Effets d’un extrait de sainfoin (Onobrychis viciifolia) et des monomères de

de nématodes abomasaux dans des explants fundiques.

BRUNET, S., JACKSON, F. et HOSTE, H.

RESUME - ARTICLE 4

Introduction

Les plantes riches en tannins représentent une alternative à l’utilisation d’anthelminthiques (AHs) de synthèse pour la maîtrise du parasitisme gastro- intestinal (GI). Cependant, le mode d’action de ces plantes sur les nématodes GIs reste mal connu.

Des études récentes ont montré que la première phase de l’invasion de l’hôte par les nématodes GIs, soit le dégainement des larves infestantes (L3s), est affectée par des extraits de plantes riches en tannins. Dans cette étude, nous avons examiné l’hypothèse selon laquelle la seconde phase de l’installation larvaire, l’association- pénétration des L3s dégainées dans la muqueuse digestive, serait également perturbée par une exposition aux tannins. Pour cela, un test in vitro utilisant des explants fundiques a été appliqué.

Les objectifs étaient de vérifier: i) si la capacité des L3s à pénétrer dans la muqueuse était affectée par le contact avec l’extrait de sainfoin ; ii) si cet effet était dépendant de la dose ; iii) si les tannins sont responsables des effets observés ; iv) si ces effets étaient spécifiques de l’espèce parasite ; v) si la structure biochimique des tannins influençait les effets.

Matériel et méthodes

Les L3s d’Haemonchus contortus ont été artificiellement dégainées par contact avec une solution d’hypochlorite de sodium. Après lavages, les L3s dégainées ont été incubées pendant 3 heures en présence : a) d’un extrait de sainfoin à différentes concentrations ; b) d’un extrait de sainfoin prétraité par du polyvinyl polypyrrolidone (PVPP), un inhibiteur de tannins ; c) d’un des monomères de tannins condensés.

Afin d’évaluer la spécificité des effets, des L3s de Teladorsagia circumcincta ont également été dégainées puis incubées en présence d’extrait de sainfoin prétraité ou non par du PVPP.

Après lavages, les L3s sont déposées sur des explants d’abomasum de chèvre. Après 3 heures de contact, les explants sont rincés pour retirer les L3s qui n’ont pas pénétrées dans la muqueuse, puis soumis à une digestion pepsique afin de récupérer les L3s ayant pénétré dans la muqueuse.

Le pourcentage de L3s ayant pénétré dans la muqueuse fundique est déterminé par le nombre de L3s récupérées dans la solution de digestion de l’explant divisé par le nombre total de L3s retrouvées, multiplié par 100.

Résultats et discussion

L’incubation des L3s avec un extrait de sainfoin à 1200µg/ml a induit une réduction significative de la capacité des L3s d’H.contortus et de T.circumcincta à pénétrer dans les explants fundiques, suggérant un effet inhibiteur sur les L3s dégainées et une absence de spécificité de l’extrait de sainfoin selon l’espèce de nématodes GIs.

Pour les deux espèces, cet effet a été totalement supprimé lorsque l’extrait de sainfoin a été prétraité avec du PVPP, ce qui suggère que les tannins sont responsables de l’effet observé.

La comparaison des résultats obtenus avec différents monomères de tannins condensés (TCs) a confirmé une relation entre la structure biochimique et l’effet inhibiteur des TCs. En effet, les monomères des prodelphinidols et les dérivés gallatés ont été plus efficaces que les monomères des procyanidols, corroborant ainsi les résultats obtenus sur la migration et le dégainement larvaires.

En conclusion, la réduction d’installation des L3s des nématodes GI chez l’hôte observée lors de la consommation de plantes riches en tannins pourrait s’expliquer par un double effet inhibiteur du dégainement larvaire et de la pénétration des L3s dégainées dans les muqueuses digestives.

