Captagon, known by its genetic name Fenethylline, is an addictive drug that complicates the War on Drugs. Captagon has a strong CNS stimulating effect than its primary metabolite, Amphetamine. However, multi-targets issues associated with the drug and metabolites as well as its underlying mechanisms have not been fully defined. In the present work, we applied our established drug-abuse chemogenomics-knowledgebase systems pharmacology approach to conduct targets/off-targets mapping (SP-Targets) investigation of Captagon and its metabolites for hallucination addiction, and also analyzed the cell signaling pathways for both Amphetamine and Theophylline with data mining of available literature. Of note, Amphetamine, an agonist for trace amine-associated receptor 1 (TAAR1) with enhancing dopamine signaling (increase of irritability, aggression, etc.), is the main cause of Captagon addiction; Theophylline, an antagonist that blocks adenosine receptors (e.g. A2aR) in the brain responsible for restlessness and painlessness, may attenuate the behavioral sensitization caused by Amphetamine. We uncovered that Theophylline’s metabolism and elimination could be retarded due to competition and/or blockage of the CYP2D6 enzyme by Amphetamine; We also found that the synergies between these two metabolites cause Captagon’s psychoactive effects to act faster and far more potently than those of Amphetamine alone. We carried out further molecular docking modeling and molecular dynamics simulation to explore the molecular interactions between Amphetamine and Theophylline and their important GPCRs targets, including TAAR1 and adenosine receptors. All of the systems pharmacology analyses and results will shed light insight into a better understanding of Captagon addiction and future drug abuse prevention.
Captagon, the trademark name for the synthetic stimulant Fenethylline1,2,3,4,5,6, was first reported by a German pharmacist in 1961 for the potential treatment of hyperactivity, depression and narcolepsy1. However, due to its addictive and hallucinogenic properties, it was listed as a controlled substance by the United States (1981) and the World Health Organization (1986), making it illegal to buy or sell Captagon in most countries. Captagon has been reported to be a central nervous system (CNS) stimulator with stronger and longer lasting effects on fighting aggression, detachment, cognitive enhancement, and alertness than one of its main metabolites, Amphetamine.
Captagon is metabolized into Amphetamine (24.5% of oral dose) and Theophylline (13.7% of oral dose), as shown in Fig. 1. Thus, the pharmacological effect of Captagon is considered the result of the combined action of these two metabolites3,6. Amphetamine, the main metabolite of Captagon, is a CNS stimulator that can increase alertness, boost concentration/physical performance, and provide a feeling of well-being, confidence, and aggression. It can be used for the treatment of obesity, narcolepsy, and attention deficit hyperactive disorder (ADHD)7. Amphetamine is reportedly an agonist on central 5-HT receptors8,9,10 and may inhibit monoamine oxidase (MAO)11, possibly causing hallucinations, violent behavior, loss of appetite, and more. Theophylline, another metabolite of Captagon, is a weak stimulator resembling caffeine12. Theophylline has a narrow therapeutic window due to its unwanted side effects including cardia dysrhythmia seizures, gastrointestinal disturbances, and drug-drug interactions13. Captagon was considered a co-drug (or mutual prodrug) of Amphetamine and Theophylline. However, the effects of Fenethylline is different from those of Amphetamine qualitatively and quantitatively. First, previous literature6,14 reported that Captagon is more lipophilic than both Theophylline and Amphetamine, resulting in easier absorption into the CNS and a faster stimulating effect than either drug2. Second, according to a recent report by Wenthur and co-workers2, “the penetration of Amphetamine into brain tissue following release from Fenethylline lags behind that of directly administered Amphetamine” which may explain why Captagon is less addictive than Amphetamine, i.e. because Fenethylline-sourced Amphetamine is less likely to accumulate in the brain. Third, unlike vasoconstrictors such as Amphetamines, Captagon does not increase the patient’s blood pressure because Theophylline possesses vasodilating properties, allowing its use by those with cardiovascular disorders6. More available information regarding Captagon/Fenethylline can be found in a recent minireview paper published by Katselou and co-workers6.
Captagon is an addictive drug that fuels conflicts in the Middle East4. Fighters on Captagon or Amphetamines may feel a sense of well-being, euphoria and invincibility. It is clear that these drugs are favored to suppress pain and increase aggression in soldiers. The use of these drugs is not limited to soldiers but also to the civilian population in areas of prevailing hopelessness or helplessness. Due to its great profitability, many Captagon seizure cases occur in the Middle East and Europe according to the United Nations Office on Drugs and Crime (UNODC). For example, Al-Imam and co-workers wrote in their article, “11,000 Captagon tablets were seized in the city of Al-Nasiryia, a city to the southeast of the Iraqi capital Baghdad.” And they also mentioned in their article that “illegal Captagon was seized by fighters of the Islamic State in Iraq and other militant groups in Syria (more than 1.4 million tablets), in Turky (107.5 kilograms of drug) and in Egypt (8.8 million pills)”15. Furthermore, in 2013 Dubai authorities seized 4.6 million Captagon pills. In May 2017, French authorities reported confiscating about 300 pounds of Captagon, seizing 750,000 pills worth an estimated $1.7 million.
Although Captagon was first synthesized in 1961, its underlying addictive mechanism, potential drug-drug interactions with Amphetamine and Theophylline, side effects such as hallucination, and detailed interactions between Captagon/Amphetamine/Theophylline and their targets have been relatively ignored by academic research and industrial studies for over 56 years.
