Interconnection between organellar functions, development and drug resistance in the protozoan parasite, Toxoplasma gondii

Inlernalional

Pergamon 0020-7519(95)00066-6

Journal

for

Parasitology, Vol. 25. No. 1 I, pp. 1293-1299, 1995 Copyright 0 1995 Australian Society for Parasitology Elsevier Science Ltd Printed in Great Britain. All rights reserved 002&7519/95 S9.50 + 0.00

Interconnection between Organellar Functions, Development and Drug Resistance in the Protozoan Parasite, Toxoplasma gondii STANISLAS

TOMAVO*

and JOHN C. BOOTHROYDT

Department of Microbiology and Immunology, Stanford University School of Medicine, Stanford CA 943055402, U.S.A.

Abstract-Tomavo S. & Boothroyd J. C. 1995. Interconnection between organellar functions, development and drug resistance in the protozoan parasite Toxoplasma gon&. International Journel for Parasitology 25: 1293-1299. The protozoan parasite Toxopla.ma gondii causes severe disease in animals and humans. In AIDS patients, for example, the encephalitis it produces is a major cause of death. Part of the very successful strategy adopted by the parasite centers on its ability to differentiate from the actively growing tachyzoite form to a chronic, almost latent state called the bradyzoite. The molecular signals and precise triggers involved in this differentiation process are not known. Drugs for treating toxoplasmosis are not capable of clearing the infection apparently because of their inability to eradicate the bradyzoites. Recently, as part of our efforts to understand the mode of action of a promising new drug, atovaquone, we have generated and analysed a mutant that is resistant to this drug. Surprisingly, we found that this mutant is predisposed to spontaneously differentiate from the tachyzoite to bradyzoite form in vitro (Tomavo & Boothroyd, submitted). Given that atovaquone is believed to act on the parasite mitochondria, we were interested to explore the relationship between mitochondrial function and differentiation. We find that atovaquone and a number of other drugs targeted to mitochondria wiU cause wild type parasites to differentiate from tachyzoites to bradyzoites suggesting some sort of adaptive response to a decrease in mitochondrial activities. The fact that atovaquone-resistant mutants are hypersensitive to clindamycin, a drug believed to work on the putative plastid of these parasites, suggests a model for how the mitochondrion and plastid interact and how they may be tied into the process and state of differentiation. This model is presented and discussed. Key words:

Toxoplusma;

INTRODUCTION is an obligate

mitochondria;

plastid;

differentiation;

protozoan that is extremely widespread in nature: 15-85% of the world’s population is infected and a variety of warm-blooded vertebrates can also serve as hosts. Traditionally, its major disease manifestation has Toxoplasma

gondii

intracellular

*Present address: Centre d’Immunologie et de Biologie Parasitaire, Inserm U. 415, Institut Pasteur de Lille, 1 Rue du Professeur Calmette, 59 045 Lille, France. tTo whom correspondence should be addressed. Fax: 415 723 6853. 1293

chemotherapy

been in the developing foetus where maternal infection during pregnancy can result in severe neurological disease and even death. Disease in healthy adults is generally mild to non-existent but in recent years T. gondii has emerged as a major opportunistic pathogen of immunocompromised individuals, especially those infected with the human immunodeficiency virus (Luft & Remington, 1992). The parasite generally exists in 2 forms in its nonfeline mammalian hosts: tachyzoites, the rapidly dividing form involved in acute infection, and the more slowly dividing bradyzoites which reproduce