Abstract

Tannin-rich forages offer an alternative to anthelmintic chemicals to control gastrointestinal nematodes. However, the mode of action of such bioactive plants still needs to be assessed. Previous studies have shown that extracts of tannin-rich plants interfere with the first phase of host invasion, i.e., the exsheathment of infective larvae (L3s). In the current study, we examined the hypothesis that exposure to tannins could also affect the second phase of larval establishment, i.e., the tissue association/penetration of the exsheathed L3s into the digestive mucosae. An in vitro direct challenge technique using fundic explants was applied in this study. The main parasite model was Haemonchus contortus. The objectives were to verify: (i) whether a modification of the association/penetration of L3s with the mucosae occurred after contact with sainfoin extract; (ii) whether this is a dose-dependent phenomenon; (iii) whether tannins were responsible for these effects; (iv) whether these effects were dependent on the parasite species; and (v) how the biochemical structure of tannins might influence these effects. Following 3 h contact with sainfoin extract at 1,200 ng/ml, the penetration of exsheathed L3s of H. contortus and Teladorsagia circumcincta into fundic explants was significantly reduced. Moreover, a dose-response relationship was found for H. contortus. For both nematodes, the changes were totally alleviated after addition of polyvinyl polypyrrolidone, an inhibitor of tannins, to the sainfoin extract, suggesting that tannins play a major role in the observed effects. Comparison of results obtained with different monomers of condensed tannins confirms a relationship between structure and activity, the prodelphinidin monomers and gal-loyl-derivatives being more effective than the procyanidin monomers. Combined with the delay or the inhibition of larval exsheathment previously shown, these effects could explain how tanniniferous plants reduce the establishment of infective larvae in small ruminants. © 2007 Australian Society for Parasitology Inc. Published by Elsevier Ltd. All rights reserved.

Keywords: Sainfoin (Onobrychis viciifolia); Tissue association; Larval establishment; In vitro direct challenge assay; Tannins; Haemonchus contortus; Teladorsagia circumcincta

1. Introduction

Due to the worldwide development of anthelmintic resistance in nematode populations, the need for alternative control strategies in small ruminant production has increased (Jackson and Coop, 2000; Kaplan, 2004).

Corresponding author. Tel.: +33 05 61 19 38 75; fax:

E-mail address:h.hoste@envt.fr (H. Hoste).

Evidence suggests that tannin-rich plants might represent an alternative solution to chemical anthelmintics (Min and Hart, 2002; Hoste et al., 2006; Waller, 2006).

Several studies have shown that the consumption of tannin-rich forages is associated with an improvement of host resistance and/or resilience. Significant reductions in egg excretion and/or worm burdens have been described in goats and sheep fed with tannin-rich forages given fresh, or as hays or silages from Sericea lespedeza

(Lespedeza cuneata) (Lange et al., 2006; Shaik et al., 2006), sainfoin

0020-7519/S34.00 © 2007 Australian Society for Parasitology Inc. Published by Elsevier Ltd. All rights reserved. -33 05 61 19 32 43.

784 S. Brunei et al. I International Journal for Parasitology 38 (2008) 783-790 (Onobrychis viciifolia) (Paolini et al., 2005a; Hoste et al., 2005;

Heckendorn et al., 2006, 2007) or sulla (Hedysarium coronarium) (Niezen et al., 1995). However, variations have been reported in the anthelmintic effects depending on the plant (nature and/or tannin content), the host and/or the parasite, i.e., the parasite species or the stage (infective larvae or adult worms) (Min and Hart, 2002; Hoste et al., 2006).

In the life-cycle of trichostrongyle nematodes, the first phase of host invasion by the L3s is represented by larval establishment. This is a key process in parasitism which is composed by two successive steps: (i) larval exsheath-ment and (ii) larval penetration into the digestive mucosa. Tissue penetration is achieved by an initial close association of the exsheathed larvae with the mucosal tissue followed by invasion of the tissue.

Previous studies have shown that utilisation of tannin-rich resources (i.e., forages or quebracho) concomitant with infection with L3s was associated with a significant decrease in larval establishment in goats infected with Tela-dorsagia circumcincta and

Trichostrongylus colubriformis (Paolini et al., 2003) but no

significant effect was observed in goats infected with Haemonchus

contortus (Paolini et al., 2005b).

In sheep, a reduction in in vivo establishment of L3s of H.

contortus has been shown in sheep fed with Sericea lespedeza (Lange

et al., 2006). Similarly, a decrease in establishment has been detected for T. circumcincta larvae in goats fed with chicory (Cichorium

intybus) or sulla {H. coronarium) (Tzamaloukas et al., 2005).