To address the above questions, we performed computational systems pharmacology analyses for Captagon and its metabolites, Amphetamine and Theophylline. First, we analyzed the signaling pathways for both Amphetamine and Theophylline. We found that Amphetamine is the main cause for Captagon addiction, and Theophylline can block adenosine receptors in the brain, attenuating Amphetamine’s behavior. Subsequently, we docked Amphetamine/Theophylline into their reported targets to explore their detailed interactions at a molecular level. Finally, to study the side effects of Amphetamine and Captagon, we carried out off-target predictions using our established hallucinogen-related chemogenomics knowledgebase and in-house computational chemogenomics tools. Overall, such drug-drug combinations could potentially be used to treat drug abuse and addiction. Our studies provided a detailed insight into the addiction mechanism of Captagon and its metabolites.
Materials and Methods
Homology Modeling of TAAR1 and Adenosine A2b Receptor
In this study, we built two GPCR models including trace amine-associated receptor 1 (TAAR1) and adenosine A2b receptor (A2bR). First, we used a β1AR crystal structure (PDB ID: 2Y00, Resolution: 2.5 Å)16 as the template to build the homology model of TAAR1 and found the sequence identity between the two proteins is about 35%. Then, a homology model of the adenosine A2b receptor was built using the crystal structure of the adenosine A2a receptor (PDB ID: 3RFM, Resolution: 3.6 Å)17 as the template, finding the sequence identity between the two proteins to be about 66%. These crystal structures were downloaded from the Protein Data Bank (http://www.pdb.org/pdb/). SYBYL-X 1.318 was used to repair all residues and minimize energy.
The full sequences of human TAAR1 receptor (TAAR1_HUMAN, 339 residues) and adenosine A2b receptor (AA2BR_HUMAN or P29275, 332 residues) were retrieved from UniProtKB/Swiss-Prot (http://www.uniprot.org/uniprot/). We then truncated the residues according to their templates for sequence alignment and homology modeling using our reported protocol19. The disulfide bridge(s) was also patched for each homology model.
Modeller 9.1820 was used to construct these homology models. Once the 3D models were generated, SYBYL-X 1.3 was used to perform the energy minimizations. Briefly, the parameters defined in SYBYL are: Gradient set to 0.5 kcal/mol, Max iterations set to 5000, force field set to MMFF94, and charge method set to MMFF94. All of these settings are the same as in our previous works21,22,23,24,25. Then, proSA-web Z-scores26 and PROCHECK Ramachandran plots27 were used to validate the models.
Docking Study of Ligand-Receptors
We used the MOLCAD module implemented in SYBYL-X 1.3 to explore the potential binding pocket for GPCR receptors. The docking program Surflex-Dock GeomX (SFXC) in SYBYL-X 1.3 was applied to construct receptor-ligand complexes in which the docking scores were expressed in −log10 (Kd)28. The main protocols or parameters of docking were addressed in our previous publications21,22,23,24,25. Briefly, the docking parameters used were: (a) number of starting conformations per ligand set to 10, max conformations per fragment set to 20; (b) maximum number of rotatable bonds per molecule set to 100; (c) flags were turned on at pre-dock minimization, post-dock minimization, molecule fragmentation, and soft grid treatment; (d) activate spin alignment method with density of search set to 9.0; and (e) number of spins per alignment set to 12.
Hallucinogen-Specific Chemogenomics Knowledgebase and Systems Pharmacology Analysis
We have constructed a Hallucinogen-Specific Chemogenomics Database29 that can be used for target, off-target, or additional identification and network systems pharmacology analysis of small molecules and their potential targets. Several in-house chemoinformatics tools were also used, including TargetHunter, HTDocking, Blood-Brain Barrier (BBB) Predictor, and more30,31. HallucinogenPlatform (http://www.cbligand.org/hallucinogen/) collected 144 hallucinogen-related target proteins and 145 chemical compounds associated with these targets in 6,721 assays and 23,598 references.
In our work, we applied our HallucinogenPlatform and established chemoinformatics tools such as HTDocking to perform network systems pharmacological analysis on Amphetamine and Captagon. First, Amphetamine and Captagon were docked into the target proteins’ pockets. We then matched these predicted target proteins to Amphetamine and Captagon according to their docking scores. Targets with higher docking scores may have higher binding affinities and therefore a greater chance of interacting with Amphetamine/Captagon. Next, we mapped out a pharmacological network of interactions between drug compounds and target proteins at the molecular level32,33. Cytoscape 3.4.034 was used to generate, analyze and visualize the network of targets and drugs/compounds as described previously33.
Molecular Dynamics (MD) Simulation
Two systems were set up for molecular dynamics simulations using the web-based tool CHARMM-GUI35,36. For the TAAR1 receptor complexed with Amphetamine, the system included 256 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) lipids, 22550 water molecules, and 60 Na+ ions and 69 Cl− ions. For the A2aR receptor complexed with Theophylline, the system included 254 POPC lipids, 22427 water molecules, 60 Na+ ions, and 67 Cl− ions. The initial configurations of protein receptors and ligands were taken from docking studies. The sizes of the initial simulation boxes were ~100 Å * 100 Å * 117 Å.
The AMBER ff14SB force field37 was applied to proteins and the AMBER Lipid14 force field38 was applied to lipids. Water molecules were treated with the TIP3P water model39. The partial atomic charges of ligands were derived by the restrained electrostatic potential (RESP) method40 to fit the HF/6–31 G* electrostatic potentials generated using the GAUSSIAN 16 software package41. Other force field parameters came from GAFF in AMBER1642. Residue topologies for ligands were prepared using the ANTECHAMBER module43.