1294

S. Tomavo and J. C. Boothroyd

within tissue cysts (Dubey, 1993). In immunocompetent individuals, tachyzoite multiplication is controlled by the immune system. Bradyzoites, however, can persist for life and are believed to be the origin of disseminated infection in AIDS patients; i.e., disease is the result of reactivation and uncontrolled growth of an infection acquired prior to the onset of AIDS. A major problem in treating the disease is that while drugs can clear the actively metabolizing tachyzoites, they typically have little effect on the bradyzoites which remain as a source of recrudescing infection (Luft & Remington, 1992). As a result, drug therapy must be maintained for the life of the patient. This is difficult, however, because of the toxicity associated with long term use of the current first-line drugs (especially sulfadiazine). There is, therefore, an urgent need for new and more effective drugs. Understanding the mechanism by which tachyzoite-bradyzoite interconversion occurs could reveal a way to influence the process. Blocking it would be invaluable in that it might prevent reactivation of the latent forms, thereby attenuating disease. Similarly, stimulating bradyzoites to differentiate back to the drug-sensitive tachyzoites could facilitate drug treatment that might completely clear the body of the Toxoplasma infection. One of the possible mechanisms for triggering the differentiation of tachyzoite to bradyzoite is suggested by our previous results (Tomavo & Boothstudying Toxoplasma mutants royd, submitted) resistant to atovaquone, an inhibitor of mitochondrial cytochrome bcl complex (Pfefferkorn, Borotz & Nothnagel, 1993). We found that at least one atovaquone-resistant mutant of Toxoplasma is predisposed to a bradyzoite pattern of gene expression suggesting that the mitochondrion may be involved in the differentiation process. We also found that this and other atovaquone-resistant mutants were hypersenstitive to clindamycin, an important therapeutic (Remington & Vilde, 1991; Filice & Pomeroy, 1991) which is thought to act on the putatitve plastid organelle of this Apicomplexan parasite (Pfefferkorn & Borotz, 1994). Like Plasmodium, Toxoplasma apparently possesses an organellar-like genome of -35 kb that is highly homologous to chloroplast DNA in plants, especially algae (Wilson et al., 1993). Although the actual sub-cellular origin of this DNA has yet to be determined in any of these parasites, candidate structures do exist (Dubremetz, in press) and the presence of a genome such as this clearly points to a novel plastid-like organelle. Our results with drug-sensitive mutants suggested that the mitochondrial and plastid functions were

interconnected and that the reliance of the parasite on the mitochondrial function differed in bradyzoites vs. tachyzoites. A corollary of this is that the parasite might respond to certain drug pressures by differentiating to a state that is less affected by that drug’s action. In this study, we present data confirming this prediction: inhibition of parasite mitochondrial functions specifically induces the differentiation of tachyzoites to bradyzoites in vitro consistent with the recently published results of Bohne, Heesemann & Gross (1994). A working model that synthesizes these results with the other observations already noted is presented for discussion. MATERIALS

AND

METHODS

Culture of Toxoplasma gondii. Unless otherwise noted, parasites (PLK strain; Sibley et al., 1992) were routinely maintained in monolayer cultures of human foreskin fibroblasts (HFF) in Dulbecco’s minimal essential medium (DMEM, Gibco) supplemented with 10% NU serum (Collaborative Biomedical Products) and incubated at 37°C with 5% COz. Growth of Toxoplasma gondii tested by [3H] uracil incorporation. Triplicate 2-cm2 wells containing confluent monolayers of HFF or Vero cells were infected with lo5 tachyzoites of PLK strain and labeled 48-60 h later using a 4 h pulse of [3H] uracil (1 uCi/ml). For drug sensitivity studies, drugs were added at the indicated concentration 6h after addition of the parasites to the host cell culture and a pulse labeling was performed 48860 h later, as above. rcss is defined as the concentration of a drug that inhibits 50% of the uracil incorporation. All drugs were obtained from Sigma Chemicals (St Louis, MI) except atovaquone which was obtained from Dr Mike Rogers, Burroughs Wellcome, North Carolina. Induction of bradyzoite gene expression in infected Vero cells. A confluent monolayer of Vero cells in a 25-cm* flask was infected with 5 x lo6 tachyzoites of the PLK strain. After 46 h, the infected cells were washed to eliminate parasites which did not invade and the medium was replaced by a fresh DMEM medium containing different concentrations of each mitochondrial inhibitor (see Results and Figure legends). After 2 days, infected cells were harvested and processed for polyacrylamide gel electrophoresis followed by western blot (see below). Induction of brndyzoite gene expression in humanfibroblast cells lacking mitochondrial DNA. 143B/260 cells were derived from the human fibroblast cell line 143B by a long term exposure to low concentrations of ethidium bromide (King & Attardi, 1989). These cells completely lack mitochondrial DNA, are unable to carry out cellular respiration and require pyruvate and uridine to grow. To test Toxoplasma growth and differentiation in these cell lines, a 25cm2 flask containing a confluent monolayer of 143B/260 or 143B cells grown in 110 ug/ml of pyruvate and 50 ug/ml of uridine were infected with 5 x lo6 tacbyzoites of the PLK strain. After invasion, the infected cells were treated with different inhibitors of mitochondrial functions and analysed as described above.

