In Vitro Anthelmintic Activity of Saponins from Medicago spp. Against Sheep Gastrointestinal Nematodes

Gastrointestinal strongyle nematodes (GIS) are included among the most important parasites of small ruminants. The widespread drug resistance and drug residues in products of animal origin have increased the interest in the search for natural compounds with anthelmintic activity as a valid alternative to current synthetic drugs. The aim of the present investigation was to test the ‘in vitro’ anthelmintic activity of saponins and prosapogenins from different Medicago species, selected for their importance as a forage crop worldwide for animal feeding. From these plants, saponin mixtures were extracted, purified and used at scalar concentrations to evaluate their anthelmintic activities against sheep gastrointestinal strongyles (GISs), by the egg hatch test. Treated and untreated controls were used as the comparison. Data were statistically analyzed, and EC50 and EC90 were also calculated. All saponins and prosapogenins showed inhibiting effects on GIS eggs in a concentration-dependent manner. At higher concentrations, most of them showed an efficacy comparable to the reference drug (Thiabendazole 3 µg/mL) (P < 0.001). With 1.72 mg/mL EC50 and 3.84 mg/mL EC90, saponin from M. polymorpha cultivars Anglona was the most active. Obtained results encourage further studies aimed at evaluating the efficacy ‘in vivo’ of saponins which resulted as most effective ‘in vitro’ in this study.


Introduction
Gastrointestinal strongyles (GIS) are considered one of the most common causes of economic losses and disease in small ruminant breeding [1]. These parasites are nematodes belonging to the Strongylida order, and localize in the gastrointestinal tract of small ruminants. Infected animals may show reduced growth, cachexia, weakness, anemia and diarrhea, that may lead to poor reproductive and productive performances and death [2]. In the last decades gastrointestinal strongyles infections have been primarily controlled with synthetic drugs belonging to different chemical classes, i.e., benzimidazoles, imidazothiazole/tetrahydroxypyrimidines, macrocyclic lactones, amino-acetonitrile derivates and spiroindoles [1]. However, the regular and sometimes excessive use of anthelmintics has contributed to the onset of drug resistance, which is now widespread worldwide [3], limiting the effectiveness of synthetic drugs for the control of gastrointestinal nematode infections [4,5]. Confirming this, there are

Saponin Composition
The compositional profile of Medicago saponin extracts used in this study differed according to the plant species [28][29][30]. Crude saponins were obtained from the Medicago species under investigation as whitish powder in a purity of about 90% and in a different yield: M. polymorpha cv. Anglona 2.1% dry matter (DM), M. polymorpha cv. Santiago 1.7% DM and M. sativa cv. Equipe 1.5% DM. Figure 1 shows the chemical structure of the most abundant saponins/sapogenins detected in the different Medicago extracts. The content of the most abundant sapogenins, obtained after acid hydrolysis of the corresponding glycosides, is reported in Table 1.
Saponins from M. sativa were characterized by a higher amount of medicagenic and zanhic acids (Figure 1), quoted as 47.2% and 25.5% of the total sapogenins, respectively. Hederagenin (Figure 1) was instead the dominant sapogenin in M. polymorpha cv. Santiago, representing 88.3% of the total aglycones, Molecules 2020, 25, 242 3 of 9 while echinocystic acid (90.1%) (Figure 1) was the dominant sapogenin detected in M. polymorpha cv. Anglona. Soyasapogenol B, the aglycone moiety of soyasaponin I, a saponin commonly present in Leguminosae, was detected in all samples, although in a different amount. From the high-performance liquid chromatography (HPLC) analyses of saponins (data not shown), and by comparison with authentic reference compounds previously identified in the Medicago spp. [30][31][32], all the saponin mixtures here evaluated were mainly constituted by bidesmosidic type saponins (70%-80%). M. sativa prosapogenins, obtained after basic hydrolysis of the corresponding saponins, were instead entirely made up by monodesmosides and possess the same sapogenin composition as M. sativa from which they were obtained.
Molecules 2020, 25, x FOR PEER REVIEW 3 of 9 Saponins from M. sativa were characterized by a higher amount of medicagenic and zanhic acids (Figure 1), quoted as 47.2% and 25.5% of the total sapogenins, respectively. Hederagenin (Figure 1) was instead the dominant sapogenin in M. polymorpha cv. Santiago, representing 88.3% of the total aglycones, while echinocystic acid (90.1%) (Figure 1) was the dominant sapogenin detected in M. polymorpha cv. Anglona. Soyasapogenol B, the aglycone moiety of soyasaponin I, a saponin commonly present in Leguminosae, was detected in all samples, although in a different amount. From the high-performance liquid chromatography (HPLC) analyses of saponins (data not shown), and by comparison with authentic reference compounds previously identified in the Medicago spp. [30][31][32], all the saponin mixtures here evaluated were mainly constituted by bidesmosidic type saponins (70%-80%). M. sativa prosapogenins, obtained after basic hydrolysis of the corresponding saponins, were instead entirely made up by monodesmosides and possess the same sapogenin composition as M. sativa from which they were obtained.

