Toxoplasma gondii infection: Novel emerging therapeutic targets

Abstract

Introduction:

Toxoplasmosis constitutes a challenge for public health, animal production and welfare. So far, only a limited panel of drugs has been marketed for clinical applications. Besides classical screening, the investigation of unique targets of the parasite may lead to the identification of novel drugs.

Areas covered:

Herein, the authors describe the methodology to identify novel drug targets in Toxoplasma gondii and review the literature with a focus on the last two decades.

Expert opinion:

During the last two decades, the investigation of essential proteins of T. gondii as potential drug targets has fostered the hope to identify novel compounds for the treatment of toxoplasmosis. Despite good efficacies in vitro, only a few classes of these compounds are effective in suitable rodent models, and none has cleared the hurdle to applications in humans. This shows that target-based drug discovery is in no way better than classical screening approaches. In both cases, off-target effects and adverse side effects in hosts must be considered. Proteomics-driven analyses of parasite- and host-derived proteins that physically bind drug candidates may constitute a suitable tool to characterize drug targets, irrespectively of the drug discovery methods.

1. Introduction

Toxoplasma gondii is an intracellular protozoan parasite belonging to the clade Alveolata of the super-group Diaphoretickes [1]. The life cycle comprises three stages, rapidly proliferating tachyzoites, slowly proliferating or resting bradyzoites, and sporozoites within oocysts [2]. Oocysts are excreted by felids, the definitive hosts. After ingestion by intermediate hosts, excysted sporozoites will invade intestinal epithelial cells and differentiate into tachyzoites, which are highly invasive and disseminate throughout the body within dendritic cells and other leukocytes. They will ultimately infect muscular tissue or the brain where they persist as bradyzoites, encapsulated within tissue cysts and surrounded by a cyst wall, which protects bradyzoites from physiological and immunological reactions on part of the host. The ingestion of an intermediate host or infected meat harboring T. gondii tissue cysts by a felid concludes the life cycle [3]. Thus, toxoplasmosis is a zoonosis that is orally transmitted. Another route of infection is via vertical transmission of tachyzoites from an infected mother to the fetus. According to recent estimations, approximately one third of the world population is infected with T. gondii, mostly via food contaminated by oocysts or bradyzoites [4] or via congenital infection [5]. In immunocompetent hosts, the infection remains, in general, without symptoms. There is, however, increasing evidence that chronical cerebral infection may cause mental disorders [6–8]. In immunocompromised hosts such as HIV-patients, cerebral toxoplasmosis is a major cause of mortality [9,10]. Vertical transmission may have severe consequences for the fetus and/or newborn [11,12], including abortion, hydrocephalus, or congenital ocular toxoplasmosis [13]. Moreover, ocular toxoplasmosis may be acquired by adults upon contact with highly virulent strains, in particular in South America [14]. Moreover, T. gondii is a major abortion-causing pathogen in farm animals, in particular in small ruminants such as goats and sheep [15,16], as well as in a variety of other mammals [17]. The closely related Neospora caninum is economically important in cattle as a major cause of abortion, stillbirth and birth of weak calves [18].

A major strategy for the prevention of toxoplasmosis in farm animals is the development of appropriate vaccines [19,20]. However, only one live-attenuated vaccine (Ovilis Toxovax^™), which is grown in tissue culture is currently marketed for use in sheep in only four countries. For human patients, no vaccine is available, and current drug treatments for toxoplasmosis therapy typically include antifolates using a combination of pyrimethamine-sulfadiazine or trimethoprim-sulfamethoxazole, and pyrimethamine combined with clindamycin, azithromycin, or atovaquone. These treatments are unspecific, have adverse effects and are not always efficacious [21,22]. In particular, the treatment of ocular toxoplasmosis is difficult and not standardized, so far [23,24].

