Rapid Identification of Novel Sirtuin-1 Inhibitors with AI-powered Reduction of Large Compound Libraries

A rapid and effective search for novel drug candidates still poses a massive challenge in drug discovery. Despite the advances in high-throughput screening (HTS) techniques, the development of new pharmaceuticals still emerges as the laborious and time-consuming stage of research programs. “Wet” screening has a number of limitations, such as the throughput of the assays, cost and commercial availability of compounds from the shelf. While artificial intelligence (AI) emerges as a powerful tool that helps to overcome bottlenecks of the HTS approach.  

In the study, we selected Sirtuin-1 as a target since the enzyme is implicated in a number of vital cell processes. It removes acetyl groups from various proteins using nicotinamide adenosine dinucleotide (NAD). Hence, not only reversing one of the major post-translational protein modifications important for various cellular functions such as chromatin remodeling, nuclear transport, and actin nucleation but also interfering with metabolic processes and energy homeostasis. As one of seven members of the sirtuin family, Sirtuin-1 was specifically shown to be relevant for protein tyrosine phosphatase 1B, a negative regulator of the insulin signaling cascade, and p53 regulation, amongst others. It is therefore not surprising that Sirtuin-1 is involved in obesity-induced diabetes, aging-associated diseases, and cancer.

A plethora of information is available that concerns the different possibilities to inhibit and regulate  Sirtuin-1, including a challenge which has as its target to determine novel and specific inhibitors to Sirtuin-1 hard, but not impossible. In the current study, we present how productive collaboration between Enamine/Bienta Ltd. and PharmAI has allowed us to discover novel Sirtuin-1 inhibitors. Enamine’s world largest chemical space, e.g. the Enamine REAL space and powerful high-throughput screening platform in collaboration with the state-of-art DiscoveryEngine allowed to identify the “needle in the haystack” within a limited timeframe with a much higher success rate compared to traditional approaches.