.png)
Target Identification, Phenotypic and Cell-based Screening
Oxford Global’s Discovery series was delighted to host some brilliant experts in the field of drug discovery for a discussion group entitled Target Identification, Phenotypic and Cell-based Screening.
Meet the Experts
Leading the conversation was Simon Bushell, Head of Structural Biology, Orbit Discovery. Bushell explained that Orbit was founded to answer the “universal question” of screening against targets in an in vitro setting, as well as the translational friction of functional screening. “A compound that you might identify and see binding effects in a test tube, does not always translate to a functional context when looking at cells.”
Orbit’s platform was therefore developed with this question in mind: “how can we come up with a universal system that we can use in both affinity and functional screening contexts?” The platform that they have developed, Bushell summarised, is a “one-bead, one-compound peptide display platform.”
Joining Bushell for the discussion was Heiko Wurdak, Group Leader and Lecturer and Head of Stem Cell and Brain Tumour Group at the University of Leeds. His lab tries to fill the gap in patient-centric disease-focused drug discovery: “we are very much inspired by what small molecules can do to change cellular behaviour.”
Nikolas Gunkel has also worked in crop science, but now heads the Drug Discovery Unit at the German Cancer Research Center (DKFZ) in Heidelberg. The DKFZ use their expertise in target validation, assay development, and synthetic chemistry to kickstart drug discovery projects independently or with partners.
Pre-identified Targets Versus Platform-Based Targeting
Bushell asked the attendees whether their work consisted of using pre-identified targets or platform-based assay.
One attendee answered that, on the one hand, they were fairly platform-based in their screening, but they also look for specific targets to screen against. Those are identified based on the literature, using the expression and essentiality data that is available, and based on their own AI and knockouts.
Bushell pointed out that this contrasted with many pharma approaches that focus on modulating function rather than knocking out completely. Gunkel added however that it did correspond to the traditional cancer therapeutic approach “where you seek cell death in certain cases.”
Gunkel said, “it’s not so much about modulation, especially as modulation can often make tumour heterogeneity much worse.” He said that an efficient kill mechanism can affect most of the tumour mass before alarming the immune system to take care of the rest. “However, if you only modulate, you only affect subpopulations in the tumour, because the others have enough buffer capacity to compensate for whatever is modulated.”
“We gave up on thinking there might be a ‘magic bullet’ that hits cancer cells and not other cells.”
Gunkel’s group also focuses on biomarkers; he said of their approach: “We gave up on thinking there might be a ‘magic bullet’ that hits cancer cells and not other cells.” Instead, they tried to find out what makes a group of patients more responsive to a given drug. Resourcefully, they also try to revive failed drugs which have a positive PK or toxic profile but that failed in the clinic “because they have potentially been used on the wrong patients.”
Bushell noted that this was something his team were paying attention to as well, as Orbit are interested in targeting peptides. “The peptide doesn’t necessarily have to have a particular effect on the cell, but it might be used as a delivery system for a particular payload. In that regard, having it focus on some sort of biomarker can be beneficial.”
There was then a short exchange between Gunkel and Wardak regarding mitochondrial targets. Gunkel recounted a conversation with a colleague where they had found that a class of ROS-inducing compounds had turned out to be too toxic – especially to the heart.
They found that if they formulated one of their drugs so that it did not induce ROS in mitochondria, they achieved high selectivity – “or very strong correlation with certain biomarkers, which are indicative of ROS sensitivity. So, when it comes to the class of ROS inducing drugs, you might want to spare mitochondria, because they are universally essential,” explained Gunkel.
- How DEL Screening is Used in Living Cells
- Target Identification: Navigating the Intricate Environment of Human Biology
Wardak said that ROS was a big concern for him as well: “it could be seen as a cancer vulnerability.” He added that they had seen zero effect on ROS or even mitochondrial membrane potential itself, which told them that they were not directly blocking any membrane activity in the mitochondria or ROS activity. “That was reassuring.”
Furthermore, Wurdak divulged that “from what [they] have seen, it is not impossible to affect mitochondrial biology in cancer cells specifically.” Wardak’s lab has a compound that gets into the cell and hits HSPD1 in the mitochondria. They originally found this compound through screening when working in the context of adult neurogenesis and stem cell differentiation and were surprised to observe selective brain cancer cell death via HSPD1 inhibition and metabolic dysfunction.
The compound was optimised for brain pharmacology in mice, and they were fascinated to see the cells dying. “I couldn’t believe what I saw because I had worked on stem cells and these compounds in many cell types and nothing like that had ever happened.”
Moreover, in in vitro folding assays, they see it inhibiting the other folding activity: “we see in cells, after compound treatment, the selective aggregation of this protein in cancer cells but not in the normal cells.”
“A Reluctance to Proceed Based on Phenotypic Changes”
Bringing a new direction to the conversation, Bushell asked if there is a reluctance to advance drug investigations based only on phenotypic changes. “Do people want to see specific interactions with specific targets?” he added, “is it a regulatory hurdle or is it just a difficult case to make to other scientists?”
Gunkel confirmed: “I think it is a regulatory challenge.” He mentioned that many academic scientists aren’t sufficiently focussed on the selectivity of the compounds they us for mechanistic studies. [But] “there is obviously now a movement where tool compounds are defined very clearly. There is a list of tool compounds that deserve that name, and the rest are ‘dirty’”
Furthermore, he said that “the regulatory process loves to see a main target and plausible drug effect markers. At the same time, it’s easy to ignore that this is not the whole truth.” Gunkel explained that “polypharmacology” is always in effect when developing a target: “if you look, you will likely find other targets.”
“Historically, when in vitro screens were invented for drug discovery, that was the moment when regulatory affairs demanded to have a clear picture of a target.” To this, Bushell concurred that there is now a growing need to “differentiate the more diffuse phenotypic effects, rather than the one-drug-one-target paradigm.”
We look forward to hosting our upcoming discussion groups, the next of which will take place in September and discuss Novel Technologies & AI Applied to Drug Design. Join our Discovery newsletter to track our latest content, events, and discussion groups.
Related Resources