Consequently, new treatment options, in particular chemotherapeutics against toxoplasmosis are required. As discussed in previous review articles [25,26], two strategies of antiparasitic drug development prevail. The first strategy is based on the initial screening of compound libraries employing suitable in vitro culture models, followed by studies in an in vivo model [27]. In cases where a specific drug development approach does not promise a substantial market return, repurposing of available drugs or drug candidates is applied as an alternative screening approach [28]. Since the beginning of the post-genomic era, a second strategy has appeared to be most attractive: target-based drug design based on the identification of potential drug targets by in silico data base mining prior to any in vitro or in vivo test [29]. However, the expected rapid breakthrough in drug development has been lacking, and the validity of this strategy was, and still is, under debate [30,31]. In the present article, we review the current status of drug targets for compounds against toxoplasmosis identified by both strategies.

2. Methodology

2.1. Definitions

Before we discuss Toxoplasma drug targets in detail, the term drug target needs to be defined. Classical definitions see a drug target as “… the specific binding site of a drug in vivo through which the drug exerts its action…” [32]. A key-lock model, the drug being a key entering a corresponding lock, the target, symbolizes this definition. In this model, the ideal compound hits a target, preferentially a protein, that is essential for the parasite and absent from the host.

However, during the genomic and post-genomic era it has become more and more evident that this model is too simplistic. Consequently, the drug-target interaction model has been extended to a drug-target network, where “promiscuous drugs” hit multiple targets [33]. In the case of anti-Toxoplasma compounds (and anti-infective compounds in general), it is unlikely that such a network is restricted to the pathogen. If it extends to the host, it should not contain elements with pivotal functions thereby minimizing side effects.

2.2. in silico selection of targets

Depending on the strategies of drug development, i.e. target or screening based, the target identification occurs via different pathways. T. gondii is an excellent molecular genetic model for intracellular apicomplexan parasites [34]. The genome of various T. gondii strains is sequenced (toxodb.org), and molecular genetic tools have been available since three decades. Using CRISPR-CAS mediated genome engineering [35,36], genome wide screens have allowed to identify 200 fitness conferring genes unique for apicomplexans [37]. This constitutes a solid base for target-based drug development strategies. The first step is the identification of a suitable target, i.e. an apicomplexan-specific essential protein. The ideal candidate is an enzyme involved in intermediate metabolism or signal transduction. The next step involves functional assays with the recombinant protein ideally resulting in the identification of inhibitors. The inhibitors with the lowest inhibitory constant (K_i), thus the highest affinity to the target will then be subjected to efficacy and cytotoxicity tests in vitro using a suitable cell culture. The compounds with the best ratio high efficacy vs. low cytotoxicity (thus a large therapeutical index T_i) are retained for further investigations using suitable in vivo models. The in vitro screening and testing is the intersection between target- and screening based drug development workflows, as illustrated by Figure 1.

Figure 1.

The workflows of target and screening based drug development strategies. Both strategies are interlinked.

2.3. Library screening and target identification

Obviously, screening-based approaches start with screening. To screen several hundred compounds of a library, the classical microscopical assessment or even PCR-based methods are too tedious, and host-cell staining methods do not distinguish between host and parasite toxicity [38]. Fortunately, transgenic T. gondii strains expressing E. coli beta-galactosidase [39,40] or yellow fluorescent protein [41] are better tools. Consequently, screenings [42], have been performed – either on open source libraries with mind of drug repurposing [43,44] or with original compounds, e.g. ruthenium complexes [45–47]. The heuristic identification of targets of compounds issuing from library screenings needs a large panel of complementary methods and appears to be less conclusive than within a target-based framework. The methodology refers to studies of effects on cell biology, e.g. ultrastructural alterations, the generation and analysis of resistant isolates and the analysis of proteins binding to effective compounds as detailed elsewhere [25,26]. All these approaches need strict controls, in particular ineffective compounds with high structural similarities to effective compounds. Of course, compounds issued from target-based identification can (and should) be subjected to the same type of investigations in order to confirm the in silico targets and (if applied to suitable host cells or organs) to identify host targets involved in drug metabolism or responsible for side effects.

