Benzimidazole and aminoalcohol derivatives show in vitro anthelmintic activity against Trichuris muris and Heligmosomoides polygyrus

Background Infections by gastrointestinal nematodes cause significant economic losses and disease in both humans and animals worldwide. The discovery of novel anthelmintic drugs is crucial for maintaining control of these parasitic infections. Methods For this purpose, the aim of the present study was to evaluate the potential anthelmintic activity of three series of compounds against the gastrointestinal nematodes Trichuris muris and Heligmosomoides polygyrus in vitro. The compounds tested were derivatives of benzimidazole, lipidic aminoalcohols and diamines. A primary screening was performed to select those compounds with an ability to inhibit T. muris L1 motility by > 90% at a single concentration of 100 µM; then, their respective IC50 values were calculated. Those compounds with IC50 < 10 µM were also tested against the adult stage of T. muris and H. polygyrus at a single concentration of 10 µM. Results Of the 41 initial compounds screened, only compounds AO14, BZ6 and BZ12 had IC50 values < 10 µM on T. muris L1 assay, showing IC50 values of 3.30, 8.89 and 4.17 µM, respectively. However, only two of them displayed activity against the adult stage of the parasites: BZ12 killed 81% of adults of T. muris (IC50 of 8.1 µM) and 53% of H. polygyrus while BZ6 killed 100% of H. polygyrus adults (IC50 of 5.3 µM) but only 17% of T. muris. Conclusions BZ6 and BZ12 could be considered as a starting point for the synthesis of further structurally related compounds. Graphical Abstract


Background
Soil transmitted helminths (STHs) are a group of human parasitic nematodes that affect around 1.5 billion of the world's population causing substantial disease and disability [1]. These infections are more common in people living in low-and middle-income countries, in areas with poor access to adequate drinking water, sanitation and hygiene [2]. Of particular importance are infections caused by roundworms (Ascaris lumbricoides), whipworms (Trichuris trichiura) and hookworms (Necator americanus or Ancylostoma duodenale) [3]. According to the latest studies, the global incidence of T. trichiura
In animals, the presence of helminth infections has an important economic impact arising directly from reductions in productive yields associated with weight gain, milk production and wool quality [5]. In Europe, the annual economic losses caused by helminth infections in livestock have been estimated at 1.8 billion euros, while those caused only by anthelmintic resistance in the group of gastrointestinal nematodes reach approximately 38 million euros annually [6].
The main control of these infections in humans is based on preventive chemotherapy, employing large-scale administration of anthelmintic drugs to populations at risk. Currently, there are four drugs on the World Health Organization (WHO) model list of essential medicines that are recommended for the treatment of soil-transmitted helminths: albendazole (ABZ), mebendazole (MBZ), levamisole (LEV) and pyrantel pamoate (PYR) [7]. However, the administration of these drugs is not always highly effective, reflecting low cure rates, especially against trichuriasis, when a single oral dose is administered [8,9]. Moreover, the efficacy of these compounds has decreased significantly over time, showing egg reduction rates dropping from 72.6% in 1995 to 43.3% in 2005 [10].
Additionally, in gastrointestinal nematodes infecting livestock, multiple cases of rapid resistance development have been reported in all continents as a consequence of the abusive use of anthelmintics, mainly benzimidazoles, such as ABZ, and macrocyclic lactones, such as ivermectin (IVM) [11][12][13].
Considering the spread of anthelmintic resistance in animals and the low efficacy of some benzimidazoles against certain STHs in humans, it is clear that there is an urgent need to search for new drugs to control helminth infections. Therefore, the present study is focused on the determination of the potential in vitro anthelmintic activity of a series of synthetic compounds from different chemical families including benzimidazoles (BZ), lipidic diamines (AA) and aminoalcohols (AO) using two rodent models of gastrointestinal nematodes: Trichuris muris and Heligmosomoides polygyrus [14,15]. For this purpose, a total of 41 compounds (15 BZ, 11 AA and 15 AO) were evaluated against two different stages of the parasites, the first stage larvae 1 (L 1 ) and adult forms.