The MD simulations were carried out using the PMEMD.mpi and PMEMD.cuda modules in the AMBER1644,45,46 package. First, several minimization steps were carried out for the systems to avoid possible steric crashes. Then, each system was gradually heated from 0 K to 300 K during the heating stage and kept at 300 K during the following equilibrium and production stages. A time step of 1 fs was used for the heating and first part of the equilibrium stage, and 2 fs was used for the remaining part of the equilibrium stage and the entire production stage. A periodic boundary condition was employed to maintain constant temperature and pressure (NPT) ensembles. Pressure was set at 1 atm and controlled by the anisotropic (x−, y−, z−) pressure scaling protocol applied in AMBER with a pressure relaxation time of 1 ps. Temperature was regulated using Langevin dynamics with a collision frequency of 2 ps−1 47,48. The Particle Mesh Ewald (PME) method49,50 was adopted to handle long-range electrostatics and a 10 Å cutoff was set to treat real-space interactions. All covalent bonds involving hydrogen atoms were constrained with the SHAKE algorithm51. The simulations time for the production stage for each system was 200 ns and the coordinates of simulated systems were saved every 100 ps.
For the saved trajectories of MD simulations, Molecular Mechanics/Generalized Born Surface Area (MM/GBSA) binding free energies were calculated52,53 and free energy decompositions were performed. The interaction energies between each residue and ligand were extracted.
Signaling Pathway Analysis for Amphetamine, a Stimulant with Addiction
Amphetamine, a dopamine inducer, has been reported to not only provide a sense of euphoria but also to contribute to addiction. Addiction to Amphetamine mainly arises from the increased release of dopamine in “the mesoaccumbens dopamine pathway, extending from the ventral tegmental area (VTA) of the midbrain to the nucleus accumbens (NAc)”54. NAc, also known as the ventral striatum (VS), has been identified to be “the critical shared substrate for the reinforcing effect of Amphetamines, cocaine, and other addictive drugs”55.
Amphetamine is reported to be an inhibitor of vesicular monoamine transporter 2 (VMAT2)56, but an agonist of Trace amine-associated receptor 1 (TAAR1)57. As shown in Fig. 2, VMAT258 trans-membrane transporter mainly delivers neurotransmitters from the cellular cytosol into synaptic vesicles59. TAAR1, a class A GPCR57,60, is primarily located in peripheral tissues61, glial cells62, and neurons63 and is responsible for regulating neurotransmission in dopamine, norepinephrine, and serotonin neurons in the CNS64. On one hand, the binding of Amphetamine to VMAT2 increases the release of dopamine from vesicles63,65. On the other hand, TAAR1 can reverse the dopamine transporter (DAT) function by activating protein kinase A (PKA) and protein kinase C (PKC), allowing DAT to transport dopamine into the intracellular region and therefore the synaptic cleft63,66. This activates the D1 dopamine receptor (D1R) on the post-synaptic neuron. D1R, a D1-like GPCR, can bind to G-proteins and increase the cAMP level, phosphorylating cAMP-response element binding protein 1 (CREB1) and increasing ΔFosB levels67,68. ΔFosB, a master control protein, governs the activity of several other transcriptional (e.g. AP-1) and epigenetic regulatory proteins and represses the c-Fos gene. c-Fos repression acts as a molecular switch that enables the accumulation of ΔFosB in the neuron. As a result, ∆FosB will persistently overexpress and accumulate in the D1-type medium spiny neurons that exist in the nucleus accumbens following repeated high-dose exposures to Amphetamines, causing the addiction.
Signaling Pathway Analysis for Theophylline, an Antagonist of Adenosine A2a Receptor for Drug Abuse
Theophylline, a weak stimulator chemically and pharmacologically resembling caffeine6, can inhibit phosphodiesterase (causing an increase in cAMP levels) and block adenosine receptors.
As shown in Fig. 3, A2aR is a GPCR whose preferred endogenous agonist is adenosine69. A2aR is an attractive therapeutic target for cardiac function70, insomnia71, inflammatory response72, pain71, Parkinson’s disease70,73, and drug addiction71,74,75,76,77. We found that blockage of A2aR by Theophylline inhibits CREB1 phosphorylation. Some literature75,76 has reported that “the inactivation/antagonists of adenosine A2aR selectively attenuates Amphetamine-induced behavioral sensitization,” implying that blockage of A2aR by Theophylline will attenuate the addiction of Amphetamine and Captagon. Jiang-Fan Chen et al. administrated the same dose Amphetamine in both WT and A2aR KO mice for 7 days, and found that the enhancement of significant locomotion induced by amphetamine was shown in the WT rats, in contrast, not in the A2aR KO mice75. Elena Bastia et al. used A2aR knockout (A2aR KO) mice model, showing that pairing daily amphetamine doses with their test A2aR antagonist (SCH58261 or KW-6002) prevented locomotor sensitization on day 8 in mice76. Although no direct pathway showed that Theophylline affects the pathway of Amphetamine addiction, it may affect the accumulation of ∆FosB (dash line in Fig. 3, our hypothesis).
Interactions of Amphetamine with TAAR1
Amphetamine directly targets TAAR1 as an agonist. We explored in detail the interactions between TAAR1 and Amphetamine. No TAAR1 crystal structure has been reported, so we built the homology model using β1AR as a template (PDB ID: 2Y00, Resolution: 2.5 Å)16, with a sequence identity of about 35%. Then the model was further validated by energy minimization, proSA-web Z-scores26, and PROCHECK Ramachandran plots27.
As shown in Fig. 4, our results showed that the potential binding pocket in TAAR1 is formed by TM3, TM5, TM6 and TM7. Several important residues formed strong hydrophobic interactions with Amphetamine, including Ile104(3.33) in TM3, Thr194(5.43) in TM5, and Phe268(6.52) in TM6, with a bond distance of 3.6 Å. Asp103(3.32) in TM3 and Tyr294(7.42) in TM7 formed strong hydrogen bonds with Amphetamine, with distances of about 2.1 Å and 3.2 Å, respectively. Interestingly, we found an additional hydrogen bond between Asp103 and Tyr294, making the previous two hydrogen bonds more stable. All of these residues in our work are supported by modeling data reported by Tan et al.78 and docking data calculated by Cichero et al.79. For further validation of the binding mode, we also carried out a molecular dynamics (MD) simulation as described below.