2.4. Differential affinity chromatography

A key step of the heuristic identification of targets of compounds issued from screening is the characterization of drug-binding proteins [48]. If we admit that a given compound uniquely or at least preferentially physically interacts with a given protein, affinity chromatography is a method of choice to identify such target proteins (Figure 2). The effective compound of interest is coupled to a suitable inert matrix. In parallel, an ineffective compound with high structural similarities to the effective compounds is coupled to the same kind of inert support. Columns filled with this material are inserted into a low-pressure liquid chromatography device. Cell-free extracts from organisms of choice are loaded to both columns in parallel. After suitable washing steps, bound proteins are eluted and identified by convenient proteomic tools. The proteomes identified in eluates from effective and ineffective compounds are compared. The differential affinoproteome, i.e. the subset of proteins binding to effective, but not to ineffective compounds, should then contain potential targets.

Figure 2.

The strategy to use affinity chromatography to identify drug binding proteins is a direct consequence of the drug-target paradigm. LC, liquid chromatography; MS, mass spectrometry; SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis.

By this approach, we have identified differential affinoproteomes of ruthenium complexes [49] and of a leucinostatin-like antimicrobial peptide [50] in T. gondii, as well as the affinoproteome of a calcium-dependent protein kinase inhibitor in tachyzoites of the closely related N. caninum [51]. Moreover, this method can be extended to host cells or tissues thereby identifying potential targets responsible for side effects; with this, differential affinity chromatography (DAC) becomes comparative differential affinity chromatography (CDAC).

2.5. Analysis of resistant strains

Drug effectiveness is measured by two parameters, namely the concentration of the drug causing 50 % proliferation inhibition (or any other percentage, e.g. 90%) as compared to an untreated control, commonly denoted as IC₅₀ (or. e.g. IC₉₀ …). IC₅₀s can be determined using the logit-log algorithm as detailed elsewhere [25]. Another parameter is the minimal inhibitory concentration (MIC), thus the concentration, at which growth cannot be detected anymore. This parameter cannot be extrapolated but must be determined by direct investigation at increasing concentrations. Resistance of a given strain is commonly defined as an increase of both IC₅₀ and MIC as compared to a corresponding wildtype [52]. Resistant strains are created by continuous breeding in the presence of increasing amounts of the drug or by mutagenesis followed by selection of resistant populations or clones in the presence of lethal concentrations of the drug. The resistant strains are compared to the corresponding non-resistant strains on the genomic, transcriptomic and/or proteomic levels.

3. Drug targets

3.1. Replication

DNA intercalating compounds such as the acridine derivative quinacrine, an early anti-malarial and the first anti-giardial drug, interact with DNA, and thereby inhibit replication [53]. Pentamidine and derivatives interact with DNA in AT-rich regions [54,55] forming a coiled-coil complex [56]. Pentamidines are effective against a variety of evolutionary distant protozoal parasites such as Plasmodium sp. and Leishmania [57]. The list of susceptible protozoans includes T. gondii [58,59] and the closely related N. caninum [60]. DNA as a target is ubiquitous. Therefore, the action of pentamidines and other DNA-intercalating agents is not excluded from the host. Treatment of prostate cancer cells with pentamidines causes reduction of mitochondrial DNA followed by ultrastructural alterations and apoptosis [61]. Similarly, apoptosis-like cell death associated with loss of mitochondrial functions occurs in pentamidine-treated L. donovani [62]. Ultrastructural alterations such as vacuolization going in parallel with the formation of lipid droplets followed by loss of mitochondria and cell death are observed in T. gondii tachyzoites treated with pentamidine derivatives [59] (Fig. 4A, B). The same study described a rapid adaptation of T. gondii to increasing concentrations of pentamidine derivatives. The nature of this adaptation has not been elucidated so far. In Leishmania sp., pentamidine resistance is correlated with a decreased uptake of the drug into the mitochondrion [63,64].

