http://orcid.org/0000-0002-7644-4084
Figures
Abstract
Piper longum (P. longum, also called as long pepper) is one of the common culinary herbs that has been extensively used as a crucial constituent in various indigenous medicines, specifically in traditional Indian medicinal system known as Ayurveda. For exploring the comprehensive effect of its constituents in humans at proteomic and metabolic levels, we have reviewed all of its known phytochemicals and enquired about their regulatory potential against various protein targets by developing high-confidence tripartite networks consisting of phytochemical—protein target—disease association. We have also (i) studied immunomodulatory potency of this herb; (ii) developed subnetwork of human PPI regulated by its phytochemicals and could successfully associate its specific modules playing important role in diseases, and (iii) reported several novel drug targets. P10636 (microtubule-associated protein tau, that is involved in diseases like dementia etc.) was found to be the commonly screened target by about seventy percent of these phytochemicals. We report 20 drug-like phytochemicals in this herb, out of which 7 are found to be the potential regulators of 5 FDA approved drug targets. Multi-targeting capacity of 3 phytochemicals involved in neuroactive ligand receptor interaction pathway was further explored via molecular docking experiments. To investigate the molecular mechanism of P. longum’s action against neurological disorders, we have developed a computational framework that can be easily extended to explore its healing potential against other diseases and can also be applied to scrutinize other indigenous herbs for drug-design studies.
Citation: Choudhary N, Singh V (2018) A census of P. longum’s phytochemicals and their network pharmacological evaluation for identifying novel drug-like molecules against various diseases, with a special focus on neurological disorders. PLoS ONE 13(1): e0191006. https://doi.org/10.1371/journal.pone.0191006
Editor: Hiroyoshi Ariga, Hokkaido Daigaku, JAPAN
Received: October 15, 2017; Accepted: December 25, 2017; Published: January 10, 2018
Copyright: © 2018 Choudhary, Singh. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: The authors received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Abbreviations: ADMET, Absorption, Distribution, Metabolism, Excretion and Toxicity; BP, Biological pathways; DA, Disease association; GPCR, G protein-coupled receptors; P. longum, Piper longum; PC, Phytochemical; PCt, Number of protein targets corresponding to a particular phytochemical; PPI, Protein protein interactions; PT, Protein target; Tt, Total number of protein targets of P. longum
Introduction
Healing with medicinal plants is an ancient idea. Secondary metabolites of various plants have been traditionally utilized for the betterment of human health. Plants belonging to genus Piper are amongst the most important medicinal plants used in various systems of medicine. More than 1,000 species belong to this genus and P. longum is one of the most well-known species amongst them, including Piper nigrum and Piper bettle. P. longum forms an active constituent of the widely used Ayurvedic poly-herbal formulation “Trikatu” [1]. The widespread use of this herb in different formulations as documented in ancient Ayurvedic manuscripts such as Caraka samhita [2], Susruta samhita [3] Vagbhata’s astanga hrdayam [4] etc. suggests its vital importance in traditional Indian medicinal system.
P. longum is an indigenously growing plant in India and is also cultivated in the tropical and subtropical regions of Asia and Pacific islands [5]. It is usually cultivated for its fruit which is dried and used as a spice. The plant grows into a shrub with large woody roots, numerous creeping and jointed stems that are thickened at the nodes. Leaves are without stipules and spreading in nature. Fruits are small, oval shaped berries and grow as spikes that are collected after maturation. Dried form of these spikes makes “pippali” while the root radix is known as “pippalimula”. The dietary piperine is known for its bioavailability and digestion enhancing properties. In vitro studies have shown the role of piperine in relieving oxidative stress by quenching free radicals and reactive oxygen species. While it is known to act as an anti-mutagenic and anti-tumor agent [6], anti-diarrheic and anti-dysenteric properties of this spice enhance its medicinal value [7]. The pharmacological properties of this plant also include anti-oxidant, anti-inflammatory, hepatoprotective, immunomodulatory, anti-microbial, anti-platelet, anti-hyperlipidemic, analgesic, anti-depressant, anti-amoebic, anti-obesity, radioprotective, cardioprotective and anti-fungal [8], [9], [10], [11]. Methanolic extract of this fruit has been reported to be involved in memory repair and improving memory performance by an in vitro model [12]. Clinical studies have revealed the efficacy of this plant in the treatment of bronchial asthma in children [13], [14]. Anti-diabetic activity of the roots has also been reported [15]. It is widely used as an important constituent in various Ayurvedic medicines to cure diseases like leprosy and tuberculosis and is also used in the treatment of cough, dyspnea, cardiac and spleen disorders, chronic-fever, gout, rheumatic pain etc. [16].
In recent years, the advancement in chemistry, pharmacology and systems biology has created a new paradigm for the drug discovery known as network pharmacology [17]. Integration of traditional knowledge of medicines with recent in silico approaches has led to the identification of novel natural drug compounds. The approach has recently gathered much attention by the research community as network pharmacology based studies have been widely used to explore the medicinal activities of herbs like Withania somnifera [18] and formulae like QiShenYiQi [19], Gegen Qinlian decoction [20] etc. to understand their molecular level effect in the treatment of syndromes or diseases.
In the present work, as the workflow (Fig 1), we firstly, reviewed the phytochemicals of P. longum as reported in the literature and public databases and attempted to cluster them in terms of their chemical and functional classes. Therapeutic relevance of these compounds was inferred through the network analysis of phytochemicals with their protein targets and their therapeutic activity was correlated with the number of proteins that a particular phytochemical may target. Further, the pharmacological action of these metabolites at biological level was explored and the potential metabolic and cellular pathways in which the target proteins are involved have been identified. We explain the disease association network that is constructed to interpret the relationship between the potential drug candidates in the human system. A subnetwork of human protein-protein interaction (PPI) network that is potentially regulated by P. longum was analyzed to identify functional modules present therein. Pharmaceutically relevant features of these phytochemicals were studied and drug likeliness of various phytochemicals was evaluated and finally, the molecular interactions of some of the potential drug-like phytochemicals with the protein targets involved in the neurological disorders were explored.
Materials and methods
Data collection
A dataset of phytochemicals present in P. longum was developed using extensive literature survey and mining of public database resources. Relevant research articles from PubMed-NCBI (https://www.ncbi.nlm.nih.gov/pubmed/) were selected and manually scrutinized. Three databases UNPD (Universal Natural Products Database) (pkuxxj.pku.edu.cn/UNPD) [21], TCMSP (Traditional Chinese Medicine Systems Pharmacology) [22] and PCIDB (PhytoChemical Interactions DB) (http://www.genome.jp/db/pcidb) [23] were screened for potentially active phytochemical present in P. longum. Chemical information of these phytochemicals was compiled from PubChem database [24] and ChEMBL [25] databases. Structural data of phytochemicals not available in PubChem and ChEMBL was derived using PubChem Sketcher v2.4 [26]. Compiled dataset was filtered to remove duplicate entries. Human proteins targeted by the phytochemicals were predicted from STITCH 5.0 (Search Tool for Interactions of Chemicals and proteins) [27], BindingDB [28], SwissTargetPrediction [29] and PCIDB.
The protein-chemical interaction reported in STITCH comes from the manually curated datasets including DrugBank [30] GLIDA [31], Matador [32], TTD (Therapeutic target database) [33] etc. In order to provide a full picture of the available data, STITCH also integrates information from several metabolic pathway databases and experimentally validated interactions from ChEMBL [25], PDB [34] and other sources. To access the high confidence targets, the interactions with a combined score of ≥0.4 were taken into account. The basic working principle of BindingDB is that similar compounds tend to bind the same proteins. It is a web-accessible database containing binding affinities of 20,000 (approx.) protein-ligand complexes. We screened the phytochemical-protein interactions having similarity search value ≥0.85. SwissTargetPrediction uses a combination of 2D and 3D similarity measures to identify the target proteins that can bind with a ligand showing the highest similarity with a library of 280,000 compounds. Top 15 protein targets for each phytochemical were selected from this resource. PCIDB returns a list of active compounds of the query herb and also enlists the possible genes involved in the interactions. We included all the phytochemical-protein interactions provided by PCIDB. UniProt IDs of all the protein targets identified from the above-mentioned four resources were used for network construction.
The biological pathways association of the identified protein targets was retrieved from KEGG database (Kyoto Encyclopedia of Genes and Genomes) [35]. A comprehensive platform of gene-disease association, called DisGeNETv4.0 [36] was used to find the disease information in which the protein targets may be involved. DrugBank database [37] was also mined to find-out the known target proteins.