Drugs

and organelles

in Toxoplusma

1295

Fig. 1. Western blotting analysis of bradyzoite (PI8 and P36) and tachyzoite (SAG 1) specific surface antigens expressed by T. gondii PLK parasites untreated (control, lane 1) or treated with rotenone (lanes 2, 3 and 4 correspond to 0.01, 0.1 and 0.5 mM, respectively), antimycin A (lanes 5, 6 and 7: 1.5, 50 and 150 nM), myxothiazol (lanes 8, 9 and 10: 0.1, 1.0 and 5 ug/ml) and CCCP (lanes 11, 12 and 13: 2, 20 and 100 PM). SDS-polyacrylamide gel electrophoresis and western blots. Infected cells were lysed in SDS-PAGE sample buffer (Laemmli, 1970) and one tenth of the material was electrophoresed through a 12% SDS-polyacrylamide gel which was transferred onto nitrocellulose filters. Western blots were blocked with 5% non-fat milk for 30 min, probed with T8 4Al2 or T8 3B1, monoclonal antibodies (mAbs) specific for the bradyzoite surface proteins P36 and Pl8, respectively (Tomavo et al., 1991) or with DG52 (Buelow & Boothroyd, 1991), a mAb specific for the major tachyzoite surface antigen (P30 or SAGl), and then probed with a sheep anti-mouse whole immunoglobulin antibody conjugated to horseradish peroxidase (Cappel). Peroxidase activity was determined by the ECL detection system (Amersham). RESULTS

studies (Tomavo & Boothroyd, submutants that were mitted), we generated Toxoplasma resistant to atovaquone, an inhibitor of the mitochondrial bcr complex. Two of the 4 mutants were predisposed to differentiate from the tachyzoite to bradyzoite form in vitro under conditions where the wild type parent does not. Since atovaquone works through inhibition of mitochondrial electron transport, this finding suggested that either bradyzoites are less dependent on mitochondrial function than tachyzoites or that differentiation is coupled with an upregulation of mitochondrial function that was needed to confer resistance to atovaquone. In

previous

A prediction of both explanations is that wild type tachyzoites treated with atovaquone might differentiate to bradyzoites to escape the drug killing (i.e., as an adaptive response). To test this in preliminary experiments, PLK parasites in normal culture conditions were exposed to atovaquone (lo-’ M); such treatment resulted in induction of a bradyzoite pattern of gene expression, as predicted (data not shown). To exclude the possibility that this effect is due to some other, unknown activity of atovaquone, other drugs that affect mitochondrial electron transport were tested to see if they also stimulated differentiation in vitro. Thus, wild type cultures were treated with rotenone, antimycin, myxothiazol or carbonyl cyanide m-chlorophenyl-hydrazone (CCCP), initially at concentrations deduced from the literature to inhibit mitochondrial function. The results (Fig. 1) show that such treatment does indeed induce the appearance of the bradyzoite-specific markers, pl8 and ~36; the greatest induction was seen with myxothiazol at 0.1 ug/ml and 1 ug/ml (lanes 8 and 9). Note that at the highest concentrations of some of the drugs, the intensity of antigen detection dropped due to a dramatic drop in parasite numbers (i.e., due to the lethal effect of the drug). To address the specificity of this induction, drugs by interfering with non-mitothat kill Toxoplasma chondrial metabolic pathways were added to the