Figure 4.

Structural alterations induced by in vitro drug treatments of T. gondii tachyzoites. Tachyzoites were grown in HFF and treated with either dicationic pentamidine derivatives such as DB745 (A, B), the bumped kinase inhibitor BKI-1748 (C, D), and mitochondrial inhibitors such as Decoquinate (E) or the endochin-like quinolone ELQ-334. Within 24 h of treatment (A), pentamidine derivatives induce the formation of cytoplasmic lipid droplets (ld) and vacuoles (vac) that contain electron dense material of unknown nature resembling autophagosomes, while the mitochondrion still appears unaffected. After 72 h (B), most tachyzoites become structurally heavily impaired demonstrating parasiticidal activity of DB745. C shows a multinucleated complex (MNC) generated after upon treatment of T. gondii tachyzoites with BKI-1748, in D the boxed area is shown at higher magnification. The MNC is located within a parasitophorous vacuole, the membrane of which is indicated with horizontal arrows and the MNC is embedded in. a granular and electron dense matrix (C). Note the presence of multiple nuclei. At higher magnification (D), the small apical complexes that eventually form the anterior part of newly formed zoites are indicated by diagonal arrows. Newly formed zoites lack the characteristic triple plasma membrane of T. gondii tachyzoites, are deficient in completing cytokinesis, and remain stuck within the host cell. The mitochondrion (mito) within these MNCs appears unaffected. In contrast, compounds such as DCQ (E)and ELQ-334(F) induce the rapid loss of the electron dense mitochondrial matrix and cristae within 6–12 h of treatment, and induce mictochondrial swelling, while other organelles such as rhoptries (rop), micronemes (mic), dense granules (dg), and the Golgi apparatus remain structurally unaffected, demonstrating that the mitochondrion is the primary target. Bars in A = 0.88 μm, B = 0.35 μm, C = 1.2 μm, D = 0.4 μm, E and F = 0.46 μm. TEM was performed as described elsewhere [92].

Pentamidines and other intercalating agents interact not only with DNA but also with DNA modifying enzymes such as topoisomerase II, as shown for Leishmania sp [62,65]. In T. gondii, the prokaryote-type apicoplast gyrase is inhibited by ciprofloxacin [66], an antibiotic effective against T. gondii in vitro as well as in vivo [67].

3.2. Gene expression

A prerequisite for transcription is the accessibility of a given region of the chromosome to RNA polymerases. Covalent modifications of the DNA itself by methylation [68] and of histones by acetylation, methylation, phosphorylation or other modifications are fundamental epigenetic mechanisms regulating gene expression [69]. Apicidin, a cyclic tetrapeptide with good efficacy against T. gondii, is a potent histone deacetylase inhibitor [70]. Based on this discovery, the potential of histone acetylation as a drug target has been investigated during the following two decades [71]. Recently, in silico investigations of T. gondii histone deacetylases have resulted in the development a novel histone deacetylase inhibitor effective against various T. gondii and Plasmodium sp. strains in vitro and in mouse models for acute and chronic toxoplasmosis [72].

The apicoplast is an organelle unique to apicomplexans and therefore offers suitable targets [73,74] that are involved in the translation machinery and in metabolic pathways. Current chemotherapy against toxoplasmosis is partly based on antibiotics such as the macrolide spiramycin, the aminoglycoside paromomycin or the lincosamide clindamycin [75,76]. These antibiotics inhibit translation within the apicoplast by binding to ribosomal RNA [77]. More recent investigations highlight another part of the translation machinery as potential drug targets, namely tRNA. In fact, some tRNAs contain thiourea derivatives of uracil at specific locations, which are essential for their function. T. gondii enzymes involved in this modification are essential and therefore constitute potential drug targets [78].

