Metabolomic profiling for the identification of drug targets in trypanosomes

1 Trypanosomes and trypanosomiasis

Trypanosomes are a group of protists which includes species responsible for serious human and animal diseases. They belong to the class Kinetoplastea, phylum Euglenozoa, that also contains the classes Diplonemea and Euglenida. Regarding trypanosomes that affect human health, they produce zoonotic diseases mainly in tropical and subtropical parts of the world where the insects transmitting the parasites between humans and other animals are endemic. Two different subspecies of Trypanosoma brucei, i.e., T b. gambiense and T b. rhodesiense, cause Human African Trypanosomiasis (HAT), a.k.a. sleeping sickness, in sub-Saharan Africa. People can become infected when tsetse flies (Glossina spp.) with parasites in their salivary glands take a bloodmeal. The trypanosomes, which throughout their life cycle remain extracellularly in their hosts, move subsequently from the subcutaneous tissue to the hemolymphatic system, where they proliferate, while also taking extravascular residence in various organs, notably adipose tissue (Trindade et al., 2016) and the skin (Capewell et al., 2016). They cause a progressive wasting disease in mammals, while at a later stage they can cross the blood-brain barrier to enter the central nervous system (CNS) (Mogk et al., 2007). In most cases, the disease is lethal if left untreated. The gambiense form of HAT is a chronic disease with for long times no major symptoms. It occurs in West and Central Africa and is responsible for 92% of the reported cases. In contrast, the rhodesiense form, which accounts for 8% of the cases, shows a more acute pattern and is found in East and Southern Africa. Three major HAT epidemies occurred in last century, the last one during the 1980s and 1990s, with 35,000 cases diagnosed each year, and estimated exceeding 300,000 people infected that went unrecorded (Barrett et al., 2024). Since then, the epidemiological situation has ameliorated considerably, by improved tsetse fly control, diagnosis and newer treatments, resulting in fewer than 600 reported cases during the last years, virtually eliminating HAT as a public health problem, the goal set for 2030 (World Health Organization, 2024a; Barrett et al., 2024). Nonetheless, more than 100 million people are still considered to be at risk of infection.

Another Trypanosoma species, T. cruzi, causes Chagas disease (CD) a.k.a. American Trypanosomiasis worldwide, but mostly in the continental Americas (Cucunubá et al., 2024). This parasite is transmitted by triatomine bugs, endemic in the (sub)tropical areas of the American continent, but can also be passed on vertically during pregnancy or birth, orally (by food and drinks), or through blood or blood products transfusion and organ transplantation (Ferreira et al., 2025). When transmitted by triatomines, people come in contact with feces or urine of infected blood-sucking bugs on their skin. The insects usually bite during the night, after which the parasites enter the body when the person instinctively smears the feces or urine into the bite, other skin breaks, the eyes or the mouth. CD presents in two phases, with upon infection an initial acute phase of about 2 months during which the number of parasites in the blood may be high, but symptoms usually absent or mild and non-specific (Rassi et al., 2012). It is followed by a chronic phase, during which the parasites are most of the time hidden intracellularly. Trypanosoma cruzi can infect virtually any tissue; however, in clinical cases it has been found mainly in the heart and digestive muscles. Approximately 30% of patients will develop symptoms, which usually appear one to 3 decades after infection. In the most common symptomatic forms, patients will suffer from cardiac disorders and up to 1 in 10 from digestive, neurological or mixed alterations. Later, damage to the nervous system and muscles of the heart and digestive system may occur in these patients, leading to cardiac arrhythmias, progressive heart failure, and ultimately sudden death. Like HAT, CD is zoonotic too, because T. cruzi can also infect animals which serve as an important wild reservoir of human infective parasites. CD continues affecting many people. More than 7 million people worldwide are estimated to be currently infected with T. cruzi, leading to more than 10,000 deaths annually (World Health Organization, 2024b).

One of the complicating factors in developing new therapies for CD is the high genetic diversity of T. cruzi. The parasite is currently classified into six discrete typing units (DTUs; TcI–TcVI), with a seventh designation, TcBat. Besides the genotypic variability reported between DTUs, these lineages also differ in their biological characteristics, virulence, susceptibility to chemotherapy, and clinical outcomes of CD (de Lima et al., 2026).

2 Current treatment of trypanosomiasis

To treat trypanosomiasis, different drugs are available. For HAT, during recent years, six treatments were recommended by the World Health Organization (WHO), to be used dependent on the stage of the disease and the T. brucei subspecies causing it. For rhodesiense HAT, suramin is used in the first stage, and melarsoprol in the second stage. For the gambiense form, pentamidine is the drug of choice to treat the first stage. For the second stage, a nifurtimox-eflornithine combination (NECT) and fexinidazole are recommended, although eflornithine can also be used as monotherapy. Vincent et al. (2012) showed that combining nifurtimox and eflornithine did not cause a synergistic trypanocidal effect in vitro, but the parasites’ metabolic changes associated with the combination are the sum of those detected in the monotherapies. However, the WHO recently revised its guidelines, making new recommendations which, however, are conditional because of limited certainty about evidence of their effectivity (Lindner et al., 2025; Barrett, 2025). The new guidelines propose that fexinidazole can be given orally in most cases for both diseases at either stage, thus replacing suramin and melarsoprol for rhodesiense HAT. It is important to note that suramin and melarsoprol have been introduced as HAT drugs a long time ago, require intravenous administration and have important adverse effects; particularly melarsoprol is quite toxic and can be lethal in 3–10% of the treated cases. Finally, a benzoxaborole compound, acoziborole, has recently been developed and is expected to become available for HAT treatment in 2026 (DNDi, 2026; Vogel, 2026). It is effective in both stage 1 and 2 of the gambiense disease with a single-dose oral treatment (Zoltner et al., 2024; Barrett, 2025).