Animals and parasites
The complete life cycles of T. muris and H. polygyrus are maintained at the Swiss Tropical and Public Health Institute (TPH). Experiments were approved by the Swiss national and cantonal authorities (permission no. 2070). Three-week-old female mice (NMRI for H. polygyrus and C57BL/6 N for T. muris) were allowed to acclimatise to the new environment for 1 week before use. During the acclimatisation period, from day 2 after arrival the animals received 0.25 mg/l dexamethasone (Sigma-Aldrich ® ) in the drinking water to immunosuppress them and facilitate parasite establishment. During the experiment, mice were kept at 22 °C, 50% humidity, with a 12-h light/dark cycle, and water and rodent food (KLIBA NAFAG, Switzerland) were available ad libitum according to Swiss Animal Welfare guidelines. Mice were orally infected with 200 embryonated T. muris eggs or 90 H. polygyrus L 3 stage.

Drug discovery strategy
To investigate the activity against T. muris, 41 compounds were first subjected to a primary screening to assess their ability to inhibit T. muris motility at the L 1 stage at a single concentration of 100 µM. Only compounds with the ability to inhibit the motility of > 90% of larvae (activity higher than 90%) progressed to subsequent dose-response evaluation to estimate their half maximal inhibitory concentration (IC 50 ) value. Then, those with IC 50 values < 10 µM were tested on the adult stage of T. muris and H. polygyrus following a similar protocol as mentioned above: initial screening at 10 µM and estimation of IC 50 values of those compounds with activities > 80%.

Evaluation of anthelmintic activity on T. muris L 1 s
The assay was performed according to Wimmersberger et al. [19]. Briefly, unembryonated eggs were collected from faeces of infected mice. After storing the eggs for 3 months at room temperature in a dark box, hatching occurred by incubation with 10 7 -10 8 cells/ml of Escherichia coli BL-21 stock. Approximately 40 L 1 per well was placed into a 96-well plate and incubated for 24 h at 37 ºC and 5% v/v CO 2 atmosphere, in the presence of 100 μl RPMI-1640 medium supplemented with antibiotic mixture (12.5 μg/ml amphotericin B, 500 U/ml penicillin and 500 μg/ml streptomycin) and 100 μM of the compound to be tested. As a positive control, LEV (Sigma-Aldrich) at a final concentration of 50 μM was used, while wells with medium and 1% DMSO served as negative control. Each compound was tested in duplicate and the assay was repeated on a different day. After 24 h incubation, larval viability was determined by using a binary scale that discriminates alive from dead larvae: "0" = no sign of motion = dead and "1" = motion observed = alive. The percentage of dead larvae was established for each well. To stimulate the movement of all live larvae, 100 µl of hot water (≈ 80 °C) was added to each well following previous protocols [19]. Each larva was observed for 3-5 s. Compounds that showed a killing effect > 90% at 100 μM were selected to determine their IC 50 . For this, L 1 s were incubated with at least six different concentrations of the compound ranging from 100 to 0.41 μM (1:3 serial dilutions).

Evaluation of anthelmintic activity on H. polygyrus and T. muris adult worms
The assay was performed according to Karpstein et al. [20]. Briefly, H. polygyrus and T. muris adults were collected from the caecum (T. muris) and colon (H. polygyrus) of the mice 2 (H. polygyrus) and 7 weeks (T. muris) after infection, respectively. All animals were humanely slaughtered by 100% CO 2 inhalation. Three to four adult worms were placed in each well of a 24-well plate and exposed to the test compounds at a final concentration of 10 μM in RPMI 1640 medium supplemented with 100 U/ ml penicillin and 100 μg/ml streptomycin in a final volume of 2.5 ml. Heligmosomoides polygyrus medium was supplemented with 12.5 μg/ml amphotericin B and T. muris medium with 5% foetal calf serum (iFCS, 100 U/ ml). Wells containing 1% (v/v) DMSO in water were included as negative control. Worms were incubated at 37 °C and 5% CO 2 up to 72 h, after which the drug effect was evaluated using a phenotypic readout. The assay was conducted in duplicate and repeated twice at different days. The condition of the worms was microscopically   evaluated according to their phenotype, using a viability scale ranging from 3 to 0 (3: good motility; 2: low motility; 1: very low motility; 0: dead). If adult worms did not move enough for a clear scoring, they were stimulated with 500 μl hot water (≈ 80 °C). Therefore, the effect of the compounds was expressed by the percentage of dead larvae, considering dead those with a score of 0 and alive those with a score ranging from 1 to 3. Compounds with efficacies > 80% at 10 µM were then selected for IC 50 determination (1:4 serial dilutions ranging from 10 to 0.039 μM).