3.3. Signal transduction

The signalome of apicomplexans including T. gondii is highly divergent from the signalome of their mammalian host cells. Therefore, it constitutes a rich source of potential targets for drug development as demonstrated for calcium-dependent protein kinases containing calmodulin-like domains (calmodulin-like domain protein kinase; CDPK). These CDPKs are homologous to kinases commonly found in plants [79] and are essential for protein secretion, invasion, and differentiation [80]. The genome of T. gondii ME49 comprises 16 proteins annotated as CDPKs (toxoDB.org; Jan 2023). During the last two decades, special attention has been paid to TgCDPK1 as a potential drug target. One of the first studies on TgCDPK1 uses a specific CDPK inhibitor to demonstrate that inhibition of TgCDPK1 correlates with a loss of motility and attachment to host cells [81]. Furthermore, TgCDPK1 is essential for exocytosis, as shown using a specific inhibitor followed by transformation with an insensitive gain-of-function mutant kinase [82]. Structure analysis of this kinase has provided the key to the development of ATP competitive inhibitors known as “bumped kinase inhibitors” (BKIs) effective against T. gondii [83–85] and N. caninum [86–88] in vitro and in vivo. Given the presence of homologs with high similarities in the genomes of both species, it is, however, unclear, whether the kinases annotated as CDPK1 in T. gondii (TGME49_301440) and as calmodulin-like domain protein kinase isoenzyme gamma, related in N. caninum (NCLIV_011980) are the only targets of these inhibitors. The homologs with the highest similarities in both species, annotated as CDPK3 in T. gondii (TGME49_305860) and as a hypothetical protein in N. caninum (NCLIV_070280) may be potential targets, as well. Knock-out of TgCDPK3 inhibits egress of tachyzoites from host cells [89]. A protein-protein interaction study with TgCDPK3 suggests that the phosphorylation of a myosin motor protein may be responsible for this phenotype [90]. In this context, it is interesting that prolonged treatment of host cells infected with T. gondii or N. caninum with various BKIs leads to the transformation of intracellular tachyzoites into multinucleated complexes (MNCs) [87]. MNCs contain multiple nuclei and newly formed zoites which are, however, deficient in completing cytokinesis and exhibit a block in tachyzoite formation (Figure 4 C, D). In the case of N. caninum, the proteome pattern of these “baryzoites” is characterized by a downregulation of intermediate metabolism and an upregulation of some bradyzoite marker proteins [91]. After removal of the drug pressure, these “baryzoites” re-differentiate into virulent tachyzoites [92]. The observed effects upon BKI treatment may thus be due to inhibition of both CDPK1 and CDPK3 or even of other homologs.

3.4. Metabolism

One of the first and most important metabolic pathways targeted in T. gondii is certainly folate biosynthesis, which is inhibited by the first line drugs pyrimethamin, sulfadiazine, sulfamethoxazole, and trimethoprim, used alone or in combination [93]. Since the dihydrofolate reductase inhibitor pyrimethamine is only effective at doses interfering with the host enzyme, the research for more specific analogs is still ongoing [94].

Like other intracellular parasites, T. gondii replenishes the pools of purine and pyrimidine base nucleosides not only by de novo biosynthesis, but to a large extent by salvage reactions comprising transfer of a free base on phosphoribose and modifications of these bases [95,96]. The transferase catalysing the transfer of guanine yielding GMP or hypoxanthine yielding IMP (hypoxanthine-guanine phosphoribosyl transferase), expressed as two alternatively spliced isoforms [97], is of particular interest as a positive or negative selection marker of transformed T. gondii tachyzoites or bradyzoites [98,99].

