1
Targeted Protein Degradation - The Story So Far

Daniel H. O Donovan Laura M. Luh and Philipp M. Cromm

Bayer AG, Research and Development, Pharmaceuticals, Drug Discovery Sciences, Müllerstraße 178, 13353, Berlin, Germany

1.1 Introduction to Targeted Protein Degradation (TPD)

Protein homeostasis is a pivotal process in cells comprising protein production and removal. The cell has two major pathways to achieve the latter; primarily, the ubiquitin proteasome system () that degrades the vast majority of all proteins, or alternatively lysosomal degradation and autophagy, capable of degrading whole organelles such as mitochondria [1, 2]. The UPS is centered around the 76 amino acid small protein ubiquitin that is attached or removed from other proteins as a post-translational modification () and encodes via a complex pattern of ubiquitination the fate of the target protein [3, 4]. Ubiquitination is achieved by the concerted actions of a series of ubiquitin ligases (E1, E2, and E3) that add ubiquitin chains to substrates, thereby marking them for proteasomal degradation. Autophagy is the second major protein degradation pathway and is a self-degrading mechanism in which a cytoplasmic material is sequestered in double-membrane vesicles and delivered to the lysosome for degradation. Using chemical biological tools to control these endogenous degradation machineries is the hallmark of targeted protein degradation (). While conventional small molecules generally block the activity of a protein, small-molecule degraders aim to eliminate or deplete the protein of interest () following the paradigm of event-driven pharmacology [5-7]. This exciting concept not only expands the drug discovery toolbox but also enables broadening of druggable space and has therefore drawn tremendous attention within both the pharmaceutical industry and academia in recent years. The TPD concept is emerging as a new therapeutic modality with many such compounds entering the clinic and a multitude of start-up companies arising with a focus on TPD.

There are essentially two classes of small-molecule protein degraders: monovalent and bivalent degraders. Destabilizers such as small-molecule hydrophobic tags (s) or selective estrogen receptor degraders (s) as well as molecular glue degraders represent monovalent molecules, while proteolysis targeting chimeras (s®) and related technologies comprise the class of bifunctional degraders (Figure 1.1) [8]. This chapter aims to provide a brief overview of the milestones in TPD achieved to date, including recent clinical advances and the enticing opportunities beyond proteasomal degradation.

Figure 1.1 Schematic representation of different TPD approaches. While the complete molecular machinery by which HyTags and SERDs induce protein degradation remains unclear, molecular glues and PROTACs act via a E3-ligase-mediated mechanism resulting in proteasomal degradation.

1.1.1 What Can Be Expected from this Book?

This book aims to provide a comprehensive overview of the field of TPD. The journey toward controlling protein abundance and function began long before the current excitement around TPD, which has seen a surge in activity during the past 5-10?years. Thus, several techniques modulating protein levels were already established in the early 2000s, consisting mostly of chemical biology tools [9].

To appreciate the rationale behind TPD, one must first familiarize oneself with the cellular machinery controlling protein degradation. The complex cellular processes driving protein degradation, i.e. the UPS and E3 ubiquitin ligases, are highlighted in the contributions of Doris Hellerschmied and Vincenzo D'Angiolella in Chapters 2 and 3, respectively. For a detailed understanding of the molecular mechanisms and structural complexity of the degrader machinery, Morgan Gadd presents Chapter 4 on ternary complexes of small-molecule degraders. Although today we still rely on structural biology to unravel the machinery of ternary complexes, computational approaches are being refined to predict the ternary interaction of a degrader and its protein targets. John Karanicolas in Chapter 5 provides an overview of the current efforts on ternary complex prediction and what we might expect in the future. Switching gears to molecular glue degraders, Cristina Mayor-Ruiz in Chapter 6 explains how the field is progressing from serendipitous discovery to the directed and rational hunt for novel molecular glue degraders. After discussing the general principles of TPD, a specialized discourse on TPD in neurodegenerative diseases and covalent protein degraders is provided by Fleur Ferguson and Xiaoyu Zhang in Chapters 7 and 8, respectively. Controlling protein levels before the protein is translated from RNA provides another means to achieve TPD. The current approaches to modulate protein abundance by addressing RNA are highlighted in Chapter 9 by Andrei Ursu. The concept of inducing novel protein-protein interactions (s) to facilitate the transfer of a ubiquitin onto a POI has sparked the creative mind of several groups to harness the concept either for additional PTMs or other forms of protein degradation. Thus, George Burslem provides an overview of heterobifunctional molecules beyond PROTACs in Chapter 10.