However, in these previous studies, the mode of action of tannins on the two successive processes comprising larval establishment was not analysed. In some recent studies, we specifically examined the effects of tannins on the larval exsheathment process. The incubation of larvae with tanni-niferous plant extracts (Bahuaud et al., 2006; Brunet et al., 2007) or with monomers of condensed tannins (Brunet and Hoste, 2006) was associated with a delay or an inhibition of the larval exsheathment of H. contortus and T. colubriformis associated with a dose-dependent response. These in vitro inhibitory effects were verified in vivo for H. contortus larvae by Brunet et al. (2007) on cannulated sheep fed with sainfoin. Altogether, these data suggest that the invasion process could be impaired at its earliest level through failure of larvae to exsheath due to the presence of tannins in the digestive environment.

In order to improve our understanding of the mechanisms of action of tannins on larval establishment, the current study examined their possible effects on the digestive mucosal association/penetration using exsheathed H. contortus larvae in an in vitro direct challenge method (IVDC) (Jackson et al., 2004).

The objectives of the current study were: (i) to test the hypothesis that sainfoin extract might affect the association of H. contortus larvae with abomasal tissue; (ii) to verify wether this is a dose- dependent phenomenon; (iii) to define the potential role of tannins in the observed effect by pre-

incubating sainfoin extract with polyvinyl polypyrrolidone (PVPP), an inhibitor of tannins; (iv) to verify the specificity of the effects by comparing data obtained on H. contortus and T. circumcincta larvae; and (v) to investigate the role of the nature of tannins by incubating larvae with monomers of condensed tannins (CTs).

2. Materials and methods

2.1. Sainfoin extract

Sainfoin (O. viciifolia) was cultivated and harvested for hay in the south-east of France in June, 2005. The method of extraction has been described previously (Brunet et al., 2007). Briefly, dried hay (500 g) was extracted with ace-tone:water (2 x 3 L; 70:30 v/v) for 24 h. The acetone was removed and the aqueous solution was washed with meth-ylene chloride (4 x 500 mL). Finally, the aqueous extract was frozen and lyophilised to obtain a dry ground extract. The tannin content of the sainfoin hay was measured according to the method of the European Pharmacopea (European Pharmacopea, 2001. Determination of tannins in herbal drugs, pp. 107) and was estimated at 3.2% of the dry matter.

2.2. Animals

Five, 5- to 6-month-old goats were used in the study. These goats had been kept indoors since birth and were therefore naive to nematode infection. One goat was used per experiment. After slaughtering, the abomasum was removed immediately, opened, the luminal contents collected and the mucosa was washed gently twice with physiological saline solution (0.9%) at 37 °C to remove any adherent digesta. The experimental design, conditions of maintenance and slaughtering of the animals were performed in accordance with the animal research ethical code.

2.3. Nematode larvae

The H. contortus L3s were obtained from donor goats infected with pure strains of H. contortus. The same batch of 1- to 2-month- old larvae was used in the IVDC assays.

The T. circumcincta L3s were obtained from donor sheep infected with pure strains of T. circumcincta. The same batch of 2- to 3- month-old larvae was used in the IVDC assays.

2.4. Larval exsheathment

H. contortus and T. circumcincta L3s were artificially exsheathed

using a 5% solution of sodium hypochlorite (2%w/v) and sodium chloride (16.5% w/v). The exsheathment process was performed at 37 °C for 15 min and 100% exsheathment was verified by microscopic observations before washing. The exsheathed larvae were washed

S. Brunei et al. I International Journal for Parasitology 38 (2008) 783-790

785

three times in PBS (0.1 M phosphate, 0.05 M NaCl, pH 7.2) and used immediately after washing.

2.5. Larval incubation

For each experiment, the exsheathed larvae were incubated for 3 h at 20 °C either in the experimental medium or, in the case of the negative controls, in PBS. After incubation, the larvae were centrifuged and washed three times in PBS.

Experiment 1. Exsheathed larvae of H. contortus were incubated

with sainfoin extract in PBS at a concentration of 1,200 (ig/mL (S-1200). Six replicates were run per treatment (PBS or S-1200). PBS was used in the assay to minimise the risk of any non- specific effects that might occur if there were pH changes.

Experiment 2. Exsheathed larvae of H. contortus were incubated

with sainfoin extract at concentrations of 300, 600 and 1,200 (ig/mL in PBS. Five replicates were used for each treatment (PBS; S-300; S-600; S-1200).