'In Vitro' Anthelmintic Activity of Saponin Mixtures
Results of the 'in vitro' anthelmintic activity of the saponin mixtures tested at different concentrations against sheep GIS eggs are reported in Figure 2. All the saponin mixtures used in this study showed inhibiting effects against GIS eggs in a concentration-dependent manner. More specifically, 10, 5 and 2.5 mg/mL M. polymorpha cv. Anglona, 10 and 5 mg/mL M. polymorpha cv. Santiago and 10 mg/mL M. sativa saponins were able to inhibit almost 100% of GIS eggs. For these

'In Vitro' Anthelmintic Activity of Saponin Mixtures
Results of the 'in vitro' anthelmintic activity of the saponin mixtures tested at different concentrations against sheep GIS eggs are reported in Figure 2. All the saponin mixtures used in this study showed inhibiting effects against GIS eggs in a concentration-dependent manner. More specifically, 10, 5 and 2.5 mg/mL M. polymorpha cv. Anglona, 10 and 5 mg/mL M. polymorpha cv. Santiago and 10 mg/mL M. sativa saponins were able to inhibit almost 100% of GIS eggs. For these saponin samples the same anthelmintic effects were recorded as for the reference drug, thiabendazole (TBZ), tested at 3 µg/mL (P < 0.001).
A very good inhibition activity (93%) was also observed for M. polymorpha cv Santiago at 2.5 mg/mL, while the efficacy of 0.5 and 0.25 mg/mL M. polymorpha cv. Anglona, 0.5 mg/mL M. polymorpha cv Santiago, 5 and 2.5 mg/mL M. sativa saponins and 10 mg/mL M. sativa prosapogenins ranged between 73% and 84% ( Figure 2). Egg hatching inhibition values comprised between 43% and 66% were observed for M. polymorpha cv. Santiago at 0.25 mg/mL, M. sativa saponins at 0.5 mg/mL and M. sativa prosapogenins in the range 5-0.5 mg/mL concentration. All other saponin solutions (0.05 mg/mL M. polymorpha cv. Anglona and cv. Santiago, 0.25 and 0.05 mg/mL M. sativa saponins and 0.25 and 0.05 mg/mL M. sativa prosapogenins) showed a very low efficacy (2%-12.5%), but statistically different (P < 0.001) from the untreated controls for which an egg hatch inhibition less than 1% was registered ( Figure 2 saponin samples the same anthelmintic effects were recorded as for the reference drug, thiabendazole (TBZ), tested at 3 µg/mL (P < 0.001).

Discussion
Results obtained in this study show that all Medicago saponins and prosapogenins evaluated in this study have 'in vitro' inhibiting effects against sheep GIS eggs, although with a different level of