Compounds classification and clustering
An automated and rapid chemical classification method “ClassyFire” [38] was used to assign a chemical class to the phytochemicals. The query is mapped to the various classes based on its features that are calculated using superstructure-search operations and other properties. Clustering toolbox from ChemMine tools [39] was used for clustering the phytochemicals. The hierarchical clustering algorithm was opted, that forms a hierarchy of clusters based on pairwise compound similarities using the atom-pair descriptors and Tanimoto coefficient.
Network construction and analysis
To investigate the pharmacological actions of the phytochemicals, various networks showing the interactions among phytochemical compounds (PC), protein targets (PT), biochemical pathways (BP) and associated diseases (AD) were constructed and analyzed using Cytoscape v3 [40].
Modularity analysis of human PPI subnetwork
STRING v10.5 [41] was used to identify first-degree interactors of all the target proteins. Only high confidence interactions (score > = 0.9) were included to construct the PPI and duplicate edges were removed. MCL (Markov Cluster) algorithm [42] was used to cluster and organize the proteins in various modules. This algorithm detects cluster structure using mathematical bootstrapping procedure and has been shown that it is efficient in the identification of modules in PPI networks [43]. GO-based functional enrichment of these clusters was carried out using BINGO [44].
Drug-likeliness prediction and molecular docking studies
Pharmacokinetic and toxicity properties of the compounds were studied using pKCSM which uses the graph-based structure signature method to predict a range of ADMET (Absorption, Distribution, Metabolism, Excretion and Toxicity) properties [45].
AutoDock 4.2 [46] was used to carry out the computational docking studies, using Lamarckian Genetic algorithm. The best binding orientation of the ligand within the protein cavity was estimated using binding energy values. The 2D representation of protein-ligand complexes for their molecular interactions was carried using Ligplot+ [47].
Results and discussion
Identification of phytochemicals
A dataset of phytochemicals present in P. longum was created from extensive literature survey and mining of natural product databases. In total 159 phytochemicals were identified and all the phytochemicals were assigned a unique ID. Details of all the phytochemicals i.e. their unique IDs, names and their references are presented in Table 1.
Compounds classification and clustering
From the chemical classification, 159 phytochemicals were found to be distributed among 26 different classes of compounds. Out of these, 11 classes contain only one phytochemical and these correspond to carboxylic acids and derivatives (PL141), cinnamic acids and derivatives (PL68), isoflavonoids (PL146), naphthalenes (PL72), organic nitrogen compounds (PL62), oxanes (Pl71), phenanthrenes and derivatives (PL147), phenol ethers (PL107), phenylpropanoic acids (PL69), pteridines and derivatives (PL67), pyridines and derivatives (PL17) and steroid and its derivatives (PL130). Adaptation of plants against various abiotic and biotic stresses over millions of years of evolution is responsible for such chemical diversity of the phytochemicals [55]. Among all these classes, the benzodioxole group constitutes the highest number of phytochemicals (31) and they are found to be clustered together. It is a widely dispersed class of compounds among natural as well as synthetic drugs [56].
The hierarchical clustering of phytochemicals is shown in Fig 2. The phylogenetic tree reveals that phytochemicals cluster with molecules that share similar scaffold. The class of chemical compounds corresponding to prenol lipids and fatty acyls are highly prominent in the dataset; a good agreement with the known fact that lipids form a large group of primary metabolites of the plant [57].
Phytochemicals were clustered on the basis of atom-pair descriptors and Tanimoto coefficient using Chemmine tool. It can be easily seen that most of the phytochemicals belonging to benzodioxoles, fatty acyl and prenol lipids category were clustered together.
Phytochemical–protein target (PC-PT) network
For understanding the interactions between small molecules and proteins, the PC-PT bipartite network was constructed by mapping 159 phytochemicals to their potential proteins targets. This resulted in the identification of 1109 unique human proteins that may be potentially targeted by the phytochemicals of P. longum. As may be seen in top two layers of the tripartite network in Figs 3A and 4A, many phytochemicals are found to interact with multiple proteins, an effect known as polypharmacology. Polypharmacological effect of the P. longum phytochemicals was evaluated using its PCt/Tt value (the average value of a number of targets for each compound). This value is calculated for each phytochemical identified, higher PCt/Tt value for a compound suggests that it may be an activator or an inhibitor of multiple proteins and may be individually or in combination serve as lead-compound. Numerous existing drugs are well known for their multi-targeting activities. One such example is Aspirin; which is usually used as an analgesic and at times also as an antipyretic [58], and its anti-inflammatory medication in treatment of various diseases like rheumatoid arthritis [59], pericarditis [60] is also well known.
(A). A tripartite phytochemical—protein target—biochemical pathway (PC-PT-BP) network of the constituents of Piper longum. Top layer (blue) represents phytochemicals (159), middle layer (red) represents the potential protein targets (1109) and the third layer (green) represents association of protein targets with six pathway classes: metabolism, genetic information processing, environment information processing, cellular processes, organismal systems and human diseases. (B) Involvement of target proteins of Piper longum in various human pathway classes. The human pathway mapping of identified target proteins were distributed among 6 classes. Genetic information processing class includes pathways belonging to transcription, translation, replication & repair, folding, sorting and degradation. Environment information processing includes membrane transport, signal transduction, signaling molecules and their interactions. Cellular processes involve pathways of cell growth, cell death and transport (endocytosis, phagosome, lysosome etc.). Organismal system includes immune, endocrine, circulatory, digestive, excretory, nervous, sensory system. Human diseases and metabolism include pathways associated with diseases and metabolic system (carbohydrates, lipids, amino acids etc.) respectively. (C) Bioactives of Piper logum affecting the Human immune system. The first layer (blue) represents 106 phytochemicals, regulating 11 immune pathways (green) by targeting 131 proteins (red).
(A). Phytochemical—protein target—disease association (PC-PT-DA) network of Piper longum. Top layer (blue) represents phytochemicals (159), middle layer (red) represents their protein targets (1109) and the bottom layer (green) represents association of protein targets with 27 disease classes obtained from DisGeNET. Size of the nodes in third layer (green) is proportional to their degree value. (B) Piper longum protein targets and their involvement in 27 disease classes. Number of potential protein targets associated with various disease classes as obtained from DisGeNETv4.0. Maximum number of targets are related with neoplasm and nervous system, while muscular dystrophy and occupational diseases have least number of targets associated with them.
PL70 (palmitic acid) being ubiquitous in nature emerges out as a compound with the highest PCt/Tt value (0.55) in the network. Next in order are the phytochemicals PL99, PL11, PL25 and PL66 with PCt/Tt value 0.068, 0.060, 0.057, 0,052 respectively. Network analysis shows that 99% of these phytochemicals are linked with more than 10 protein targets, indicating towards the multi-target properties of this herb. However, sometimes the concept of polypharmacology could be a double-edged sword and may cause adverse effects when the idea is not fully understood [61]. In that case, the approach may result in identifying off-targets for a new drug.
We also explored the specificity of these phytochemicals to hit a unique target. Among 159 phytochemicals, PL59 (2, 2-Dimethoxybutane (C6H14O2)) is one such compound, that is shown to interact with a single target (UniProt ID—P08575). P08575 (receptor-type tyrosine-protein phosphatase C, CD45) plays a critical role in receptor-mediated signaling in both B and T-cells [62], [63]. Since altered signaling process is one of the causes that leads to several disorders including SCID (Severe combined immunodeficiency syndrome), in such cases the explicit behavior of PL59 and its reliability to target specific protein may be evaluated.
Among 159 phytochemicals, 109 target P10636. Such a high degree corresponding to this node refers its ability to interact with other phytochemicals. In other words, it is the most commonly targeted protein in the network. P10636 is a “microtubule-associated protein tau (MAPT)” encoded by the MAPT gene. A system level investigation of neurodegenerative dementia reveals the accumulation of this protein in a diseased state, including frontotemporal dementia and supranuclear palsy [64]. Hence, it can be assumed that these 109 phytochemicals may contribute to the P. longum’s effect in the treatment of mental diseases. Although previous studies on treatment of mental disorders by the herbal extracts are known [12], piperine (PL25) and piperlongumine (PL68) have got special attention against Parkinson’s disease [65]. Detailed study on the pharmacophore properties of additional phytochemicals identified in P. longum will help in deciphering its detailed mechanism and highlighting their key features that may be helpful in designing synthetic drugs. Additionally, the positive synergistic effect of the compounds can be explored for better affinity and efficacy.
Other proteins with high degree centrality include Q9NUW8, Q9NR56, Q5VZF2, Q9NUK0, P22303 and P06276 with a value corresponding to 64, 60, 59, 57, 56 and 55 respectively. These proteins may be given special attention in aspects of behaving as key targets, specifically in light of the fact that interventions at specific proteins can be weak in terms of binding affinity, yet they may be highly effective in combinations [66].
