Ex Vivo Anthelmintic Activity of Albendazole-Sulphoxide Enantiomers
The antiparasitic activity of racemic albendazole-sulphoxide (Ricobendazole = racRBZ) and its (+) and (−) enantiomers was tested in an ex vivo murine model for Trichinella spiralis infection. Larvae were isolated from the muscle of infected mice and exposed to concentrations between 0.01 and 1 μg/ml of the racemic mixture or to each of its enantiomers. The activity of each compound was then assayed by measuring the ability of the treated larvae to infect naive mice (larval viability). At a concentration of 0.5 μg/ml, all 3 compounds were highly effective in reducing the viability of the larvae, achieving reductions of 91.26% (racRBZ), 96.7% (+), and 89.2% (−), when compared with untreated controls. At lower concentrations (0.1 μg/ml), only treatment with (+)RBZ rendered a significant reduction in larval viability in comparison with controls (84.3%; P < 0.01), whereas at 0.01 μg/ml, none of the compounds altered larval viability (P > 0.05).
Albendazole (ABZ), methyl (5-(propyl-thio)-1-H-benzimidazole-2yl) carbamate, is a broad-spectrum drug that acts against the most important animal and human helminth parasite species (Horton, 2000). After administration, ABZ is rapidly transformed by 2 distinct hepatic microsomal enzyme systems: the flavin-containing monooxygenase system (Lanusse et al., 1993) and the cytochrome P-450 chain (Souhaili-El Amri et al., 1987). These modifications produce albendazole-sulphoxide (ABZSO) and albendazole-sulphone, the main metabolites that can be recovered from the plasma of sheep (Lanusse et al., 1995), cattle (Sanchez et al., 1997), mice (Rueda-Polo et al., 1998), and humans (Prochanzkova et al., 2000). Because of their affinity for the parasite's beta- tubulin, both ABZ and ABZSO exhibit antiparasitic activity (Lacey, 1988). However, ABZ can be further metabolized and is thus only found at low levels in plasma. Hence, the ABZSO product is thought to be the more active of the 2 in combating tissue-dwelling parasites (Marriner and Bogan, 1980). The chiral core in the sulfur atom of ABZSO permits 2 enantiomers to be generated, (+)ABZSO and (−)ABZSO. The pharmacokinetic disposition of ABZSO enantiomers after oral administration of ABZ in distinct animal species and humans (Delatour, Benoit et al., 1991; Delatour, Garnier et al., 1991; Garcia et al., 1999) reveals differences that possibly reflect the selective metabolism of ABZ and ABZSO enantiomers (Marques et al., 1999; Solana et al., 2000; Virkel et al., 2002). Indeed, even sex-related differences in the pharmacokinetic behavior of ABZSO enantiomers have been documented in sheep (Capece et al., 2000). Furthermore, the selective uptake of ABZSO enantiomers and their selective binding to cytosolic proteins isolated from different helminth parasites has also been demonstrated (Alvarez et al., 2000; Solana et al., 2002). Thus, these factors must be taken into account because they may contribute significantly to the pharmacological properties of this chiral molecule.
However, despite the interest in these agents, the antiparasitic activity of the (+)ABZSO and (−)ABZSO enantiomers has never been directly tested. Using optimized chiral high-performance liquid chromatography (HPLC), we have been able to obtain very pure samples of (+)ABZSO and (−)ABZSO. This has enabled us to test the anthelmintic activity of these enantiomers in an ex vivo model system, using the well-established murine model of Trichinella spiralis infection.