The apicoplast, a specific organelle that is a plant chloroplast homolog, is found in T. gondii and other apicomplexans. Since these plastids harbour specific metabolic pathways absent from mammalian cells, it is natural that not only plastid-borne protein biosynthesis (see above), but also metabolic pathways are regarded as suitable anti-Toxoplasma drug targets [74,100]. Since lipid biosynthesis in plants is plastid-borne and can be targeted by various herbicides, it is straightforward to tests such compounds against T. gondii. In vitro tachyzoite proliferation and activity of recombinant Acetyl-Coenzyme A- carboxylase, the first key enzyme of lipid biosynthesis, is inhibited by aryloxyphenoxypropionate herbicides [101]. Moreover, fatty acid synthase II is inhibited by the herbicide haloxyfop [102]. These studies, and the generation of a conditional null mutant of the apicoplast acyl carrier protein reveal that apicoplast borne fatty acid biosynthesis is essential for the survival of T. gondii in vitro as well as in vivo [103]. In detail, apicoplast fatty acid synthesis seems to be essential to compounds required for the final step of parasite division [104].

Other apicoplast-specific pathways investigated as potential drug targets include carbohydrate metabolism [105], the ferredoxin redox system, which plays a central role in various metabolic pathways [106], as well as transporters [107]. A ubiquitous metabolomic pathway, which has gained some attention as a potential target in T. gondii is the degradation of methylglyoxal by glyoxalases. The isoenyzme Glo1 as well and tachyzoite proliferation in vitro is inhibited by curcumin, but in the 10⁻⁵ M range only [108]. Moreover, curcumin has positive effects on neuroglirogenesis in primary neuronal cell cultures infected with T. gondii [109].

3.5. Proteases

Like other organisms [110], T. gondii harbors serin, cysteine, metallo and aspartylproteases involved in a variety of physiological processes. Serine proteases, in particular subtilisin and rhomboid family enzymes, are involved in invasion and egress [111] as evidenced by inhibitor studies [112]. Moreover, the microneme rhomboid protease TgROM1 is essential for intracellular proliferation, as evidenced by conditional knock-out [113]. Other ROM proteases, in particular ROM4, are involved in host cell invasion [114]. The secreted microneme proteins MIC2, MIC4 and M2AP are processed by the subtilisin TgSUB1, as shown by knock-out plus complementation [115]. Other proteases involved in microneme and rhoptry protein maturation are cysteine proteases belonging to the family of cathepsins [116]. Therefore, it is not surprising that serine and cysteine protease inhibitors have detrimental effects on microneme secretion and host cell invasion [117,118]. Other proteases include the rhoptry protease toxolysin-1, a metalloprotease [119] and aspartylproteases. The aspartylproteases of T. gondii belong to five families. TgASP1 is associated to the inner membrane complex, but is not essential, as shown by knock-out [120]. The situation is different with ASP5. This protease is located in the Golgi and processes effector proteins excreted to the host cell. Consequently, the knock-out of ASP5 decreases virulence [121]. ASP5 is essential for the cleavage of the GRA family proteins 16, 19, and 20 and involved in the correct targeting of GRA16 and 24 to the host cell nucleus [122,123]. Recent investigations suggest that treatment of mice with aspartylprotease inhibitors developed against the HIV enzyme reduced the parasite burden in a mouse model for chronic T. gondii infection [124].