Phytochemical, protein target and biochemical pathway (PC-PT-BP) network
To obtain a global view of pathways targeted by P. longum, a tripartite network was constructed using its phytochemicals, their protein targets and associated biochemical pathways Fig 3A. 279 unique human pathway maps were classified into 6 broad categories i.e. metabolism, genetic information processing, environmental information processing, cellular processes, organismal systems and human diseases. The detailed mapping of target proteins into different pathways is given in S1 Table.
As suggested in Fig 3B, the highest numbers of target proteins (455, 27%) are associated with pathways belonging to organismal systems, followed by environmental information processing category (384, 23%) and human diseases (371, 22%). These findings suggest that the mode of action of these phytochemicals may be largely via regulating organismal systems (which include immune, endocrine, circulatory, digestive, nervous and excretory system). An earlier in vivo study on P. longum also support this finding, in which it was shown that immunomodulatory properties of the herbal extract lead to increase in white blood cells (WBC) in Balb/c mice [9]. The endocrine effect is explained via the process of ovulation which includes interrelationship between the endocrine and cytokine system. By modulating the inflammatory mediators like cytokines, reactive oxygen species etc., phytochemicals of this plant exert their antifertility properties [67]. Similarly, its effect on cardiovascular [15], digestive [68], [69], nervous [12], [65] and excretory [70] systems has also been studied. Protein targets such as P28482, Q9Y243, P31751, P31749, P42336, P27986, and P27361 are found to be involved in many pathways. Thus it may be hypothesized that these proteins may be important targets, as their modulation may lead to regulation of multiple pathways.
To explore the basic principle of the herb in relation to the human immune system, a sub-network of immune pathways being regulated by P. longum was created. 106 phytochemicals out of 159 are shown to regulate 19 human immune pathways via 131 proteins. 57.25% of these proteins are affecting the immune system via chemokine (hsa04062) and interleukin (hsa04657) signaling pathways. The role of piperine in reducing Th2 cytokines and regulating cytokine in asthma models have been explored earlier [71].
This sub-network identifies Prostaglandin G/H synthases (PTGS) as common targets. PTGS-1 (P23219) and PTGS-2 (P35354) are the targets of 28 and 22 phytochemicals respectively. This enzyme is involved in the conversion of arachidonate to prostaglandin H2, which are the major components that induce inflammation and pain [72]. The high rate of edema inhibition by herb-oil in comparison to the standard anti-inflammatory drug ‘Ibuprofen’ is also reported [73].
Further, the network data provides that NOD-like receptors and Toll-like receptors (TLRs) are regulated through 33 and 29 targets respectively. The receptors play a key role in innate immunity. The pathogen invasion caused by bacterial lipopolysaccharide (LPS) induces signaling pathways which further lead to the activation of macrophages via TLRs. An earlier report on the herb also shows that the root area possesses anti-amoebic properties [69]. Thus, to explore the effect of this herb on the innate immunity, the sequence of events which lead to the interaction of receptor proteins with these phytochemicals (especially present in root region) may be specifically focused upon. Various studies have supported the anti-inflammatory behavior of P. longum. Our work reveals that this effect is not due to a limited number of phytochemicals rather a vast number of phytochemicals are involved in this property. Among 105 phytochemicals involved in immunomodulation, palmitic acid (PL70), demethoxycurcumin (PL11), bisdemethoxycurcumin (PL10), 1,2-benzene dicarboxylic acid (PL66), sesamin (PL96) and piperine (PL25) are the top-immunomodulators with 76, 15,12, 11, 11, 10 protein targets, respectively. Thus, combining the effect of other phytochemicals reported in our study will help to provide a wholistic view of the herb’s immunomodulatory potency. Additionally, the use of analytical and structural chemistry of the phytochemicals and phytochemical-protein target complexes will help in understanding the molecular mechanism in detail. The detailed information of the immune pathways considered and the number of target proteins involved in each class is presented in Table 2.
Phytochemical, protein target and disease association (PC-PT-DA) network
The data of protein targets and the diseases in which they are involved was collected and a tripartite network were drawn (Fig 4A). The diseases were classified into 27 classes and distribution of proteins among each of these classes is shown in Fig 4B. It can be easily seen that majority of protein targets have their association with neoplasm and nervous system diseases, that are 908 and 770 proteins, respectively. Numerous studies have shown the effect of long pepper plant in the treatment of different cancers including prostate, breast, lung, colon, lymphoma, leukemia, primary brain tumors and gastric cancer. A recent study states that the plant’s anti-cancerous property is due to the inhibitory mechanism of Glutathione S-transferase pi 1 (GSTP1) by its compound piperlongumine (PL29). GSTP1 is overexpressing protein in cancerous cells and the reactive olefins in the piperine (PL25) attenuate the cancer-cell proliferation by blocking its active site [74].
According to the network data, the association between P. longum and nervous system diseases is mainly by the regulation of 434 protein targets. Among these, the majority of proteins are involved in signaling transmission and developmental pathways. Out of total 434, 29 protein targets are found to be interacting with piperine (PL25) that is known to have anti-epileptic [75], analgesic and anti-convulsant [76] nature. Anti-depression like activity of this compound on animal samples suggests that the compound may act as a potential functional food to improve brain function [77]. Analysis of degree distribution shows that 1,2-Benzenedicarboxylic acid (PL66), demethoxycurcumin (PL11) and D-camphor (PL99) are having a higher number of protein interactors in comparison to piperine (PL25) i.e 36, 34 and 33, respectively. Thus, it may be inferred that these phytochemicals may also have an effect in neuroprotection and may act as new lead compounds for neurodegenerative disorders.
The prioritization of the target proteins of P. longum was performed by mapping them to the approved drug targets of DrugBank with the intent of selecting known drug targets for identifying their novel regulators from this herb’s constituents. For this, common target proteins among all the four target prediction software were selected. Five proteins (P04150, P37231, Q8NER1, P21397, and P27338) among these are FDA approved drug-targets, reflecting their importance for the identification of their novel regulators. The position of these potential targets in the PC-PT-DA network shows that 77 phytochemicals are involved in their regulation (Fig 5). Further, these key targets are associated with a diverse array of diseases that means they are involved at multiple levels in the biological system. Twenty eight protein targets were predicted by three of the four softwares used for phytochemical-protein interaction prediction. These may be considered as potential targets and may be further explored with respect to the phytochemicals involved in their regulation (S3 Table). The network data provides a scope to explore the interrelationship of the target proteins with phytochemicals using computer-aided drug discovery approaches, which will be helpful in understanding the medicinal and multi-targeting potential of the herb in detail. Further, pleiotropic nature of the genes at the system level may also be investigated.
Middle layer (red) represents the 5 key protein targets, which are regulated by 77 phytochemicals of Piper longum, represented in the top layer (blue). Third layer (green) represents the mapping of these targets in 825 diseases.
Module and GO enrichment analysis of human PIN
First-degree interactors of all the target proteins of P. longum were used to construct a subnetwork of the PIN of Homo sapiens (Fig 6). Topological analysis of the network shows that its degree distribution follows a power-law with y = 975.99x-1.245. The PIN was analysed using MCL clustering algorithm for module identification. Modules are shown in supplementary material (S1 Fig). Functional enrichment analysis of the modules with dense connection shows that P. longum exerts its effect mainly through regulating cell cycle, signal transduction, genetic information processing and metabolism machinery (Table 3).
(A). Protein-protein interaction subnetwork of Homo sapiens targeted by phytochemicals of P. longum. First neighbours of all the targets proteins were mapped into the human PPI as obtained from STRING having high confidence level (score ≥ 900). Green highlighted nodes in the network represent the location of the target proteins of P. longum. (B) Node degree distribution of the PPI subnetwork. The neighbourhood connectivity of each node is represented using node degree distribution graph, analysed using Cytoscape. Both axes of the graph are represented in the logarithmic scale.
Replication and repair mechanism (Module-1) involves proteins such as ATM, BLM and BRCA2. ATM is serine/threonine kinase gene and acts as an important cell-cycle checkpoint kinase [78]. BLM gene is a Bloom syndrome RecQ like helicase; protein encoded by this gene is involved in suppression of inappropriate recombination event in the cell [79]. BRCA gene helps in maintaining the stability of the genome. BRCA2 is involved in double-stranded DNA repair by regulating the homologous recombination pathway [80]. Mitotic cell cycle (Module-3) contains genes like ANAPC10, CDC 20 etc. ANAPC10 is a core subunit of anaphase-promoting complex and it is known that APC genes get altered in human colon cancer [81]. CDC20 (cell division cycle 20) acts as a regulatory protein and it has been shown in vitro that it is a promising therapeutic target for cancer treatment [82]. This shows that P. longum anti-cancerous activities are mainly due to the regulation of cell-cycle events and DNA-repair mechanisms.
