In silico and in vitro investigations of antihelmintic activities of selected approved drugs

Objective: We investigated the binding a ﬃ nities of some approved drugs to Ascaris suum Mitochondrial Rhodoquinol Fumarate Reductase (MRFR), an essential enzyme for ascaris survival, and the possibility of repurposing these drugs as antihelmintic agents using in silico molecular docking and in vitro paralysis and mortality times of ﬁ fteen selected front runners. Method: Two hundred approved drugs were selected from ZINC® database based on bioactivity scores while MRFR (PDB code, 3vra) was obtained from the Protein Data Bank (PDB). Both were prepared using AutoDock tools v.1.5.6 and Chimera v.1.9.The docking protocol was validated by computationally reproducing the binding of atpenin to MRFR. The selected approved drugs and the receptor were docked using AutoDockVina v. 4.0. The docking results were analyzed using PyMoL v. 1.4.1.The paralysis and mortality times of the identi ﬁ ed frontrunners against Pheretima posthuma were determined in vitro and synergistic testings were done by the checkerboard method. Result: Fifteen drugs had binding free energies between -7.825 to -11.025 kcal/mol while four of these drugs (me ﬂ oquine, doxycycline, mepacrine and proguanil) emerged as major frontrunners by both in silico and in vitro assessments. The paralysis and mortality times of the four drugs were between 0.33-0.50 hr as against 1.80-2.36 hr for albendazole. They were therefore predicted to have ability to affect MRFR in the same manner as atpenin hence, suggestive of potential antihelmintic activity. Conclusion: The antihelmintic potentials of me ﬂ oquine, doxycycline, proguanil and mepacrine have been demonstrated. In vivo investigation of these frontrunner drugs is strongly recommended.


Introduction
The fortuitous nature, enormous cost and huge time invested in traditional drug discovery and development is a major factor that has fueled the inertia in most pharmaceutical companies to engage in research into new chemical entities. This is a consequence of the unpredictability of these researches hence results often do not justify the effort. This scenario has mostly affected the neglected tropical and orphan disease domains. It is so due to the limited number of sufferers and the demography that stratifi es them to the resource-poor nations [1]. As a fall out of this, pharmaceutical majors located in regions with the technology and wherewithal do not fi nd it attractive to direct research into these areas due to marginal profi t prospects.
Helminthiasis, a major Neglected Tropical Disease (NTD), has long been identifi ed by the World Health Organization (WHO) as a disease with very high morbidity rate and cognitive defi cit especially in school-aged children [2,3]. The prevalence is mostly confi ned to the tropical and subtropical regions where there still exists infrastructural defi cit; poor sanitation; use of untreated fecal matter as fertilizers and subsisting bare soil defecation [4]. In addition, no new antihelmintic agents have been added to the list in the past decade with gradual development of resistance to the standard treatments being reported [2,5].
Drug repurposing has recorded major milestones in several treatment landscapes hence this technique has been considered a veritable tool to achieve this objective in antihelmintic sphere.
It is a well-known fact that observed pharmacological activity of drugs is premised on their ability to form stable complexes with their receptors while the magnitude of activity is a function of avidity of binding. The application of computational techniques in addition to in vitro tools in modern day drug design has afforded the benefi ts of specifi city and druglikeness optimization hence mitigating cost and time investment.
Molecular docking simulations represent the classical method in computational studies and the successes recorded using this technique both in antihelmintic and other drug discovery landscapes gave impetus for the employment of this technique in this study. Prominent among such studies is the work of Uzochukwu, et al. [6] which revealed the potentials of some approved drugs as antihelmintic agents.
This study investigated the antihelmintic potential of some approved drugs using Ascaris suum mitochondrial rhodoquinol fumarate reductase enzyme as target. The in vitro paralysis and mortality times of the frontrunners using the ascaris surrogate, Pheretima posthuma, were determined.

Selection and preparation of receptor
Bioinformatic mining of the Protein Data Bank (PDB) was done to identify the suitable ascaris MRFR 3D structure for the study. The crystal structure of the MRFR(PDB code, 3vra) was obtained from the Research Collaboration Standard Bioinformatics (RCSB) database.
The Flavin Adenine Dinucleotide (FAD) and hememolecules present in the structure and the extra subunits were deleted using Chimera v.1.9. The rest of the structure (A to D) were Graphical abstract.

Worm collection and preparation
The earthworms were harvested from swampy soil in Agulu, Anambra State, Nigeria and were stabilized in the soil marsh from where they were scooped and kept under cold chain till time of investigation. The worms were identifi ed by Mrs. Olue Annastasia of department of Parasitology and Entomology, Nnamdi Azikiwe University, Awka, Nigeria.

Drugs: Purchase and preparation
The drugs used for the bioassay were obtained from registered pharmacies in Nigeria and a few from Boots, United Kingdom. Patent holders brands were used or brands from reputable manufacturers and seven assay points were chosen between 0.078 to 5.0 mg/ml and were prepared as stock\ solutions.

