a natural hotspot which our Smart Allostery platform has uncovered. We are rapidly advancing the chemistry required modulate its activity.
Transcription Factors are very different, necessitating fresh thinking to uncover new biological real estate present on the protein and complimentary chemical matter that will affect a desired behavior on those protein hotspots. Thanks to advances in technology – and the refusal of a group of determined scientists to accept the conventional wisdom that Transcription Factors are “undruggable” – new classes of drug targets to be prosecuted are being uncovered every day. As we continue to learn more about the unique composition of these proteins and refine the process of predictably targeting and treating them, we have an opportunity to bring an entirely new target class of modulators into human clinical trials for important diseases.