Drugs of allosteric modulators derived from natural products

Drugs of allosteric modulators derived from natural products. Drugs that are allosteric modulators of natural product origin. Drug molecular design is the sway of thinking in drug discovery. Let us taste the works of drug molecular designers together. It is necessary to have time to market a patent drug, and the original design is wonderful.

Finding active leads with clear targets is the source of innovation in new drug development. In this process, in addition to clarifying the type of compound target protein and its regulatory pathway (ie what and how), it is also necessary to clarify the interaction site of the compound and protein (where).

At present, most of the drugs on the market are aimed at the orthosteric/catalytic site (orthosteric/catalytic site) responsible for enzyme catalytic activity/receptor-ligand binding site/protein-protein interaction (Protein-Protein interaction, PPI). site). However, because the orthomorphic sites are relatively conservative in evolution, orthomorphic drugs are often not selective for similar proteins.

In contrast, allosteric sites are highly specific, providing new options for drug development. In addition to the advantages of stronger targeting, allosteric drugs can also be used in combination with orthomorphic drugs to improve the effectiveness of orthomorphic drugs and other problems.

In addition, allosteric drugs also have unique advantages in PPI modulators. In recent years, significant progress has been made in the development of allosteric modulators. According to statistics from the Global Allosteric Resource Center ASD (http://mdl.shsmu.edu.cn/ASD), since 2000, the number of allosteric compounds has grown from less than 1,000 at the beginning to more than 80,000. Its targets involve more than 1300 proteins.

This is due to people’s gradual realization of the potential of allosteric modulators. On the other hand, it is also due to the development of multiple disciplines such as chemical biology, bioinformatics, and structural biology. Among them, more than 70% of allosteric modulators are screened by a combination of kinetics and experiments, followed by functional screening at the cellular level, and a few studies have used structural biology methods.

Currently, only half of the drugs on the market are directly or indirectly derived from natural products, suggesting that natural products also have great research potential in the field of allosteric drugs. Using ASD to study allosteric modulators can identify more than 200 allosteric modulators from natural product sources. Data analysis shows that the discoveries of these natural allosteric modulators are mostly based on random discoveries, and only a small part of them are screened for mature targets.

Take ASD (including 79,290 allosteric modulators, including 461 allosteric drugs under research and 19 on the market), UNPD-ISDB (containing 208 240 natural products), DrugBank (including 11452 compounds, containing 2255 Based on databases such as listed drugs) and iPPI-DB (containing 1854 non-peptide PPI inhibitors), the structural features of natural allosteric modulators (such as atomic polarization) were further analyzed using Principal Component Analysis (PCA). Rate, molecular connectivity, π system, number of heteroatoms, molecular aromaticity, etc.).

It was found that compared with the traditional Drugbank data, the distribution of allosteric modulators (lilac dots, derived from ASD data) was mainly on the outer edge of the “drug group” (picture B) or Q4 quadrant (picture C), suggesting allosteric Modulators have different structural characteristics from traditional orthomorphic drugs. The above results further support the rich chemical diversity of natural products, which not only have the characteristics of marketed drugs and normal compounds, but also show structural characteristics similar to allosteric compounds, and are valuable resources for drug research.

According to the source structure of natural products included in ASD, 221 natural allosteric modulators can be divided into 15 structural types, and representative structures are listed. It can be seen from the figure that natural allosteric modulators mostly contain aromatic rings or polycyclic ring systems and have a relatively rigid structure.

In addition, most natural allosteric modulators are highly oxidized. Only 10% of natural allosteric modulators contain nitrogen atoms. This is consistent with the traditional understanding of allosteric drugs. Interestingly, although allosteric modulators located in the Q3 quadrant account for only 1% of all allosteric modulators, about 20% of natural allosteric modulators are located in this quadrant (mostly triterpenoids), reflecting natural The uniqueness of the product.

At the same time, other structural fragments such as halogen also made important contributions to the affinity and pharmacokinetic parameters of these natural allosteric modulators.

