Final Report Summary - TARGET-PPIS (Targeting the ubiquitin-proteasome system and ubiquitin-like protein conjugation pathways for non-genotoxic therapy of cancer)
The principal objective of our research has been to determine structural attributes and to develop small-molecule compounds for new cancer-related protein-protein interactions in the ubiquitin-proteasome system (UPS) and the ubiquitin-like protein (UBL) conjugation pathways. We use nuclear magnetic resonance (NMR) and X-ray crystallography to structurally characterize such protein-protein interactions (PPI) and to develop small-molecule inhibitors for these PPIs. The proteins studied by us include: the oncogenic E3 ligase Mdm2, which regulates levels of p53, and the deubiquitinating protease USP2. The E3 ligases and USP2 are important components of the ubiquitin conjugation system - this is because they directly bind to their target proteins and thus control substrate specificity. The tumour suppressor p53 protein, "the guardian of the genome", has an overarching role in protecting the organism from cancer. In order to escape the safeguard system mediated by p53, nearly all human cancers have compromised the effectiveness of the p53 pathway. The restoration of the impaired function of the single gene, p53, by disrupting the Mdm2-p53/Mdmx-p53 interactions, offers a fundamentally new avenue for anticancer therapy across a broad spectrum of cancers. We have also studied the non-canonical UBL protein Hub1 and the immune checkpoint proteins: the programed cell death protein-1 (PD-1) and the programed cell death protein ligand-1 (PD-L1). Hub1 has been recently shown to control alternative splicing by non-covalent binding to the Snu66 spliceosomal protein. Targeting the protein-protein interaction between PD-1 and PD-L1 by antibodies comprises a game changer in oncology - in many cases rendering formerly lethal cancers into a well treatable disease and even cure in hard-to-treat cancers.
In the past few years, new drugs have been developed that can improve cancer patient survival. For example, Herceptin specifically blocks the oncogenic proteins that drive tumour growth. But oncogenes are only one part of the tumorigenic driven system. It is believed that all human cancers also have defects in tumour suppressor genes that would normally contain cancer development. The tumour suppressor genes are thus increasingly targeted for possible therapies that work by restoring their activity.
All specific aims of the subproject on the development of the low-molecular weight inhibitors of MDM2-p53 AND MDMX-p53 have been met. Some highlights of the research in this area are:
• The co-crystal structures of two complexes of Mdm2 with inhibitors based on novel scaffolds were presented by us. In contrast to other structurally characterized antagonists, which mimic three amino acids of p53, the new compounds induced an additional hydrophobic pocket on the Mdm2 surface and unveiled a four-point binding mode. This work shows that the inhibitors of the Mdm2-p53 interaction can be designed to interact with a transiently folded alpha-helical segment of the Mdm2 region. Thus targeting transient protein states in PPI inhibitor design could be a promising strategy to improve affinity and/or selectivity profiles.
• A potential application of drugs blocking the p53-Mdm2 interaction is acute myeloid leukemia (AML) due to the occurrence of wild type p53 (wt-p53) in the majority of patients. We have reported the design, synthesis and optimization of potent p53-Mdm2 antagonizing and apoptosis inducing agents characterized by a number of leukemia cell lines as well as patient derived AML blast samples. The structural basis of the interaction between Mdm2 and these inhibitors was elucidated by a co-crystal structure. The inhibitors act as prodrugs and are the most potent compounds which induce apoptosis in AML cells and patient samples. This observed superior activity compared to reference compounds provides the preclinical basis for further investigation and progression of of these small-molecule inhibitors.
• We also presented the synthesis, activities and structures on two other novel classes of the Mdm2-p53 inhibitors that are based on the 3-pyrrolin-2-one and 2-furanone scaffolds. The crystal structures of the complexes formed by these inhibitors and Mdm2 reveal the dimeric protein molecular organization that has not been observed in the small-molecule/Mdm2 complexes described until now. In particular, the 6-chloroindole group does not occupy the usual Trp-23 pocket of Mdm2, but instead is engaged in the dimerization. This entirely unique binding mode of both types of compounds opens new possibilities for optimization of the Mdm2-p53 interaction inhibitors.
