INTRODUCTION

Top Abstract Introduction Materials and methods Results Discussion References

Leishmania donovani is the major causative agent of visceral leishmaniasis. It exists either as the extracellular promastigote found in the alimentary tract of sandflies or as the obligatory intracellular amastigote found in the phagolysosomes of mammalian macrophages (7, 8). Historically, promastigotes have been readily cultured in cell-free media, while amastigotes were maintained either in animals or in macrophage cell lines. Thus, most of the in vitro studies on leishmanial cell biology as well as antileishmanial drug action or resistance have been performed with promastigotes. During the last few years, several laboratories have succeeded in culturing L. donovani amastigotes axenically (12, 13, 23, 30, 33). This technique allows the direct evaluation of cell biological and biochemical processes in the amastigote, which is devoid of the host macrophage.

The treatment of choice of human visceral leishmaniasis is the administration of a pentavalent antimony (SbV)-containing drug, sodium stibogluconate (Pentostam; The Wellcome Foundation Ltd., London, United Kingdom) or meglumine antimoniate (Glucantime; Rhone-Poulenc, Paris, France). Despite the extensive use of these compounds over the last decades, their mechanism of action remains unclear. Furthermore, clinical resistance to these drugs is being increasingly reported (3, 15, 16, 21, 25, 27).

On the basis of data showing that trivalent antimony (SbIII) is much more toxic to Leishmania than SbV (5, 9, 28), it has been hypothesized that SbV is accumulated by the macrophage and is reduced to SbIII in the phagolysosome, the site where the antileishmanial activity of antimony occurs (28). Recent studies have suggested that at least in the case of New World species, this hypothesis is not applicable because axenic amastigotes of Leishmania mexicana are as susceptible to SbV as intracellular amastigotes (6). An additional possibility is that SbV is directly and specifically toxic to amastigotes (14).

The host and parasite roles in these phenomena have yet to be defined. Cross-resistance between SbIII and SbV has been shown in nonpathogenic Leishmania tarentolae (11, 18), in which trypanothione plays a major role in parasite resistance to antimony. SbV is reduced to SbIII, which then forms complexes with trypanothione (10, 17, 26). However, this mechanism is apparently different in Leishmania spp. pathogenic to humans (24).

To date, Pentostam-based studies have only indirectly assessed the antileishmanial activity of SbV because of two problems: chlorocresol toxicity to both macrophage and parasite and the fact that amastigotes were studied in macrophage cell culture (21, 28, 29). In order to delineate the direct effect of SbV on both stages of L. donovani, the activities of sodium stibogluconate (without chlorocresol) and meglumine antimoniate were studied with an axenic culture system (30).

	MATERIALS AND METHODS

Top Abstract Introduction Materials and methods Results Discussion References

Materials. Meglumine antimoniate was a gift from Rhone-Poulenc. Sodium stibogluconate was a gift from The Wellcome Foundation Ltd. Ornithine, leupeptine, and E64 were obtained from Sigma Chemical Co. (St. Louis, Mo.). Medium 199 and fetal calf serum were obtained from Biological Industries, Inc. (Beit Haemek, Israel). All other chemicals were of analytical grade.

Parasites. A cloned line of L. donovani 1SR (13, 30) and the Pentostam-resistant mutant L. donovani Ld1S.20 (14) were used.

Cell culture. Promastigotes were grown at 26°C in medium 199 supplemented with 10% fetal calf serum. In vitro culture of amastigotes was performed as described by Saar et al. (30).

Assessment of effect of pH and temperature on Leishmania susceptibility to meglumine antimoniate. The effect of acidic pH was determined by growing promastigotes at a pH of 5.5 and a temperature of 26°C and assessing their susceptibility to meglumine antimoniate, as described previously (14), periodically for up to 2 weeks. The effect of temperature was determined by growing promastigotes at a temperature of 37°C and a pH of 7.4 and assessing their susceptibility to meglumine antimoniate during a similar period.

Enzymatic assay of ODC activity. Quantification of organisms was accomplished by measuring ornithine decarboxylase (ODC) activity and calculating relative cell density as described previously (14). In control experiments, the levels of ODC in wild-type (1SR) and mutant (Ld1S.20) L. donovani strains were compared and found to be the same in both promastigotes and axenic amastigotes.

