The application of PROTACs targeted protein degradation in drug design (II)

2.3 Highly selective

Most protein families contain many subtypes with highly similar structures, such as HDAC, BET and other families. Different subtypes have different functions and expressions in various tumors. Pan-inhibitors will produce greater toxic and side effects, and clinical applications are limited. Therefore, the development of subtype-selective inhibitors has always been a hot spot in drug development.

So far, most of the published PROTACs have been developed based on highly selective inhibitors, such as PROTACs targeting RIPK2 and BRD4. A large number of experiments prove that PROTACs

degradation not only retains the selectivity of the parent inhibitor, but also surpasses protein. The binding selectivity of the inhibitor. Bondeso et al. selected foretinib, an inhibitor of 54 kinases, and designed two types of degradation agents with VHL and CRBN as the E3 ligase. The results showed that the degradation of PROTACs is more selective than the inhibition of foretinib. Among these 54 kinases, 9 are degraded by VHL PROTAC and 14 are degraded by CRBN PROTAC, of ​​which 6 kinases are the same. The research team also found that PROTACs based on the Her family inhibitor lapatinib can degrade EGFR while retaining HER2, indicating that PROTACs have an affinity for the target protein and E3 ligase not enough to degrade the protein, which may be accompanied by more complex dynamics process. The selective degradation of PROTACs depends on the stable ternary complex formed between the protein: PROTAC: E3 ligase. Even though PROTACs have weak affinity for the target protein, they can still form stable trimers for effective degradation. At the same time, different types of E3 ligases have different subcellular locations, which may provide new ideas for selective protein degradation.

1. The development history of protein degradation agents

2.1 The idea of artificially degrading proteins Artificially induced protein degradation was first reported in 1995. The Hynes research group designed a series of modified E2 binding enzymes to fuse with Ig binding motifs. The result is that a series of fusion proteins less than 250 amino acids can induce direct transfer of ubiquitin from E2 to a specific substrate. In addition, the resulting ubiquitinated protein substrate is recognized by the proteasome in the cell lysate, leading to protein degradation, which indicates that the protein ubiquitination process can be artificially induced. In 2000, Zhou et al. constructed two engineered E3 ligases CDC4 and -TrCP, which successfully degraded intracellular proteins in yeast and SAOS-2 osteosarcoma cells, and these proteins were not the physiological substrates of CDC4 or p-TrCP. It shows that theoretically E3 can knock out any protein of interest.

2.2 The beginning of PROTAC technology---peptide PROTACs

The development of PROTACs technology started with the use of peptides to recruit E3 ligase. In 2001, Sakamoto et al. linked Ovalicin, a small molecule inhibitor of methionine aminopeptidase-2 (MetAP2), and 10 aa-sized IκBα phosphor-peptide to obtain the first protein-targeting chimeric molecule Protac-1. IκBα can bind to β-TRCP in the SCF ^β-TRCP (SKP1-CUL1-F-box protein) E3 ubiquitin ligase complex, and small ubiquitin molecules are pulled to the surface of MetAP2 to cause polyubiquitination, and Degraded in a Protac-1 dependent manner. So far, the concept of PROTACs has been formally proposed. However, due to the lack of cell permeability of these early PROTACs, they cannot be applied to the field of medicine, and the covalent combination of Ovalicin and MetAP2 limits the unique catalytic properties of PROTACs. Soon thereafter, using the same E3 ubiquitin ligase ligand, the group designed an estradiol-based phosphor-peptide Protac-2 targeting the estrogen receptor (ERα), which successfully induced degradation of ERα. In order to solve the problem of lack of cell permeability, Sakamoto et al. introduced the dihydrotestosterone (DHT)-based phosphor-peptide Protac-3 into HEK293 cells by microinjection, which proved that PROTACs degrade androgen receptors in mammalian cells capabilities.

The breakthrough of peptide PROTACs is the use of smaller VHL E3 ligase to recruit fragments. Hypoxia-inducible factor 1 (HIFα) can be recognized by Von Hippel Lindau tumor suppressor protein (VHL), the substrate of E3 ligase complex CRL2^VHL. The P564 residue of HIFld is hydroxylated by proline hydroxylase, resulting in continuous HIFα Recognized by CRL2^VHL and polyubiquitinated. In 2004, Schneekloth et al. cited the 7 amino acid sequence of HIFα (ALAPYIP, VHL recognition sequence) instead of the IκBα phosphor-peptide component, and conjugated it with the ligand AP21998 of the recombinant human receptor protein FK-BP12 (F36V) to obtain Protac-4 Effectively degrade FKBP12 (F36V) in cells. This is the first PROTACs that can act on intracellular proteins without the need for external force. Subsequently, Rodriguez Gonzale et al. changed the ligand part of Protac2/3 E3 ligase targeting ER/AR, and designed ProtacB/A based on CRL2^VHL, which greatly improved the membrane permeability of the degradant.