USP2 is a DEUBIQUITINATING PROTEASE that reverses protein ubiquitination and rescues its target proteins from destruction by the proteasome. USP2 is overexpressed in prostate cancer, shows oncogenic properties in vivo. Targeting USP2a is a promising strategy for protein-directed therapies in the treatment of cancer. Within this subproject, we have discovered small molecules that bind to USP2 using NMR-based fragment screening and biophysical binding assays. Iterations of fragment combination and structure-driven design identified a series of 5-(2-thienyl)-3-isoxazoles as the inhibitors of the USP2-ubiquitin protein-protein interaction. The affinity of these molecules for the catalytic domain of USP2 parallels their ability to interfere with the USP2 binding to ubiquitin in vitro and with the inhibition of the growth of the USP2a-dependent HCT116 cells.
We also reported that the derivatives of lithocholic acid (LCA) are potent inhibitors of USP2. They inhibit the growth of colorectal carcinoma cell line HCT116 independently of its p53 status. LCA’s significantly decreased the expression of cyclin D1, but did not change expression of p27.
Different from canonical UBIQUITIN-LIKE PROTEINS, HUB1 does not form covalent conjugates with substrates but binds proteins non-covalently. In Saccharomyces cerevisiae, Hub1 associates with spliceosomes and mediates alternative splicing of SRC1, without affecting pre-mRNA splicing generally. Human Hub1 is highly similar to its yeast homolog, but its cellular function remained unexplored. We showed that human Hub1 binds to the spliceosomal protein Snu66 as in yeast; however, unlike its S. cerevisiae homolog, human Hub1 is essential for viability. Human Hub1 is not a canonical spliceosomal factor needed generally for splicing, but rather a modulator of spliceosome performance and facilitator of alternative splicing.
Targeting the PD-1/PD-L1 IMMUNOLOGIC CHECKPOINT with monoclonal antibodies has provided unprecedented results in cancer treatment in recent years. Development of chemical inhibitors for this pathway lags the antibody development mainly because of insufficient structural information. We have therefore undertaken the task of structural characterization of the proteins and ligands involved in the PD-1/PD-L1 interactions using X-ray crystallography and NMR spectroscopy. Within this work, we have recently published the first crystal structure of human PD-1/PD-L1 and described structural bases for small molecule targeting of PD-L1.
Monoclonal antibodies (mAbs) are the fastest growing segment within biotech and pharmaceutical research. Global sales for mAbs are in billions. Despite excitement over this efficacy (especially for the recent advancement in immunotherapy of the PD-1/PD-L1 system), oncologists have been raising concerns over the cost of antibody drugs, especially as the field moves towards multidrug regimens. Targeted cancer therapies based on small-molecule drugs are substantially cheaper and simpler to deliver to the patient. Modulating tumorigenesis through a small-molecule approach offers several unique advantages that are complementary to, and potentially synergistic with, biologic modalities of the mAbs. Thus the identification of a small molecule drug should be of the highest importance as this could lead to an inexpensive cancer therapeutic.
In the past few years, new drugs have been developed that can improve cancer patient survival. For example, Herceptin specifically blocks the oncogenic proteins that drive tumour growth. But oncogenes are only one part of the tumorigenic driven system. It is believed that all human cancers also have defects in tumour suppressor genes that would normally contain cancer development. The tumour suppressor genes are thus increasingly targeted for possible therapies that work by restoring their activity.
All specific aims of the subproject on the development of the low-molecular weight inhibitors of MDM2-p53 AND MDMX-p53 have been met. Some highlights of the research in this area are:
• The co-crystal structures of two complexes of Mdm2 with inhibitors based on novel scaffolds were presented by us. In contrast to other structurally characterized antagonists, which mimic three amino acids of p53, the new compounds induced an additional hydrophobic pocket on the Mdm2 surface and unveiled a four-point binding mode. This work shows that the inhibitors of the Mdm2-p53 interaction can be designed to interact with a transiently folded alpha-helical segment of the Mdm2 region. Thus targeting transient protein states in PPI inhibitor design could be a promising strategy to improve affinity and/or selectivity profiles.