The uniqueness of natural products has fostered the good research potential and development value of these molecules. Although natural products are often criticized as having the disadvantage of being difficult to obtain continuously, this problem can be solved by structural simplification or chemical synthesis/biosynthesis.

In addition to resource issues, natural products can also circumvent their potential drug metabolism problems by changing the way of administration. For example, microtubule stabilising agents (MSAs), such as paclitaxel and vincristine in the fourth quadrant of PCA analysis, can be injected intravenously to cope with their poor oral bioavailability. In fact, the famous “five principles of drug-like drugs” are only for oral drugs, and cannot be used to exclude biologically effective compounds.

In contrast, what really needs special attention in natural medicine research is the so-called “pan-assay interference (PAINS) compounds.” Pan-assay interference (PAINS) compounds refer to those high-throughput screening in multiple models. In the experiment, they are considered to be “seeding compounds”, but there is not much potential for modification in the subsequent structural optimization, and they are not suitable for chemical probe compounds.

In fact, leading pharmaceutical companies at home and abroad have invested a lot of manpower and material resources to develop a “de-pan-positive compound system” to optimize their compound library, and to conduct a second evaluation of the signs of compounds discovered in their previous studies. The 221 natural allosteric modulators mentioned above were pan-positively screened by using a network database. 76 compounds (or their substructures) were marked as pan-positive compounds, and they were mainly chromones, chalcones and quinones. . When the seed compounds are labeled as pan-positive compounds, a series of methods can be used to verify these so-called “pan-positive compounds”: First, we have to distinguish whether these compounds are specifically bound to the target.

Ideally, specific binding should be further confirmed by structural biology methods, such as obtaining a protein-ligand complex crystal structure. Subsequently, the seed compounds can be derivatized by medicinal chemistry and structure-activity relationship (SAR) analysis can be performed. Compounds that specifically bind generally have a clear structure-activity relationship.

In addition, we can also test the binding ability of the seed compound and other proteins to determine the compound’s selectivity and off-target effect on the protein. Especially for allosteric modulators, because allosteric sites are usually non-conserved during evolution, this method can well judge whether the candidate molecule is a pan-positive compound.

Using the natural allosteric regulators discovered in the 10 years from 2008 to 2018 as probes, scientists have identified a total of 81 proteins involved in different life processes, and divided them into enzymes, G protein coupled receptors, and membrane transporters. There are five types of transcription factors and other proteins. Among them, enzymes account for 77% of all target proteins. This is partly because the current discovery of natural allosteric agonists is too dependent on kinetic methods.

These enzymes can be further divided into four categories: oxidoreductases (such as cytochrome P450), transferases (mainly various kinases), hydrolases (73% of which are phosphatase, protease and glycosylase family) and others Enzymes (such as lyase and isomerase). In fact, allosteric modulators can act on a series of non-enzymatic protein targets, and these target proteins also have great research potential.

The effects of the aforementioned natural allosteric modulators on their target proteins can be divided into three categories:

(1) Change the local conformation of the normal/catalytic site;
(2) Change the dynamic characteristics of the protein;
(3) Regulate the effect of PPI through allosteric sites.

Example 1: Allosteric modulators can cause conformational changes in the catalytic site region (PTP1B)

The protein tyrosine phosphatase family (PTPases) represented by PTP1B plays an important role in the regulation of multiple pathways. Although there are many types of PTPases, their active sites are highly conserved. Therefore, the development of selective PTPases inhibitors has important clinical significance. In recent years, with the reports of natural allosteric inhibitors of protein phosphatase, scientists have ignited new hopes for the development of PTPase drugs. Most of the reported natural allosteric PTPase inhibitors have IC50 values ​​in the submicromolar range, making them one of the most effective PTPase inhibitors.

There are mainly three allosteric pockets of PTPases discovered by natural products, which are located in:

1) Exosite;
2) C-terminal disordered non-catalytic domain;
3) Allosteric sites near the catalytic site (as shown below).

