Drug sensitivities

The potential of the apicoplast as a drug target reflects its algal origin, with many proteins and pathways not shared by the human host. Furthermore, many apicoplast proteins are enzymes, which bind small molecules and hence have a higher possibility of being druggable (Hopkins and Groom, 2002). Indeed, several of the prokaryotic-like features of apicoplast functions can be inhibited by existing compounds. These have been important research tools, and some are important clinically as well. The ability of inhibitors of organellar DNA replication, transcription, and translation to kill T. gondii and P. falciparum indicate that organellar functions are essential. For example, the rifamycin S antibiotics, such as rifampicin and rifabutin, inhibit eubacterial RNA polymerases like those encoded by apicoplast genomes, and are active against both T. gondii and Plasmodium species (Alger et al., 1970; Divo et al., 1985; Strath et al., 1993; Araujo et al., 1994; Olliaro et al., 1994; Pukrittayakamee et al., 1994). Both the A and B subunits of P. falciparum DNA gyrase bear apicoplast targeting sequences (Khor et al., 2005), and inhibitors of DNA gyrase are toxic to T. gondii (Fichera and Roos, 1997).

T. gondii is sensitive to the macrolide antibiotic clindamycin, an inhibitor of prokaryotic-like translation, and a T. gondii mutant resistant to clindamycin was cross-resistant to the macrolide antibiotics azithromycin and spiramycin (Pfefferkorn and Borotz, 1994). Two other clindamycin-resistant T. gondii mutants were found to have a point mutation in the plastid large subunit rRNA that mapped close to known clindamycin specificity determinants (Camps et al., 2002). Macrolide interaction sites are known to be restricted to a small region of the ribosome's peptidyl transferase domain, and co-crystallizations of these drugs with ribosomes have shown that they all block the peptide exit tunnel (Hermann, 2005). Consequently, the observed cross-resistance is unsurprising.

Other inhibitors of prokaryotic-like translation include thiostrepton, which acts against P. falci-parum (McConkey et al., 1997; Clough et al., 1997; Rogers et al., 1997) but not T. gondii. The differential sensitivity is likely due to an alternate nucleotide at the critical position in the T. gondii apicoplast large subunit rRNA (Clough et al., 1997). The antibiotic actinonin inhibits the peptide deformylase that removes the formyl group from the initiator methionine in eubacterial proteins. It shows some activity against P. falciparum in vitro, likely acting against either or both the apicoplast or mitochondrion (Rohrich et al., 2005). Tetracyclines, which inhibit translation by prokaryotic and prokaryotic-like ribosomes, decrease growth of both T. gondii (Tabbara et al., 1982; Chang et al., 1990, 1991) and malaria parasites (Tabbara et al., 1982; Geary and Jensen, 1983; Divo et al., 1985). These antibiotics bind to the small subunit of the ribosome (Brodersen et al., 2000; Anokhina et al., 2004), and it has been suggested that, in apicomplexans, the mitochondrial ribosome is the main site of action, based on its effects on mitochondrial functions and pattern of protein synthesis inhibition in T. gondii and P falciparum (Kiatfuengfoo et al., 1989; Beckers et al., 1995). However, Camps et al. (2002) reported that the kinetics of T. gondii growth inhibition by clindamycin and tetracycline were quite similar, including a delayed-death phenotype (see below). Thus tetracyclines appear to inhibit protein translation in the apicoplast in addition to or instead of the mitochondrion.

Although an effect on the mitochondrion often cannot be ruled out, many of the drugs discussed above clearly affect the apicoplast. They show an interesting effect, called delayed death (Fichera et al., 1995). An example is the case of clindamycin (Figure 9.9A). After drug is added, parasite multiplication continues vigorously within the first vacuole. However, when establishing the second

FIGURE 9.9 Delayed-death phenotype.

(A) Kinetics of delayed death. T. gondii intracellular tachyzoites were treated with ciprofloxacin (25 |M, squares) or clindamycin (1 |M, circles) and their growth was compared to untreated parasites (crosses). The left-hand graph shows the growth within the first vacuole, where no significant effect was seen. After lysis and entry into the next host cell (right-hand graph), the drug-treated parasites die whether or not the drug ciprofloxacin is present (closed squares) or removed (open squares). Image courtesy of M.E. Fichera and D.S. Roos (1997) Nature 390, 407, reprinted by permission from Macmillan Publishers Ltd.

(B) Replication without apicoplast division. T. gondii were transiently transfected with ACP-GFP-mROP1 and inoculated into an HFF cell monolayer. This image, taken 48 h after transfection, shows a vacuole with about 64 parasites, only 1 of which contains an apicoplast as shown by the fluorescence of the GFP reporter. Bar = 5 |im. Image courtesy of C.Y. He and D.S. Roos (2001), EMBO J 20,330, reprinted by permission from EMBO J. This figure is reproduced in color in the color plate section.

FIGURE 9.9 Delayed-death phenotype.

(A) Kinetics of delayed death. T. gondii intracellular tachyzoites were treated with ciprofloxacin (25 |M, squares) or clindamycin (1 |M, circles) and their growth was compared to untreated parasites (crosses). The left-hand graph shows the growth within the first vacuole, where no significant effect was seen. After lysis and entry into the next host cell (right-hand graph), the drug-treated parasites die whether or not the drug ciprofloxacin is present (closed squares) or removed (open squares). Image courtesy of M.E. Fichera and D.S. Roos (1997) Nature 390, 407, reprinted by permission from Macmillan Publishers Ltd.

(B) Replication without apicoplast division. T. gondii were transiently transfected with ACP-GFP-mROP1 and inoculated into an HFF cell monolayer. This image, taken 48 h after transfection, shows a vacuole with about 64 parasites, only 1 of which contains an apicoplast as shown by the fluorescence of the GFP reporter. Bar = 5 |im. Image courtesy of C.Y. He and D.S. Roos (2001), EMBO J 20,330, reprinted by permission from EMBO J. This figure is reproduced in color in the color plate section.

vacuole, the parasites fail. This effect occurs even if the drug is removed at the second cycle. The delayed-death phenomenon was further explored using an unusual system for generating apicoplast-deficient cells (He et al., 2001a). Transient transfec-tion of construct containing ACP-GFP fused to the rhoptry-targeting sequences of ROP1 was found to interfere with division of the apicoplast. After a series of divisions, the parasitophorous vacuole contained many parasites, but only one with an apicoplast. Within the original vacuole the cells remained healthy whether they contained an apicoplast or not (Figure 9.9B), and all were capable of invading a new host cell. However, only those with an apicoplast could proliferate in the new host. Hence, it appears that some molecules produced directly or indirectly by the apicoplast in the preceding cycle are required in the next round of infection. At least one molecular defect related to delayed death has been identified. The level of apicoplast DNA is dramatically reduced upon clindamycin treatment (Fichera et al., 1995; Fichera and Roos, 1997). Evidently the inability to translate certain proteins encoded by the apicoplast genome prevents the proper replication of the DNA, either directly or, more likely, indirectly.

Several of the antibiotics that target apicoplast functions are in use clinically. The antibiotic clindamycin is used as a second-line drug for toxoplas-mosis, while spiramycin is chosen for toxoplasmosis in pregnancy. However, some of the agents, while effective in vitro, are not sufficiently antiparasitic for use in treatment of disease. A recent review has extensively discussed the apicoplast as a drug target in apicomplexan diseases (Wiesner and Seeber, 2005), and the topic is also covered in Chapter 19 of this volume.

0 0

Post a comment