Interactions of Theophylline with A2aR and A2bR
Using the same protocol, we studied interactions between A2aR (and A2bR) and Theophylline at a molecular level. The crystal structure of A2aR (PDB ID: 3RFM, Resolution: 3.6 Å)17 was used to construct the homology model of A2bR, with a sequence identity of about 66%.
We first docked the reported compound caffeine back to the crystal structure of A2aR as a validation of our docking protocol as shown in Supporting Figure S1a. The binding pocket of A2aR is formed by Helices I, II, III, V, VI, and VII, and surrounded by important residues including Asn253 (hydrophilic), His278 (hydrophilic), Val84 (hydrophobic), Phe168 (hydrophobic), Met177 (hydrophobic), Leu249 (hydrophobic), Met270 (hydrophobic, not shown), and Ile274(hydrophobic). We found that Asn253 formed strong hydrogen bonding (3.3 Å) with caffeine, indicating this residue is very important for ligand recognition. Moreover, six other residues including Val84 (3.8 Å), Phe168 (3.6 Å), Met177 (3.3 Å), Leu249 (3.8 Å), Met270 (3.9 Å, not shown), and Ile274 (4.1 Å) formed strong hydrophobic interactions with caffeine. Moreover, our docking results showed that the docked caffeine (yellow sticks) overlapped well with the crystallized compound (cyan) with a root mean square deviation (RMSD) of 0.3 Å, indicating our docking protocol is reliable.
We then docked Theophylline into A2aR using the same protocol and compared the binding mode of Theophylline (salmon) with that of caffeine (cyan) with their co-crystal structures, as shown in Fig. 5. Our docking data showed that Theophylline (salmon) overlapped with caffeine (cyan) very well, with an RMSD of 0.15 Å. Moreover, these two compounds shared almost the same interactions. Briefly, Asn253 (2.8 Å) formed hydrogen bonding interactions with Theophylline (salmon), while Val84, Phe168, Met177, Leu249, Met270, and Ile274 interacted with Theophylline through strong hydrophobic interactions.
For comparison, we also docked caffeine and Theophylline into A2bR as shown in Supporting Figure S1b. Our docking results showed that both caffeine and Theophylline shared similar binding modes and interactions with these two receptors.
Molecular Dynamics (MD) Simulation of Amphetamine and Theophylline with Target Receptors
To study the dynamics and mechanisms of protein-ligand binding between Amphetamine, Theophylline, and related receptors, we carried out 200 ns-MD simulations for the TAAR1-Amphetamine complex and the A2aR-Theophylline system.
Detailed interactions after the MD simulations and the RMSDs of the receptors and ligands compared to their initial configurations were plotted in Fig. 6. Our MD results showed that the RMSDs of both protein receptors were small (around 2.5 Å) during 200 ns simulations, indicating the systems were stable and reasonable. The RMSDs of both ligands during the MD simulations were more stable than that of the receptors (less than 1 Å), demonstrating that our docking results and poses of the ligands were reliable. Next, Fig. 6a shows the detailed interactions between Amphetamine and TAAR1. The data showed that the hydrogen bond between Amphetamine and Asp103 was stable (1.8 Å). Importantly, an additional hydrogen bond was formed between Amphetamine and Ser107 (2.2 Å). Moreover, Amphetamine interacted with Val184 in the ECL2 (The Extracellular Loop 2) domain and Met158 in TM4, forming strong hydrophobic interactions. Finally, Fig. 6b shows that Theophylline had a slight rotation during the MD simulation, forming two strong hydrogen bonds with Asn253 (2.7 Å/3.0 Å). The hydrophobic interactions kept stable during the simulations.
We further characterized the protein-ligand binding by calculating the interaction energy (IE) between the ligand and each residue of the receptor using MM-GB/SA free energy decomposition analysis, as shown in Table 1. We found that the MD results correlated very well with our previously described docking studies. For example, Asp103 (−5.46 kcal/mol, red in Fig. 6a), Ile104 (−4.19 kcal/mol, red in Fig. 6a), and Val184 (−5.46 kcal/mol, light pink in Fig. 6a) in TAAR1 contributed greatly to the binding of Amphetamine, while Asn253 (−6.24 kcal/mol, red in Fig. 6b), Phe168 (−6.02 kcal/mol, red in Fig. 6b), and Leu249 (−3.00 kcal/mol, white in Fig. 6b) in A2aR played important roles in the recognition of Theophylline.
Drug-Drug Interactions between Amphetamine and Theophylline
Some literature reported that Theophylline is the substrate of CYP2D680,81, an enzyme involved in many metabolic processes. Moreover, Amphetamine is reported to be a substrate and inhibitor of CYP2D681,82. As a result, Theophylline will be eliminated more slowly due to competition and/or blockage of CYP2D6 by Amphetamine. Thus, some side effects of Theophylline like nausea, diarrhea, and an increase in heart rate may last longer if its plasma concentration is high enough. Because Theophylline only constitutes 13.7% of the oral dose of Captagon, we suggest that the risk of overdose due to Theophylline is low. Importantly, blockage of A2aR by Theophylline should last longer due to this prolonged elimination time. In turn, the effects of Theophylline on decreasing Amphetamine addiction will last longer than in a scenario without this competitive elimination.
Knowledgebase-guided Off-target Predictions for Amphetamine/Captagon and Associated Side Effects
Previously, we constructed a Hallucinogen-Specific Chemogenomics Knowledgebase29 that can be used for target, off target, or additional identification and network systems pharmacology analysis of small molecules and their potential targets. In our current work, we adapted a small but specific dataset with twelve 5-HT receptors in our knowledgebase to explore the side effects (e.g. hallucinations) for both Amphetamine and Captagon. All of these 5-HT or serotonin receptors are GPCRs widely distributed throughout the CNS, reportedly associated with depression, anxiety, and drug addiction83. Here, we docked both Amphetamine and Captagon into 5-HT receptors to predict potential off-targets.