Experiment 3. The aim of this experiment was to determine the

role played by tannins in interfering with larval association with the fundic mucosa. Based on gravimetric methods, PVPP is deemed to be the most appropriate inhibitor of tannins. PVPP is able to bind to tannins in the sainfoin extract and the resulting PVPP-tannin complexes were removed by centrifugation (Makkar et al., 1995; Makkar, 2006). A 1,200 (ig/mL sainfoin extract was pre-incubated overnight at 4 °C with PVPP, at a PVPP:extract ratio of 50:1, in order to deplete the extract in tannins (S1200- PVPP). A non-treated 1,200 (ig/mL sainfoin extract (S-1200) and PBS sample were maintained under similar conditions to provide control material. The next day, the two sainfoin extracts (with or without PVPP) and PBS were centrifuged. The superna-tants were immediately removed and used to incubate the exsheathed larvae, as previously described. Six replicates were run per treatment (PBS; S-1200; S1200-PVPP).

Experiment 4. In order to investigate parasite specificity of the

effects, experiment 3 was repeated using exsheathed T.

circumcincta larvae. Six replicates were run per treatment (PBS;

S-1200; S1200-PVPP).

Experiment 5. The aim of this experiment was to investigate

whether the nature of the flavan-3-ols, i.e., the monomers of CTs, might influence larval association with the mucosa. Flavan-3-ols differ in their biochemical structure by the number of phenolic groups on the B-ring, the 2,3 stereochemistry of the C-ring and the presence of a galloyl group, attached to the C-ring (Fig. 1). There are two main classes of CTs: the procy-anidin polymers (PCs) and the prodelphinidin polymers (PDs). In the PD monomers, a third phenolic group is added on the B-ring compared with PC monomers. The PCs comprise catechin, epicatechin or their galloyl

derivatives and the PDs comprise gallocatechin, epigal-locatechin or their galloyl derivatives (Fig. 1). The IVDC assay was used to examine the effects of different CT monomers, i.e., catechin (C), epicatechin (EC), gallocatechin (GC) and epigallocatechin (EGC), which are the flavan-3-ols naturally present in sainfoin (Kou-pai- Abyazani et al., 1993; Marais et al., 2000; Barrau et al., 2005), and two flavan-3-ol gallates: epicatechin gallate (ECg) and epigallocatechin gallate (EGCg). The commercial monomers used were obtained from Sigma, St. Louis, Mo, USA. Exsheathed H.

contortus larvae were incubated with one of the six molecules at

the concentration of 150 (ig/mL in PBS. Four replicates were run per molecule tested (C; EC; GC; EGC; ECg; EGCg). In parallel, six replicates were run for the negative control (PBS).

2.6. IVDC

The IVDC assay used in this study was based on the tissue explant method developed by Jackson et al. (2004) with T.

circumcincta. In the present study, the IVDC method was used with

exsheathed larvae of H. contortus and T. circumcincta (Section 2.4). One piece of tissue was excised from the fundic region of the goat abomasum (approximately 4 cm2_{), and was placed in each well of a}

six-well plate (Becton Dickinson and Co., USA). A cylinder was placed on the centre of each explant to provide an isolation chamber which contained the exsheathed larvae and kept those in contact with the mucosa. A challenge solution (approximately 1 mL) composed of Hank's medium (20 mL; Sigma, USA), 1M HEPES buffer (2mL; Sigma, USA), phenol red (0.2 mL; Sigma, USA) and completed with sterile water to 100 mL (Jackson et al., 2004) at 37 °C was added around the explants and care was taken not to submerge those. Approximately 1,500 exsheathed larvae were deposited in each isolation chamber (in

786 S. Brunei et al. I International Journal for Parasitology 38 (2008) 783-790

0.5 mL of the 37 °C PBS solution) (Section 2.5). Sufficient pressure was applied on the cover of the six-well plates, and indirectly on the isolation chambers, to ensure an effective seal between the isolation chambers and the tissue samples. Since the isolation chambers were longer than the depth of the six-well plates, the lids of the plates were raised to maintain the explants in an oxygen-rich environment. The six-well plates were incubated in darkness at 37 °C for 3 h. The time from animal slaughter to the incubation of the plates never exceeded 40 min.