Two drugs are in use for treatment of CD, benznidazole and nifurtimox (World Health Organization, 2024a; 2024b). Both medications are fully effective in curing the disease in the early, acute phase. However, their efficacy to kill the parasites decreases in the later, chronic phase. Moreover, the drugs cause adverse effects, particularly in patients of older age, and should not be administered to pregnant women or people with kidney or liver failure. Nifurtimox is also contra-indicated for people with a background of neurological or psychiatric disorders (Berenstein et al., 2021). These adverse effects and contra-indications frequently compromise patient adherence, often leading to treatment interruption or discontinuation and ultimately reducing therapeutic effectiveness (Jackson et al., 2020).

Since 2019, several studies have been initiated to investigate different regimens of benznidazole or nifurtimox, or both, evaluating lower doses administered for shorter or longer periods, with the aim of improving the efficacy or tolerability, or both, of the regimens currently considered the standard treatment (Pinazo et al., 2024). Notably, the BEnznidazole New Doses Improved Treatment and Associations (BENDITA) phase II clinical trial demonstrated that shorter benznidazole regimens, either alone or in combination with fosravuconazole (E1224, a broad-spectrum antifungal triazole), may reduce treatment duration while reducing and clearing parasitaemia in adults with chronic indeterminate CD (Torrico et al., 2021). Furthermore, the development of a pediatric formulation of benznidazole has improved dosing accuracy and facilitated drug administration in infants and young children, with a high treatment response rate and infrequent, mild adverse events (Altcheh et al., 2023).

3 Necessity to develop new drugs for trypanosomiases and to identify the mode of action of drugs and drug candidates

Although treatments are currently available for trypanosomiasis, development of new drugs remains important (Field et al., 2017). As described, several of the drugs currently in use are not adequate, either because they are only partially efficacious, toxic and/or difficult to administer, particularly in rural settings. Moreover, development of drug resistance remains a threat for infectious diseases, including trypanosomiasis (Fairlamb et al., 2016; Kasozi et al., 2022). Experimental studies have shown that trypanosomes can develop reduced susceptibility to trypanocidal drugs through multiple mechanisms, including alterations in drug uptake, drug activation, efflux, energy metabolism, and stress response pathways (Ospina et al., 2025; Ochoa-Martínez et al., 2026). These mechanisms may compromise drug efficacy and further limit the already restricted therapeutic options available for the treatment of trypanosomatid infections.

Although the prospects of eliminating, in the coming years, HAT as a public health problem, it is important to realize that the disease, being a zoonosis with both domestic and wild animals as reservoirs of parasites and flies as vectors, may flare up again if tsetse control and public health facilities collapse due to environmental or societal upheavals, as previously occurred (Smith et al., 1998). Moreover, arguments have been provided that asymptomatic infections of humans form an overlooked reservoir of skin-dwelling parasites that may contribute to transmission of HAT, even without an animal reservoir (Capewell et al., 2019). Likewise, sustained vector surveillance and control remain fundamental for the long-term control of CD. However, because non-vectorial transmission also contributes substantially to the epidemiology of this disease, effective disease control requires integrated strategies that extend beyond vector control (World Health Organization, 2024b).

To develop new drugs, or improve existing ones, knowledge of the targets of the drugs and candidate drugs will be instrumental. Different approaches are being used to achieve it. They all involve measuring the susceptibility of a phenotypic feature to a (potential) drug after changing the expression or structure/function of the possible target(s)) by (reverse) genetics (e.g., Alsford et al., 2011; 2012) or chemoproteomics and chemogenomics (i.e., using a reactive analog of the (potential) drug). The phenotypic analyses generally used comprise following proliferation and development (Alsford et al., 2011; 2012), cytological assays (Thomas et al., 2018), or metabolomics in which one follows (time-dependent) changes in the metabolome upon administration of a compound. However, the reverse genetics methods described above were primarily developed in T. brucei; genetic manipulation of T. cruzi has been more challenging. Despite progress in setting up reverse genetics in T. cruzi (Chiurillo and Lander, 2021), drug discovery for CD continues to rely heavily on phenotypic and target-based approaches, increasingly complemented with tools of genetics, omics, and artificial intelligence for target identification and validation (Francisco et al., 2020).

These approaches may not only identify the target but also help to unravel the mode of action (MOA) of the compounds. In this review, we will present metabolomics studies specifically performed to identify targets of drugs and potential drugs of the Trypanosoma species causing the human diseases HAT and CD.

