INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

The inability of any presently available therapeutic regimen to eliminate Toxoplasma gondii from the infected host requires lifelong maintenance therapy for patients with human immunodeficiency virus-induced immunosuppression who develop reactivated toxoplasmosis. Standard treatment with pyrimethamine-sulfadiazine is often associated with significant toxicity, requiring discontinuation of the drugs (13, 16). The consequent need for new therapies led to evaluation of a number of known and newly developed drugs and drug combinations, which resulted in increasing recognition that combination therapies may be necessary to treat toxoplasmosis in this setting (5).

Clindamycin (CLI) is an alternative drug widely used as a single agent or combined with pyrimethamine (7, 8). It has remarkable but delayed in vitro anti-T. gondii activity, achieved at low drug concentrations (22). In addition to the well-established anti-T. gondii activity of CLI as a single agent in animal models of infection (1, 10, 19, 29), doses of CLI that were ineffective when used alone were shown to afford protection in acute murine toxoplasmosis if combined with rifabutin (5).

The impetus to combine CLI with the hydroxynaphthoquinone atovaquone (ATO) was provided by reports on the activity of ATO against both tachyzoites and bradyzoite-containing T. gondii cysts (2, 3, 12). In addition, simultaneous administration of ATO has been shown to enhance the anti-T. gondii activity of pyrimethamine and sulfadiazine (4), as well as that of rifabutin (5, 27). Moreover, it has recently been suggested that a combination of ATO and CLI could act synergistically on both asexual life stages of T. gondii (28). Based on the proposed mechanisms of action of the two drugsATO inhibits the parasite's mitochondrial function (23), and CLI, which inhibits protein synthesis on prokaryotic ribosomes (11), is thought to act on the T. gondii prokaryote-type plastid-like organelle (24)this hypothesis predicts that simultaneous administration of ATO and CLI would cause parasites resistant to ATO to differentiate to bradyzoites, which should be hypersensitive to clindamycin. An end point of this joint action may be parasite elimination.

Thus, we examined in an in vivo experimental model of acute toxoplasmosis whether the addition of ATO to CLI would be beneficial for its anti-T. gondii effect and, moreover, if this combination is capable of eliminating the parasite, as suggested.

(The results of this study were presented in part at the 9th European Congress of Clinical Microbiology and Infectious Diseases [ECCMID] in Berlin, Germany, 21 to 24 March 1999.)

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Mice. Female Swiss Webster mice (Medical Military Academy Animal Research Facility, Belgrade, Yugoslavia) weighing 18 to 20 g at the beginning of each experiment were used. Mice were housed six to a cage and offered drinking water ad libitum.

T. gondii. Tachyzoites of the virulent RH strain maintained through serial intraperitoneal (i.p.) passages were used. For experimental infections, tachyzoites were harvested from mouse peritoneal fluids 72 h postinfection and purified by centrifugation, cotton wool filtration, and needle extraction. The parasites were counted in a hemocytometer, and their numbers were adjusted to 2 × 10⁶/ml with saline. Suspensions were serially 10-fold diluted, and 0.5-ml aliquots of 2 × 10⁴/ml and 2 × 10²/ml dilutions were inoculated i.p. into fresh mice.

Drugs. CLI (CLI hydrochloride powder, lot 353YS; Upjohn Co.; supplied by Yusapharm, Belgrade, Yugoslavia) was administered at 25, 50, and 400 mg per kg of body weight per day.

Experimental design. Mice injected i.p. with 10² or 10⁴ parasites were arbitrarily assigned to one of the 12 treatment groups according to the treatment given, as follows: CLI alone at 25, 50, or 400 mg/kg/day, ATO alone at 5, 25, 50, or 100 mg/kg/day, or ATO plus CLI at 5 plus 25, 25 plus 25, 25 plus 50, 50 plus 50, and 100 plus 400 mg/kg/day. Each treatment group comprised 6 to 12 animals, depending on the experiment. Treatment was initiated 24 h following parasite inoculation and was continued for 14 consecutive days. A group of untreated animals served as the negative control. Since a preliminary experiment showed lethal events in mice treated with the combined drugs beyond the usual observation period for acute toxoplasmosis models of 4 weeks, we chose to continue the observation period to 7 weeks postinfection. Mouse survival was monitored daily. Peritoneal exudates of mice that succumbed were examined at random for the presence of T. gondii. All experiments were performed three times, with similar results, and the data shown represent their cumulative results.

Bioassay. Brains were homogenized, and 0.5-ml saline suspensions were inoculated into two fresh mice per sample, one each by the i.p. and intraesophageal routes. Mice were monitored daily over 4 weeks; peritoneal fluids of those succumbing were examined for the presence of T. gondii.