Determination of paralysis and mortality times of frontrunners
The determination of paralysis and mortality times of frontrunners among the selected approved drugs was evaluated as described by Ajaiyeoba, et al. [11]. Five worms of average weight were rinsed with distilled water and placed in 20 ml solution of each drug in a standard petri dish according to the labeled concentrations.
The petri dishes were mechanically swirled to ensure the entire worm bodies were covered in the drug solution. The worms were then monitored for paralysis and mortality times and the observations were recorded. This same procedure was replicated in the synergistic testing. Albendazole served as standard reference while distilled water was negative control.
Paralysis was assessed as a situation when the worm loses muscular tone and unable to move its body except with vigorous shaking or when pricked with an object while mortality was considered when the worm does not move its body even when placed in water bath at 50ºC.

Preparation of drugs for synergistic testing
Doxycycline and mepacrine were selected for combination studies from among the four best frontrunners using effectbased strategy [12]. The combination of both drugs were prepared in 9 ratios of 5 mg/ml assay concentration (9:1, 8:2, 7:3, 6:4, 5:5, 4:6, 3:7, 2:8, 1:9). Paralysis and mortality effect of both drugs separately and when combined were used for the calculation of their combination index using the formula.

Statistical analysis
Data collected were presented as mean ± SEM and analyzed for one-way ANOVA statistics with SPSS version 16

Selection and preparation of approved drugs
The bioactivities of the reference compound Atpenin (A5) were determined using Molinspiration online tool (www. molinspiration.com). An in-house database of approved drugs was sorted and the best two bioactivity scores of the reference compound were used to query the in-house database of the approved drugs in order to select drugs with similar bioactivity scores to the reference compound. In this instance, G-Protein Coupled Receptor (GPCR) and Enzyme Inhibitor (EI) were used for selection of approved drugs. In addition, bioactivity scores in the lower range served as negative controls.

Validation of molecular docking protocol
Validation of the docking protocol was implemented by reproducing the experimental complex of the probe compound with its receptor in silico. The receptor in complex with the probe was obtained from the RCSB database [7] by the use of bioinformatic mining and prepared for docking simulation.
The probe and all hetero-molecules were deleted with Chimera v.1.9 [8]; polar hydrogen, Kollman charges were deleted then the grid box sizes and grid space centre of 10Å were determined with MGL tools v.0.1.5.6 [9]. The probe coordinates were obtained from the 3D structure in ZINC® database [10] to determine the conformation of the non-complexed atpenin prior to docking with the target.

Molecular docking simulations
The molecular docking stimulations of the approved drugs

In silico antihelmintic predictions
On the basis of bioactivity scores similarity of the approved drugs to atpenin, the in-house drug database was sorted and one hundred drugs were selected. And the docking protocol was validated by superimposition of the experimental atpenin A5 on the atpenin from docking which showed a near perfect fi t ( Figure 1).
The post docking analysis yielded two hundred drugs (including their isomers) with binding energies within the range of atpenin (-7.825 kcal/mol). Out of this list, fi fteen drugs with the closest binding energies to atpenin were selected as frontrunners. This result is presented in Table 1. Hydrogen bonding was found to have predominated among molecular interactions between the receptor and the frontrunner drugs especially amongst the polar ligands though hydrophobic interactions were also present and these hydrogen bond networks helped to strengthen the binding effect between the target and the ligands (Figures 3,4a,b,c,d).
Analogous to H-bonds are halogen bonds which are as well specifi c and have been found to play a major role in establishment of strong anchor points between different subunits during ligand-protein binding which leads to enhanced drug effi cacy. The generous amount of fl uorine atoms in the molecular structure of mefl oquine resulted in high halogen bond formation with consequent optimization of the bonding of mefl oquine to the receptor ( Figure 5).

In vitro antihelmintic activity
The in vitro antihelmintic evaluation of the frontrunners across the seven assay points was dose-dependent and the activities of the frontrunners were signifi cantly different from the reference standards at p<0.05 across all assay points.   using the one-way Anova statistics when compared to the means of the frontrunners with mefl oquine and doxycycline respectively producing about six-fold activity as against the positive control, albendazole.
However, proguanil and mepacrine produced a fourfold in vitro activity against albendazole though at lower concentrations such as 0.078 mg/ml the mean paralysis and mortality times for doxycycline, mepacrine and proguanil were still signifi cantly different from other drugs but the activities dropped to a double-fold.

Drug combination interactive effect
The synergistic testing produced the greatest activity (optimal fi xed dose) at the 6:4 ratio of doxycycline and        (Figures 8,9). This combination produced supraadditive effect (synergism) with paralysis combination index of 2.35 and mortality combination index of 2.23.