1296

S. Tomavo and J. C. Boothroyd

analyses: i.e., sulfadiazine, clindamycin and adenosine arabinoside (AraA). First, the approximate rcso was determined for each drug using the [3H]-uracil incorporation assay in vitro. [This assay is highly specific for the parasite because the host cell lacks the required salvage enzyme uracil phosphoribosyltransferase (Pfefferkorn, 1991).] The effect on bradyzoite gene expression was then determined at 3 concentrations of each drug, O.l-, l- and lo-fold the rc5a, 4860 h after addition of the drug to infected cultures. The data, summarized in Table 1, show that only the inhibitors of the mitochondrial functions are able to significantly induce the bradyzoite markers although AraA does cause a slight induction. As with any experiment dealing with an obligate intracellular parasite such as ToxopZusma, there is a chance that the effect is being mediated directly or indirectly via the host cell. In this case, it is important to show that the drugs are not working to induce the differentiation through inhibiting the bci complex of the host cell’s mitochondria. To control for this, we used the recently derived line of human fibroblasts that are completely deficient in mitochondrial electron transport (143B/206 cells) because of the absence of a mitochondrial genome (King & Attardi, 1989). First, however, it was necessary to check if the lack of a respiratory chain in the host cell would prevent parasite growth (i.e., even in the absence of drug). As shown in Fig. 2, the parasites do grow in the 143B/206 cells but at a reduced rate compared with the parental cell line (143B) cells). This difference was seen both by comparison of [3H] uracil incorporation (Fig. 2) and by direct counting of intracellular parasites released by forcing through a syringe (data not shown). This result indicates that

Fig. 2. Comparison of growth of T. gondii PLK within human fibroblast cells lacking mitochondrial DNA (mt DNA) (143B/206 cells, line A) or within the parental 143B cells possessing functional mitochondria (line B). [3H] uracil labelling was used to measure parasite growth. Each point is the average of 3 independent experiments (f S.D.) with each being determined in triplicate.

Table l--Effect of various concentrations of rotenone, antimycin, myxothiazol, CCCP, sulfadiazine, clindamycin and adenosine arabinoside on the expression of bradyzoitespecific proteins (~18 and ~36) by wild type PLK parasites Inhibitors

Go

Expression of bradyzoite specific proteins, ~18 and p36

Rotenone Myxothiazol Antimycin A Sulfadiazine Clindamycin Ara A

0.42 mM 0.1 ug ml 12.5 nM 10 uM 5 rig/ml 3W

++ +++ + +I-

the host cell mitochondria are important but not essential for normal growth of T. gondii. The relevance of this is further discussed below. No expression of bradyzoite marker proteins was observed when the parasites were grown in the 143B/ 206 host cells without drug (Fig. 3A, lane 2) suggesting that the induction seen with the mitochondria-inhibiting drugs in normal cells was due to an effect on the parasite’s rather than host cell’s mitochondria. Confirming this, when mitochondrial inhibitors were added to infected cultures of 143B/ 206 (Fig. 3A, lanes 335) the induction of a bradyzoite pattern of gene expression was essentially the same as seen with the 143B parental cell line (Fig. 3B, lanes 2, 3 and 4). Since the 143B/206 host cells have no mitochondrial respiration to inhibit, this result strongly suggests that the effect of the mitochondrialinhibiting drugs is on the parasite mitochondrion. DISCUSSION

The precise mechanism which mediates the interconversion of tachyzoite to bradyzoite is poorly understood. Since this is an important clinical concern in immunocompromised patients there is a serious need for understanding the mechanism that triggers this aspect of the life cycle of T. gondii. Differentiation of tachyzoite to bradyzoite is accompanied by a stage-specific alteration of gene expression: using monoclonal antibodies generated against both stages, common surface antigens or antigens exclusively expressed by each stage have been characterized (Kasper, 1989; Tomavo et al., 1991; Bohne, Heesemann & Gross, 1993a; Soete et al., 1993). Bohne, Heesemann & Gross (1993b) and Soete et al., (1994) have recently reported reliable and efficient methods for inducing the transformation of tachyzoite to bradyzoite in vitro by incubating the intracellular tachyzoites in either y-interferon or alkaline pH medium. Similarly, Gazzinelli et al. (1993) have shown that IFN gamma is able to inhibit the multiplication of tachyzoites in vivo and induces