Although in these examples, these allosteric small molecules and their sites of action are different, they have a similar mechanism of action: the WPD region responsible for catalysis in PTPase is very flexible, and there are at least two conformations. The open conformation is related to the inactive conformation, while the closed conformation is critical to the catalytic activity of PTPase, and this conformation tends to be stable after binding to the substrate. The above-mentioned allosteric inhibitor stabilizes the open conformation of WPD, making PTP1B in an inactive state.

Trodusquemine is a representative compound of this mechanism of action. This compound was reported in 2000, and subsequent phenotypic studies showed that the compound has extensive and significant pharmacological activities. Further studies have shown that PTP1B is the main target of the compound. Subsequent studies have shown that the interaction between PTP1B and trodusquemine occurs at two different allosteric sites, and they interact through positive synergy, with Kd of 0.3 μM and 1 μM, respectively.

Trodusquemine can bind to the first site of PTP1B and can cause conformational changes in the C-terminal disordered extension region, and the PTP1B residue located on the alpha-helix 9’plays an important role in this binding. The study by Krishnan et al. showed that trodusquemine can bind to the first site formed by α helices 7 and 9, and then cause the conformational changes of α helices 7, 3 and 6, to form a second binding pocket-the pocket and the outer The exosite partly overlaps and eventually affects the WPD loop region of PTP1B, making the protein in an inactive conformation (as shown in the figure above).

In this process, the binding of the compound to the C-terminal (ie the first binding site) is critical. At present, trodusquemine has passed phase I and phase Ib clinical trials, and phase II clinical trials are mainly for diabetes, obesity and other indications.

The example of Trodusquemine illustrates how natural products can bring new ideas to the field of drug development. In this example, PTP1B, as an “old target” that has been studied for more than 20 years, still lacks selective and cell permeability inhibitors. Through natural product chemistry, chemical biology, structural biology and other multidisciplinary cooperation, a new role of PTP1B has been discovered. Use complex conformational changes induced by ligands to stabilize the protein in an inactive state, thereby exerting an inhibitory effect.

Example 2: Allosteric modulators can cause overall changes in protein dynamics (20S proteasome)

Protease is a key enzyme for maintaining cell homeostasis and is associated with many diseases. Proteases have a common active site, so it is difficult to selectively target proteases. At present, most protease inhibitors are non-selective competitive inhibitors and have obvious side effects. The 20S proteasome is a 700 kDa polyproteolytic complex and an N-terminal hydrolase responsible for the non-enzymatic degradation of intracellular proteins.

At present, three proteasome inhibitors, bortezomib, carfizomib and ixazomib, are on the market, all of which are polypeptides with high electrophilic activity. However, the aforementioned drugs have relatively large side effects and poor permeability to solid tumors, which restricts the clinical application of these drugs. Therefore, looking for allosteric modulators with new chemical frameworks and better selectivity is expected to provide ideas for the next generation of protease inhibitors.

The 20S core particle (CP) is a cylindrical structure composed of two outer α rings and two inner β rings. Each β ring is composed of 7 subunits. In eukaryotes, the 20S proteasome has 6 catalytic sites, all of which have similar N-terminal hydrolase characteristics, but have different substrate and inhibitor preferences. The 20S proteasome is regulated by 11S and 19S regulatory proteins. These two proteins are responsible for specific substrate recognition and can open the cylinder channel, allowing the substrate protein to enter the catalytic center and degrade it.

In addition, the 20S nucleus has 3 internal chambers, two are formed between the α-β loop interface at the entrance of the proteasome CP, and one central chamber is formed within 6 β subunits. They show adjacent small pockets that can be bound by small molecules. In addition, the kinetic study of 20S proteasome CP has confirmed its importance for 20S CP regulation and catalysis. Proteasome dynamics can be allosterically regulated by regulatory proteins, substrates, and catalytic site inhibitors.

Studies have shown that the natural products of rapamycin and chloroquine can act as allosteric inhibitors of the proteasome by regulating the conformational dynamics of the proteolytic complex. The allosteric pathway can change the balance of proteasome conformation in many ways, making CP in an inactive conformation.

