In this work we aim to show how Genome Scale Metabolic Models (GSMMs) can be used as tools for drug design. By comparing the chemical structures of human metabolites (obtained using their KEGG indexes) and the compounds contained in the DrugBank database, we have observed that compounds showing Tanimoto scores higher than 0.9 with a metabolite, are 29.5 times more likely to bind the enzymes metabolizing the considered metabolite, than ligands chosen randomly. By using RNA-seq data to constrain a human GSMM it is possible to obtain an estimation of its distribution of metabolic fluxes and to quantify the effects of restraining the rate of chosen metabolic reactions (for example using a drug that inhibits the enzymes catalyzing the mentioned reactions). This method allowed us to predict the differential effects of lipoamide analogs on the proliferation of MCF7 (a breast cancer cell line) and ASM (airway smooth muscle) cells respectively. These differential effects were confirmed experimentally, which provides a proof of concept of how human GSMMs could be used to find therapeutic windows against cancer. By using RNA-seq data of 34 different cancer cell lines and 26 healthy tissues, we assessed the putative anticancer effects of the compounds in DrugBank which are structurally similar to human metabolites. Among other results it was predicted that the mevalonate pathway might constitute a good therapeutic window against cancer proliferation, due to the fact that most cancer cell lines do not express the cholesterol transporter NPC1L1 and the lipoprotein lipase LPL, which makes them rely on the mevalonate pathway to obtain cholesterol.

Essential oils from plants are a potential source of molecules having anti-inflammatory, anticancer, cardiotropic, and other activities. However, most of these effects lack mechanistic explanations and structure-activity relationship testing. In the present study, we: 1) identified the nutmeg essential oil (NEO) composition; 2) using molecular docking, we determined the putative regulatory binding sites on the connexin 43 (Cx43) that is responsible for gap junction-dependent intercellular communication (GJIC) in the majority of tissues; 3) examined the effect of NEO and its three constituents - sabinene, α-pinene, and α-copaene - on GJ conductance and gating in Novikoff cells expressing endogenous Cx43; and 4) verified whether NEO effects on GJIC correlated with its action on Novikoff cell viability, proliferation, and colony formation capability. Our results revealed NEO and its constituents as potent and efficient Cx43 GJ inhibitors acting by slow gating mechanism. In addition, NEO reduced Novikoff hepatoma cell viability, proliferation, and colony formation capability; however, this was achieved at higher doses and was unrelated to its effects on GJIC.

In our recent study, we have demonstrated that short carbon chain n-alcohols (up to octanol) stimulated while long carbon chain n-alcohols inhibited the conductance of connexin (Cx) 36 (Cx36) gap junction (GJ) channels. In contrast, GJ channels composed of other types of Cxs all were inhibited by n-alcohols independent of their carbon chain length. To identify the putative structural domains of Cx36, responsible for the dual effect of n-alcohols, we performed structural modeling of Cx36 protein docking with hexanol and isoflurane that stimulated as well as nonanol and carbenoxolone that inhibited the conductance of Cx36 GJs and revealed their multiple common docking sites and a single pocket accessible only to hexanol and isoflurane. The pocket is located in the vicinity of three unique cysteine residues, namely C264 in the fourth, and C92 and C87 in the second transmembrane domain of the neighboring Cx36 subunits. To examine the hypothesis that disulphide bonding might be involved in the stimulatory effect of hexanol and isoflurane, we generated cysteine substitutions in Cx36 and demonstrated by a dual whole-cell patch-clamp technique that in HeLa (human cervix carcinoma cell line) and N2A (mouse neuroblastoma cell line) cells these mutations reversed the stimulatory effect of hexanol and isoflurane to inhibitory one, typical of other Cxs that lack respective cysteines and a specific docking pocket for these compounds. Our findings suggest that the stimulatory effect of hexanol and isoflurane on Cx36 GJ conductance could be achieved by re-shuffling of the inter-subunit disulphide bond between C264 and C92 to the intra-subunit one between C264 and C87.

Gap junctions (GJs) made of connexin-43 (Cx43) are necessary for the conduction of electrical impulses in the heart. Modulation of Cx43 GJ activity may be beneficial in the treatment of cardiac arrhythmias and other dysfunctions. The search for novel GJ-modulating agents using molecular docking allows for the accurate prediction of binding affinities of ligands, which, unfortunately, often poorly correlate with their potencies. The objective of this study was to demonstrate that a Quantitative Structure-Activity Relationship (QSAR) model could be used for more precise identification of potent Cx43 GJ inhibitors. Using molecular docking, QSAR, and 3D-QSAR, we evaluated 16 known Cx43 GJ inhibitors, suggested the monocyclic monoterpene d-limonene as a putative Cx43 inhibitor, and tested it experimentally in HeLa cells expressing exogenous Cx43. The predicted concentrations required to produce 50% of the maximal effect (IC₅₀) for each of these compounds were compared with those determined experimentally (pIC₅₀ and eIC_50, respectively). The pIC₅₀ies of d-limonene and other Cx43 GJ inhibitors examined by our QSAR and 3D-QSAR models showed a good correlation with their eIC₅₀ies (R = 0.88 and 0.90, respectively) in contrast to pIC₅₀ies obtained from molecular docking (R = 0.78). However, molecular docking suggests that inhibitor potency may depend on their docking conformation on Cx43. Searching for new potent, selective, and specific inhibitors of GJ channels, we propose to perform the primary screening of new putative compounds using the QSAR model, followed by the validation of the most suitable candidates by patch-clamp techniques.

The effectiveness of lipid-lowering therapies may be insufficient in high-risk cardiovascular patients and depends on the genetic variability of drug-metabolizing enzymes. Customizing statin therapy, including treatment with atorvastatin, may improve clinical outcomes. Currently, there is a lack of guidelines allowing the prediction of the therapeutic efficacy of lipid-lowering statin therapy. This study aimed to determine the effects of clinically significant gene variants of CYP2C19 on atorvastatin therapy in patients with acute coronary syndromes. In total, 92 patients with a confirmed diagnosis of ST-elevation myocardial infarction (STEMI) or non-ST-elevation myocardial infarction (NSTEMI) were sequenced for target regions within the CYP2C19 gene on the Illumina Miniseq system. The CYP2C19 poor metabolizer phenotype (carriers of CYP2C19*2, CYP2C19*4, and CYP2C19*8 alleles) was detected in 29% of patients. These patients had significantly lower responses to treatment with atorvastatin than patients with the normal metabolizer phenotype. CYP2C19-metabolizing phenotype, patient age, and smoking increased the odds of undertreatment in patients (∆LDL-C (mmol/L) < 1). These results revealed that the CYP2C19 phenotype may significantly impact atorvastatin therapy personalization in patients requiring LDL lipid-lowering therapy.