The targets and MOAs of only a few drugs currently in use against trypanosomiasis have been identified, while for most they are under debate or still unknown. A recently published review provides the current knowledge in great detail (Barrett, 2025), but here we will briefly summarize it. Eflornithine is an inhibitor of the enzyme ornithine decarboxylase (ODC), an enzyme of the polyamine biosynthetic pathway. Benzoxaboroles such as the new drug acoziborole were shown to promote the degradation of the RNA cleavage and polyadenylation specificity factor 3 (CPSF3). Metabolomic profiling that contributed to revealing the MOAs of these latter two drugs will be discussed below. The precise targets of pentamidine, fexinidazole, nifurtimox and benznidazole are still unknown, but appear to be multiple. Pentamidine binds strongly to DNA, within the minor groove of the double helix, possibly a major cause of its trypanocidal activity. Fexinidazole, nifurtimox and benznidazole are pro-drugs which, upon entry in the parasites, are activated by a prokaryotic type I nitroreductase present in trypanosomatids. This generates highly reactive molecules which will bind to, or chemically modify a large variety of trypanosomal components such as proteins, nucleic acids, lipids, metabolites, resulting in killing of the parasites. The precise MOA of melarsoprol is unknown, but multiple activities have been reported for it, such as interactions with cellular thiols and inhibition of MAP kinase-dependent pathways, affecting processes such as DNA synthesis and mitosis. Suramin is an efficient drug, but has also multiple adverse effects. Its MOA has been the topic of many studies, and equally many potential targets have been proposed (reviewed by Wiedemar et al., 2020). Below, we will address a likely target and the MOA, based on recent metabolomic profiling studies.

4 Metabolism of the human-pathogenic life-cycle stages of trypanosomes

Trypanosomes have complex life cycles, involving both mammals (humans amongst them) and insects as hosts. Between both types of hosts, but also within each of them, trypanosomes undergo successive developmental transitions, involving changes of both morphology and metabolism (reviewed in Silvester et al., 2017; Walsh and Hill, 2021; Zuma et al., 2021). For drug treatment, only the trypanosomes in the human host are relevant. For T. brucei, these are the proliferating long-slender bloodstream forms (BSFs) and the non-proliferating short-stumpy forms. The long-slender forms develop from growth-arrested metacyclic trypanosomes injected by a tsetse fly into the skin from where they migrate to the hemolymphatic system. Some of the long-slender BSFs differentiate into non-proliferating short-stumpy BSFs which are trypanosomes pre-adapted to differentiate into procyclic forms (PCFs) in the digestive tract of a tsetse fly after it has taken them up in a blood meal. Furthermore, BSFs also enter extravascular spaces of various organs from where they may re-enter the blood/lymph circulation system (Trindade et al., 2016). In the case of T. cruzi, non-replicative metacyclic trypomastigote forms from the insect vector enter the mammal via wounds in the skin or exposed mucosa and then invade cells where, in the cytosol, they differentiate into replicative amastigote forms. After several proliferation cycles, the amastigotes develop into transient replicative intracellular epimastigote-like forms – similar to the epimastigotes living in the insect’s gut – which, in turn, differentiate in non-proliferating trypomastigotes (Alves and Colli, 2007). These latter forms burst from the infected cells into the extracellular environment from where they can infect neighboring cells or enter the bloodstream to infect cells of other tissues, or can be taken up by a new triatomine bug taking a bloodmeal (Zuma et al., 2021).

The metabolism of T. brucei and T. cruzi in different life-cycle stages has been reviewed elsewhere (e.g., Michels et al., 2021; Marchese et al., 2018; Parreira de Aquino et al., 2021, for carbohydrate, amino-acid and lipid metabolism, respectively). For T. brucei, only the metabolism of long-slender forms taken from the blood of infected animals and cultured in vitro, and cultured PCFs has been analyzed in great detail. However, some insight in the metabolism of short-stumpy BSFs in blood and trypanosomes residing in adipose-tissue has been inferred from proteomics and transcriptomics data (Gunasekera et al., 2012; Dejung et al., 2016; Briggs et al., 2021; Trindade et al., 2016; 2022), revealing some changes compared to the long-slender BSFs. T. brucei takes up many nutrients from the mammalian blood, including monosaccharides, lipids and amino acids, the latter by devoted transporters or as proteins via endocytosis. Glucose, abundantly present in blood, is imported and, through a high rate of aerobic glycolysis, is responsible for the major part of the trypanosome’s ATP production as well as providing metabolites for other processes. Unique to kinetoplastids and related diplonemids is the compartmentalization of the first 6 or 7 of the 10 glycolytic enzymes in peroxisome-related organelles called glycosomes (Figure 1) (Opperdoes and Borst, 1977; Andrade-Alviárez et al., 2022). Particularly in BSF T. brucei, which for their free energy supply are entirely dependent on glycolysis, the glycosomal protein composition is remarkable: about 90–95% of the glycosomal matrix proteins are glycolytic enzymes, which appear to be assembled as a tight complex (Misset and Opperdoes, 1984; Misset et al., 1986; Michels and Gualdrón-López, 2023). Each trypanosome contains 60 - 65 glycosomes which together comprise only 4–5% of the cellular volume. Correct compartmentalization of the glycolytic enzymes is vital for the trypanosomes (Haanstra et al., 2014). Oxygen-dependent reoxidation of the glycolytically produced NADH inside the glycosomes involves a shuttle mechanism to transfer the electrons to a mitochondrially located glycerol oxidase system. The tricarboxylic acid (TCA) cycle is highly repressed and oxidative phosphorylation (OxPhos) absent. Yet, some ATP can be formed by substrate-level phosphorylation in an acetate-producing pathway in the mitochondrial matrix from glycolytically formed pyruvate or threonine taken up by the cells. The acetate is used for fatty-acid synthesis (Mazet et al., 2013). By far, the major part of ATP produced by the proliferating trypanosomes is used for protein synthesis (proteome doubling and turnover) (Nascimento et al., 2023). Part of glucose enters the pentose-phosphate pathway (PPP) to provide NADPH for biosynthetic processes and dealing with antioxidant mechanisms.