As shown in Fig. 7, additional potential Amphetamine target proteins were predicted and ranked by their docking scores (shown in Table 2) in the form of an interaction network (green one-known target, purple one-predicted target). Our results showed that most of the docking scores were not high due to the huge size of the binding pocket in the receptor and the small size of the Amphetamine molecule. However, our predicted results correlated well with the known therapeutic targets of Amphetamine. For example, the docking scores of Amphetamines with 5-HT2a and 5-HT2c were 4.97 and 5.14, which is well correlated with their experimental data. The docking score of Amphetamine with 5-HT1a is 5.12 with experimental data of 7.66 µM. Importantly, we predicted two potential targets for Amphetamine, including 5-HT1f (docking score of 5.26) and 5-HT1e (docking score of 6.14), as shown in Table 2.
In terms of Captagon lipophilicity6,14, we also calculated the docking scores of Captagon before metabolism within these 5-HT receptors, as shown in Table 2. All the docking scores of target proteins for Captagon were higher than those of Amphetamine, due to the latter’s more flexible structure and stronger interactions with the receptor. In our present work, we showed the detailed interaction of Captagon with its most probable target, 5-HT7. For comparison, we aligned the predicted binding mode of Amphetamine (purple) and Captagon (slate) with 5-HT7, as shown in Fig. 8. Our docking data showed that Amphetamine and Captagon overlapped very well and that Asp162 (D3.32) formed strong hydrogen bonds with ligands. For further validation, we will carry out future experiments.
Insight into Captagon
The data analyses are summarized: (1) Captagon when orally dosed will be metabolized into 24.5% Amphetamine and 13.7% Theophylline; (2) Captagon is more lipophilic than both Theophylline and Amphetamine, resulting in easier absorption into the CNS; (3) Amphetamine, a TAAR1 agonist that enhances dopamine signaling (causing increased irritability, aggression, etc.), is the main cause of Captagon addiction; (4) Theophylline, an A2aR antagonist that blocks adenosine receptors in the brain (causing restlessness and painlessness), may attenuate the behavioral sensitization caused by Amphetamine; (5) the drug-drug interactions between Amphetamine and Theophylline slow the metabolism and elimination of Theophylline through competition and/or blockage of CYP2D6 by Amphetamine; and, (6) Theophylline and Amphetamine act synergistically to augment Captagon’s psychoactive effects beyond those caused by Amphetamines alone.
Moreover, A2aR reportedly has antagonistic interactions with the D2 receptor through heterodimer formation84. Poleszak E et al. reported that Adenosine A2a receptor antagonist, DMPX (3 and 6 mg/kg ip) can attenuate the amphetamine-induced stereotypy in male Wistar rats85,86. This complex exists in neurons found in the nucleus accumbens and ventral and dorsal striatopallidal GABAergic neurons, reportedly the main areas involved in addiction87. Theophylline, a known A2aR antagonist specifically targets the A2aR protomer, increasing the affinity of dopamine for the D2 receptor and facilitating D2-mediated Gi/o signaling. D2 neurons are associated with suppressing addictive drug rewards, opposite to D1-expressing neurons reinforcing rewards88. Thus, the A2aR-D2 receptor heterodimer theory helps to explain the lessened addictiveness of Captagon versus Amphetamine.
Furthermore, counterfeit Captagon may contain other types of Amphetamines (3,4-methylenedioxymethAmphetamine-MDMA, 3,4-MethylenedioxymethAmphetamine-ecstasy).
On one hand, Amphetamine, the main Captagon metabolite, is an addictive CNS stimulator. On the other hand, Theophylline, another Captagon metabolite that blocks A2aR in the brain, may attenuate the behavioral sensitization caused by Amphetamine. These metabolites make Captagon more potent but less addictive than Amphetamine. The further experiment is on-going, and the data will be published elsewhere.
In our present work, we systematically analyzed the addiction mechanism of Captagon and its metabolites, Amphetamine and Theophylline, using our established drug abused chemogenomics knowledgebase systems pharmacology methods. Our data mining analysis of signaling pathways for both Amphetamine and Theophylline, reveals that Amphetamine is the main cause of Captagon addiction and Theophylline can attenuate this behavioral sensitization. In addition, we explored the detailed interactions of Amphetamine and Theophylline with their reported targets using molecular docking studies and molecular dynamics simulation studies. Furthermore, using our established hallucinogen-related chemogenomics knowledgebase and in-house computational chemogenomics tools, we studied the side effects of Amphetamine and Captagon through off-target prediction. Overall, such potential drug-drug combinations are a promising method for developing novel medications to treat drug abuse and addiction.
Kristen, G., Schaefer, A. & Von Schlichtegroll, A. Fenetylline: therapeutic use, misuse and/or abuse. Drug and alcohol dependence 17, 259–271 (1986).
Wenthur, C. J., Zhou, B. & Janda, K. D. Vaccine-driven pharmacodynamic dissection and mitigation of fenethylline psychoactivity. Nature 548, 476–479 (2017).
Alabdalla, M. A. Chemical characterization of counterfeit captagon tablets seized in Jordan. Forensic science international 152, 185–188 (2005).
Van Hout, M. C. & Wells, J. Is Captagon (fenethylline) helping to fuel the Syrian conflict? Addiction 111, 748–749 (2016).
Al-Hemiary, N. J., Al-Diwan, J. K., Hasson, A. L. & Rawson, R. A. Drug and alcohol use in Iraq: findings of the inaugural Iraqi Community Epidemiological Workgroup. Substance use & misuse 49, 1759–1763 (2014).