PCR. For DNA extraction, mouse brains and lungs were dilacerated and suspended in a lysis buffer containing 200 mM Tris-HCl, 1 mM EDTA, 0.5 M NaCl, 1% sodium dodecyl sulfate, and 5 mg of proteinase K/ml. The extraction procedure was conducted according to a classical protocol (18) including two purification steps with phenol-chloroform-isoamyl alcohol (25:24:1). The DNA obtained was precipitated with cold ethanol, and after centrifugation, DNA pellets were washed in 70% alcohol, centrifuged, and resuspended in sterile water. The DNA concentration was determined by spectroscopy (Pharmacia Biotech, Saint-Quentin en Yvelines, France), and 1 µg of DNA was used for PCR. The gene target was a 129-bp sequence from the 35-fold-repeated B1 gene (6), amplified by using the primers 5'-CCGCCTCCTTCGTCCGTCGTA and 5'-TGAAGAGGAAACAGGTGGTCG. The amplification was carried out in a 50-µl final volume by using the PCR enzyme-linked immunosorbent assay (ELISA) DIG Labeling Plus kit (Boehringer Mannheim, Meylan, France) yielding digoxigenin (DIG)-labeled PCR products. Negative controls were brain DNA of a noninfected mouse and leukocyte DNA of a noninfected patient, and positive controls were brain DNA of a mouse infected with a chronic strain of T. gondii isolated from the amniotic fluid of a congenitally infected human fetus and two quantitated controls (i.e., low [1 to 10 parasites] and high [100 parasites]) consisting of T. gondii RH strain DNA from tachyzoites obtained by peritoneal lavage of a mouse infected 4 days earlier. Following 35 amplification cycles, PCR products were detected by an adapted protocol supplied with the PCR ELISA DIG Detection kit (Boehringer Mannheim) by using a biotin-labeled probe (5'-GCAAGAGAAGTATTTGAGGTC) and a detection system with streptavidin-coated microwells. Hybridized PCR products were revealed with an anti-DIG conjugate labeled with peroxidase, then incubated with the specific substrate. Appropriate technical controls were included (probe and substrate blank), according to the manufacturer's recommendations. The optical density (OD) was determined on an ELISA reader, and the threshold was obtained by a threefold multiplication of the OD of a negative patient. The technique was validated when all the controls were correct. In particular, special attention was given to the positivity of the low DNA control, corresponding to 1 to 10 parasites. The sensitivity was therefore estimated as between 1 and 10 parasites per sample. The specificity of this PCR technique was previously evaluated and determined to be 100% (26). Carryover contamination was prevented by use of uracil DNA glycosylase (17).

Statistical analysis. The rates of survival in particular treatment groups were estimated by the Kaplan-Meier product limit method and compared by the log-rank test (two curves) and multiple sample (three or more curves) tests. The level of statistical significance was 0.05.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

The effects of treatment with ATO and CLI combined in five different doses on the survival of mice infected with 10² or 10⁴ RH strain T. gondii tachyzoites were compared to those of the same drugs given alone. Peritoneal exudates of mice dying during the 7-week observation period were examined at random; T. gondii tachyzoites were seen in all.

Drug activity in infection model with 10² parasites. Untreated infected mice died between days 8 and 11 (mean ± standard deviation [SD], 9.4 ± 0.9 days), compared to which an effect of treatment was shown in all treatment groups (Fig. 1A). ATO alone was effective compared with no treatment, since as little as 5 mg/kg/day significantly prolonged time to death (P = 0.0000), and when it was given at 50 or 100 mg/kg/day, it resulted in 7-week survival of 18 or 13% of mice, respectively. CLI alone at 25, 50, or 400 mg/kg/day led to survival of 60, 67, or 86% of mice, respectively.

View larger version (36K): [in this window] [in a new window]   FIG. 1.   Survival rates of mice treated with ATO (5, 25, 50, or 100 mg/kg/day) and CLI (25, 50, or 400 mg/kg/day) alone or combined (5 plus 25, 25 plus 25, 25 plus 50, 50 plus 50, or 100 plus 400 mg/kg/day) in infection models induced by i.p. inoculation of 102 (A) or 104 (B) T. gondii RH strain tachyzoites. , control mice; , ATO; , CLI; , ATO plus CLI; *, P < 0.05 for ATO plus CLI versus ATO alone; x, P < 0.05 for ATO plus CLI versus CLI alone.

View larger version (26K): [in this window] [in a new window]   FIG. 2.   Estimated probabilities of survival (Kaplan-Meier product limit method) for mice treated with ATO (25 or 50 mg/kg/day) or CLI (25 or 50 mg/kg/day) alone or combined (25 plus 25, 25 plus 50, or 50 plus 50 mg/kg/day) in infection induced by i.p. inoculation of 102 (A) or 104 (B) T. gondii RH strain tachyzoites. Peroral treatment was initiated 24 h following challenge and continued for 14 days.