As the main goal of this book is to provide a summary of the TPD field from all angles (including both academia and the industry), the transition of small-molecule protein degraders from chemical biology tools to drug discovery and clinical applications is an essential part of the story. Proof-of-concept () studies that often originated within academic groups provide the foundation to generate excitement and ultimately produce convincing preclinical data. However, to develop viable clinical candidate degraders, many additional parameters need to be optimized and other challenges must be addressed. To shed light on the path towards clinical candidate development, a team from Boehringer Ingelheim provides an overview of TPD in drug discovery in Chapter 11, while a team from Roche digs deep into the drug metabolism and pharmacokinetic () properties and hurdles associated with small-molecule protein degraders in Chapter 12. Finally, in Chapter 13, a team from Arvinas shares their views on their exciting journey of TPD from a chemical biology tool to a clinical candidate.

1.2 Development of Targeted Protein Degradation: Chronology and Milestones

Despite the recent explosion in interest in TPD, the first protein degrader drugs appeared more than 30?years ago with the development of selective estrogen receptor degraders (s) for the treatment of breast cancer (Figure 1.2). Shortly before the launch of fulvestrant as the first SERD drug, the first paper describing PROTACs was published in 2001. However, it was not until 2004 with the advent of cell-permeable PROTACs that the current "gold rush" slowly began to take shape. Since the late 2000s, advances have been reported with increasing frequency, beginning with the recruitment of new E3 ligases (cereblon [], von Hippel-Lindau [], and cellular inhibitor of apoptosis protein [] binding warheads) and the burgeoning development of new modalities for TPD such as HyTags and IMiD molecular glues. In 2015, PROTACs took a leap forward toward drug-like therapeutics with the first in vivo efficacy experiments. Just four years later in 2019, ARV-110 became the first PROTAC to enter human clinical trials for degradation of the androgen receptor () to treat prostate cancer. The following year, the first non-oncology PROTAC entered the clinic (KT-474), an IRAK4 degrader for immune-mediated inflammatory diseases, and today, there are more than 20 PROTACs and molecular glues in ongoing clinical trials. It is clear from the ever-increasing pace of development that we are still in the expansion phase for this promising class of therapeutic agents, and we expect many more advances in the coming years.

1.2.1 Milestone 1: Early Monovalent Protein Degraders - The Surprising Biology of Fulvestrant

In the early 1990s, researchers discovered a strange phenomenon while studying the anti-estrogen breast cancer drug today called fulvestrant (known then as ICI 164,384, Figures 1.2 and 1.3). In a radioligand displacement experiment, fulvestrant showed twofold weaker binding when compared to the endogenous agonist ligand estradiol (Kd values of 1.9 and 0.9?nM, respectively). Despite this fact, cells treated with fulvestrant showed a profound reduction in their response to stimulation with estradiol [10]. From the standpoint of classical pharmacology, this posed a conundrum; as an antagonist, fulvestrant would be expected to exert its effect through an occupancy-driven mechanism. A competitive antagonist would occupy the binding site and therefore block the binding and stimulatory effect of estradiol, while an allosteric modulator could bind to an orthosteric site to modify protein dynamics and modulate the pharmacological response. However, fulvestrant did something else; somehow, the compound elicited a...