These approaches include: regulating the opening/closing of the CP cylinder, hindering the interaction between CP and its agonist, and avoiding further reactions between the substrate polypeptide and the corresponding catalytic subunit. TROSY NMR spectroscopy analysis results show that chloroquine binds to the α-β interface of the proteasome, changing the equilibrium state of 20S CP from multiple isomers to inactive states, thereby regulating the function through allosteric pathways. The natural product rapamycin can also dynamically inhibit the proteasome.

Osmulski and Gaczynska analyzed the CP of the 20S proteasome using an atomic force microscope (AFM) and found that after rapamycin was incubated with the CP, the pore size of the 20S CP changed, making the cylindrical channel open.

Example 3: Allosteric protein-protein interaction modifier (CBP/p300)

During the transcription process, transcription factors regulate their biological functions by interacting with coactivator and corepressor. CBP/p300 is a co-activator protein that interacts with a variety of transcription factors, such as CREB, Myb, and p53, to regulate their DNA binding and transcription activity. CBP/p300 co-activator has 5 protein interaction domains.

Among them, the GACKIX domain of CBP/p300 plays a key role in transcription factor activation and is an important interface for its interaction with targeted transcription factor transcription activation domains (TADs). The GACKIX domain interacts with the TADs of p53, MILL, c-Jun, CREB, c-Myb and Tax. These interactions are important in some biological processes and diseases. The GACKIX domain of CBP/p300 shows two binding sites, which are alternatively and cooperatively combined with TADs. According to their topological properties, they are called “deep” and “shallow” sites, respectively.

In 2012, Majmudar et al. planned to find compounds that inhibit the GACKIX-MLL TAD interaction by fluorescence polarization analysis. After a lot of screening work, the team selected 2 active samples from 65,000 samples (50,000 of which are synthetic compounds), both of which were natural product extracts.

Through a series of chromatography techniques, they obtained 3 active natural products, including sekikaic acid. Through fluorescence experiments and nuclear magnetic resonance studies, the researchers confirmed that the compound can indeed interact with the protein. In addition, they also performed cell function analysis and molecular dynamics simulations.

Studies have shown that the interaction between sekikaic acid and GACKIX domain occurs at the “deep” site where MLL (the orange part in the figure below) binds to the c-Jun transcription activator. The binding of small molecules to the GACKIX domain changes the protein conformation, which in turn affects the binding of the GACKIX domain to other co-agonists (such as CREB, c-Myb, the yellow part in the figure below).

At the same time, Lys662 (the blue part in the figure) in the shallow layer of GACKIX has also shifted, and this position is related to the interaction of CREB with pSer133. In summary, phenolic acid compound sekikaic acid can bind to the GACKIX domain of CBP/p300, selectively destroying the interaction between CBP/p300 and its transcriptional activators MLL and KID (CREB). Interestingly, sekikaic acid does not interact with other co-agonists (such as Med15 and VP2) and shows good selectivity for the GACKIX domain of CBP/p300.

In addition, the researchers also studied the effects of lecanoric acid and lobaric acid on GACKIX. Lobaric acid exhibited TAD binding inhibition, while lecanoric acid did not exhibit any activity. This may be due to the more flexible structure of the latter. Flexibility hinders stable binding of the compound to the GACKIX domain.

Prospects and prospects

Just like the development of orthomorphic drugs, natural products are also expected to be developed as new allosteric modulators. However, because many natural products tend to have pan-positive characteristics, they need to be carefully excluded in research. The development of chemical biology and biophysics methods can help eliminate some natural allosteric modulators with non-specific binding, and provide a reliable preliminary basis for further in-depth research.

The current research on natural allosteric modulators is still very limited, but considering the uniqueness and complexity of natural products, if effective and high-quality target biological data can be obtained, these natural allosteric inhibitors will become revealing new types of allosteric inhibitors. A powerful weapon for the construction mechanism.

(source:internet, reference only)

Disclaimer of medicaltrend.org