Drugs and organelles in Toxoplasma

1297

A

B 4s I

8

1 I

23 1

t

4 I

*

*

Fig. 3. Western blotting analysis of bradyzoite-specific protein, P36, expression in parasites grown on 143B/206 cells or 143B cells. A. Growth in 143B/206 cells. Lane 1, control bradyzoites isolated from mouse brain; lanes 2, 3, 4 and 5: tachyzoites grown in untreated 143B/206 cells or cells treated with rotenone (0.1 mM), myxothiozol (1 pg/ml) and antimycin A (1.5 nM), respectively. B. Growth in 143B cells. Lanes 1, 2, 3 and 4 correspond to tachyzoites grown in 143B cells untreated or treated with rotenone, myxothiozol and antimycin A, respectively. Drug concentrations were as for panel A.

possibly by means of NO which is known as an inhibitor of iron-sulphur proteins involved in the mitochondrial respiration (Stamler, Singe1 & Loscalzo, 1992). Together with our data on atovaquone-resistant mutants, these data further suggest a connection between mitochondrial function and the differentiaton of tachyzoites to bradyzoites. In this paper, we have used 4 well characterized inhibitors of mitochondrial functions and have shown that they are able to induce the expression of bradyzoite specific proteins, p36 and ~18. The fact that AraA induces a slight expression of bradyzoite encystation,

proteins may indicate that the trigger of the switch is in the ATP level instead of an interference with de nova synthesis of pyrimidine which is also linked to mitochondrial metabolism in at least some Apicomplexa. The fact that the induction occurs in the human fibroblasts grown in excess uridine (as needed for the 143B/206 cells) further argues against the pyrimidine pathway being involved in the trigger. While this manuscript was in preparation, similar results using mitochondrial inhibitors were reported by Gross and colleagues (Bohne et al., 1994). While details differed, the same trends were observed:

1298

S. Tomavo

and J. C. Boothroyd

inhibition of mitochondrial function induces differentiation to the bradyzoite in vitro. The only notable difference was that the respiration-deficient host cell line used by these latter investigators sustained essentially normal growth of the parasite. In our hands, such host cells allow growth but at a reduced rate. This difference could be due to any of a number of variables, including the use of different parasite strains or medium. However, the key observation, that treatment of tachyzoites grown in such host cells still results in differentation to a bradyzoite pattern of gene expression was seen by both groups. Invasion of host cells by Toxoplasma involves the creation of a parasitophorous vacuole which is quickly surrounded by closely abutting host cell mitochondria. The mechanism by which this occurs and its function are not known. Using rhodamine 123, it has been reported that the mitochondria of intracellular parasites do not maintain a pH gradient and thus may not be fully functional (Tanabe & Murakami, 1984). This is difficult to reconcile with the effect of mitochondrial inhibitors on the induction of differentiation to the bradyzoite form. One possible explanation is that the parasite has a mitochondrial respiration which is difficult to detect using rhodamine 123. HYPOTHESIS

The results presented here and elsewhere suggest a model that may serve to stimulate our thinking about tachyzoite to bradyzoite differentiation and the interconnection of the parasite organelles to this process. We stress that this is only one of many alternative hypotheses but it appears to fit with the available data and has many testable predictions. Hence, we offer it for discussion and as a target for critical experiments that should help refine our understanding of this complex process. The model (Fig. 4) proposes that the mitochondrion and putative plastid have functional convergence in some unknown product “X” whose concentration is critical for optimal growth. “X” could be a key metabolite or something more general such as pH, ATP or calcium concentration. In tachyzoites, the concentration of X needed for optimal growth is provided equally from pathways operating in the mitochondrion and in the plastid. Hence, the model predicts that tachyzoites will be moderately sensitive to inhibition of either pathway. In bradyzoites, the model supposes that only 10% of the necessary concentration of X comes from the mitochondrion, with 90% being provided by the plastid (e.g., possibly due to a decreased need for X coupled with a reduction in mitochondrial production but no change in the amount made by the