It is commonly observed that disease occurrence is associated with the signal transduction failure, but the degree of its association varies greatly depending on the severity of disease. Wnt signaling pathways (Module-8) contain genes such as APC2, WNT7A and FRAT-1. Mutation in Wnt signaling pathway genes like APC (Adenomatous polyposis coli) is particularly evident in memory deficit cases. A gene knockout study in mice sample has shown that this gene plays an important role in the regulation of spinal locomotor activity and memory performance [83]. Its involvement in ocular, bone density disorders [84], [85] and colorectal cancer [86] are also well studied. Proteins of WNT gene family like WNT7A (Wnt family member 7A) encode secreted signaling proteins [87]. An earlier report in the developmental biology shows that Wnt7a is a highly conserved gene and plays an important role in early development of midbrain and telencephalon regions of the human brain [88]. Target data analysis shows that “P12644” encoded by BMP4 (Bone morphogenetic protein-4) is the target gene of WNT/beta-catenin signaling pathway [89]. Thus, it may be hypothesized that the herb’s nootropic effect is mainly associated with the proteins constituting this module.
P. longum’s anti-inflammatory quality may be linked to G-protein coupled signaling process (Module-4). This module contains genes like GRK2 and GRK6. Inflammatory mediators modulate GRKs signaling either by transcription regulation or its degradation. GRK2 act as a mediator in the pathway that causes inflammatory pain [90]. Thus, targeting GRKs by P. longum could be the reason of its anti-inflammatory properties. Pain relieving properties of this herb can be easily linked to the proteins in Module-4.
Proteins of module 10 and 14 are involved in genetic information processing by regulating transcription and translation machinery respectively. Genes like HES1, RUNX1, TGFB1, EP300 constitute module 10. HES1 is a hes family bHLH transcription factor-1. RUNX-1 is runt-related transcription factor 1 which participates in hematopoiesis. Its involvement in leukemia conditions is well documented [91]. TGFB1 is a transforming growth factor beta-1 and it regulates cell proliferation and differentiation, but shows an unregulated response in the tumor cells [92]. EP300 encodes E1-A associated cellular p300 transcriptional co-activator protein. It also helps in stimulating hypoxia-induce genes. A defect in the gene leads to Rubinstein-Taybi syndrome [93]. Thus, P. longum may be effective in improving transcriptional errors or the diseases associated with it.
Module 14 constitutes ribosomal proteins like RPL36, RPS3, RPL8, RPS7, RPS13 and RPL3. These proteins help in maintaining the structural integrity of the ribosomal assembly. The extra-ribosomal function of ribosomal proteins includes regulation of gene expression, cell-cycle control, regulation of apoptosis, modulation of DNA repair, regulation of development and differentiation, modulation of cell migration and invasion and regulation of angiogenesis [94]. Dysfunction of ribosome leads to a condition known as ribosomopathies. Although no data support the herb’s effectiveness in treating such dysfunction, but we believe that this area needs a detailed investigation. P. longum phytochemicals may have a regulatory effect on ribosomopathies.
Proteins of module-7 are present in the metabolic machinery which is reported to regulate Electron transport chain (ETC) and forms the basis of the energy production in the cell. These proteins are MT-CO1, UQCRC1 and NDUFA6. MT-CO1 (mitochondrial cytochrome c oxidase subunit-1), UQCRC1 (ubiquinol-cytochrome c reductase core protein) and NDUFA6 (NADH ubiquinone reductase subunit A6) are associated with the redox processes of ETC [95], [96], [97] and aid in an essential aspect of ATP generation. Abnormalities in the ETC chain are a characteristic feature of Alzheimer’s [98] and Parkinson’s [99] diseases of the brain. Thus, the chemical compounds produced in P. longum are involved in the crucial steps of ETC.
Glucose metabolism (Module-11) includes enzymes that are crucial for the conversion of glucose into pyruvate. This process is carried out by the cell to meet its energy requirements. The module contains gene like PGK2 (Phosphoglycerate kinase 2), ENO1 (Enolase 1), PKLR (Pyruvate kinase) and ALDOA (Aldolase, fructose-bisphosphate A). This shows that the herb may also be helpful in regulating energy metabolism. Additionally, high rate of glycolysis have been reported in cancerous cells [100]. In such condition, P. longum’s anti-cancerous property may be correlated with its interaction with the proteins of module involved in glucose metabolism.
Drugability analysis and docking studies
The estimation of ADMET and other drug-like properties are important to consider at an early stage of drug-discovery process, as the majority of drug candidates fail in clinical trials due to poor ADMET properties [101]. We evaluated the complete pharmacokinetic and toxicity profile of each phytochemical of P. longum for characterizing its drug likeliness (Fig 7) (S2 Table).
The graph shows the plots of ADMET variables corresponding to octanal-water partition coefficient (log_P), water solubility (Water_sol.), skin permeability (Skin_perm.), blood-brain permeability (BBB_per), central nervous system permeability (CNS_perm.), total clearance (Total_clear.), maximum tolerated dose-humans (MTD_humans), oral rat acute toxicity (ORA_tox.), oral rat chronic toxicity (ORC_tox.), T. Pyriformis toxicity (T_pyri. tox), minnow toxicity (Minnow_tox), caco2 cell permeability (Caco2_per.) and intestinal absorption (Intestinal_abs.).
In silico estimation shows that the percentage intestinal absorption of the phytochemicals is more than 90% for 136 phytochemicals. 93% of the phytochemicals are likely to permeate the Caco2 cells as they have a permeability value greater than 1. Longumosides B (PL13), piperazine adipate (PL63), 2-amino-4-hydroxypteridine-6-carboxylic acid (PL67) and 4-[(1-Carboxy-2-methylbutyl) amino]-2(1H)-pyrimidinone (PL141) are predicted to have the least permeability among all other phytochemicals. The possible distribution of compounds through various compartments of the body was accessed using their blood-brain barrier (BBB) penetration and central nervous system (CNS) penetration coefficient. One of the criteria for a successful CNS drug is that the compound should not have binding affinity for P-glycoprotein [102]. The result shows that 111 phytochemicals may act as non-substrate of P-glycoprotein, thus possess the ability to be a potential CNS drug. Piperine (PL25) is known to have CNS acting power [103], the BBB and CNS penetration values were found to be -0.131 and -1.932 respectively. 115 phytochemicals of P. longum were showing the BBB and CNS penetration values similar to or greater than the piperine (PL25) values. This indicates that CNS targeting potential of other phytochemicals is noteworthy and should be explored further.
Lipinski’s “rule of five” criterion was adopted to estimate the likeliness of the phytochemicals to act as drug molecule [104]. 105 phytochemicals are shown to maintain the criteria of molecular weight less than 500 Dalton, number of hydrogen bond donors less than 5, hydrogen bond acceptors less than 10 and logP value (octanol-water partition coefficient) less than 5. Toxicity risk was also evaluated by checking Ames toxicity, oral-rat acute toxicity, oral-rat chronic toxicity, hepatotoxicity, cardiotoxicity, T. pyriformis toxicity and Minnow toxicity. 13% of the compounds are predicted to be positive for hepatotoxicity. The cardiotoxicity was evaluated with hERG (human ether-a-go-go-related gene) inhibition. Interestingly, not a single phytochemical is positive for hERG1 inhibition, which reflects the cardioprotective nature of P. longum.
The final screening of all the parameters in phytochemical dataset resulted in identification of 20 phytochemicals that possess a high probability of acting as effective lead molecules (S2 Table). For identifying the novel regulatory molecules to the previously discussed 5 key protein targets, these 20 phytochemicals were back-mapped in the PC-PT network. This resulted in the selection of 7 phytochemicals that were forming 16 phytochemical-protein target pairs. To estimate the binding affinity of these potential phytochemical-protein target pairs, docking studies were performed. This is essential to identify the best fit between the phytochemical and protein molecule, both in terms of energy and geometry. Binding energy calculations of each pair is represented in Table 4. Chemical features of all these phytochemicals can be explored further to design their synthetic analogs with optimised pharmacological activity.
A case-study on the Piper longum’s action on neurological diseases and disorders
To further explore the neuromodulatory prospectives of the P. longum, curated gene-disease associations corresponding to nervous system diseases and mental disorders were selected from PC-PT-DA network. This resulted in the identification of 384 protein targets, out of these 215 were FDA-approved protein targets as listed in Drugbank. These proteins were back-mapped to PC-PT-BP network for selecting their interactions with previously shortlisted 20 potential drug-like phytochemicals.