Discussion
Results obtained in this study show that all Medicago saponins and prosapogenins evaluated in this study have 'in vitro' inhibiting effects against sheep GIS eggs, although with a different level of efficacy, with M. polymorpha saponins as the most active. A dose dependent inhibition effects on egg hatch and development was also observed ( Figure 2). Moreover, the EC 90 and EC 50 values found for the most active saponins in this study are similar to those reported for other active plant compounds [33]. These results are in agreement with previously reported data, confirming a high ovicidal activity of saponins from Medicago spp. [19]. As observed by [19], when tested at the 10 mg/mL concentration against donkey GIS eggs, saponins from Medicago spp. show a high activity (80%-100% egg hatch reduction) with M. polymorpha cv. Anglona and M. sativa as the most active. Differences in saponin activity between sheep and donkey GIS eggs can probably be ascribed to a different susceptibility of sheep and donkey GIS species to tested saponins. The available data from literature also confirm the 'in vitro' anthelmintic activity of this class of compounds against ruminant GIS. The saponins aescin and digitonin [17] and saponin fractions contained in Zizyphus joazeiro [16], Phytolacca icosandra [34] and Agave sisalana [18] showed 'in vitro' ovicidal activity against the nematodes of small ruminants. Similarly, the 'in vitro' ovicidal action of Combretum molle against eggs of H. contortus from sheep and of Ipomoea chiliantha, Tocoiena formosa and Aspilia latissima against the eggs of Haemonchus placei from cattle, have been attributed to their saponin content [33,35]. Biological effects of saponins are normally ascribed to their specific interaction with the cell membrane, causing changes in the cell permeability [20,36,37]. By this way, saponins may penetrate inside the GIS eggs, altering some biological functions and preventing the normal development of eggs, thus leading to the inhibition of the development of eggs definitively, as proposed by [16]. It has been also hypothesized that these compounds may be able to interfere with enzymatic pathways involved in larval development, which results in larval death [16].
Data obtained in this study on the structure-activity relationship showed that all the Medicago saponins are active compounds against GIS eggs, independently of the involved genins. Comparing results between saponins and related prosapogenins, data here presented indicated that M. sativa prosapogenins (monodesmosides) resulted less active than the related saponins (bidesmosides). These results agree with previously reported data on GIS eggs from donkey [19].
The 'in vitro' efficacy of saponins from Medicago spp. in inhibiting the hatching of GIS eggs of sheep and donkey [19], encourage further studies aimed at evaluating their efficacy 'in vivo' as new anthelmintic compounds, as nutraceuticals or as a means to inhibit the environmental development of GIS eggs in order to lower pasture contamination. However, various aspects related to their potential toxicity should be considered. Saponins, in fact, as well as tannins, if ingested in large quantities, are potentially toxic, and may diminish the digestibility of feed [38,39]. However, if taken in moderate concentrations, they can improve nutritional effects [40], and at the same time can reduce the parasitic burden [34]. A previous study [41] showed that a diet containing 1.5% of saponins from Quillaja saponaria bark may reduce by 38.8% the sheep fecal egg count (FEC) of GIS eggs when compared to untreated animals. The high 'in vitro' activity of the Medicago saponins evaluated in the present study against sheep GIS eggs may suggest their potential 'in vivo' efficacy at non-toxic dosages.

Plant Material, Extraction, Purification and Characterization of Saponin Mixtures
Medicago plants used in this study were grown at the Research Center for Animal Production and Aquaculture (CREA-ZA, Lodi, Italy). Aerial parts from M. polymorpha cv. Santiago, M. polymorpha cv. Anglona, and M. sativa cv. Equipe were utilized for saponin processing. Leaves were separated from stems and dried at 40 • C to a constant weight. All samples were ground and used for the successive extractions. Saponins were extracted and purified following general procedures, as previously reported [30][31][32]. Powdered plant materials (150 g) were separately defatted with CHCl 3 in a Soxhlet apparatus for 24 h. Defatted material (100 g) was separately extracted with 80% MeOH under reflux for 24 h. The solvent was removed under reduced pressure, and the residue was re-suspended in 30% Molecules 2020, 25, 242 6 of 9 MeOH. The solution was applied onto a 100 × 60 mm RP-18 (40-63 µm) column, preconditioned with 30% MeOH. Elution was carried out with 30% MeOH (500 mL) to remove sugars and some phenolics.
Total saponins were then eluted with 90% MeOH (400 mL) and dried under vacuum. 2.06 g of saponins were obtained from M. polymorpha cv. Anglona (2.1% yield), 1.72 g of saponins were obtained from M. polymorpha cv. Santiago (1.7% yield), while 1.48 g of saponins were obtained from M. sativa cv. Equipe (1.5% yield). In addition, saponins from M. sativa were subjected to basic hydrolysis [23] to obtain the related prosapogenins, which were also evaluated in this study. All samples were dissolved in H 2 O-5% dimethyl sulfoxide (DMSO), solutions were properly diluted with H 2 O and used in the bioassay at different final concentrations from 0.05 to 10.0 mg/mL. The saponin mixtures, obtained as whitish powders of about 90% purity, were analyzed by thin layer chromatography (TLC), as previously described [30]. In addition, extracted and purified saponins were characterized for their qualitative and quantitative aglycone composition by gas chromatography (GC) and gas chromatography/mass spectrometry (GC/MS) analyses of derivative sapogenins obtained after acid hydrolysis, as already reported [42]. To obtain information on saponin composition and purity, the saponin mixtures were then analyzed by high-performance liquid chromatography (HPLC), using an external standard method [30], and the results compared with standards of previously purified and identified saponins and data from literature [30][31][32].