Data analysis
IC 50 values of both assays were calculated based on median effect principle, using the CompuSyn software (CompuSyn, version 3.0.1). These values were defined as the concentration of a drug required to decrease the mean worm's motility by 50%. The "r" value is the linear correlation coefficient of the median-effect plot; it illustrates the goodness of fit and thus the accuracy of the IC 50 value.

Cytotoxicity assays and selectivity indexes
Cytotoxicity assays for most compounds were carried out in a previous study on two different cell lines, the human colorectal adenocarcinoma Caco-2 (ATCC ® HTB-37 ™ ) and the human hepatocarcinoma HepG2 (ATCC ® HB-8065 ™ ), using the Alamar Blue staining method, to estimate their toxicity [18,21]. In the present study, the cytotoxicity of only two compounds, AO14 and AO15, was performed following the protocols mentioned above. Selectivity indexes (SIs) were calculated by dividing the CC 50 values obtained in the cytotoxicity assays by the IC 50 values of the in vitro assays. The greater the SI value, the more selective the compound is in inhibiting T. muris activity and the less in inhibiting mammalian cell growth (general cytotoxicity).

Results
Tables 1, 2 and 3 display the basic scaffold of the different classes of compounds tested and the results of the in vitro assays performed against T. muris L 1 together with cytotoxicity data and SIs. AO and AA compounds are arranged according to the type and size of substituents present on R 1 , R 2 and R 3 and to the length (n) of the alkylside-chain. BZ compounds are distributed first (R 1 ) by the type of substituents at the C-5 (C-6) position of the benzimidazole system and second (R 2 ) by the substituents on the 2-phenyl ring.