3.6. Mitochondrial integrity as a target

Three decades ago, it was still under debate whether the single, ramified mitochondrion of T. gondii was functional or not [125,126]. According to the present knowledge, it is, however, clear that the T. gondii mitochondrion not only imports the full set of tRNAs [127] and a variety of enzymes [128,129], but also possesses unique structural features [130] and active energy and intermediary metabolisms [131,132]. It is therefore not surprising that compounds interfering with the membrane potential and therefore with the electron transport chain such as monensin affect the mitochondrial integrity, most likely by induction of oxidative stress [133]. Treatment with artemisinin derivatives such as artemisone or artemiside results in destruction of T. gondii mitochondria, but without formation of reactive oxygen species [134]. Another class of compounds affecting membrane integrity are leucinostatins. DAC performed with a leucinostatin with pronounced effects on T. gondii mitochondria has led to the identification of a variety of binding proteins with homologies to cytochrome c oxidase subunits [50]. This suggests that besides unspecific effects such as uncoupling, mitochondrial compounds may interact with protein targets. This is further evidenced by investigating the mode of action of quinones such as the naphthoquinone buparvaquone inducing mitochondrial alterations in the closely related Besnoitia besnoitii [135]. The analysis of resistant strains of Theileria annulata suggests that the cytochrome bc1 complex is the direct target [136]. Moreover, the cytochrome bc1 complex is considered as the main target of decoquinate and related endochin-like quinolones [137,138] Inhibition of this complex results in ultrastructural alterations of mitochondria, as shown in the closely related N. caninum and B. besnoitii, as well as in T. gondii (Figure 4 E, F), all with a highly conserved bc1 complex [139,140]. Other compounds affecting the mitochondrial integrity are ruthenium complexes with various substituents [45,46]. DAC with one class of these complexes yields a hypothetical protein with structural homologies to an inner membrane protein of human mitochondria in T. gondii, as well as ATPase subunits in Trypanosoma brucei, an evolutionary distant protozoan parasite [49].

4. Conclusions

The drugs currently used for the treatment of T. gondii infections originate from whole organism screening approaches and target apicoplast-borne protein biosynthesis and folate metabolism. Fostered by genome sequencing efforts and by the availability of suitable molecular genetic tools, T. gondii specific proteins with essential functions for vitality and interactions with the host have been identified. These proteins constitute suitable drug targets, and, in some cases, small molecular inhibitors are available or can be designed. In other studies, libraries with novel compound classes have been screened in vitro. However, despite promising results obtained in vitro, these inhibitors identified through screening or target-based approaches, with few exceptions, did only rarely exhibit in vivo activity in suitable rodent models.

5. Expert opinion

The availability of powerful in silico tools and the proficiency of molecular genetics has rendered T. gondii a most suitable model for the identification of drug targets. Moreover, the last 20 years have witnessed the development of many novel classes of compounds with anti-Toxoplasma activities in vitro, but only few have been shown to be also sufficiently active in vivo (see Table 1 for a summary). The reasons for this discrepancy between promising activities in vitro and poor or non-existent activities in vivo can be explained by different examples. For instance, inhibitors can be created via structural modelling that fits them into the active site of an essential parasite protein. The inhibition is confirmed by functional assays, and the compounds are active in vitro or even in a rodent model. However, this does not mean that this protein represents the sole target of the inhibitors. Other targets may exist in T. gondii and/or in the host. If the compounds in question interfere e.g. with functional activities of host immune cells or with regulators of embryonic development, treatment failures or abortions may result when the compounds are investigated in animal models. On the other hand, compounds stimulating innate immune responses such as silver nanoparticles for instance may be effective against T. gondii as well as against other pathogens [141]. In another example, a class of compounds identified by library screening exhibits promising in vitro activities, and structural alterations in the mitochondrion of T. gondii are demonstrated, indicating that the targets may be components of the respiratory chain and/or extramitochondrial proteins essential for mitochondrial maintenance (e.g. via transport processes) and/or the membrane and/or nucleic acids. In rodent models, the compounds may fail simply because T. gondii has switched to a dormant state upon treatment and recovers afterwards, or because of unfavourable pharmacokinetics, or because of detrimental effects on the host immune system. These examples illustrate that the classic drug-target paradigm is too simplistic to predict the success of in vitro inhibitors in animal models. Modern approaches such as DAC coupled to MS and proteomics combined with carefully designed in vitro studies are powerful tools to eliminate potential “lead compounds” prior to embarking on costly animal experimentation, can contribute to the 3R concept of laboratory animal welfare.

Table 1.

Summary of identified Toxoplasma gondii targets and effective compounds. Additional information and references are presented in section 3.