Katselou, M. et al. Fenethylline (Captagon) Abuse–Local Problems from an Old Drug Become Universal. Basic & clinical pharmacology & toxicology 119, 133–140 (2016).
Heal, D. J., Smith, S. L., Gosden, J. & Nutt, D. J. Amphetamine, past and present–a pharmacological and clinical perspective. Journal of Psychopharmacology 27, 479–496 (2013).
Dowd, C. S. et al. 1-[4-(3-Phenylalkyl) phenyl]-2-aminopropanes as 5-HT2A partial agonists. Journal of medicinal chemistry 43, 3074–3084 (2000).
Glennon, R. A., McKenney, J., Lyon, R. A. & Titeler, M. 5-HT1 and 5-HT2 binding characteristics of 1-(2, 5-dimethoxy-4-bromophenyl)-2-aminopropane analogs. Journal of medicinal chemistry 29, 194–199 (1986).
Glennon, R. A., Raghupathi, R., Bartyzel, P., Teitler, M. & Leonhardt, S. Binding of phenylalkylamine derivatives at 5-HT1C and 5-HT2 serotonin receptors: evidence for a lack of selectivity. Journal of medicinal chemistry 35, 734–740 (1992).
Vilches-Herrera, M. et al. Naphthylisopropylamine and N-benzylamphetamine derivatives as monoamine oxidase inhibitors. Bioorganic & medicinal chemistry 17, 2452–2460 (2009).
Sun, H.-w, Qiao, F.-x & Liu, G.-y Characteristic of theophylline imprinted monolithic column and its application for determination of xanthine derivatives caffeine and theophylline in green tea. Journal of Chromatography A 1134, 194–200 (2006).
Shannon, M. Predictors of major toxicity after theophylline overdose. Annals of internal medicine 119, 1161–1167 (1993).
Nickel, B., Niebch, G., Peter, G., Von Schlichtegroll, A. & Tibes, U. Fenetylline: new results on pharmacology, metabolism and kinetics. Drug and alcohol dependence 17, 235–257 (1986).
Al‐imam, A. et al. Captagon: use and trade in the Middle East. Human Psychopharmacology: Clinical and Experimental 32, e2548 (2017).
Warne, T. et al. The structural basis for agonist and partial agonist action on a β1-adrenergic receptor. Nature 469, 241 (2011).
Doré, A. S. et al. Structure of the adenosine A 2A receptor in complex with ZM241385 and the xanthines XAC and caffeine. Structure 19, 1283–1293 (2011).
SYBYL-X 1.3, Tripos International, 1699 South Hanley Rd., St. Louis, Missouri, 63144, USA (2010).
Xie, X. Q., Chen, J. Z. & Billings, E. M. 3D structural model of the G-protein-coupled cannabinoid CB2 receptor. Proteins: Structure, Function, and Bioinformatics 53, 307–319 (2003).
Martí-Renom, M. A. et al. Comparative protein structure modeling of genes and genomes. Annu. Rev. Bioph. Biom. 29, 291–325 (2000).
Chen, J.-Z., Wang, J. & Xie, X.-Q. GPCR structure-based virtual screening approach for CB2 antagonist search. J. Chem. Inf. Model. 47, 1626–1637 (2007).
Feng, Z. et al. Modeling, Molecular Dynamics Simulation, and Mutation Validation for Structure of Cannabinoid Receptor 2 Based on Known Crystal Structures of GPCRs. J. Chem. Inf. Model. 54, 2483–2499 (2014).
Feng, Z. et al. Design and activity of AP endonuclease-1 inhibitors. J. Chem. Biol. 8, 79–93 (2015).
Feng, Z. et al. Structural Insight into Tetrameric hTRPV1 from Homology Modeling, Molecular Docking, Molecular Dynamics Simulation, Virtual Screening and Bioassay Validations. J. Chem. Inf. Model. 54, 2483–2499 (2015).
Feng, Z. et al. Structural Insight into Tetrameric hTRPV1 from Homology Modeling, Molecular Docking, Molecular Dynamics Simulation, Virtual Screening and Bioassay Validations. J. Chem. Inf. Model. (2015).
Wiederstein, M. & Sippl, M. J. ProSA-web: interactive web service for the recognition of errors in three-dimensional structures of proteins. Nucleic Acids Res. 35, W407–W410 (2007).
Laskowski, R. A., MacArthur, M. W., Moss, D. S. & Thornton, J. M. PROCHECK: a program to check the stereochemical quality of protein structures. J. Appl. Crystallogr. 26, 283–291 (1993).
Jain, A. N. Scoring noncovalent protein-ligand interactions: a continuous differentiable function tuned to compute binding affinities. J. Comput. Aided-Mol. Des. 10, 427–440 (1996).
Xu, X. et al. Chemogenomics knowledgebase and systems pharmacology for hallucinogen target identification—Salvinorin A as a case study. Journal of Molecular Graphics and Modelling 70, 284–295 (2016).
Wang, L. et al. TargetHunter: an in silico target identification tool for predicting therapeutic potential of small organic molecules based on chemogenomic database. The AAPS journal 15, 395–406 (2013).
Ma, C., Wang, L. & Xie, X.-Q. GPU accelerated chemical similarity calculation for compound library comparison. Journal of chemical information and modeling 51, 1521–1527 (2011).
Xue, Y. et al. The efficacy and safety of cilostazol as an alternative to aspirin in Chinese patients with aspirin intolerance after coronary stent implantation: a combined clinical study and computational system pharmacology analysis. Acta Pharmacologica Sinica (2017).