To purify ricobendazole enantiomers from a racemic ABZSO mixture (Ricobendazole = RBZ), a modular liquid chromatography apparatus was used, equipped with: a Jasco PU-1580 isocratic pump (Tokyo, Japan); an automatic sampler (Gilson 231 XL, Villers le Bels, France) fitted to a 100-μl sampler loop (Rheodyne, Rohnert Park, California); a variable wavelength detector (UV-1575, Jasco); and a PC integrator (Borwin 1.5, JMBS developments [Jasco]). Before injection, the samples were filtered through a polyvinylidene fluoride Durapore® 0.45-μm filter (Millipore Co., Boston, Massachusetts). Subsequently, the racemic ricobendazole (racRBZ) sample mix was injected into the HPLC system at a concentration of 0.1 mg/ml, and chiral HPLC separation was performed according to a modified version of the method described previously by Delatour et al. (1990) (Garcia et al., 1999). Briefly, a chiral– alpha 1-acid glycoprotein column (100 × 4 mm, 5 μm) and a mobile phase containing sodium phosphate buffer (8 mM, pH 7.0), at a flow rate of 0.9 ml/min, were used. The samples were analyzed at 290 nm, and under these conditions, the retention times for (−) and (+)RBZ were 3.1 and 10.5 min, respectively. The liquid samples were collected in accordance with the retention times of the enantiomers. The samples were concentrated to dryness under vacuum at 70 C using a Savant Speedvac® concentrator (Holbrook, New York). The samples recovered were quantified by the same HPLC method described above but included 2-propanol (1 ml/L) as an additive in the mobile phase. Under these conditions, the retention times after injection were 2.6 min for (−)RBZ and 4.6 min for (+)RBZ. For use in biological assays, the concentrated samples were dissolved in acidified deionized water (pH 2) and then diluted to the appropriate concentrations in culture medium.
Trichinella. spiralis L1 larvae (MFEL/SP/62/GM-1 ISS48 strain; Trichinella Reference Centre, Istituto Superiore di Sanitá, Rome, Italy) were obtained from the carcasses of experimentally infected mice by standard artificial digestion. After isolation, the larvae were separated using a Baermann apparatus and then washed 10 times by passive sedimentation in phosphate-buffered saline containing added antibiotics. The larvae were then suspended in culture media (500 ± 50 larvae/ml), which consisted of a basic Hank's balanced salt solution (HBSS) with added antibiotics (50 units/ml penicillin and 50 μg/ml streptomycin; Sigma Chemical Co., St. Louis, Missouri). Under these conditions, 66.8% of incubated larvae retain their capacity to infect mice when compared with nonincubated controls (Table I). Interestingly, we noted that the addition of l-glutamine or N-2-hydroxythylpiperazine-N′-2-ethane-sulfonic acid buffer significantly affected further larval viability (data not shown).
To test the different compounds, 0.2 ml of the larval suspension was added to each well of a 24-well culture plate (Costar, Corning, Albany, New York) and mixed with 1.8 ml of either racRBZ or one or other of the purified enantiomer solutions. Different concentrations of each solution were tested in 6 wells of each plate, whereas, as controls, the larvae in the remaining 6 wells were exposed to medium alone. After 24 hr at 37 C in a conventional 5% CO2 incubator, the larvae were observed using an inverted light microscope. Thereafter, 1.5 ml of medium was carefully removed from each of the wells, and the larvae in the remaining 0.5 ml (100 ± 10) were used to orally infect 6 Swiss- CD-1, 8-wk-old male mice. As controls, another group of 6 age- matched mice was infected 24 hr beforehand with freshly isolated larvae. Five days after infection (6 days for the fresh larval controls), the mice were sacrificed with an excessive dose of chloroform anesthesia, and their small intestines were removed and opened longitudinally. Each intestine was placed on sanitary gauze and incubated in saline solution for 4 hr at 37 C. The tissues were removed, and the adult worms released were allowed to settle before being counted using a stereomicroscope. The mean number of worms from the animals infected with ex vivo–treated larvae was compared with those obtained from animals infected with incubated but untreated larvae using Student's t-test. Significance was set at P < 0.05 and P < 0.01.
After the 24-hr incubation in medium alone, there was a 33.9% reduction in larval viability, as seen in Table I. Whereas 66.8 worms were recovered from animals infected with fresh isolated larvae, this figure dropped to 44.1 when the animals were infected with control larva that had been incubated for 24 hr in medium alone. This reduction may reflect the noxious effects of maintaining the larvae in an atmosphere of 5% CO2 because an earlier study has shown that totally anaerobic conditions are necessary to retain the full capacity of infection of Trichinella larvae cultured in conventional cell culture media (Bolas-Fernandez, 2002). However, under these anaerobic conditions, larvae are not sensitive to drugs (probably because of a hypobiotic condition); therefore, some degree of aerobiosis is needed to be able to detect anthelmintic activity.