	RESULTS

Top Abstract Introduction Materials and methods Results Discussion References

The stage specificity of L. donovani susceptibility to SbV in the form of sodium stibogluconate was evaluated. Figure 1A shows the dose-response of L. donovani to sodium stibogluconate (free of the preservative chlorocresol). Amastigotes are susceptible (50% inhibitory concentration [IC₅₀], 19.3 ± 2.3 µg of SbV/ml). For promastigotes, in contrast, the IC₅₀ is 5,230 ± 980 µg of SbV/ml, and promastigotes still show a 27% relative cell density at an SbV concentration of 10 mg of SbV/ml. As shown in Fig. 1B, the dose-response of L. donovani to SbV in the form of meglumine antimoniate resembles the parasite's response to sodium stibogluconate. Amastigotes are susceptible (IC₅₀, 80 ± 18 µg of SbV/ml). For promastigotes, in contrast, the IC₅₀ is 5,830 ± 1,000 µg of SbV/ml, and promastigotes still show a 38% relative cell density at an SbV concentration of 10 mg/ml.

View larger version (16K): [in this window] [in a new window]   FIG. 1.   Stage-specific susceptibility of L. donovani to SbV. Promastigotes () and amastigotes () were incubated in the presence of increasing concentrations of sodium stibogluconate (A) and meglumine antimoniate (B) for 48 h and were assayed for ODC activity as described in Materials and Methods. The results are expressed as means ± standard deviations (n = 6).

SbIII in the form of potassium antimonyl tartrate is more toxic than SbV to both promastigotes and amastigotes (28, 29). In order to elucidate the differential effects of antimony compounds (SbV and SbIII) on the two stages of L. donovani 1SR, the activity of SbIII was determined. As opposed to SbV, both axenic amastigotes and promastigotes are susceptible to low concentrations of SbIII (IC₅₀s, of 3.6 ± 0.29 and 13.0 ± 2.4 µg of SbIII/ml, respectively, a 3.5-fold difference as opposed to the 73- and 271-fold differences for meglumine antimoniate and sodium stibogluconate, respectively, described above; Fig. 2). These findings further emphasize the marked stage specificity of SbV as opposed to that of SbIII.

View larger version (14K): [in this window] [in a new window]   FIG. 2.   Dose-response curve for L. donovani to SbIII. Promastigotes () and amastigotes () were incubated in the presence of increasing concentrations of potassium antimonyl tartrate for 48 h and were assayed for ODC activity as described in Materials and Methods. The results are expressed as means ± standard deviations (n = 3).

In order to show that extrinsic conditions of the promastigote growth medium do not affect SbV activity, the independent effects of pH and temperature on the susceptibility of promastigotes to SbV were evaluated. Figure 3 shows that the resistance of promastigotes to SbV remains unchanged over a period of up to 2 weeks when they are subjected to changes of either pH or temperature alone. These data imply that extracellular reduction of SbV to SbIII under the experimental conditions described does not occur. Together with the data in Fig. 1, it seems that SbV possesses intrinsic antileishmanial activity which is stage specific.

View larger version (12K): [in this window] [in a new window]   FIG. 3.   Dose-response curve for L. donovani promastigotes and meglumine antimoniate: effects of pH and temperature over time. Promastigotes were maintained at 26°C and pH 7.4 (), at 37°C and pH 7.4 (), or at 26°C and pH 5.5 (). After ~2 weeks, the parasites were incubated and were assayed for ODC activity.

In order to show that SbV resistance is common to both commercial SbV preparations (sodium stibogluconate [Pentostam] and meglumine antimoniate [Glucantime]), the activity of SbV against L. donovani Ld1S.20, a Pentostam-resistant mutant of L. donovani 1SR, was assessed. Table 1 shows that this mutant is resistant to both sodium stibogluconate and meglumine antimoniate. Compared to the wild type, amastigotes of L. donovani Ld1S.20 are 64 and 19 times more resistant to SbV in the form of sodium stibogluconate and meglumine antimoniate, respectively.

Arsenite (As)-resistant mutants of L. tarentolae show cross-resistance to SbIII and SbV (5, 11). Furthermore, it was recently shown that sodium stibogluconate-resistant mutants of this species show cross-resistance to SbV, SbIII, and As (18). Given the previously mentioned difference in the mechanism of resistance between nonpathogenic L. tarentolae and pathogenic Leishmania species (L. major, L. donovani, and L. mexicana), the possibility of cross-resistance between SbV and SbIII was investigated in Pentostam-resistant L. donovani mutant Ld1S.20 (14). As shown in Table 1, both promastigotes and amastigotes of wild-type and Ld1S.20 L. donovani strains were equally susceptible to SbIII but not to SbV. This indicates that in L. donovani, resistance to SbV can occur by a mechanism that is not related to SbIII susceptibility.