Recent research has shown that cultured BSF T. brucei has also the capacity to grow on glycerol, performing gluconeogenesis to produce the hexose-phosphates necessary for the PPP and the synthesis of carbohydrate moieties of glycoproteins (Bringaud et al., 2021). However, the importance of this capacity remains to be demonstrated for trypanosomes in the infected hosts where the glycerol concentrations are much lower than those used in the in vitro studies.

Besides uptake of lipids, the trypanosomes cells have also the capacity for fatty-acid and sterol synthesis. Many enzymes of this core metabolic network have been shown to be important or essential for the viability of the BSF trypanosomes. In PCF T. brucei and other kinetoplastid species, where glycolysis plays a less prominent role in metabolism and the OxPhos system of the mitochondrion is dominant, the contribution of glycolytic enzymes to the glycosomal protein content is considerably lower, with other metabolic processes in these organelles upregulated.

The proteomic and transcriptomic data for short-stumpy BSF cells showed not only a decrease of enzymes producing metabolites important for the synthesis of nucleic acids, in line with the arrest of proliferation, but also a remarkable increase of mitochondrial proteins, such as those involved in OxPhos, and a decrease of some glycolytic components (Gunasekera et al., 2012; Dejung et al., 2016; Briggs et al., 2021). Furthermore, in BSF trypanosomes adapting to residence in adipose tissue, a lower growth rate and overall protein production were observed. But transcriptome and proteome analysis showed an upregulation of enzymes of purine metabolism, glycolysis, the PPP, sterol and lipid metabolism, the TCA cycle, fatty-acid ß-oxidation, the ATP producing acetate pathway in the mitochondrial matrix and glutamine/proline degradation, while enzymes involved in fatty-acid synthesis were downregulated (Trindade et al., 2016; 2022). Whether ß-oxidation in these cells serves for ATP production, or, for example, only to trim long-chain fatty acids to the 14-carbon myristate for use in the formation of glycosylphosphatidylinositol anchors, remains uncertain, because of the lack of detection of any component of the OxPhos system.

By constructing genome-scale models based on transcriptomic data, and validating the models by flux assays – involving measurements of the rates by which metabolic end products are formed and the activity of representative enzymes of different pathways of central metabolism – Shiratsubaki et al. (2020) succeeded to make realistic predictions of the metabolism of the different life-cycle stages of T. cruzi. The results are in agreement with those from previously performed focused studies of the metabolic network (e.g., Cazzulo, 1992; Atwood et al., 2005). In the proliferative epimastigotes, the glycolytic/gluconeogenic pathway is active, while somewhat lower activity occurs in the PPP and the TCA cycle, the latter being required for proliferation based on the consumption of amino acids forming directly or indirectly glutamate, such as proline, glutamine, histidine or glutamate itself. Moreover, earlier research had already shown that these cells have a flexible metabolism, with the ability to switch from consumption of carbohydrates (mainly glucose) as nutrients to amino acids (mainly proline) (Cazzulo, 1992; Barros-Álvarez et al., 2014a; Barisón et al., 2017), while a recent study demonstrated that also serine and threonine, taken up from the environment, play several roles in the biology of epimastigotes, including sustaining ATP levels (Alencar et al., 2025). In addition, when glucose becomes exhausted, fatty acids can also be taken up, oxidized and used for OxPhos as well as being used for anabolic processes (Souza et al., 2021). Also in the proliferative amastigotes, the PPP and TCA cycle are active, and high activity is found for glycolytic/gluconeogenic enzymes, despite earlier studies showing that glycolysis is not essential in this stage and no glucose transporter and glucose transport activity could be detected (Atwood et al., 2005; Silber et al., 2009). The cells may either rely on gluconeogenesis – when using proline and glutamate as carbon and free energy substrates – or import glycolytic intermediates from the host cell. In contrast, in the non-proliferative trypomastigotes the PPP is strongly, and the glycolytic/gluconeogenic pathways are partially down-regulated, whereas most TCA cycle enzymes are upregulated.