Drug activity in infection model with 10⁴ parasites. All untreated control mice infected with 10⁴ parasites died between days 7 and 9 (mean ± SD, 7.4 ± 0.6 days). As in infection caused with 10² parasites, an effect of treatment was shown in all treatment groups infected with this higher parasite load relative to the control animals (Fig. 1B), e.g., treatment with 5 mg of ATO/kg/day significantly prolonged survival (P = 0.0000). However, in contrast with results for animals infected with fewer parasites, treatment with both drugs combined was much more effective than treatment with either drug given alone.

Detection of residual infection. The results of the procedures performed to assess the presence of residual infection are presented in Table 1. Subinoculation in fresh mice of brain tissue of 4-week survivors infected with 10² and 10⁴ parasites treated with the 100-plus-400-mg/kg/day combination resulted in mortality due to acute toxoplasmosis of 75% (mean survival ± SD, 14.3 ± 4.2 days) and 50% (mean survival ± SD, 10 ± 2.8 days) of mice, respectively. Interestingly, lethal events were recorded following both i.p. and peroral inoculation of brain tissue, suggesting the presence of both the tachyzoite and bradyzoite life stages of the parasite in some survivors at that time. However, residual infection was harbored by fewer long-term survivors from the respective groups of animals, since parasite DNA was detected in 50 and 0% of brains 7 weeks postinfection. At this time, parasite DNA was detected in 50 and 100% of animals treated with lower doses of the combined drugs. No parasite DNA was detected in any of the lungs examined.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

This is the first report to date on the anti-T. gondii activity of CLI combined with ATO. The results presented demonstrate that the combination of CLI and ATO acts synergistically against T. gondii, since the combined drugs exhibit a significantly higher activity than that expected from the simple additive effects of both drugs. In our murine model of RH strain-induced acute toxoplasmosis, the effect of CLI and ATO was both drug dose and parasite inoculum dependent. The synergistic effect was remarkable in infections with 10⁴ parasites, in which 14-day treatment with combinations of 25 and 50 mg of CLI and ATO/kg/day protected 20 to 78% of animals by week 7, whereas the same doses of either CLI or ATO alone could not prevent mortality. Moreover, the effect of a combination of 50 mg of CLI/kg/day with either 25 or 50 mg of ATO/kg/day was similar to the effect of 400 mg of CLI/kg/day alone. On the other hand, in infections with 10² parasites, which were well controlled by any dose of CLI used, the combinations of ATO and CLI exhibited an additive effect compared to that of either drug alone.

It is difficult to put these results in perspective by comparing them with the efficacy of these or other drugs, alone or in combination, obtained in other animal models, since the infection and treatment protocol characteristics vary widely. One important characteristic of our model is the length of the observation period, since, having observed lethal events following the standard observation period of 4 weeks, we continued to monitor survival to 7 weeks following infection.

This late mortality (between weeks 4 and 7 postinfection) may explain why less residual infection is detected in 7-week than in 4-week survivors. The absence of residual infection, determined by a sensitive PCR method, in the brains of four of eight 7-week survivors (and in the lungs of four of four) treated with ATO and CLI at various dose combinations demonstrates the potential of the combined drugs to eliminate the parasite. This potential is emphasized by the fact that the mere presence of T. gondii DNA does not necessarily indicate the presence of viable parasites able to induce infection. If it does, it implies immune control of infection. Thus, in the treated, apparently healthy mice in which T. gondii DNA was detected, drug treatment, by reducing the initial parasite burden, may have provided the time for protective immunity against an otherwise lethal parasite infection to develop; a delayed specific T-cell response has been associated with treatment (20).

The results presented show ATO and CLI to be a promising drug combination with the potential to eliminate the parasite, therefore warranting further investigation. The low doses used to obtain high protection and cure rates may allow prolonged administration; 4- and 8-week courses of ATO were shown to be effective in chronic murine toxoplasmosis (3, 9), and we have previously shown that the effect of CLI critically depends on treatment duration (21). Furthermore, adverse side effects of CLI (7, 8) may be reduced by the use of lower doses, and ATO is generally well tolerated (14, 15). However, to appreciate the potential of the combination of CLI and ATO in the treatment of human disease, studies of the levels of the drugs achieved in serum and tissues with the regimens used, relative to the doses feasible in human therapy, are needed. Nevertheless, while one should bear in mind the limitations of extrapolating data from animal models to the human situation, the observation that at both 25 and 50 mg of CLI/kg/day an increase in the dose of ATO significantly enhanced the effect of the combined drugs may offer direction for future clinical trials of the anti-T. gondii potential of this drug combination.