The axenic amastigote culture system also allows the evaluation of changes in parasite susceptibility to SbV during the process of differentiation. Relative cell density was measured every 24 h for 6 to 7 days beginning with the induction of differentiation. In order to ensure good discrimination between susceptibility and resistance, concentrations of meglumine antimoniate highly toxic to amastigotes but essentially nontoxic to promastigotes were chosen, 1,000 and 2,000 µg of SbV/ml were used because even 1,000 µg of SbV/ml is a concentration well above the IC₉₀ for amastigotes (Fig. 1).

Figure 4 demonstrates that when promastigotes are transformed to amastigotes, the development of SbV susceptibility occurs within 4 to 5 days. Previous studies have shown that 4 to 5 days is required for the completion of differentiation of promastigotes to amastigotes (1, 2, 22, 30). These results indicate that the development of SbV susceptibility parallels the differentiation process.

View larger version (13K): [in this window] [in a new window]   FIG. 4.   Development of L. donovani susceptibility to meglumine antimoniate during in vitro transformation of promastigotes to amastigotes. Promastigotes were transformed to amastigotes as follows: mid-logarithmic-phase promastigotes were transferred to 37°C for 24 h and then shifted to pH 5.5 as described in Materials and Methods. Meglumine antimoniate dose-response measurements were performed daily over 5 additional days: 1 day at 37°C and pH 7.4 () and 1 day (), 2 days (), 3 days (), 4 days (), and 5 days () at 37°C and pH 5.5. The results are expressed as means ± standard deviations (n = 3).

In the process of transformation from amastigotes back to promastigotes, previous studies have shown that more than 24 and up to 48 h is required for full parasite differentiation (30). Figure 5 shows that resistance to SbV was acquired within 24 h of the onset of transformation, prior to the completion of full differentiation from amastigotes to promastigotes.

View larger version (12K): [in this window] [in a new window]   FIG. 5.   Loss of L. donovani susceptibility to meglumine antimoniate during in vitro transformation of amastigotes to promastigotes. Amastigotes () were transformed to promastigotes as follows: amastigotes were centrifuged, suspended in promastigote medium, and incubated at 26°C and pH 7.4 as described in Materials and Methods. Meglumine antimoniate dose-response measurements were performed daily for 5 additional days: day 1 (), day 2 (), day 3 (), and day 5 (). The results are expressed as means ± standard deviations (n = 3).

	DISCUSSION

Top Abstract Introduction Materials and methods Results Discussion References

Axenic amastigote and promastigote culture systems were used to evaluate the direct effects of SbV on the two developmental forms of L. donovani. Susceptibility to SbV is stage specific. Promastigotes are relatively resistant (meglumine antimoniate IC₅₀, 5,830 µg of SbV/ml; sodium stibogluconate IC₅₀, 5,230 µg of SbV/ml), although at high concentrations they are somewhat susceptible (27 and 38% growth with 10 mg of SbV/ml as sodium stibogluconate and meglumine antimoniate, respectively). In contrast, amastigotes are highly susceptible to both forms of SbV, with the meglumine antimoniate IC₅₀ being 73 times lower (80 µg SbV/ml) and the sodium stibogluconate IC₅₀ being 271 times lower (19.3 µg SbV/ml) for amastigotes than for promastigotes.

In agreement with previous data (12, 28) and as opposed to SbV, SbIII is highly toxic to both amastigotes and promastigotes. For both forms of the pathogenic species L. donovani, SbIII IC₅₀s are similar, with only a 3- to 4-fold difference between the IC₅₀s for promastigotes and amastigotes, whereas the difference for SbV is 73- to 271-fold, indicating that the activity of SbIII is not stage specific. Because these data are derived from experiments with axenic cultures, the difference in activities between SbV and SbIII is clearly macrophage independent. Furthermore, the possibility that SbV (either as sodium stibogluconate or as meglumine antimoniate) is reduced to SbIII by the growth medium has been ruled out. Therefore, we conclude that SbV enters the parasite cells and subsequently, either directly or after intracellular reduction to SbIII, exerts its antileishmanial effect.