5 Metabolomics methodologies

Metabolomic analysis can be done using two different approaches: targeted and untargeted metabolomics. In targeted metabolomics, selected metabolites of a biological sample are accurately quantified, while non-targeted metabolomics involves the relative quantification of many, usually small (<1,500 Da) metabolites without a priori selection. In this review, we will focus on untargeted metabolomics studies performed on trypanosomes treated with (potential) drugs, and how from specific changes in the intracellular metabolome conclusions were drawn about the cellular target(s) of the compounds. In such studies, a drug is added to in vitro cultured trypanosomes, and samples taken from the culture at different periods of time after the initiation of treatment. The metabolism of the cells in the samples is immediately quenched, and cell extracts prepared. The metabolome content of the extracts is then analyzed by either mass spectrometry (MS) or nuclear magnetic resonance (NMR). Description of details of the successive steps of this experimental design is beyond the scope of this paper, but can be found in several reviews and protocol papers (e.g., Creek et al., 2012; 2013; Creek and Barrett, 2014; Vincent and Barrett MP, 2015; Fall et al., 2022; 2024a; MacKenzie et al., 2026).

Important to mention here is that perturbations of trypanosome metabolism must be performed at one or different sublethal doses of compounds resulting in measurable growth retardation (e.g., at EC₅₀ and/or EC₉₀ – the effective concentration resulting in 50%/90% growth inhibition) for different periods of time. This will enable to distinguish metabolic changes due to alteration of the activity of the primary target from secondary effects, the latter being changes by which the metabolic network attempts to adapt to the new situation and/or stress responses resulting the metabolic assault.

6 Metabolomic profiling for identification of (potential) drug targets

6.1 Identification of targets in Trypanosoma brucei

6.1.1 Existing drugs and drug candidates

6.1.1.1 Suramin

Suramin has been used as drug for treatment of multiple diseases, parasitic and viral ones as well as cancer, snakebite and autism (reviewed in Wiedemar et al., 2020). It has been introduced already more than a century ago for treatment of the rhodesiense form of HAT, but only for its stage 1; it does not cross the blood-brain barrier. Multiple targets for it have been proposed. As a drug, it is a remarkably peculiar molecule, not compliant with the so-called Lipinski rule of 5: it is very large, with a molecular weight of 1,297 Da, very low logP value, and six sulphonic acid groups giving it a highly negative charge. This charge is the reason for its many proposed targets, as well as for its side effects; it easily binds electrostatically to many proteins and other macromolecules. Trypanosomes internalize it by endocytosis, for which two routes have been demonstrated: bound to low-density lipoprotein (LDL) in blood and bound to the trypanosome’s invariable surface glycoprotein ISG75, with the latter route probably being the predominant one (Vansterkenburg et al., 1993; Alsford et al., 2012). The complexes are then routed to the lysosome from which the suramin is released into the cytosol through facilitator transporter MFST.

Willson and coworkers showed already in 1993 that glycolytic enzymes purified from isolated glycosomes are inhibited by suramin at IC₅₀ values of 3–100 μM (Willson et al., 1993). Indeed, a characteristic of trypanosomatid glycosomal enzymes is their high isoelectric point (Misset et al., 1986), explaining the affinity for the drug, while its sulphonic acid groups may mimic the phosphate group of glycolytic intermediates, allowing blockage of the active sites. However, the IC₅₀ values for enzyme inhibition differed considerably from the EC₅₀ values of ∼0.015–0.070 μM (dependent on strain and culture conditions) reported in various papers for in vitro cultured BSF T. brucei, suggesting that these enzymes may not be the in vivo targets of suramin. Despite this discrepancy, two research groups performed metabolomic profiling on in vitro cultured BSF T. brucei treated with suramin, providing nonetheless strong support for the notion that glycosomal glycolytic enzymes form a primary target of the drug. First, Zoltner et al. (2020) showed, using ³H-suramin, that the drug rapidly (within 1 h) accumulated in the trypanosomes, depletes the cytosolic ATP within 12 h after addition, while the mitochondrial ATP level is increased, possibly by an increased flux through the acetate producing pathway. After 12 h, levels of only a limited number of metabolites were shown to be affected, such as an increase of some amino acids. In contrast, after 21 h more significant changes were observed, such as an increase of several glycolytic intermediates, most drastically intracellular pyruvate but also fructose 6-phosphate (F6P), fructose 1,6-bisphosphate (FBP) and glycerol 3-phosphate (Gly3P), while phosphoenolpyruvate (PEP) was decreased. The drug did not significantly alter the proteome within the first 24 h, while specifically was shown that levels of selected glycolytic enzymes (phosphofructokinase (PFK), phosphoglycerate kinase (PGK) and pyruvate kinase (PYK)) were not affected, and the morphology and volume of glycosomes only minimally changed.