We tested concentrations of racRBZ ranging from 0.06 to 1 μg/ml. Visual observation of the larvae at the light microscopic level after incubation showed no difference between treated and untreated larvae at any of the concentrations analyzed. The larvae exhibited a semicoiled “serpentine” shape and were actively motile. However, in the larval viability assay, it was apparent that racRBZ was highly active in reducing the viability of the larva at concentrations of 0.25, 0.5, and 1 μg/ml. At these concentrations, the viability and capacity of infection of the larvae was reduced by 92.5–99.6%, when compared with untreated controls (Table I). Lower, but still significant, activity was recorded at 0.12 μg/ml (74.3%), whereas at 0.06 μg/ml, no significant reduction was observed (P > 0.05).
We compared the antilarval activity of racRBZ, (+)RBZ, and (−)RBZ at concentrations of 0.5 μg/ml (experiment 1) and 0.1 and 0.01 μg/ml (experiment 2). All 3 compounds were highly active at 0.5 μg/ ml, producing a significant reduction (P < 0.01) in larval viability when compared with controls (Table II). The (+)RBZ was still highly active at 0.1 μg/ml, producing a reduction of 84.4% in larval viability (P < 0.01), which was greater than that observed for racRBZ (69.3% reduction, P < 0.05). At this concentration, (−)RBZ did not produce a significant reduction in larval viability (30.1%). None of these 3 compounds was active at 0.01 μg/ml, provoking reductions of 0.00, 33.4, and 26.61% for racRBZ, (+)RBZ, and (−)RBZ, respectively (P > 0.05). Nevertheless, we did observe some evidence that (+)RBZ at 0.01 μg/ml may provoke a significant response, and this is being analyzed further (data not shown).
These results suggest that the (+) enantiomer of ABZSO is likely to be responsible for it's activity against helminth parasites. However, more definitive proof of this hypothesis would require the total separation of these enantiomers or their chemical synthesis, which is now in progress. Nevertheless, we believe that pending this achievement, it is of interest to advance these findings from our ex vivo model system of anthelmintic activity. Total purification of the enantiomers was not achieved in the HPLC separation because each contained ≈15% of the opposite isomeric moiety. Thus, it seems likely that the activity shown by (−)RBZ at 0.5 μg/ml may actually be due to the (+)RBZ contamination that was present (equivalent to 0. 075 μg/ml).
The suitability of our ex vivo model for testing anthelmintic activity is supported by the fact that helminths, unlike mammals, oxidize ABZ in a nonenantiomer-selective fashion (Solana et al., 2001). On the other hand, it seems unlikely that T. spiralis larvae might convert the racRBZ and the (+) or (−) enantiomers to ABZ, as has been shown for the cestode Moniezia expansa with ABZSO (Solana et al., 2001), because no activity was recorded for the (−) enantiomer. Although the selective retroconversion of (+)RBZ to ABZ in detriment of (−)RBZ cannot be ruled out, the weak antilarval activity of (−)RBZ would again be confirmed. As mentioned above, variability in the pharmacokinetic disposition of ABZ and its metabolites is well documented in different animal species. Thus, a few minutes after oral administration of ABZ, (−)ABZSO becomes the dominant enantiomer in mice (reaching > 60%), and by 1,400 min, the proportion of this enantiomer increases to 75% (García et al., 1999). This pharmacokinetic enantiomer selectivity may explain why high doses of ABZ are necessary to achieve an effect in mice (50–100 mg/kg/day; García et al., 2003), when taken into consideration along with the results obtained in our study. In contrast, in other species such as sheep and cattle (Fetterer, 1982), in which the (+) enantiomer is the dominant form (Delatour, Garnier et al., 1991), a dose of 5 mg/kg is effective for treatment of tissue-dwelling parasites. However, another reason may relate to the pharmacokinetic differences between monogastric and ruminant species, which may influence the solubility and absorption rates of ABZ. For instance, it has been demonstrated that the rumen may act a reservoir, prolonging the duration of benzimidazole absorption and its outflow down to the gastrointestinal tract in sheep (Hennessy et al., 1994), thus favoring a sustained action. Moreover, in addition to host pharmacokinetic variability, variation in the uptake and accumulation of ABZSO and its enantiomers by different helminth species has also been demonstrated (Alvarez et al., 2000, 2001), and this variation is also relevant to the efficiency of anthelmintic activity.
In summary, knowledge of pharmacokinetic behavior of ABZSO and its enantiomeric forms in production animals and in the parasites they host, together with the data provided in this study on pharmacological activity of these compounds, will be of use for optimizing chemotherapy of parasitic diseases.
We thank C. Picornell (Chemo Ibérica, Madrid, Spain) for kindly supplying us with racemic ricobendazole.