Zhang, H. et al. Cardiovascular Disease Chemogenomics Knowledgebase-guided Target Identification and Drug Synergy Mechanism Study of an Herbal Formula. Scientific reports 6 (2016).
Shannon, P. et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome research 13, 2498–2504 (2003).
Jo, S., Kim, T., Iyer, V. G. & Im, W. CHARMM-GUI: a web-based graphical user interface for CHARMM. Journal of computational chemistry 29, 1859–1865 (2008).
Wu, E. L. et al. CHARMM-GUI Membrane Builder toward realistic biological membrane simulations. Journal of computational chemistry 35, 1997–2004 (2014).
Maier, J. A. et al. ff14SB: improving the accuracy of protein side chain and backbone parameters from ff99SB. Journal of chemical theory and computation 11, 3696–3713 (2015).
Dickson, C. J. et al. Lipid14: the amber lipid force field. Journal of chemical theory and computation 10, 865–879 (2014).
Jorgensen, W. L., Chandrasekhar, J., Madura, J. D., Impey, R. W. & Klein, M. L. Comparison of simple potential functions for simulating liquid water. The Journal of chemical physics 79, 926–935 (1983).
Bayly, C. I., Cieplak, P., Cornell, W. & Kollman, P. A. A well-behaved electrostatic potential based method using charge restraints for deriving atomic charges: the RESP model. The Journal of Physical Chemistry 97, 10269–10280 (1993).
Frisch, M. et al. Gaussian Inc 16, revision A. 03; Gaussian Inc. Wallingford, CT (2016).
Wang, J., Wolf, R. M., Caldwell, J. W., Kollman, P. A. & Case, D. A. Development and testing of a general amber force field. Journal of computational chemistry 25, 1157–1174 (2004).
Wang, J., Wang, W., Kollman, P. A. & Case, D. A. Automatic atom type and bond type perception in molecular mechanical calculations. Journal of molecular graphics and modelling 25, 247–260 (2006).
Götz, A. W. et al. Routine microsecond molecular dynamics simulations with AMBER on GPUs. 1. Generalized born. Journal of chemical theory and computation 8, 1542–1555 (2012).
Salomon-Ferrer, R., Götz, A. W., Poole, D., Le Grand, S. & Walker, R. C. Routine microsecond molecular dynamics simulations with AMBER on GPUs. 2. Explicit solvent particle mesh Ewald. Journal of chemical theory and computation 9, 3878–3888 (2013).
Case, D. et al. (San Francisco, 2016).
Loncharich, R. J., Brooks, B. R. & Pastor, R. W. Langevin dynamics of peptides: The frictional dependence of isomerization rates of N‐acetylalanyl‐N′‐methylamide. Biopolymers 32, 523–535 (1992).
Izaguirre, J. A., Catarello, D. P., Wozniak, J. M. & Skeel, R. D. Langevin stabilization of molecular dynamics. The Journal of chemical physics 114, 2090–2098 (2001).
Darden, T., York, D. & Pedersen, L. Particle mesh Ewald: An N⋅ log (N) method for Ewald sums in large systems. The Journal of chemical physics 98, 10089–10092 (1993).
Essmann, U. et al. A smooth particle mesh Ewald method. The Journal of chemical physics 103, 8577–8593 (1995).
Ryckaert, J.-P., Ciccotti, G. & Berendsen, H. J. Numerical integration of the cartesian equations of motion of a system with constraints: molecular dynamics of n-alkanes. Journal of Computational Physics 23, 327–341 (1977).
Wang, J. & Hou, T. Develop and test a solvent accessible surface area-based model in conformational entropy calculations. Journal of chemical information and modeling 52, 1199–1212 (2012).
Hawkins, G. D., Cramer, C. J. & Truhlar, D. G. Parametrized models of aqueous free energies of solvation based on pairwise descreening of solute atomic charges from a dielectric medium. The Journal of Physical Chemistry 100, 19824–19839 (1996).
Ikemoto, S. Brain reward circuitry beyond the mesolimbic dopamine system: a neurobiological theory. Neuroscience & biobehavioral reviews 35, 129–150 (2010).
Pierce, R. C. & Kumaresan, V. The mesolimbic dopamine system: the final common pathway for the reinforcing effect of drugs of abuse? Neuroscience & biobehavioral reviews 30, 215–238 (2006).
Sulzer, D., Sonders, M. S., Poulsen, N. W. & Galli, A. Mechanisms of neurotransmitter release by amphetamines: a review. Progress in neurobiology 75, 406–433 (2005).
Bunzow, J. R. et al. Amphetamine, 3, 4-methylenedioxymethamphetamine, lysergic acid diethylamide, and metabolites of the catecholamine neurotransmitters are agonists of a rat trace amine receptor. Molecular pharmacology 60, 1181–1188 (2001).
Surratt, C. K. et al. A human synaptic vesicle monoamine transporter cDNA predicts posttranslational modifications, reveals chromosome 10 gene localization and identifies TaqI RFLPs. FEBS letters 318, 325–330 (1993).
Eiden, L. E., Schäfer, M. K.-H., Weihe, E. & Schütz, B. The vesicular amine transporter family (SLC18): amine/proton antiporters required for vesicular accumulation and regulated exocytotic secretion of monoamines and acetylcholine. Pflügers Archiv 447, 636–640 (2004).
Borowsky, B. et al. Trace amines: identification of a family of mammalian G protein-coupled receptors. Proceedings of the National Academy of Sciences 98, 8966–8971 (2001).
Grandy, D. K. Trace amine-associated receptor 1—Family archetype or iconoclast? Pharmacology & therapeutics 116, 355–390 (2007).
Cisneros, I. E. & Ghorpade, A. Methamphetamine and HIV-1-induced neurotoxicity: role of trace amine associated receptor 1 cAMP signaling in astrocytes. Neuropharmacology 85, 499–507 (2014).