Discussion
In recent years, the number of new anthelmintics compounds introduced to the market to control the infections produced by gastrointestinal nematodes has been limited, mainly because of economic difficulties in the development and marketing of new drugs [22]. In the last 2 decades, only four drugs have been introduced on the market: emodepside [23], monepantel [24], derquantel [25] and tribendimidine [26]. Therefore, there is a clear need to develop novel anthelmintic drugs for the control of these parasitic worms in humans and farm animals.
One of the approaches proposed to alleviate the severe scarcity of anthelmintics is the synthesis of new derivatives of known drugs. Although BZ resistance is present in many gastrointestinal species infecting livestock, the synthesis of novel BZ derivatives may lead to compounds with improved properties such as better solubility and pharmacokinetic profile, resulting in increased effectiveness [27]. Some promising compounds, such as tenvermectin [28], diisopropylphenyl-imidazole [29] and mebendazole hydrochloride [30], have been developed in recent years following this approach. Based on these assumptions, in the present study, a total of 15 AO and 11 AA derivatives, both structurally related to sphingosine, and 15 benzimidazole derivatives were tested against L 1 of T. muris and adult stages of T. muris and H. polygyrus. The anthelmintic activity of most of these compounds was previously tested in vitro against the gastrointestinal nematode infecting sheep Teladorsagia circumcincta [18,21] and some of them were also tested against Leishmania spp. [31,32], Trypanosoma spp. [17,33] and Strongyloides venezuelensis [34].
The L 1 assay has proven to be a good tool to screen new potential candidate compounds before carrying out adult motility assays, the in vitro assay of choice, which is more expensive, labour intensive and time-consuming, and it requires the use of live animals [19]. Moreover, the results obtained with the motility assay based on L 1 seem to correspond to the findings observed with adult T. muris [35]. However, some studies showed that L 1 appears to be more sensitive to drugs than older stages of T. muris [36,37], which can facilitate the discarding compounds with no activity.
In the present study, 10 out of the 41 compounds tested showed activity > 90% against the L 1 stage of T. muris at 100 µM, and only three, namely AO14, BZ12 and BZ6, reached an IC 50 < 10 µM. The screening performed at a single final concentration of 10 µM on adults showed that only BZ12 and BZ6 had significant activity against the adult stage of T. muris and H. polygyrus, respectively.
Thus, BZ6 seems to be the only compound reaching IC 50 values < 10 µM in both eggs and L 1 of T. circumcincta (IC 50 = 6.54 µM in eggs and IC 50 = 5.01 µM in L 1 ) and also in L 1 of T. muris (IC 50 = 4.17 µM), with values quite close to each other). However, BZ6 did not have any affect against the adult stage of T. muris at a concentration of 10 µM (17.2% of activity), but it was effective against H. polygyrus adults (100% of activity) displaying an IC 50 of 5.3 µM. On the other hand, BZ12 did not produce an effect against any of the stages of T. circumcincta, eggs, L 1 or L 3 , but it showed activity against T. muris L 1 with an IC 50 of 8.89 µM. Moreover, this BZ12 reached an efficacy of 53.3 and 81.7% on the adult stage of H. polygyrus and T. muris at 10 µM, respectively, presenting an IC 50 of 8.1 µM in the latter.
In terms of the relationship between the structure and efficacy of the compounds and focusing on the benzimidazole derivatives, the only group of compounds that has shown significant efficacy on the adult stage of the parasites in this study, we can observe that the presence of a mild basic group such as the NH 2 group on R 1 (BZ15) did not induce any measurable effect on the nematode viability, while the combinations of 5-Me-4'-OMe/Cl (BZ1 and BZ2), 5-Cl-4'-Cl (BZ6) and 5-NO 2 -4'-Cl/diMe (BZ12 and BZ13) produced a deadly effect > 90% on the initial screening of T. muris L 1 . Regarding the substituent present on the B-phenyl ring (R 2 ), 4'-Cl − is required for the anthelmintic effect since all compounds with this substituent at this position showed anthelmintic activity on T. muris L 1 (BZ2, BZ6 and BZ12), including here the two most potent compounds (BZ6 and BZ12), while double substitutions on this ring, such as 3'-NO 2 4'-OMe (BZ4, BZ9 and BZ14) or 3'-NH 2 4'-OMe (BZ10), led to inactivity. However, a di-substitution in position 2' and 6' with electron donating groups such as 2' ,6'-diMe in addition to a polar group in ring A such as 5-NO 2 (BZ13), gave good anthelmintic inhibitory activity in T. muris L 1 (99.40 inhibition at 100 µM), although its IC 50 was > 10 µM.
Comparing the results of the adult motility assay of the present study with previous experiments using the marketed human drugs (ABZ, MBZ, LEV and PYR) showed that the IC 50 values obtained are much lower (8.1 µM for BZ12) since BZ compounds showed a lack of activity on T. muris adults and LEV and PYR displayed IC 50 values around 68 and 57 µM, respectively [19].
All compounds tested against L 1 had a possible toxic potential, as their SIs were very close to one, except BZ6 and AO14, which reached values > 4 in both cell lines. Regarding the SIs obtained in the adult assays, although they were in any case > 1, BZ6 seems to be a safer candidate than BZ12, as it had SI values of 4.3 for Caco2 cells and 3.1 for HepG2 cells.

Conclusions
Compounds BZ6 and BZ12 could represent a starting point for the synthesis of further structurally related compounds, as they showed activity against the adult stage of H. polygyrus and T. muris, respectively.