plastid). This predicts that differentiation to bradyzoites will be accompanied by a marked decrease in sensitivity to drugs that act on the mitochondrion and an increase in sensitivity to drugs that work on the plastid (since the plastid is now almost the exclusive provider of “X”). The model is consistent with the data currently available: (i) tachyzoites are sensitive both to antimitochondrials and proposed anti-plastid drugs (such as clindamycin; Pfefferkorn, Nothnagel & Borotz, 1992); (ii) inhibition of mitochondrial activity stimulates differentiation to bradyzoites (Bohne et al., 1994 and this paper; i.e., by differentiating to a form not dependent on mitochondrial activity the parasite is able to escape the drug pressure); (iii) as for item ii, selection for atovaquone-resistant mutants results in some lines that grow as bradyzoites under normal tachyzoite culture conditions (Tomavo & Boothroyd, manuscript submitted); (iv) parasites resistant to atovaquone are hyper-sensitive to clindamycin (Tomavo & Boothroyd, submitted); (v) bradyzoites are less sensitive to the action of atovaquone than are tachyzoites (Araujo, Huskinson & Remington, 1991). Two key predictions now being tested are that wild type bradyzoites (induced by pH or other stress) should be hypersensitive to clindamycin and that atovaquone and clindamycin should show synergistic toxicity on tachyzoites. The results presented here and the model they suggest begin to shed light on possible inter-connecTachyzoites

Bradyzoites

t,L t I t t t ml tt X

X

50%

50%

t

mitoch.

‘t

plastid

10%

90%

0

mitoch.

plastid

Fig. 4. Model for interconnection between the mitochondrion (mitoch), the putative plastid and differentiation. See text for full details. Percentages contributed by each compartment are shown, relative to the amount necessary for maximal growth in that developmental stage. Product X is undefined and could be a metabolite, ATP, pH, Ca2+ or any other physiological function or entity.

Drugs

and organelles

tion of the two key organelles in this parasite. Assuming the model is correct, the challenge becomes finding the identity of “X” and determining why Apicomplexan parasites should need a plastid to produce it. Regardless of whether our model proves right or wrong, we hope it will help focus efforts to understand the contribution of 2 key organelles to parasite metabolism and the exact mechanism by which 2 key therapeutics (clindamycin and atovaquone) work. Acknowledgements-We are grateful to Faust0 Araujo, Elmer Pfefferkorn, Jack Remington, Joe Schwartzman and our colleagues in the lab for fruitful discussions. We thank Uwe Gross for exchange of information before publication and Drs Kami Kim and Frank Seeber for comments on the manuscript. This work was supported by an NCDDG and Center for Aids Resarch grants from the NIH (AI 30230 and AI 27762, respectively). REFERENCES Araujo F. G., Huskinson J. & Remington J. S. 1991. Remarkable in vitro and in viva activities of the hydroxynapthoquinone 566C80 against tachyzoites and tissue cysts of Toxoplasma gondii. Antimicrobial Agents and Chemotherapy 35: 293-299. Bohne W., Heesemann J. & Gross U. 1993a. Induction of bradyzoite-specific Toxoplasma gondii antigens in gamma interferon-treated mouse macrophages. Injection and Immunity 61: 1141-l 145. Bohne W., Heesemann J. & Gross U. 1993b. Coexistence of heterogeneous populations of Toxoplasma gondii parasites within parasitophorous vacuoles of murine macrophages as revealed by a bradyzoite-specific monoclonal antibody. Parasitology Research 79: 485487. Bohne W., Heesemann J. & Gross U. 1994. Reduced replication of Toxoplasma gondii is necessary for induction of bradyzoite-specific antigens: a possible role for nitric oxide in triggering stage conversion. Infection and Immunity 62: 1761-1767. Buelow R. & Boothroyd J. C. 1991. Protection ofmice from fatal Toxoplasma-gondii infection by immunization with p30 antigen in liposomes. Journal of Immunology 147: 34963500. Dubey J. P. 1993. Toxoplasma, Neospora, Sarcosystis, and other tissue cyst-forming coccidia of humans and animals. In: Parasitic Protozoa (Edited by Krier J.P.), Vol 6, pp. l-158. Academic Press, New York. Dubremetz J. F. in press. Toxoplasma gondii: Cell biology update. In: Molecular Approaches to Parasitology (Edited by Boothroyd J. C. and Komuniecki R.). Wiley-Liss, New York. Filice G. A. & Pomeroy C. 1991. Effect of clindamycin on pneumonia from reactivation of Toxoplasma gondii infection in mice. Antimicrobial Agents and Chemotherapy 35: 78&782. Gazzinelli R. T., Eltoum I., Wynn T. A. & Sher A. 1993. Acute cerebral toxoplasmosis is induced by in viva neutralization of TNF-alpha and correlates with down