Miller, G. M. The emerging role of trace amine‐associated receptor 1 in the functional regulation of monoamine transporters and dopaminergic activity. Journal of neurochemistry 116, 164–176 (2011).
Jing, L. & Li, J.-X. Trace amine-associated receptor 1: a promising target for the treatment of psychostimulant addiction. European journal of pharmacology 761, 345–352 (2015).
Eiden, L. E. & Weihe, E. VMAT2: a dynamic regulator of brain monoaminergic neuronal function interacting with drugs of abuse. Annals of the New York Academy of Sciences 1216, 86–98 (2011).
Ledonne, A. et al. Electrophysiological effects of trace amines on mesencephalic dopaminergic neurons. Frontiers in systems neuroscience 5, 56 (2011).
Renthal, W. & Nestler, E. J. Chromatin regulation in drug addiction and depression. Dialogues in clinical neuroscience 11, 257 (2009).
Robison, A. J. & Nestler, E. J. Transcriptional and epigenetic mechanisms of addiction. Nature reviews neuroscience 12, 623 (2011).
Fredholm, B. B., IJzerman, A. P., Jacobson, K. A., Klotz, K.-N. & Linden, J. International Union of Pharmacology. XXV. Nomenclature and classification of adenosine receptors. Pharmacological reviews 53, 527–552 (2001).
Sun, B. et al. Crystal structure of the adenosine A2A receptor bound to an antagonist reveals a potential allosteric pocket. Proceedings of the National Academy of Sciences, 201621423 (2017).
Ferré, S. et al. Adenosine A 2A receptors in ventral striatum, hypothalamus and nociceptive circuitry: implications for drug addiction, sleep and pain. Progress in neurobiology 83, 332–347 (2007).
Ohta, A. & Sitkovsky, M. Role of G-protein-coupled adenosine receptors in downregulation of inflammation and protection from tissue damage. Nature 414, 916–920 (2001).
Morelli, M. et al. Role of adenosine A 2A receptors in parkinsonian motor impairment and l-DOPA-induced motor complications. Progress in neurobiology 83, 293–309 (2007).
Hack, S. P. & Christie, M. J. Adaptations in adenosine signaling in drug dependence: therapeutic implications. Critical Reviews™ in Neurobiology 15 (2003).
Jiang-Fan, C. et al. Inactivation of adenosine A2A receptors selectively attenuates amphetamine-induced behavioral sensitization. Neuropsychopharmacology 28, 1086 (2003).
Bastia, E. et al. A crucial role for forebrain adenosine A2A receptors in amphetamine sensitization. Neuropsychopharmacology 30, 891 (2005).
Cunha, R. A., Ferré, S., Vaugeois, J.-M. & Chen, J.-F. Potential therapeutic interest of adenosine A2A receptors in psychiatric disorders. Current pharmaceutical design 14, 1512–1524 (2008).
Tan, E. S. et al. The molecular basis of species-specific ligand activation of trace amine-associated receptor 1 (TAAR1). ACS chemical biology 4, 209 (2009).
Cichero, E. & Tonelli, M. New insights into the structure of the trace amine‐associated receptor 2: Homology modelling studies exploring the binding mode of 3‐iodothyronamine. Chemical biology & drug design 89, 790–796 (2017).
Ha, H. R., Chen, J., Freiburghaus, A. & Follath, F. Metabolism of theophylline by cDNA‐expressed human cytochromes P‐450. British journal of clinical pharmacology 39, 321–326 (1995).
Preissner, S. et al. SuperCYP: a comprehensive database on Cytochrome P450 enzymes including a tool for analysis of CYP-drug interactions. Nucleic acids research 38, D237–D243 (2009).
Dring, L. G., Smith, R. & Williams, R. The metabolic fate of amphetamine in man and other species. Biochemical Journal 116, 425–435 (1970).
Hoyer, D. et al. International Union of Pharmacology classification of receptors for 5-hydroxytryptamine (Serotonin). Pharmacological reviews 46, 157–203 (1994).
Fuxe, K. et al. Adenosine A2A and dopamine D2 heteromeric receptor complexes and their function. Journal of molecular neuroscience 26, 209 (2005).
Poleszak, E. & Malec, D. Influence of adenosine receptor agonists and antagonists on amphetamine-induced stereotypy in rats. Polish journal of pharmacology 52, 423–429 (2000).
Chen, J.-F. et al. Inactivation of adenosine A 2A receptors selectively attenuates amphetamine-induced behavioral sensitization. Neuropsychopharmacology 28, 1086 (2003).
Mach, R. H. In Seminars in Nuclear Medicine. 524–535 (Elsevier).
Lobo, M. K. et al. Cell type–specific loss of BDNF signaling mimics optogenetic control of cocaine reward. Science 330, 385–390 (2010).
Authors would like to acknowledge the funding support to the Xie laboratory from the NIH NIDA (P30 DA035778A1), NIH (R01 DA025612) and the Department of Defense (W81XWH-16-1-0490), the funding support to J.M. Wang laboratory from the National Institutes of Health of USA (R01-GM079383, R21-GM097617). We give thanks to Dr. Rao Rapaka from NIDA for his enlightenment of this project. Computational support from the Center for Research Computing of University of Pittsburgh, and the Extreme Science and Engineering Discovery Environment (CHE090098, MCB170099), is acknowledged.
The authors declare no competing interests.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
About this article
Cite this article
Wu, N., Feng, Z., He, X. et al. Insight of Captagon Abuse by Chemogenomics Knowledgebase-guided Systems Pharmacology Target Mapping Analyses. Sci Rep 9, 2268 (2019). https://doi.org/10.1038/s41598-018-35449-6
ACS Chemical Neuroscience (2020)
Briefings in Bioinformatics (2020)