In vitro activity of commercial formulation and active principle of trypanocidal drugs against bloodstream forms of Trypanosoma brucei gambiense

The in vitro trypanocidal activities of 4 commercial formulations Ornidyl®, Pentamidine isethionate®, Germanin® and Lampit® and their corresponding active principles (Dl- α -difluoromethylornithine, pentamidine isethionate, suramine and 5-nitrofuran) were compared against Trypanosoma brucei gambiense . Differences of minimum inhibitory concentration (MIC) were observed between Ornidyl® and Dl- α -difluoromethylornithine and between Lampit® and 5-nitrofuran. For RO 15 strain and the comparison of Ornidyl®/ DFMO, the MIC when using the commercial drug was more than twice the MIC value obtained with the active principle. For all 3 trypanosome strains, MICs were identical for Lampit® and 5-nitrofuran but the MIC with the commercial formulation was twice the MIC obtained with the active principle. The active principles, rather than commercial formulations, should be used for standardization of in vitro assay protocols.


INTRODUCTION
Several in vitro methods have been developed for studying the drug sensitivity of Trypanosoma brucei gambiense, the agent of the chronic form of Human African Trypanosomiasis (HAT) or sleeping sickness. The chemotherapy of HAT is based on few drugs (W.H.O., 1998). Cases of treatment failure have been reported in Central Africa (Ollivier and Legros, 2001) (Kaminsky and Brun, 1993). The active principle is a pure compound, and its use is probably more reliable than commercial formulations, especially for long-term or multicentric in vitro studies. For drugs obtained from commercial sources, storage and interference of excipients could influence the in vitro activity.
In the present study, a comparison of the use of commercial drugs versus active principles for in vitro drug sensitivity assay on T. b. gambiense is described. In our knowledge, this is the first study about such a comparison for Human African Trypanosomiasis. We tested 3 reference strains (STIB 894, STIB 891, RO15) exposed to Ornidyl®, Pentamidine isethionate®, Germanin® and Lampit® versus their corresponding active principles Dl-α-difluoromethylornithine (DFMO), pentamidine isethionate sodium salt, suramine sodium salt and 5-nitrofuran, respectively.

MATERIALS AND METHODS
Reference strains of T. b. gambiense STIB 894, STIB 891, RO15 were isolated in Omugo, northwest Uganda and kindly provided by the Swiss Tropical Institute in Basel, Switzerland (Matovu et al. 2001). Parasites were cultured at 37°C in a 5% CO 2 incubator. The medium used was a 1:1 mixture of RPMI 1640 and MEM, supplemented with 15% heat-inactivated human serum and 5% heat-inactivated fetal calf serum. Additional supplements were 1% 2-mercaptoethanol, 1% L-glutamine, and 1% of a mixture containing 1 mM sodium pyruvate, 0.5 mM hypoxanthine, 0.05 mM bathocuproine sulfonate, 1.5 mM L-cysteine, as previously described (Hirumi and Hirumi, 1994). Cultures were monitored daily with medium change in order to maintain the trypanosomes in the exponential growth phase.
The long-term viability assay was used to determine the MIC for both the active principle and the corresponding commercial formulation in the same 24-well plates (Kaminsky and Brun, 1993). The trypanosomes (10 5 parasites/ml) were exposed for 10 days to twofold serial drug dilutions ranging from 0.462 to 7.39 µg/ml DFMO and 0.5 to 8 µg/ml Ornidyl®, 0.2 to 3.2 µg/ml for both suramine sodium salt and Germanin®, 0.0005 to 0.008 µg/ml for both Pentamidine isethionate® and pentamidine isethionate sodium salt, and 0.625 to 10 µg/ml for both Lampit® and 5-nitrofuran. Cultures were monitored and evaluated daily with appropriate medium replacement and addition of fresh drug or active principle every 48 h. The MIC was defined as the lowest concentration at which no Likeufack et al. 475 trypanosomes of normal morphology and motility could be detected microscopically. Each test was repeated 6 times for each comparison of drug pair.

RESULTS AND DISCUSSION
The MIC was identical at each repetition of test (6) for a given strain and a given active principle or commercial drug. Results are summarized in Table 1. For the 3 trypanosome strains, MICs of two drug pairs, Germanin®/suramine sodium salt and Pentamidine isethionate®/pentamidine isethionate sodium salt, were similar. Therefore, the purity of the drug powder and active principle seemed to be identical. For these two drug pairs, the sodium salt forms and commercial drug/active principle have identical molecular weights. For the comparison of Ornidyl®/DFMO, the MICs obtained for both STIB 894 and STIB 891 strains were 4 and 3.69 µg/ml, respectively. Using Ornidyl®, Matovu et al. (2001) found similar results within experimental error. For both STIB 894 and STIB 891 strains, the difference of molecular weight between DFMO (MW 219) and Ornidyl® (MW 237), which is due to the presence of a water molecule in the latter, could explain the slight difference observed when MIC was expressed in µg/ml but the MIC was identical (16.9 µM) when expressed as molar concentrations. However, for RO 15 strain, the MIC when using the commercial drug (4 µg/ml = 16.9 µM) was more than twice the MIC value obtained with the active principle (1.85 µg/ml = 8.45 µM). For this strain, Matovu et al. (2001) observed a similar MIC using Ornidyl® within experimental error. Despite the absence of excipients mentioned by the manufacturer and the high solubility of both commercial formulation and active principle in water, the possible reasons underlying the discordant result were not found. Commercial drug is formulated to deliver a stable and reliably absorbed compound. This may well require additives.
For all 3 trypanosome strains, MICs were identical for Lampit® (2.5 µg/ml) and 5-nitrofuran (1.25 µg/ml). Again, the MIC with the commercial formulation was twice the MIC obtained with the active principle. The active principle 5-nitrofuran was provided as micro-fine granules, while Lampit® is tablet. When we dissolved them in DMSO, a slight precipitation was observed with Lampit® but not with 5-nitrofuran. The solubility of Lampit® and 5-nitrofuran seemed to be different. Therefore, 5-nitrofuran may have a higher solubility than Lampit®, which could explain a higher in vitro trypanocidal activity. However, there is need to use sensitive and accurate methods (e.g. high pressure liquid chromatography) to investigate this problem of solubility.
Several hypotheses could explain the differences in MIC when using active principle and commercial formulation, such as solubility and interference of Afr. J. Biotechnol. excipients in the commercial formulations with the components of culture medium, such as serum. As far as we know, there is no previous study on the comparison of in vitro trypanocidal activities between different pharmaceutical formulations and their corresponding active principles. Although MIC determination was considered for many years as a reliable technique, these results must be confirmed using a more accurate technique such as 3H-hypoxanthine incorporation assay (Kaminsky and Brun, 1993). These preliminary results suggest that standardization of in vitro assay protocols is required and should be based on the use of active principles to ensure drug quality and purity that permits rationale and therefore to allow comparison of MIC between laboratories.