To date, a direct effect of SbV on amastigotes has been shown, but it has not been fully characterized. For example, in L. mexicana, axenic amastigotes are 370 times more susceptible than promastigotes (6). It has been hypothesized that in vivo, SbV activity is indirect and results from macrophage-dependent reduction of minimally toxic SbV to highly toxic SbIII (4, 29). If this, in fact, occurs, it is probably not due to the spontaneous reduction of SbV in the acidic environment of the lysosome but might be via an enzymatic reaction. The data in Fig. 3 show that SbV exerts an effect on axenic amastigotes of L. donovani which is macrophage independent and exclude the possibility of spontaneous reduction since neither acidic pH nor elevated temperature alone resulted in increased toxicity of SbV to promastigotes. Hence, SbV is specifically toxic to Leishmania amastigotes in axenic culture. The possibility that enzymatic reduction of SbV to SbIII occurs in vivo by the macrophage cannot be excluded.

Roberts et al. (28) showed stage-specific differences in the accumulation of SbV in the form of sodium stibogluconate in Leishmania panamensis. In order to reach the IC₅₀, promastigotes needed to be exposed to a concentration ~90 times greater than that to which amastigotes were exposed. This difference was required to achieve equivalent intracellular concentrations of SbV at the IC₅₀. These data, as well as those presented here, suggest that differential SbV accumulation is a major factor in the determination of amastigote and promastigote susceptibility to SbV.

The kinetics of the development of SbV susceptibility was evaluated during the process of parasite differentiation. As shown both in vivo (7, 22) and in vitro (23, 30), full differentiation from promastigotes to amastigotes occurs over a period of 4 to 5 days. In axenic culture, the full differentiation of promastigotes to amastigotes requires changing both pH and temperature (1, 2, 12, 19, 23, 30, 31). When promastigotes are grown at pH 5.5 and 26°C or pH 7.4 and 37°C, parasite susceptibility to SbV does not develop. Only the combination of elevated temperature and acidic pH will result in full parasite differentiation, and only fully differentiated axenic amastigotes are susceptible to SbV.

Not all stage-specific functions of L. donovani require both pH and temperature changes. For example, the expression of proline transport systems is regulated by extracellular pH (32). Also, the expression of amastigote-specific heat shock protein 100 is regulated by temperature (20). It seems, therefore, that the phenotypic expression of SbV susceptibility can serve as a marker of complete parasite differentiation in vitro.

As shown in vitro, full differentiation from amastigotes to promastigotes occurs over a period of about 48 h (30). When axenic amastigotes are exposed to growth conditions that induce their differentiation to promastigotes, the loss of SbV susceptibility occurs by 24 h and thus is not dependent on completion of the transformation process.

A previously isolated Pentostam-resistant L. donovani mutant (Ld1S.20) is resistant to chlorocresol and, by inference, sodium stibogluconate (14). Current data show that this mutant is resistant to both sodium stibogluconate and meglumine antimoniate, indicating that resistance is related to the SbV component of both formulations. The results obtained with this mutant, isolated from promastigotes grown continuously in Pentostam (14), together with the data presented in Fig. 1, suggest that promastigotes are somewhat susceptible to SbV. These data also support the idea that L. donovani Ld1S.20 is resistant to both SbV and chlorocresol. Furthermore, no cross-resistance was observed between SbIII and SbV since promastigotes and amastigotes of both the wild-type strain 1SR and the mutant strain Ld1S.20 were similarly susceptible to SbIII.

Previous studies have shown that in nonpathogenic L. tarentolae, cross-resistance between SbIII and SbV exists (11, 18). However, this is not necessarily the case for L. donovani, in which both promastigotes and amastigotes of L. donovani Ld1S.20 remain highly susceptible to low concentrations of SbIII, despite their resistance to SbV. This further supports the hypothesis that mechanistic differences in drug susceptibility and resistance exist between pathogenic and nonpathogenic Leishmania strains. It is therefore possible that SbV and SbIII act differently on L. donovani than on nonpathogenic Leishmania, and furthermore, significant differences in parasite susceptibility and resistance patterns (quantitative if not qualitative) may exist between different species of pathogenic Leishmania, e.g., L. donovani and L. mexicana. Therefore, the relationship between the transport of SbV and SbIII in L. donovani must be further delineated.