• A potential application of drugs blocking the p53-Mdm2 interaction is acute myeloid leukemia (AML) due to the occurrence of wild type p53 (wt-p53) in the majority of patients. We have reported the design, synthesis and optimization of potent p53-Mdm2 antagonizing and apoptosis inducing agents characterized by a number of leukemia cell lines as well as patient derived AML blast samples. The structural basis of the interaction between Mdm2 and these inhibitors was elucidated by a co-crystal structure. The inhibitors act as prodrugs and are the most potent compounds which induce apoptosis in AML cells and patient samples. This observed superior activity compared to reference compounds provides the preclinical basis for further investigation and progression of of these small-molecule inhibitors.
• We also presented the synthesis, activities and structures on two other novel classes of the Mdm2-p53 inhibitors that are based on the 3-pyrrolin-2-one and 2-furanone scaffolds. The crystal structures of the complexes formed by these inhibitors and Mdm2 reveal the dimeric protein molecular organization that has not been observed in the small-molecule/Mdm2 complexes described until now. In particular, the 6-chloroindole group does not occupy the usual Trp-23 pocket of Mdm2, but instead is engaged in the dimerization. This entirely unique binding mode of both types of compounds opens new possibilities for optimization of the Mdm2-p53 interaction inhibitors.
USP2 is a DEUBIQUITINATING PROTEASE that reverses protein ubiquitination and rescues its target proteins from destruction by the proteasome. USP2 is overexpressed in prostate cancer, shows oncogenic properties in vivo. Targeting USP2a is a promising strategy for protein-directed therapies in the treatment of cancer. Within this subproject, we have discovered small molecules that bind to USP2 using NMR-based fragment screening and biophysical binding assays. Iterations of fragment combination and structure-driven design identified a series of 5-(2-thienyl)-3-isoxazoles as the inhibitors of the USP2-ubiquitin protein-protein interaction. The affinity of these molecules for the catalytic domain of USP2 parallels their ability to interfere with the USP2 binding to ubiquitin in vitro and with the inhibition of the growth of the USP2a-dependent HCT116 cells.
We also reported that the derivatives of lithocholic acid (LCA) are potent inhibitors of USP2. They inhibit the growth of colorectal carcinoma cell line HCT116 independently of its p53 status. LCA’s significantly decreased the expression of cyclin D1, but did not change expression of p27.
Different from canonical UBIQUITIN-LIKE PROTEINS, HUB1 does not form covalent conjugates with substrates but binds proteins non-covalently. In Saccharomyces cerevisiae, Hub1 associates with spliceosomes and mediates alternative splicing of SRC1, without affecting pre-mRNA splicing generally. Human Hub1 is highly similar to its yeast homolog, but its cellular function remained unexplored. We showed that human Hub1 binds to the spliceosomal protein Snu66 as in yeast; however, unlike its S. cerevisiae homolog, human Hub1 is essential for viability. Human Hub1 is not a canonical spliceosomal factor needed generally for splicing, but rather a modulator of spliceosome performance and facilitator of alternative splicing.
Targeting the PD-1/PD-L1 IMMUNOLOGIC CHECKPOINT with monoclonal antibodies has provided unprecedented results in cancer treatment in recent years. Development of chemical inhibitors for this pathway lags the antibody development mainly because of insufficient structural information. We have therefore undertaken the task of structural characterization of the proteins and ligands involved in the PD-1/PD-L1 interactions using X-ray crystallography and NMR spectroscopy. Within this work, we have recently published the first crystal structure of human PD-1/PD-L1 and described structural bases for small molecule targeting of PD-L1.
Monoclonal antibodies (mAbs) are the fastest growing segment within biotech and pharmaceutical research. Global sales for mAbs are in billions. Despite excitement over this efficacy (especially for the recent advancement in immunotherapy of the PD-1/PD-L1 system), oncologists have been raising concerns over the cost of antibody drugs, especially as the field moves towards multidrug regimens. Targeted cancer therapies based on small-molecule drugs are substantially cheaper and simpler to deliver to the patient. Modulating tumorigenesis through a small-molecule approach offers several unique advantages that are complementary to, and potentially synergistic with, biologic modalities of the mAbs. Thus the identification of a small molecule drug should be of the highest importance as this could lead to an inexpensive cancer therapeutic.