In this way, we could derive a sub-network specific to neurological diseases and consisting of druggable phytochemical—protein target pairs from the PC-PT-BP network of P. longum (S2 Table). In this sub-network, it is observed that metabolic pathways (path:hsa01100) and neuroactive ligand-receptor interaction (path:hsa04080) were highly enriched, implying that multiple proteins from the dataset exert their biological functions mainly through these pathways. This result is in confirmation with the earlier findings showing the atypical role of impaired metabolic pathway processes in various neurological diseases and disorders like Alzheimer etc. [105].
Identification and inquiry of neuroactive pathways are crucial for the design and development of improved therapeutic strategies against nervous system disorders like Schizophrenia [106] and Parkinson’s disease [107]. Therefore, the derived sub-network is analyzed in detail with specific focus to highlight the position and role of protein targets in the neuroactive ligand-receptor interaction pathway. We could map 11 proteins from 6 gene classes to this pathway and are shown as yellow rectangles in Fig 8. To figure out if these proteins participate in a specific biological process, these were back mapped to the module classes of the human PIN (detailed in section corresponding to module and GO enrichment analysis). 6 proteins out of 11 were showing their involvement in “G-protein coupled receptor protein signaling pathway” corresponding to Module 4 (S2 Fig). This illustrates that these proteins (corresponding to the gene class CHRM, ADRA, DRD, PTGER1) mutually contribute to the G-protein coupled signaling process. The alteration of the signaling process, especially dysfunction of GPCRs, has been a cause for the pathological changes within a brain region. In recent years, targeting potential of GPCR heteroreceptor complexes specific to CNS region is being explored to provide new insights towards the field of neuropharmacology. They have become impressive targets for neurological and mental disorders like schizophrenia, anxiety, depression and Parkinson’s disease [108].
The mapped genes corresponding to CHRM, ADR, DRD, PTGER, CHRN and TRPV1 are marked in yellow.
Phytochemical mapping of the 11 proteins highlights the role of 14 potential drug-like phytochemicals in their regulation. It was observed that sesamol (PL95) has a one-to-one regulatory relationship with PTGER1 (P34995) in this pathway, while three phytochemicals (PL6, PL104 and PL152) have multi-targeting potential for regulating more than one protein from the rest of 5 gene classes (Fig 9).
DPC (blue rhombus) stands for Drugabble PhytoChemicals from P. longum and PTN (red triangle) stands for Protein Targets selected from Neuroactive-ligand interaction pathway (path:hsa04080). 11 potential protein targets (covering 6 gene classes) for neurological disease and disorders from “path:hsa04080” are regulated by 14 potential drug-like phytochemicals of P. longum. Edge of the network represents the possible regulatory phytochemical partner for each protein target. Proteins are grouped according to their gene class. Three gene classes corresponding to adrenergic receptor (ADA), Muscarinic acetyl choline receptor (CHRM) and Dopamine receptor (DRD) constitute multiple proteins, while remaining three corresponding to Transient receptor potential cation channel subfamily V member 1 (TrpV1), Prostaglandin E receptor 1 (PTGER1) and Cholinergic receptor nicotinic alpha7 subunit (CHRNA7) are specific for a particular protein.
PTGER1 is a prostaglandin E receptor 1 (EP1), one of the four G-protein coupled receptors for PGE2 (Prostaglandin E2) that are EP1, EP2, EP3 and EP4. PGE2 is the most abundant eicosanoid in the human system that shows a range of paracrine and autocrine effects by binding to these GPCRs [109], [110], [111] and PGE2 pathway is also known to be acting as a novel biomarker against antipsychotic treatment [106]. The possible molecular interaction between PL95 and PTGER1 was studied using Autodock Vina that estimated the binding affinity to be -4 kcal/mol. As only few inhibitors are available for the PGE2 receptors, the inhibitory activity of PL95 can be explored further for developing novel chemoprotective and antipsychotic agents.
Multi-targeting phytochemical screening identifies PL6, PL152 and PL104 as a common ligand for two different receptors. Selection of such phytochemical that targets multiple genes, adds an extra advantage towards their choice as a drug candidate. To evaluate the molecular interactions of these phytochemicals and their interacting protein partners from the given pathway, complexes of enzymes and ligands were obtained using Autodock Vina. Binding affinity calculation for each pair is shown in Table 5.
Molecular interactions of PL6 that is a potential regulator of protein targets from two gene classes, DRD (Dopamine receptor) and TrpV1 (Transient receptor potential cation channel subfamily V member 1), are studied in detail. Altered functioning of the DRD2 receptor (corresponding to P14416 in our dataset) has been a cause for schizophrenia and search of novel antipsychotics corresponding to DRD2 receptor antagonists are still in process [98]. Similarly, TrpV1 (corresponding to Q8NER1) present in the brain is a key protein for microglia and neuron communication. Neurobehavioral studies show their involvement in neurological and psychiatric disorders like epilepsy, depression etc. [111]. There has been a continuous search of the antagonists for TRPV1 as anti-inflammatory agents like capsaicin and resiniferatoxin (RTX) for the treatment of neuropathic pain [112]. Ligplot+ analysis of the PL6 and Q8NER1 complex highlights the role of Glutamic acid (Glu:392) in the interaction. Similarly, the analysis of docking complex of PL6 and its interacting protein corresponding to other class of membrane receptors (P14416: Dopamine receptor: DRD2) shows that the compound is able to form hydrophobic interactions with the protein molecule. Further, Histidine residue (His:303) of the protein is involved in the formation of a hydrogen bond with PL6 while for P35462 (Dopamine receptor: DRD3) Isoleucine (Ile:183) is hydrogen bonded (Fig 10).
Two dimensional representation of interaction observed between PL6 and its interacting proteins listed in Table 5. (A) Ligplot+ analysis of the docked complex of PL6 and P14416: Dopamine receptor DRD2. (B) Ligplot+ analysis of the docked complex of PL6 and P53462: Dopamine receptor DRD3. (C) Ligplot+ analysis of the docked complex of PL6 and Q8NER1 (TrpV1). Protein residues involved in hydrophobic interactions are represented as arcs and hydrogen bonding with dashed lines.
Using the similar strategy, crucial pathways associated with other disease classes can be analyzed and explored for the detailed information about their regulatory phytochemicals. We hope that methodology developed in this work will open a new way to explore the drug-like molecules from natural herbs for disease management.
Summary
The traditional Indian medicine (TIM) system commonly known as Ayurveda is a more than four thousand years old heritage of the Indian subcontinent and is a huge repository of information about multiple natural medicines for their therapeutic potential. P. longum is an important constituent of many Ayurvedic formulations and is most widely used as a part of “Trikatu”. However, the multi-targeting potential of this herb and underlying mechanism of its cellular-level action are still unexplored. In the current study, we have examined the medicinal effects of P. longum using network pharmacology, an approach that has emerged in recent years as a key route to investigate the healing potential of the traditional herbs for drug discovery and drug development procedures. Our methodology involved literature survey, database mining, drug-likeliness prediction, phytochemical-protein target, and protein target-disease relationship study to examine the multifaceted potential of this herb. Among 159 phytochemicals, 20 are estimated to be the potential lead molecules based on the drug-likeliness filter. 14 of these phytochemicals were found to be regulators of protein targets involved in the nervous system related diseases and disorders. These phytochemicals affect the signaling process of the neuroactive ligand interaction pathway via regulation of 11 proteins. Specific and multi-targeting potential of 4 of these compounds were further explored via molecular docking studies for their possible usage in neuro pharmacotherapy. We expect that this work based on systems-level network assisted studies of P. longum will offer a new way to look upon the hidden potentials of this herb. The data obtained from docking analysis may be taken for in vitro studies, which may eventually be helpful in identification of novel and effective lead molecules from a natural pool of compounds present in P. longum. We are hopeful that this study will prove to be an important basis for understanding the phytochemical-protein level coordination in various Ayurvedic formulae that use P. longum as an integral part. The phytochemical-protein target and disease relationship represented in the form of interaction networks will be helpful in understanding the underlying molecular mechanism in detail. We believe that the comprehensive computational approach developed in this work involving multi-level studies of phytochemical-protein target interaction identification will be helpful to screen natural lead like compounds against various diseases.
Supporting information
S1 Fig. Modules identified in the subnetwork of the PIN of Homo sapiens, constructed using the first-degree protein interactors of all the target proteins of P. longum.
The modules were identified using MCL (Markov Cluster) algorithm.
https://doi.org/10.1371/journal.pone.0191006.s001
(TIF)
S2 Fig. Mapping of the proteins associated with neuroactive–ligand interaction pathway in the module 4 corresponding to “G-protein coupled receptor protein signaling pathway”.
https://doi.org/10.1371/journal.pone.0191006.s002
(TIF)
S1 Table. Information of the 1109 protein targets of P. longum with their KEGG pathway mapping.
https://doi.org/10.1371/journal.pone.0191006.s003
(XLSX)
S2 Table. ADMET values of the phytochemicals of P. longum (including the listing of 20 putative lead phytochemicals).
https://doi.org/10.1371/journal.pone.0191006.s004
(XLSX)
S3 Table. List of potential protein targets (identified by 3 of the 4 protein target prediction softwares) with their interacting phytochemical information.
https://doi.org/10.1371/journal.pone.0191006.s005
(XLSX)
Acknowledgments
N.C. and V.S. thank the Central University of Himachal Pradesh for providing computational infrastructure.