After 3 h, the six-well plates were removed from the 37 °C incubator. The isolation chambers were retrieved and vigorously washed in 35 mL of 0.9% physiological saline solution into centrifuge tubes (tubes 'Wl'). The wells were washed with physiological saline solution and the washings added to the tube 'Wl'. The tissue explants were vigorously washed in 35 mL of 0.9% physiological saline solution (tubes 'W2') to collect larvae which had not become closely associated with the mucosa. The tissue was then transferred to a third centrifuge tube (tubes 'Dig') containing 35 mL of pepsin/HCl solution (1% pepsin (Sigma, USA) and 1% HC1 solution (Sigma, USA)) to digest the tissue and recover the larvae that had become closely associated with the mucosa. The tissue was incubated in the pepsin solution overnight at 37 °C. All tube contents were fixed with 0.5 mL of an iodine solution (4 g potassium iodide; 2 g sublimed iodine; in 100 mL distilled water). The digest contents were counted immediately and the washings were placed at 4 °C until required.

To count the larvae, the volume of each tube was adjusted to 40 mL by adding physiological saline solution. The number of larvae present in a 4% aliquot was determined at lOx magnification under a stereomi-croscope. The percentage of larvae associated with the mucosa was calculated as the number of larvae obtained from the tissue digest divided by the total number of larvae obtained from the two washings and the digest, multiplied by 100.

2.7. Statistical analyses

In order to meet the parametric test assumption of normality, the percentage of larvae recovered from the 'Dig' tubes were subjected to arcsin ^J transformation. The transformed data were then subjected to a one-way analysis of variance with post-hoc comparisons using the Bonfer-roni test. In Section 3, the statistical differences are given by focusing on comparison with control values.

3. Results

Overall, in the four experiments with H. contortus, a mean of 50.19 ± 7.22% of the larvae had become associated with the fundic mucosae in the negative controls (PBS) (Fig. 2A and B, Fig. 3A and Fig. 4). For the T. cir-cumcincta experiment, a similar pattern was observed with 47.9 ± 6.49% of the exsheathed larvae associated with the fundic tissue (Fig. 3B).

Fig. 2. Percentage of Haemonchus contortus L3s recovered from the two washings, Wl (black bars) and W2 (grey bars) and from the digestion, Dig (white bars) of the fundic explants in experiment 1 (A) and in experiment 2 (B), after treatment (3 h larval incubation with PBS: negative control; or with sainfoin extract at 300 ug/mL: S-300; at 600 ug/mL: S-600; at 1,200 ug/mL: S-1200). Statistical significance in the digestions compared with negative control are indicated with asterisks (***P < 0.001).

S. Brunei et al. I International Journal for Parasitology 38 (2008) 783-790 787

3.1. Reduction of tissue association of H. contortus larvae after incubation with sainfoin extract

Compared with the negative control, the 3 h incubation of H.

contortus larvae with sainfoin extract (1,200 (ig/mL) led to a

significant (76.93%) reduction (7><0.01) in the

association of larvae with the fundic tissue (Fig. 2A). Similar results were found with the sainfoin extract at 1,200 (ig/ mL in experiments 2 and 3, which presented, respectively, 87.17% and 79.86% reduction in larval tissue association (Figs. 2A and 3B).

3.2. Dose-dependent effect of the sainfoin extract on the tissue association of H. contortus larvae

In experiment 2, the incubation with sainfoin extract at different concentrations led to a significant reduction (P < 0.05) in tissue association by H. contortus larvae (Fig. 2B). In addition, this appears to be a dose-dependent response.

At 300 (ig/mL (37.83% reduction), the sainfoin extract had no effect on the association of larvae with tissue. In contrast, at 600 (ig/mL (78.36% reduction) and 1,200 (ig/ mL (87.17% reduction), a significant difference (P < 0.001) in tissue association was observed between the negative control and treatment with sainfoin extract.

3.3. Role of the tannins in the sainfoin effects

In experiment 3, as observed in experiment 1, the sainfoin extract at 1,200 (ig/mL caused a significant (79.86%) reduction (P < 0.01) in larval association with tissue, compared with the PBS control (Fig. 3A). The Pre-treatment with PVPP (S1200-PVPP) removed the inhibitory effects observed with the untreated extract (S-1200) since no significant differences were observed between the controls (PBS) and the extract pre-treated with PVPP (SI200-PVPP).

3.4. Comparison results obtained with T. circumcincta larvae

In experiment 4, exposure to sainfoin extract (S-1200) produced a significant reduction (7><0.01) of 60.35% in

788 S. Brunei et al. I International Journal for Parasitology 38 (2008) 783-790

the numbers of T. circumcincta larvae associated with tissues compared with the PBS control (Fig 3B). Pre-incubation of the sainfoin extract with PVPP (S1200-PVPP) restored tissue association to control values since no significant differences were observed