in Toxoplasma

1299

regulated expression of inducible nitric oxide synthase and other markers of macrophage activation. Journal of Immunology 151: 3672-3681. Kasper L. H. 1989. Identification of stage-specific antigens of Toxoplasma gondii. Injection and Immunity 57: 6688 672. King M. P. & Attardi G. 1989. Human cells lacking mtDNA: repopulation with exogenous mitochondria by complementation. Science 246: 500-503. Laemmli U. K. 1970. Cleavage of structural proteins during assembly of bacteriophage T4. Nature London 277: 686685. Luft B. J. & Remington J. S. 1992. Toxoplasmic Encephalitis in AIDS. Clinical Infectious Diseases 15: 21 l-222. Pfefferkom E. R. 1991. Cell biology of Toxoplasma gondii. In: Cell Biology, Molecular Biology and Immunology of Parasites (Edited by Wyler D. J.), pp. 2650. W. H. Freeman Press, New York. Pfefferkorn E. R., Nothnagel R. F. & Borotz S. E. 1992. Parasiticidal effect of clindamycin on Toxoplasma gondii grown in cultured cells and selection of a drug-resistant mutant. Antimicrobial Agents and Chemotherapy, 36: 109ll1096. Pfefferkorn E. R., Borotz S. E. & Nothnagel R. F. 1993. Mutants of Toxoplasma gondii resistant to atovquone (566C80) or decoquinate. Journal of Parasitology, 79: 5599564. Pfefferkorn E. R. & Borotz S. E. 1994. Comparison of mutants of Toxoplasma gondii selected for resistance to azithromycin, spiramycin, or clindamycin. Antimicrobial Agents and Chemotherapy 38: 31-37. Remington J. S. & Vilde J. L. 1991. Clindamycin for Toxoplasma encephalitis in AIDS. Lancet 338: 1142-l 143. Sibley L. D., LeBlanc A., Pfefferkorn E. R. & Boothroyd J. C. 1992. Generation of a restriction fragment length polymorphism linkage map for Toxoplasma gondii. Gene 132: 1003-1015. Soete M., Fortier B., Camus D. & Dubremetz J.-.F. 1993. Toxoplasma gondii: Kinetics of bradyzoite-tachyzoite interconversion in vitro. Experimental Parasitology 76: 2599264. Soete M., Fortier B., Camus D. & Dubremetz J. F. 1994. Experimental induction of bradyzoite-specific antigen expression and cyst formation by the RH strain of Toxoplasma gondii in vitro. Experimental Parasitology 78: 361-370. Stamler J. S., Singe1 D. J. & Loscalzo J. 1992. Biochemistry of nitric oxide and its redox-activated forms. Science 258: 189881902. Tanabe K. & Murakami K. 1984. Reduction in the mitochondrial membrane potential of Toxoplasma gondii after invasion of host cells. Journal of Cell Science 70: 73-8 1. Tomavo S., Fortier B., Soete M., Ansel C., Camus D. & Dubremetz J.-F. 199 1. Characterization of bradyzoitespecific antigens of Toxoplasma gondii. Infection and Immunity 59: 375&3753. Wilson I., Gardner M., Kaveri R. & Williamson D. 1993. Extrachromosomal DNA in the Apicomplexa. NATO ASZ Series H78: 5160.