Fall et al. (2024b) also detected an important increase in the levels of some glycolytic intermediates, in their analysis already after 12 h: F6P, 2-phosphoglycerate (2PG) and pyruvate (Figure 1), as well as of glucose and other monosaccharides, some TCA intermediates, several amino acids and various metabolites directly derived from glycolytic intermediates (ATP and ADP levels were not determined in this study). They explained the increase in intracellular concentration of amino acids in both studies by an inhibition of an early common step in protein synthesis, downstream of the amino-acyl tRNA synthetases which are distinct for each amino acid. Amino acids are likely taken up from the rich medium by transporters, rather than synthesized de novo by the BSF trypanosomes. The increase of several glycolytic intermediates and metabolites derived from them, also observed in both studies, strongly suggested inhibition of glycolytic enzymes. The decrease of cytosolic/total cellular ATP content reported by Zoltner and coworkers is in agreement with inhibition of the glycolytic flux. A possible explanation for the effect on glycolysis, despite the important discrepancy between the IC₅₀ and EC₅₀ values is a two-step accumulation of suramin into glycosomes to reach the IC₅₀ values. Zoltner et al. (2020) showed that the drug rapidly accumulated 9-fold in trypanosomes, to 1.8 μM at an external concentration of 250 nM. If the suramin will predominantly be sequestered within the organelles, which occupy ∽1/20th of the cellular volume, a maximally 9 x 20 = 180-fold accumulation may be achieved, implying an intraglycosomal concentration ∼3 μM would be reached at the external concentration of 17 nM used by Fall and coworkers, sufficient to partially inhibit several glycosomal glycolytic enzymes. It is conceivable that suramin, which due to its large size and high negative charge would be unable to diffuse passively through biological membranes, enters glycosomes bound to the positively charged matrix proteins, which are synthesized in the cytosol and are imported in folded, oligomeric form into the organelles. This would result in a slow build-up of the intraglycosomal concentration of suramin, in line with its slow trypanocidal action. But with a growth doubling time of ∼6 h, >90% of new glycosomal protein content (i.e., ∽340 mg/mL; Misset et al., 1986) will have been formed after 24 h, sufficient time for accumulating suramin to low micromolar concentration in the organelles. Two different – not necessarily mutually exclusive – ways are conceivable as to how glycolysis within glycosomes might be affected by suramin; (i) by non-specific binding suramin may affect protein-protein interactions, disrupting the multi-enzyme complex and consequently decreasing the glycolytic flux, and (ii) by binding to ligand-binding sites of individual enzymes inhibiting their activities.

Zoltner et al. (2020) also demonstrated that, after 2 days of suramin exposure, expression of most proteins involved in glycolysis was still unchanged, except an 1.5-fold increase of the glucose transporter. Furthermore, proteome changes observed were similar to those seen, although to a larger extent, during the differentiation of BSF to PCF trypanosomes. These included 1 – 4-fold increases of pyruvate phosphate dikinase (PPDK, an alternative enzyme for interconverting PEP and pyruvate), enzymes of the succinic branch of glucose metabolism (mostly glycosomal enzymes converting PEP into succinate), all TCA-cycle enzymes, and all enzymes of proline catabolism. The authors attributed the increased pyruvate level observed after 21 h to the increased PPDK expression. However, as argued by Fall et al. (2024b), this is unlikely because PYK in BSF T. brucei is in large excess and exerts no control on the glycolytic flux (Albert et al., 2005; Haanstra et al., 2017), while the expression of PPDK is almost undetectable, implying that its 3-fold increase would still keep its cellular activity negligible. A more likely explanation is inhibition of a process downstream of PYK, such as pyruvate’s excretion or further metabolism. Although direct evidence for it is lacking, the observation that also PEP and phosphoglycerates increase in levels do support the notion.

Finally, it should be noted that the experimental set-up in the two studies differed significantly in some respects, for example, in cell density and glucose concentration (1 × 10⁸ cells/mL, in medium with ∼60 mM glucose and 1–2 x 10⁶ cells/mL with ∼20 mM glucose in the work by Zoltner et al. and Fall et al., respectively), we don’t think it affected the outcome importantly. The results indeed show considerable similarity. In both cases, the glucose concentration is well above the K _m of the glucose transporter (∼1 mM; Tetaud et al., 1997) and the average blood concentration about 5 mM, thus the uptake rate occurs in all cases close to the V_max. Moreover, the glucose transporter has nearly all control on the glycolytic flux (e.g., Bakker et al., 1999). The difference in the extracellular glucose concentration will thus have had little effect on the intracellular glucose concentration, the glycolytic flux and the rate of ATP production. With regard to the calculation of the intracellular suramin concentration, it is relevant that suramin is a slow, late acting inhibitor. No significant effect on the growth rate is observed during the first 20 h after the addition of the compound at around the EC₅₀ value. Therefore, the calculation, although a rough estimation, seems not unrealistic.

6.1.1.2 Eflornithine

Eflornithine was initially developed for treatment of cancer. It is a suicide inhibitor of ODC, an enzyme of the pathway for the biosynthesis of the polyamines putrescine, spermidine and spermine. Interestingly, the compound appeared to be a very useful drug against African trypanosomes (Bacchi and Yarlett, 1993; Barrett, 2025). These trypanosomes have, in contrast to mammalian cells and T. cruzi (Reigada et al., 2016), no plasma-membrane transporter for polyamines and thus cannot accumulate them from the environment, but rely entirely on their biosynthesis. Polyamines are crucial for trypanosomatids, because spermidine is linked to two glutathione molecules to form trypanothione that performs a similar role as played by glutathione in other organisms, such as in defense against oxidative stress. Eflornithine enters trypanosomes via the amino-acid transporter TbAAT6. metabolomic profiling performed by Vincent et al. (2012) on BSF T. brucei treated with eflornithine confirmed ODC as its target, in agreement with the role for which the drug had been developed and earlier gene-knockout studies. Within hours after adding a dose of eflornithine that causes division arrest after 48 h and is lethal between 48 and 55 h, ornithine accumulates and the levels of the polyamines putrescine and spermidine decrease markedly, although the level of trypanothione disulfide did not during the first 24 h (Figure 2 ). Within 24 h, accumulation was also observed for molecules putatively identified as N-acetyl spermidine, N-acetyl lysine and N5-(L-1-carboxyethyl)-L-ornithine, indicating an upregulation of derivatisation of accumulated ornithine and other polyamine metabolites. Over 24 h, also major decreases were observed among phospholipids, later followed by a decrease of the majority of cellular metabolites. Since polyamines have previously been shown to be important for membrane stability, the authors interpreted these latter changes as damage to the cell membrane followed by leakage of metabolites. Eflornithine appeared to be trypanostatic during 48 h, with the subsequent killing attributed to the delayed occurrence of membrane damage and general loss of metabolites.