References
- 1. Johri RK and Zutshi U (1992). An Ayurvedic formulation ‘Trikatu’ and its constituents. Journal of Ethnopharmacology 37: 85–91. pmid:1434692
- 2.
Das and Sharma (2002). Caraka Samhita. Chowkhamba Sanskrit Series, Varanasi.
- 3.
Srikantha Murthy KR (2012). Susruta Samhita. Chaukhamba Orientalia, Varanasi.
- 4.
Srikantha Murthy KR (2000). Vagbhata’s Astanga Hrdayam. Chaukhamba Orientalia, Varanasi.
- 5. Tripathi DM, Gupta N, Lakshmi V, Saxena KC and Agrawal AK (1999). Antigiardial and immunostimulatory effect of Piper longum on giardiasis due to Giardia lamblia. Phytotherapy Research 13: 561–565. pmid:10548746
- 6. Srinivasan K (2007). Black pepper and its pungent principle-piperine: a review of diverse physiological effects. Critical Reviews in Food Science and Nutrition 47: 735–748. pmid:17987447
- 7. Srinivasa Reddy P, Jamil K, Madhusudhan P, Anjani G and Das B (2001). Antibacterial activity of isolates from Piper longum and Taxus baccata. Pharmaceutical Biology 39: 236–238.
- 8. Kumar S, Kamboj J and Sharma S (2011). Overview for various aspects of the health benefits of Piper longum linn. fruit. Journal of Acupuncture and Meridian Studies 4: 134–140. pmid:21704957
- 9. Sunila ES and Kuttan G (2004). Immunomodulatory and antitumor activity of Piper longum Linn. and piperine. Journal of Ethnopharmacology 90: 339–346. pmid:15013199
- 10. Sharma AK, Kumar S, Chashoo G, Saxena AK and Pandey AK (2014). Cell cycle inhibitory activity of Piper longum against A549 cell line and its protective effect against metal-induced toxicity in rats. Indian Journal of Biochemistry & Biophysics 51(5): 358–364.
- 11. Jalalpure SS, Patil MB, Prakash NS, Hemalata K and Manvi FV (2003). Hepatoprotective activity of the fruits of Piper longum linn. Indian Journal of Pharmaceutical Sciences 65: 363.
- 12. Hritcu L, Noumedem JA, Cioanca O, Hancianu M, Kuete V and Mihasan M (2014). Methanolic extract of Piper nigrum fruits improves memory impairment by decreasing brain oxidative stress in amyloid beta (1–42) rat model of Alzheimer’s disease. Cellular and Molecular Neurobiology 34: 437–449. pmid:24442916
- 13. Clark CE, Arnold E, Lasserson TJ and Wu T (2010). Herbal interventions for chronic asthma in adults and children: a systematic review and meta-analysis. Primary Care Respiratory Journal, 19: 307–314. pmid:20640388
- 14. Dahanukar SA, Karandikar SM and Desai M (1984). Efficacy of Piper longum in childhood asthma. Indian Drugs, 21(9): 384–388.
- 15. Nabi SA, Kasetti RB, Sirasanagandla S, Tilak TK, Kumar MVJ and Rao CA (2013). Antidiabetic and antihyperlipidemic activity of Piper longum root aqueous extract in STZ induced diabetic rats. BMC Complementary and Alternative Medicine 13: 37. pmid:23414307
- 16. Khushbu C, Roshni S, Anar P, Carol M and Mayuree P (2011). Phytochemical and therapeutic potential of Piper longum Linn a review. International Journal of Research in Ayurveda and Pharmacy 2: 157–161.
- 17. Hopkins AL (2008). Network pharmacology: the next paradigm in drug discovery. Nature Chemical Biology 4: 682–690. pmid:18936753
- 18. Chandran U and Patwardhan B (2017). Network ethnopharmacological evaluation of the immunomodulatory activity of Withania somnifera. Journal of Ethnopharmacology 197: 250–256. pmid:27487266
- 19. Li X, Wu L, Liu W, Jin Y, Chen Q, Wang L, et al. (2014). A network pharmacology study of Chinese medicine QiShenYiQi to reveal its underlying multi-compound, multi-target, multi-pathway mode of action. PLoS One, 9: 95004.
- 20. Li H, Zhao L, Zhang B, Jiang Y, Wang X, Guo Y, et al. (2014). A network pharmacology approach to determine active compounds and action mechanisms of ge-gen-qin-lian decoction for treatment of type 2 diabetes. Evidence-Based Complementary and Alternative Medicine 495840.
- 21.
UNPD–Universal Natural Products Database pkuxxj.pku.edu.cn/UNPD/
- 22. Ru J, Li P, Wang J, Zhou W, Li B, Huang C, et al. 2014. TCMSP: a database of systems pharmacology for drug discovery from herbal medicines. Journal of Cheminformatics 6: 13. pmid:24735618
- 23.
PCIDB- PhytoChemical Interactions Database http://www.genome.jp/db/pcidb
- 24. Bolton EE, Wang Y, Thiessen PA and Bryant SH (2008). PubChem: integrated platform of small molecules and biological activities. Annual reports in Computational Chemistry 4: 217–241.
- 25. Gaulton A, Bellis LJ, Bento AP, Chambers J, Davies M, Hersey A, et al. (2011). ChEMBL: a large-scale bioactivity database for drug discovery. Nucleic Acids Research 40: D1100–D1107. pmid:21948594
- 26. Ihlenfeldt WD, Bolton EE and Bryant SH (2009). The PubChem chemical structure sketcher. Journal of Cheminformatics 1: 20. pmid:20298522
- 27. Szklarczyk D, Santos A, von Mering C, Jensen LJ, Bork P and Kuhn M (2015). STITCH 5: augmenting protein–chemical interaction networks with tissue and affinity data. Nucleic Acids Research 44: D380–D384. pmid:26590256
- 28. Liu T, Lin Y, Wen X, Jorissen RN and Gilson MK (2006). BindingDB: a web-accessible database of experimentally determined protein–ligand binding affinities. Nucleic Acids Research 35 (suppl_1): D198–D201.
- 29. Gfeller D, Grosdidier A, Wirth M, Daina A, Michielin O and Zoete V (2014). SwissTargetPrediction: a web server for target prediction of bioactive small molecules. Nucleic Acids Research 42: W32–W38. pmid:24792161
- 30. Wishart DS, Knox C, Guo AC, Shrivastava S, Hassanali M, Stothard P, et al. (2006). DrugBank: a comprehensive resource for in silico drug discovery and exploration. Nucleic Acids Research 34 (suppl_1): D668–D672.
- 31. Okuno Y, Yang J, Taneishi K, Yabuuchi H and Tsujimoto G (2006). GLIDA: GPCR-ligand database for chemical genomic drug discovery. Nucleic Acids Research 34 (suppl_1): D673–D677.
- 32. Günther S, Kuhn M, Dunkel M, Campillos M, Senger C, Petsalaki E, et al. (2007). SuperTarget and Matador: resources for exploring drug-target relationships. Nucleic Acids Research 36 (suppl_1): D919–D922.
- 33. Chen X, Ji ZL and Chen YZ (2002). TTD: therapeutic target database. Nucleic Acids Research 30: 412–415. pmid:11752352
- 34. Kouranov A, Xie L, de la Cruz J, Chen L, Westbrook J, Bourne PE, et al. (2006). The RCSB PDB information portal for structural genomics. Nucleic Acids Research 34 (suppl_1): D302–D305.
- 35. Kanehisa M and Goto S (2000). KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Research 28: 27–30. pmid:10592173
- 36. Pinero J, Bravo A, Queralt-Rosinach N, Gutiérrez-Sacristán A, Deu-Pons J, Centeno E, et al. (2017). DisGeNET: a comprehensive platform integrating information on human disease-associated genes and variants. Nucleic Acids Research 45: D833–D839. pmid:27924018
- 37. Wishart DS, Knox C, Guo AC, Shrivastava S, Hassanali M, Stothard P, et al. (2006). DrugBank: a comprehensive resource for in silico drug discovery and exploration. Nucleic Acids Research 34 (suppl_1): D668–D672.