Cellular function	Target	Compound	Investigative status
Replication	DNA, Topoisomerase	Pentamidines	In vitro, in vivo, clinical (cancer, leishmaniasis, malaria)
	Gyrase	Ciprofloxacin	In vitro, in vivo, clinical (toxoplasmosis)
Gene expression	Histone acetylation	Apicidine, novel compounds	In vitro, in vivo
	Translation (apicoplast)	Macrolide antibiotics	In vitro, in vivo, clinical (toxoplasmosis)
	tRNA modification	none	In vitro, in vivo (knock-out studies)
Signal transduction	Calmodulin-like domain protein kinase	Bumped kinase inhibitors	In vitro, in vivo
Metabolism	Glyoxalase	Curcumin	In vitro
	Folate biosynthesis	Sulfonamides, dehydrofolate reductase inhibitors	In vitro, in vivo, clinical (toxoplasmosis)
	Lipid biosynthesis (apicoplast)	Herbicides	In vitro
Protein degradation	Aspartyl, cysteine, serine proteases	Inhibitors specific for each class	In vitro, in vivo
Mitochondrial integrity	Membrane potential	Monensin	In vitro, in vivo, veterinarian (coccidian and related)
	Membrane potential, respiratory chain	Leucinostatin derivatives	In vitro
	Respiratory chain	Naphthoquinones, quinolones	In vitro, in vivo, veterinarian (coccidian and related)
	Unknown	Artemisinin derivatives	In vitro, in vivo, clinical (malaria)
	Unknown	Ruthenium complexes	In vitro

Taken together, we believe that the ‘theoretical’ advantages of target based antiprotozoal drug design over library screening must be re-considered when compounds issuing from either approach are assessed in a more realistic in vivo situation. Here, “trial and error” continue to prevail.

To further proceed with a compound into the domestic pet or food animal area, are into applications in humans, economic aspects concerning the development of drug candidates into market ready drugs are paramount. The current hurdles for approval of novel drugs are incredibly high. Consequently, pharmaceutical companies will always focus on the development of drugs that promise a high market return. Toxoplasmosis has certainly an impact on human and animal health, but is not a major threat to modern civilization. Since effective treatments in particular for human patients exist, it is safe to forecast that upcoming anti-Toxoplasma drugs will be incrementally improved versions of existing compounds rather than something completely new.

Figure 3.

Structural hallmarks of T. gondii tachyzoites visualized by TEM. A and B show T. gondii tachyzoites were cultured in human foreskin fibroblast monolayers during 36 h (A, B) and 72 h. The boxed area in A is shown at higher magnification in B. Tachyzoites reside and proliferate in the host cell cytoplasm inside a parasitophorous vacuole, surrounded by a parasitophorous vacuole membrane (small arrows). In rapidly proliferating cultures, tachyzoites are often seen still attached to the residual body (rb) (B). Parasites exhibit a single nucleus and an apically located conoid (con), which in conjunction with secretory organelles such as micronemes (mic), rhoptries (rop) and dense granules (dg) forms the apical complex. Tachyzoites harbor a single mitochondrion (mito) of which, depending on the section plane, only parts are visible. The mitochondrion contains an electron dense matrix and clearly discernible cristae. After 72 h (C), large vacuoles containing numerous newly formed tachyzoites are visible, which will eventually undergo egress and infect neighboring host cells. Bars in A = 2.2 μm, B = 0.35 μm, C = 1.5 μm. TEM was performed as described elsewhere [134].

ARTICLE HIHGLIGHTS.

• The apicomplexan protozoan Toxoplasma gondii has a very broad host range and a high prevalence in humans.

• The genome has been sequenced, and reverse genetic molecular tools are available.

• Consequently, during the last two decades, T. gondii has become a suitable model system for anti-protozoal drug development.

• Despite the increasing efforts on target-based drug development, only a limited number of drugs is available against human toxoplasmosis.

• State-of-the art proteomics may provide suitable tools to fill this gap.