6.1.1.3 Benzoxaboroles

Benzoxaboroles are a group of boron-containing compounds that, in the last 2 decades, have shown promising results in treating a wide variety of diseases (Zoltner et al., 2024). In addition, different enzymes have been proposed as targets for inhibition by these compounds. In 2010, two benzoxaboroles with trypanocidal effect were also identified by screening of a library of benzoxaborole 6-carboxamides using in vitro cultured T. brucei (Nare et al., 2010). The same study showed that these compounds, when orally administered, were able to cure both stage 1 (hemolymphatic) and stage 2 (CNS) trypanosomiasis in mice. Subsequent studies provided detailed pharmacokinetic data about treatment of trypanosomiasis with oxaboroles. A genomic analysis of resistance of trypanosomes against oxaborole SCYX-7158/AN5568 (compound later renamed to acoziborole) (Jones et al., 2015) detected multiple changes, including gene copy amplification of the RNA splicing and polyadenylation specificity factor subunit 3 (CPSF3). This factor with endonuclease activity was subsequently identified as a benzoxaborole target in apicomplexan parasites (Sonoiki et al., 2017; Palencia et al., 2017). Begolo et al. (2018) studied the MOA of AN7973, a benzoxaborole structurally similar to AN5568. The toxic effect on BSF T. brucei was first shown to result from arrest of protein synthesis that could subsequently be attributed to inhibition of trans-splicing and coupled polyadenylation involved in the mRNA maturation as occurs in these organisms (reviewed in Clayton, 2019). During trans-splicing, a methylated 39 nt leader sequence is added to precursor transcripts to create mature mRNAs. This short leader sequence is derived from a methylated, 139 nt-long precursor form. Because methylation of the spliced leader precursor RNA, involving the methyl donor S-adenosyl-L-methionine (AdoMet), is known to be important for trans-splicing, it was initially considered a possible target for AN7973; however, the compound did not appear to affect this process. After several hours of AN7973 treatment, trypanosomes were still motile indicating that splicing inhibition was unlikely due to ATP depletion. But a metabolomic analysis of samples taken 5 h after AN7973 addition showed significant increases in AdoMet and 5′-methylthioadenosine (5′-MTA), and modified lysines (methyl lysine, dimethyl lysine, trimethyl lysine and acetyl lysine) (Figure 3).

A metabolomic analysis of BSF T. brucei treated with AN5568 was also performed by Steketee et al. (2018) who similarly observed increases in the levels of AdoMet and 5′-MTA as the most profound changes, as well as that of adenine that is formed as a byproduct of AdoMet degradation. Other significant alterations after the treatment of BSF trypanosomes included a range of metabolites involved in carbohydrate and lipid metabolism, but mostly amino acid metabolism, including the modified lysines, while similar changes were observed in treated PCF trypanosomes, although then higher concentrations and a longer exposure time to AN5568 were required. These data demonstrate that the benzoxaborole disturbs methionine metabolism; specifically the accumulation of AdoMet, an essential methyl group donor, seemed relevant with regard to the findings about interference of the benzoxaborole with trans-splicing and the role of CPSF3. This is corroborated by the observation of similar metabolic phenotypes when trypanosomes were treated with sinefungin, an AdoMet analog that is a known non-specific methylation inhibitor and has previously been shown to prevent RNA trans-splicing. When Begolo et al. (2018) tested a variety of benzoxaboroles which different EC₅₀ values on trypanosomes, they observed that those causing faster killing induced the AdoMet/5′-MTA pattern. Furthermore, this experiment revealed a remarkably good correlation between splicing inhibition and the average increase in AdoMet and 5′-MTA accumulation. In a control experiment with the mRNA synthesis inhibitor actinomycin D, also large increases were seen in the levels of methylated amino acids and significant, although smaller, in those of AdoMet and 5′-MTA. Together, these data suggested that the increases in methylated and acetylated lysine, and possibly also those of AdoMet and 5′-MTA could be an indirect consequence of a decrease in the production of mRNAs.