- 38. Feunang YD, Eisner R, Knox C, Chepelev L, Hastings J, Owen G, et al. (2016). ClassyFire: automated chemical classification with a comprehensive, computable taxonomy. Journal of Cheminformatics, 8: 61. pmid:27867422
- 39. Backman TW, Cao Y and Girke T (2011). ChemMine tools: an online service for analyzing and clustering small molecules. Nucleic Acids Research 39: W486–W491. pmid:21576229
- 40. Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, et al. (2003). Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Research 13: 2498–2504. pmid:14597658
- 41. Szklarczyk D, Franceschini A, Wyder S, Forslund K, Heller D, Huerta-Cepas J, et al. (2014). STRING v10: protein–protein interaction networks, integrated over the tree of life. Nucleic Acids Research 43: D447–D452. pmid:25352553
- 42. Enright AJ, Van Dongen S and Ouzounis CA (2002). An efficient algorithm for large-scale detection of protein families. Nucleic Acids Research 30: 1575–1584. pmid:11917018
- 43. Brohee S and Van Helden J (2006). Evaluation of clustering algorithms for protein-protein interaction networks. BMC Bioinformatics 7: 488. pmid:17087821
- 44. Maere S, Heymans K and Kuiper M (2005). BiNGO: a Cytoscape plugin to assess overrepresentation of gene ontology categories in biological networks. Bioinformatics, 21: 3448–3449. pmid:15972284
- 45. Pires DE, Blundell TL and Ascher DB (2015). pkCSM: predicting small-molecule pharmacokinetic and toxicity properties using graph-based signatures. Journal of Medicinal Chemistry 58: 4066–4072. pmid:25860834
- 46.
Morris GM, Huey R, Lindstrom W, Sanner MF, Belew RK, Goodsell DS, et al. (2014). AutoDock Version 4.2; Updated for version 4.2. 6. SL: The Scripps Research Institute.
- 47. Wallace AC, Laskowski RA and Thornton JM (1995). LIGPLOT: a program to generate schematic diagrams of protein-ligand interactions. Protein Engineering, Design and Selection 8: 127–134.
- 48. Jiang ZY, Liu WF, Huang CG and Huang XZ (2013). New amide alkaloids from Piper longum. Fitoterapia 84: 222–226. pmid:23232323
- 49. Liu W, Jiang Z, Chen J, Zhang X and Ma Y (2009). Chemical constituents from Piper longum. China Journal of Chinese Materia Medica 34: 2891–2894. pmid:20209953
- 50. Jiang ZY, Liu WF, Zhang XM, Luo J, Ma YB and Chen JJ (2013). Anti-HBV active constituents from Piper longum. Bioorganic & Medicinal Chemistry Letters 23: 2123–2127.
- 51. Sahi S, Tewatia P and Ghosal S (2012). Leishmania donovani pteridine reductase 1: comparative protein modeling and protein–ligand interaction studies of the leishmanicidal constituents isolated from the fruits of Piper longum. Journal of Molecular Modeling 18: 5065–5073. pmid:22752544
- 52. Li K, Zhu W, Fu Q, Ke Y, Jin Y and Liang X (2013). Purification of amide alkaloids from Piper longum L. using preparative two-dimensional normal-phase liquid chromatography× reversed-phase liquid chromatography. Analyst 138: 3313–3320. pmid:23596626
- 53. Das J, Jha DK, Policegoudra RS, Mazumder AH, Das M, Chattopadhyay P, et al. (2012). Isolation and characterization of antidermatophytic bioactive molecules from Piper longum L. leaves. Indian Journal of Microbiology 52: 624–629. pmid:24293721
- 54. Chatterjee A and Dutta CP (1967). Alkaloids of Piper longum Linn—I: Structure and synthesis of piperlongumine and piperlonguminine. Tetrahedron 23: 1769–1781. pmid:6047519
- 55. Dias DA, Urban S and Roessner U (2012). A historical overview of natural products in drug discovery. Metabolites 2: 303–336. pmid:24957513
- 56. Fang ZZ, Zhang YY, Wang XL, Cao YF, Huo H and Yang L (2011). Bioactivation of herbal constituents: simple alerts in the complex system. Expert Opinion on Drug Metabolism & Toxicology 7: 989–1007.
- 57.
Busia K (2016). Fundamentals of Herbal Medicine: History, Phytopharmacology and Phytotherapeutics (Vol. 1). Xlibris, UK.
- 58. Vane JR and Botting RM (2003). The mechanism of action of aspirin. Thrombosis Research 110: 255–258. pmid:14592543
- 59. Preston SJ, Arnold MH, Beller EM, Brooks PM and Buchanan WW (1989). Comparative analgesic and anti‐inflammatory properties of sodium salicylate and acetylsalicylic acid (aspirin) in rheumatoid arthritis. British Journal of Clinical Pharmacology 27: 607–611. pmid:2788004
- 60. Imazio M and Adler Y (2013). Treatment with aspirin, NSAID, corticosteroids, and colchicine in acute and recurrent pericarditis. Heart Failure Reviews 18: 355–360. pmid:22661042
- 61. Reddy AS and Zhang S (2013). Polypharmacology: drug discovery for the future. Expert Review of Clinical Pharmacology 6: 41–47. pmid:23272792
- 62. Desai DM, Newton ME, Kadlecek T and Weiss A (1990). Stimulation of the phosphatidyl-inositol pathway can induce T-cell activation. Nature 348: 66. pmid:2234059
- 63. McFarland EDC and Thomas ML (1995). CD45 protein-tyrosine phosphatase associates with the WW domain-containing protein, CD45AP, through the transmembrane region. Journal of Biological Chemistry 270: 28103–28107. pmid:7499298
- 64. Caberlotto L and Nguyen TP (2014). A systems biology investigation of neurodegenerative dementia reveals a pivotal role of autophagy. BMC Systems Biology 8: 65. pmid:24908109
- 65. Bi Y, Qu PC, Wang QS, Zheng L, Liu HL, Luo R, et al. (2015). Neuroprotective effects of alkaloids from Piper longum in a MPTP-induced mouse model of Parkinson’s disease. Pharmaceutical Biology, 53: 1516–1524. pmid:25857256
- 66.
Mason JS (2012). Designing Multi-Target Drugs: In Vitro Panel Screening–Biological Fingerprinting. In Designing Multi-Target Drugs (Editors: Richard Morphy J., John Harris C). pp. 66–85, Royal Society of Chemistry.
- 67. Lakshmi V, Kumar R, Agarwal SK and Dhar JD (2006). Antifertility activity of Piper longum Linn. in female rats. Natural Product Research 20: 235–239. pmid:16401553
- 68. Ghoshal S, Prasad BK and Lakshmi V (1996). Antiamoebic activity of Piper longum fruits against Entamoeba histolytica in vitro and in vivo. Journal of Ethnopharmacology 50: 167–170. pmid:8691851
- 69. Ghoshal S and Lakshmi V (2002). Potential antiamoebic property of the roots of Piper longum Linn. Phytotherapy Research 16: 689–691. pmid:12410557
- 70. Meena AK, Bansal P and Kumar S (2009). Plants-herbal wealth as a potential source of ayurvedic drugs. Asian Journal of Traditional Medicines 4: 152–170.
- 71. SeungHyung K and YoungCheol L (2009). Piperine inhibits eosinophil infiltration and airway hyperresponsiveness by suppressing T cell activity and Th2 cytokine production in the ovalbumin-induced asthma model. Journal of Pharmacy and Pharmacology 61: 353–359. pmid:19222908
- 72. Ricciotti E and FitzGerald GA (2011). Prostaglandins and inflammation. Arteriosclerosis, Thrombosis, and Vascular Biology 31: 986–1000. pmid:21508345
- 73. Kumar A, Panghal S, Mallapur SS, Kumar M, Ram V and Singh BK (2009). Antiinflammatory activity of Piper longum fruit oil. Indian Journal of Pharmaceutical Sciences 71: 454–456. pmid:20502557
- 74. Harshbarger W, Gondi S, Ficarro SB, Hunter J, Udayakumar D, Gurbani D, et al. (2017). Structural and Biochemical Analyses Reveal the Mechanism of Glutathione S-Transferase Pi 1 Inhibition by the Anti-cancer Compound Piperlongumine. Journal of Biological Chemistry 292: 112–120. pmid:27872191
- 75. Pattanaik S, Hota D, Prabhakar S, Kharbanda P and Pandhi P (2006). Effect of piperine on the steady-state pharmacokinetics of phenytoin in patients with epilepsy. Phytotherapy Research 20: 683–686. pmid:16767797
- 76. Bukhari IA, Alhumayyd MS, Mahesar AL and Gilani AH (2013). The analgesic and anticonvulsant effects of piperine in mice. Journal of Physiology and Pharmacology 64: 789–794. pmid:24388894
- 77. Wattanathorn J, Chonpathompikunlert P, Muchimapura S, Priprem A and Tankamnerdthai O (2008). Piperine, the potential functional food for mood and cognitive disorders. Food and Chemical Toxicology 46: 3106–3110. pmid:18639606
- 78. Abraham RT (2001). Cell cycle checkpoint signaling through the ATM and ATR kinases. Genes & Development 15: 2177–2196.