Discussion
The choice of the 3vra receptor out of the four Ascaris protein structures deposited in PDB was informed by the fact it is the only that is bound to an experimental compound which offered a guide as to the wet laboratory binding conformation and orientation while the successful superimposition of the atpenin experimental structure on the x ray crystallographic structure validated the docking protocol that was implemented.
When drugs bind to their receptors the conformation with the lowest binding energy is considered the most energetically favourable spatial arrangement and thus is speculated to have better binding affi nity. An analysis of the binding free energies of the selected drugs docked with the Ascaris MRFR showed the binding free energy of atpenin to be -7.825 kcal / mol which was taken as the benchmark value for assessment of the antihelminthic activity for the drugs investigated.
Seventy compounds produced binding energies far less than that of the reference compound, atpenin and fi fteen among these were chosen as frontrunners which included mefl oquine, proguanil, mepacrine, doxycycline, chloroquine, mebendazole and albendazole. The ligands were observed to have exploited the same binding pocket and interacted with same amino acid at the active site and these observations are suggestive of these  Ranking of Activity: *: Third in ranking; **: Next in activity; ***: Highest activity RANKING OF ACTIVITY: *: Third in ranking; **: Next in activity; ***:Highest activity      which is expected to affect its pharmacokinetic properties in vivo.
However, Verma, et al. [15]  Uzochukwu, et al. [6] reported similar results while targeting same fumarate reductase enzyme with some approved drugs and interestingly their research produced many antimalarials with good antihelmintic activities as was the case in this study. Among these antimalarials were mefl oquine, proguanil, quinine, chloroquine, pyrimethamine and mepacrine. There is evidence of cross activity between antimalarials and antihelmintics for which reason the employment of oil of chenopodium and santonin both in malarial and helminth chemotherapy is speculated to stem from an identical mode of action of the two drug classes wherein they form heme complexes in parasite food vacuoles obstructing glucose uptake [19]. Mepacrine and chloroquine share structure similarity and have for years been employed as treatments for teniasis and N.
americanus infections hence further strengthening the cross activity theory [20].
The action of antihelmintics could be by any of disruption of parasite metabolism or destruction of cuticle/cytoskeleton which eventually leads to paralysis and eventual death [21]. Ekeanyanwu and Etienjirhevwe [22] reported the interference with energy generation in helminth parasites by phenolic compounds and postulated that the mechanism of action seemed to be connected to the uncoupling of specifi c reductase-mediated reactions. Coincidentally, the target in this study, also a reductase, is an essential enzyme for adult helminth anaerobic metabolism where it catalyzes conversion of fumarate to succinate therefore ability of a chemical entity to block the activity of this enzyme would ultimately lead to parasite mortality hence predicting signifi cant antihelmintic capacity. This novel mode of action of these approved drugs could hopefully address the developing parasite resistance to the current human antihelmintic agents. The works of Adeniran and Sonibare [23] corroborated this mode of action that is associated with phenolic compounds when extracts of Dioscorea bulbifera, Mondora myristica and Xylopia aethiopia were found to interfere with oxidative phosphorylation reactions.
Roy [24] reported a central nervous system action of some alkaloids on earthworms for which reason it was speculated that mepacrine, an alkaloid could possibly have yet another mechanism of action. Raghavanma and Rama [25] also reported a dose-dependent antihelmintic activity of different extracts of Nauclea orientalis leaves which they opined to be a consequence of their tannin and saponin content. Notwithstanding that the antihelmintic activities observed in the present study compared and correlated very well with most of these studies, our results were viewed as superior since they were obtained at much lower assay points thus presupposing higher activities at higher concentration. This therefore gives hope for development of more effective and tolerable chemotherapy as drug safety is optimized when effective at lower concentrations.
The activities of drugs have been shown to be affected by interaction with other drugs which could be benefi cial or adverse. Certain pharmacokinetic factors such as solubility, bioavailability, metabolism (which can result from enzyme induction or inhibition) do impact on the activity of a drug [26]. Many drug preparations have provided better therapeutic benefi ts through this process and combination therapies are fast becoming a norm in most therapeutic landscapes and in the opinion of Portsmouth [27], "drug combinations are considered the ideal therapy for treatment of important infectious diseases including HIV, malaria and TB". Hu, et al. [28] noted that of all the antihelmintics currently in use, only tribendimidine (a nicotinic acetycholine receptor agonist (nAchR) and albendazole -a benzoimidazole are adequate for single dose mass drug administration. This in their opinion is because both have excellent activity against ascaris, moderate against hookworms and poor activity against threadworm (Strongyloides) and whipworm. But a major drawback however is that tribendimidine is still under trial and unapproved except in China hence, raises urgent need to explore combination therapies against helmintic infections. The recorded supra-additive interaction (synergism) between doxycycline and mepacrine may be as a result of the documented inhibition of the metabolism of mepacrine by doxycycline and this is of great interest given the inadequacies of single molecule therapy of antihelmintics currently in addition to the increasing wave of resistance to available treatments. The works of Keiser, et al. [4] confi rmed these inadequacies and the effectiveness as well as the desirability of combination therapy in antihelmintic chemotherapy. It is therefore hoped that this strong positive interaction observed with the combination would be replicated in clinical studies and lead to the full repurposing of these two approved drugs as new antihelmintic agents.