2.3 The development of PROTAC technology-the emergence of small molecule PROTAC

The research of peptide PROTACs laid the foundation of this technology, but due to the poor membrane permeability and low degradation efficiency of these molecules, their application has been greatly restricted. The breakthrough of PROTACs technology is mainly focused on the change of the type of E3 ligase, which makes PROTACs transform from peptides to small molecules. Currently, the four main types of small molecule E3 ligases are MDM2, cIAP, VHL and CRBN.

In 2008, the first small molecule PROTACs appeared in a complete sense. Using the known MDM2 protein ligand Nutlin as the E3 ligase recruitment part, Nutlin was coupled to the non-steroidal androgen SARM, and the produced PROTACs (Figure 5a) were induced in prostate tumor cells at a concentration of 10umol·L Degradation of AR. Although the effectiveness of this degrading agent is not as good as its peptide analogs, it is the first time to prove the feasibility of small molecule PROTACs.

In 2010, Hashimoto et al. used Bestatineaters (MeBS) in combination with all-trans retinoic acid (ATRA) to design PROTACs based on cIAP1 E3. ATRA is an endogenous retinoic acid receptor that can recruit cellular retinoic acid. Binding proteins (CRABP-I and CRABP-II), MeBS can be directly combined with the BIR3 domain of cIAP1 E3 ligase selectively to cause the ubiquitination process to induce the degradation of the target protein. In addition to retinoic acid receptors, these MeBS-based PROTACs can also induce AR and ERα degradation. Unfortunately, MeBS is a common aminopeptidase inhibitor. It can also inhibit arginyl aminopeptidase and leukotriene A4 hydrolase. It is not specific to clAP1and often produces off-target effects. In addition, MeBS also induces the autoubiquitination and degradation of cIAP1, which limits the effectiveness of PROTACs. Subsequently, the research group improved the PROTACs based on cIAP1 by replacing the MeBS part with cIAP1/clAP2/XI-AP pan-ligand MV1.

At present, the design and optimization of most PROTACs are mainly guided by the binary interaction between the target protein and PROTACs, usually by selecting existing proteins. The inhibitor of the ligand co-crystal structure is designed. The more common design idea is to select a specific inhibitor, determine that it is on the surface of the protein pocket (solvent exposure) and does not affect the inhibitory function, and extend a flexible chain to connect the E3 ligase ligand to ensure that the inhibitor is in The ability to bind proteins in PROTACs molecules is not affected, thereby recruiting ubiquitinated molecules to the protein surface.

Another important part of PROTACs is the linker. The composition, length and connection site of the linker also affect the degradation of the protein. When the linker length is 16 atoms, the VHL-based ER degradation agent shows the best degradation performance. When the Linker is derived from the C-7a position of estradiol instead of being connected to O-17 through an ester bond, the degradation effect is also enhanced. Among the PROTACs that degrade TANK-binding kinase 1 (TBKI), they show significant degradation differences due to different linker lengths.

PROTACs have the characteristics of high efficiency, strong specificity, sub-stoichiometric function, and unique advantages of being able to target traditional "undruggable" proteins. Although there are still many problems to be overcome in the research field, such as further reducing the relative molecular weight, finding suitable target protein ligands and more diverse E3 ligases, improving bioavailability and tissue distribution, etc., all need to be improved. Extensive and more in-depth research.

References

1. PAIK YK，JEONG SK，OMENN GS，et a1．The chromosome-centric human proteome project for cataloging proteins encoded in the genome[J]．Nat Biotechnol。2012，30(3)：221—223．


2. PEER D，LIEBERMAN J．Special delivery：targeted therapy with small RNAs[J]．Gene Ther，201l，18(12)：1127-1133


3. CONG L，RAN FA，COX D，et a1．Multiplex genome engineering using CRISPR／Cas systems[J]．Trends Genet，201 3，32 (12)：819—823


4. BOBBIN ML，ROSSI JJ．RNA interference(RNAi)．based therapeutics：delivering on the Promise? [J]．Annu Rev Pharmacol Toxic01．2016。56(I)：103-122


5. FEDOROV Y．ANDERSON EM，BIRMINGHAM A，et a1．0iftarget effects by siRNA can induce toxic phenotype[J]．RNA， 2006，12(7)：1188-1196


6. NEKLESA TK，WINKLER JD，CREWS CM．Targeted protein degradation by PROTACs[J]．Pharmacol Therapeut，2017,174：138-144