Nematode Egg Collection, Purification and Suspension
Purified GIS eggs were obtained from fecal samples of naturally infected sheep from an organic farm in Tuscany (central Italy). For transport to the laboratory, fecal samples were placed in sealed and refrigerated bags. Fecal microscopic analysis was performed using the Mini-FLOTAC technique [43], with a sensitivity of 20 eggs per gram of feces (EPG). Recovery and suspension of GIS eggs were performed within 3 h of collection using previously reported methods [44] with some modifications. More specifically, 30 g of feces were mixed with distilled water and then centrifuged in 50 mL tubes at 900× g for 5 min. The supernatant was than eliminated and an NaCl saturated solution was added to the sediment and centrifuged at 170× g for 5 min. The supernatant was collected and centrifuged for the last time with distilled water in 15 mL tubes at 900× g for 5 min. The supernatant was eliminated in order to obtain the sediment containing the purified eggs that were inspected microscopically to exclude any embryonation that had begun, then they were counted, diluted in distilled water to the final concentration of about 400 eggs/mL and used immediately in the bioassay. Finally, fecal cultures were also performed by using the same pooled fecal samples employed for obtaining gastrointestinal strongyles eggs, to the aim of identifying GIS nematodes at the genus level. Fecal samples were cultured in an incubator at 25 • C for seven days, and larvae were recovered by the Baermann technique and identified according to the key and description given by [45].

Evaluation of the 'In Vitro' Anthelmintic Activity of Saponins
The Egg Hatch Test (EHT) was performed to evaluate the 'in vitro' anthelmintic activity of saponin and prosapogenin samples. In EHT the effectiveness of the substances is evaluated based on their ability to inhibit the development and hatching of parasite eggs [44,46]. In the present work the EHT was carried as in the previously reported methods [35,44,46], with some minor modifications. In detail, 24-well, flat-bottomed microplates were used, and 0.5 mL of distilled water containing approximately 200 eggs were placed in each well. All the test samples were treated with 0.5 mL of the different saponin solutions to obtain the reported final saponin concentration. Positive controls contained the same amount of GIS eggs and the anthelmintic drug TBZ (2-(4-Thiazoly) Benzimidazole (Thiabendazole) minimum 99%, Sigma S.r.l., Milan, Italy) at the final concentration of 3 µg/mL in 1% DMSO. Two different negative controls were prepared by adding to the egg suspension 0.5 mL of distilled water and 0.5 mL of 1% DMSO. The plates were then incubated at 25 • C in darkness and 80% humidity for 48 h, and the number of eggs and of the first-stage larvae (L1) in each well was microscopically counted. The percentage of egg hatch reduction was calculated using the following formula: number of eggs/(number of L1 + number of eggs) × 100 (1) All experiments were performed in triplicate in three independent assays.

Statistical Analysis
Results from all experiments were statistically analyzed and compared. Statistical analysis was performed using the Statistical Analysis System (SAS) program. To perform the statistical analysis of the obtained data, a one-way analysis of variance (ANOVA) test with 5% significance (P < 0.05), was used. Significant results were further tested with the Tukey post-hoc test (P < 0.05). The determination of EC 50 and EC 90 concentrations was done through non-linear regression analysis by using XLSTAT ® .

Ethical Declaration
This study has not included animal experiment. Sheep fecal samples used for the evaluations performed in this study have been collected with the consensus of the farm owner. Authors declare that the work has been carried out in adherence to a high standard of veterinary care.