- 79. Hickson ID (2003). RecQ helicases: caretakers of the genome. Nature Reviews Cancer 3: 169–178. pmid:12612652
- 80. Karran P (2000). DNA double strand break repair in mammalian cells. Current Opinion in Genetics & Development 10: 144–150.
- 81. Wang Q, Moyret-Lalle C, Couzon F, Surbiguet-Clippe C, Saurin JC, Lorca T, et al. (2003). Alterations of anaphase-promoting complex genes in human colon cancer cells. Oncogene 22: 1486–1490. pmid:12629511
- 82. Wang Z, Wan L, Zhong J, Inuzuka H, Liu P, Sarkar FH, et al. (2013). Cdc20: a potential novel therapeutic target for cancer treatment. Current Pharmaceutical Design 19: 3210–3214. pmid:23151139
- 83. Koshimizu H, Fukui Y, Takao K, Ohira K, Tanda K, Nakanishi K, et al. (2011). Adenomatous polyposis coli heterozygous knockout mice display hypoactivity and age-dependent working memory deficits. Frontiers in Behavioral Neuroscience, 5: 85. pmid:22347851
- 84. Canalis E (2013). Wnt signalling in osteoporosis: mechanisms and novel therapeutic approaches. Nature Reviews Endocrinology 9: 575–583. pmid:23938284
- 85. MacDonald BT, Tamai K and He X (2009). Wnt/β-catenin signaling: components, mechanisms, and diseases. Developmental Cell 17: 9–26. pmid:19619488
- 86. Schneikert J and Behrens J (2007). The canonical Wnt signalling pathway and its APC partner in colon cancer development. Gut 56: 417–425. pmid:16840506
- 87. Smolich BD, McMahon JA, McMahon AP and Papkoff J (1993). Wnt family proteins are secreted and associated with the cell surface. Molecular Biology of the Cell 4: 1267–1275. pmid:8167409
- 88. Fougerousse F, Bullen P, Herasse M, Lindsay S, Richard I, Wilson D, et al. (2000). Human–mouse differences in the embryonic expression patterns of developmental control genes and disease genes. Human Molecular Genetics 9: 165–173. pmid:10607827
- 89. Kim JS, Crooks H, Dracheva T, Nishanian TG, Singh B, Jen J, et al. (2002). Oncogenic β-catenin is required for bone morphogenetic protein 4 expression in human cancer cells. Cancer research 62: 2744–2748. pmid:12019147
- 90. Sun L and Richard DY (2012). Role of G protein-coupled receptors in inflammation. Acta Pharmacologica Sinica 33: 342–350. pmid:22367283
- 91. Huang H, Yu M, Akie TE, Moran TB, Woo AJ, Tu N, et al. (2009). Differentiation-dependent interactions between RUNX-1 and FLI-1 during megakaryocyte development. Molecular and Cellular Biology 29: 4103–4115. pmid:19470763
- 92. Cohen MM (2003). TGFβ/Smad signaling system and its pathologic correlates. American Journal of Medical Genetics 116(A): 1–10.
- 93. Roelfsema JH, White SJ, Ariyürek Y, Bartholdi D, Niedrist D, Papadia F, et al. (2005). Genetic heterogeneity in Rubinstein-Taybi syndrome: mutations in both the CBP and EP300 genes cause disease. The American Journal of Human Genetics 76: 572–580. pmid:15706485
- 94. Wang W, Nag S, Zhang X, Wang MH, Wang H, Zhou J, et al. (2015). Ribosomal proteins and human diseases: pathogenesis, molecular mechanisms, and therapeutic implications. Medicinal Research Reviews 35: 225–285. pmid:25164622
- 95. Herrero‐Martín MD, Pineda M, Briones P, López‐Gallardo E, Carreras M, Benac M, et al. (2008). A new pathologic mitochondrial DNA mutation in the cytochrome oxidase subunit I (MT‐CO1). Human Mutation 29(8): E112–122. pmid:18484665
- 96. Cadenas E, Boveris A, Ragan CI and Stoppani AO (1977). Production of superoxide radicals and hydrogen peroxide by NADH-ubiquinone reductase and ubiquinol-cytochrome c reductase from beef-heart mitochondria. Archives of Biochemistry and Biophysics 180: 248–257. pmid:195520
- 97. Zickermann V, Wirth C, Nasiri H, Angerer H, Mankowska M, Wittig I, et al. (2014). Structural analysis of mitochondrial complex I and essential function of accessory subunit NB4M/NDUFA6. BBA-Bioenergetics 40: 902–914.
- 98. Parker WD, Parks J, Filley CM and Kleinschmidt-DeMasters BK (1994). Electron transport chain defects in Alzheimer's disease brain. Neurology 44: 1090–1090. pmid:8208407
- 99. Parker WD, Boyson SJ and Parks JK (1989). Abnormalities of the electron transport chain in idiopathic Parkinson's disease. Annals of Neurology 26: 719–723. pmid:2557792
- 100. Elstrom RL, Bauer DE, Buzzai M, Karnauskas R, Harris MH, Plas DR, et al. (2004). Akt stimulates aerobic glycolysis in cancer cells. Cancer Research 64: 3892–3899. pmid:15172999
- 101. Van De Waterbeemd H and Gifford E (2003). ADMET in silico modelling: towards prediction paradise?. Nature reviews Drug discovery 2: 192–204. pmid:12612645
- 102. Pajouhesh H and Lenz GR (2005). Medicinal chemical properties of successful central nervous system drugs. NeuroRx 2: 541–553. pmid:16489364
- 103. Lee SA, Hong SS, Han XH, Hwang JS, Oh GJ, Lee KS, et al. (2005). Piperine from the fruits of Piper longum with inhibitory effect on monoamine oxidase and antidepressant-like activity. Chemical and Pharmaceutical Bulletin 53: 832–835. pmid:15997146
- 104. Lipinski CA, Lombardo F, Dominy BW and Feeney PJ (1997). Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Advanced Drug Delivery Reviews 23: 3–25.
- 105. Kaddurah-Daouk R, Zhu H, Sharma S, Bogdanov M, Rozen SG, Matson W, et al. (2013). Alterations in metabolic pathways and networks in Alzheimer's disease. Translational Psychiatry 3: 244.
- 106. Adkins DE, Khachane AN, McClay JL, Aberg K, Bukszar J, Sullivan PF, et al (2012). SNP-based analysis of neuroactive ligand-receptor interaction pathways implicates PGE2 as a novel mediator of antipsychotic treatment response: data from the CATIE study. Schizophrenia Research 135: 200–201. pmid:22099390
- 107. Kong Y, Liang X, Liu L, Zhang D, Wan C, Gan Z, et al. (2015). High throughput sequencing identifies MicroRNAs mediating α-synuclein toxicity by targeting neuroactive-ligand receptor interaction pathway in early stage of drosophila Parkinson's disease model. PLoS One 10: e0137432. pmid:26361355
- 108. Borroto-Escuela DO, Carlsson J, Ambrogini P, Narváez M, Wydra K, Tarakanov AO, et al (2017). Understanding the role of GPCR heteroreceptor complexes in modulating the brain networks in health and disease. Frontiers in Cellular Neuroscience 11:37. pmid:28270751
- 109. Lee BC, Kim HS, Shin TH, Kang I, Lee JY, Kim JJ, et al. (2016). PGE2 maintains self-renewal of human adult stem cells via EP2-mediated autocrine signaling and its production is regulated by cell-to-cell contact. Scientific Reports 6: 26298. pmid:27230257
- 110. Dohadwala M, Batra RK, Luo J, Lin Y, Krysan K, Põld M, et al. (2002). Autocrine/paracrine prostaglandin E2 production by non-small cell lung cancer cells regulates matrix metalloproteinase-2 and CD44 in cyclooxygenase-2-dependent invasion. Journal of Biological Chemistry 277: 50828–50833. pmid:12393872
- 111. Marrone MC, Morabito A, Giustizieri M, Chiurchiù V, Leuti A, Mattioli M, et al. (2017). TRPV1 channels are critical brain inflammation detectors and neuropathic pain biomarkers in mice. Nature Communications 8: 15292. pmid:28489079
- 112. Pal M, Angaru S, Kodimuthali A and Dhingra N (2009). Vanilloid receptor antagonists: emerging class of novel anti-inflammatory agents for pain management. Current pharmaceutical design 15: 1008–1026. pmid:19275664