Homopharma: a new concept for exploring the molecular binding mechanisms and drug repurposing

Abstract

Background: Drugs that simultaneously target multiple proteins often improve efficacy, particularly in the treatment of complex diseases such as cancers and central nervous system disorders. Many approaches have been proposed to identify the potential targets of a drug. Recently, we have introduced Space-Related Pharmamotif (SRPmotif) method to recognize the proteins that share similar binding environments. In addition, compounds with similar topology may bind to similar proteins and have similar protein-compound interactions. However, few studies have focused on exploring the relationships between binding environments and protein-compound interactions, which is important for understanding molecular binding mechanisms and helpful to be used in discovering drug repurposing.

Results: In this study, we propose a new concept of "Homopharma", combining similar binding environments and protein-compound interaction profiles, to explore the molecular binding mechanisms and drug repurposing. A Homopharma consists of a set of proteins which have the conserved binding environment and a set of compounds that share similar structures and functional groups. These proteins and compounds present conserved interactions and similar physicochemical properties. Therefore, these compounds are often able to inhibit the proteins in a Homopharma. Our experimental results show that the proteins and compounds in a Homopharma often have similar protein-compound interactions, comprising conserved specific residues and functional sites. Based on the Homopharma concept, we selected four flavonoid derivatives and 32 human protein kinases for enzymatic profiling. Among these 128 bioassays, the IC50 of 56 and 25 flavonoid-kinase inhibitions are less than 10 μM and 1 μM, respectively. Furthermore, these experimental results suggest that these flavonoids can be used as anticancer compounds, such as oral and colorectal cancer drugs.

Conclusions: The experimental results show that the Homopharma is useful for identifying key binding environments of proteins and compounds and discovering new inhibitory effects. We believe that the Homopharma concept can have the potential for understanding molecular binding mechanisms and providing new clues for drug development.

Figure 1

Overview of identification of Homopharma using serine/threonine-protein kinase Pim-1 and quercetin complex as query. (A) Search similar pharma-interface candidates of serine/threonine-protein kinase Pim-1 and quercetin complex. (B) The candidates with bound compounds and RMSD ≤ 3Å are used to generate multiple sequence alignment and the protein-compound interaction profile using iGEMDOCK. The interaction profile is clustered into several Homopharma groups based on interaction similarity scores. (C) Some superimposed complex structures of Homopharma 3. (D) The flavonoid derivative (JMY65) is tested against 32 selected protein kinases based on Homopharma concept. Among 32 bioassays, the IC₅₀ of 11 and 18 kinase-JMY65 interactions are less than 1 μM (red) and 10 μM (blue), respectively. The green and dark denote the inhibition efficiency from high to low.

Figure 2

The receiver operating characteristic curves of different similarities. The receiver operating characteristic curves using protein similarity (blue), compound similarity (black), and protein-compound interaction similarity (red) based on 176 protein-compound groups.

Figure 3

The pharma-interface and Homopharma groups of thymidine kinase and deoxythymidine complex. (A) The pharma-interface of thymidine kinase (TK) and deoxythymidine (THM) complex consists of five pharma-motifs. The protein-compound interactions were grouped into four Homopharma groups based on interaction similarity scores. (B) The superimposed structures of Homopharma 1 and conserved interacting residues (residue numbering of PDB code: 1OI3). (C) The superimposed structures of Homopharma 2 and conserved interacting residues (residue numbering of PDB code: 1KIM).

Figure 4

The pharma-interface and Homopharma groups of serine/threonine-protein kinase Pim-1 and quercetin complex. (A) The pharma-interface of serine/threonine-protein kinase Pim-1 (PIM1) and quercetin complex consists of four pharma-motifs. The protein-compound interactions are grouped into six Homopharma groups. (B) The superimposed structures of Homopharma 3 and conserved interacting residues (residue numbering of PDB code: 3BV2). (C) The superimposed structures of Homopharma 4 and conserved interacting residues (residue numbering of PDB code: 3GUB).

Figure 5

The in vitro enzymatic profiling results. (A) Quercetin and its target kinases. (B) JMY51 and its target kinases. (C) JMY65 and its target kinases. (D) JMY84 and its target kinases. The red and blue points mean the bioactivities are ≤ 1 μM and ≤ 10 μM, respectively. Kinome tree representation is prepared using Kinome Mapper (http://www.reactionbiology.com/apps/kinome/mapper/LaunchKinome.htm). (E) The in vitro enzymatic profiling results of 32 protein kinases and four flavonoid derivatives. The green and dark denote the inhibition efficiency from high to low.

Figure 6

The superimposed structures and enzymatic profiling results. (A) The 16 protein kinases inhibited by four flavonoids (quercetin, JMY51, JMY65, and JMY84) are superimposed (residue numbering of PDB code: 2O3P). Kinases and flavonoids are colored by green and pink, respectively. (B) The structures of 16 protein kinases, which are not inhibited by four flavonoids, are superimposed (residue numbering of PDB code: 3OP5). Kinases are colored by blue.