To provide further proof that CPSF3 is the primary target of benzoxaboroles in T. brucei, Begolo et al. (2018) overexpressed it, resulting in a three-fold increase of its EC₅₀ value. A similar conclusion was drawn by Wall et al. (2018) who identified it as target of the benzoxaboroles based on screening for clones with resistance against drugs in an overexpression library of T. brucei. Knockdown of CPSF3 provided its genetic validation as essential, and GFP tagging confirmed its nuclear localization. Molecular docking using a homology model of the protein and CRISPR-Cas9-based editing indicated that blocking of the trypanosome’s CPSF3 active site by AN5568 and inhibition of mRNA processing by the parasite seem feasible, but not for the host CPSF3. Furthermore, recent unpublished data suggest that the high susceptibility of trypanosomes to benzoxaboroles may also result from a high turnover of the CPSF3-drug complex triggered by its SUMOylation (Zoltner et al., 2024).

6.1.2 Compounds in research

6.1.2.1 Pyrrolopyridines

Because glycolysis is essential for BSF T. brucei, as it is for some host-type cells, studies have been performed to develop parasite versus host selective inhibitors for several enzymes of this pathway (reviewed by Verlinde et al., 2001; Barros-Alvarez et al., 2014b). In one of these research lines, trypanosomatid PFK-specific potent inhibitors, capable of killing in vitro cultured BSF T. brucei, were identified in a compound library screen (Brimacombe et al., 2013). Trypanosomatid PFK is a glycosomal enzyme that differs considerably from its cytosolic human counterparts, both structurally, with only 23–26% sequence identity, and kinetically (Fernandes et al., 2020). Structure-activity relationship (SAR)-guided optimisation of the library hits yielded compounds with a pyrrolopyridine scaffold that were shown to bind in a trypanosomatid PFK-specific hydrophobic allosteric pocket of the active site, thus locking the enzyme in its inactive T-state (McNae et al., 2021). The best compounds from the series (among which 1-[(3,4-dichlorophenyl)methyl]-5-[2-(dimethylamino)ethyl] pyrrolo [3,2-c] pyridin-4-one, a.k.a. CTCB405) showed IC₅₀ values of 0.18 ± 0.03 μM on purified T. brucei PFK and killed in vitro cultured BSF parasites very quickly after their administration with an EC₅₀ of 0.37 ± 0.03 μM and a selectivity index (SI) 56 to human HepG2 cells. The rapid killing was expected from the complete reliance of the parasites on glycolysis. The compounds did not cause any apparent adverse effects in toxicity studies with mice and were able to cure the stage 1 disease in T. brucei-infected mice upon one-day oral administration.

To demonstrate that PFK is indeed the in vivo target of the pyrrolopyridines, Nare et al. (2023) performed metabolomic profiling. Already 5 min after its addition to the culture could compound CTCB405 be detected in the BSF trypanosomes at a level that hardly further increased. Among the most prominent changes observed during the 60 min sampling period was, also after 5 min, a large increase in the F6P level and that of PEP gradually increasing over time, while pyruvate showed a gradually decreasing trend (Figure 4). These changes confirmed PFK as the primary target of the pyrrolopyridines since F6P is the PFK substrate. Moreover, the increase in PEP and decrease in pyruvate can be attributed to a decreased PYK activity, because it is likely that F1,6BP (not detected in the dataset), the product of the PFK reaction and an allosteric activator of PYK (Morgan et al., 2014), has its level decreased. The ATP level dropped quickly after CTCB405 addition, reaching a lowered plateau or after a while increasing slightly again, dependent on the trypanosome cell line (even though cells continued to lose motility). The lack of drastic depletion of cellular ATP, still allowing motility, and even a slight recovery, suggests that part of the pyruvate can be channeled to the mitochondrion for substrate-level phosphorylation in the acetate synthesis, to compensate for the decrease in glycolytic ATP production. Threonine metabolism may also contribute.

Other notable changes were (i) an increase in the level of L-carnitine and a decrease of that of O-acetylcarnitine, and (ii) glycerophospholipids including phosphatidylethanolamine, phosphatidylcholine, phosphatidylserine and sphingomyelin, although not in a detectably consistent pattern. The change in L-carnitine and O-acetylcarnitine levels suggests an increased export from the mitochondrion of acetyl - CoA – produced in the above-mentioned flux from pyruvate and/or threonine – by the carnitine shuttle across the mitochondrial membrane (Figure 4). The changes in lipid compounds are possibly secondary. BSF trypanosomes acquire their fatty acids and lipids by two routes: endocytosis and de novo synthesis (Nascimento et al., 2023). Both routes require ATP. The lower ATP level may thus affect the fatty-acid and lipid levels. Decreased protein expression and cell damage due to the ATP drop may also contribute.

6.1.2.2 3-O-p-(Z/E)-coumaroyltormentic acid

Leclercq and coworkers isolated a mixture of 3-O-p-(Z/E)-coumaroyltormentic acids (CTA) from the West Africa-native tree Vitellaria paradoxa which killed BSF T. brucei with EC₅₀ value < 1 mg/L and a SI > 20 towards human cells (Catteau et al., 2021). To determine its target, they then performed a metabolomic analysis, using the same compounds isolated from an in vitro-cultivated apple cell extract which allowed more accessible production with higher yield (Mamede et al., 2025). After exposure of trypanosomes for 3 h at 5 and 10 x EC₅₀, by far the most prominent changes were observed in levels of metabolites that must have been derived from tryptophan (Figure 5). Although tryptophan metabolism in trypanosomatids remained so far been poorly studied, with only a few enzymes identified in databases, the metabolome analysis clearly indicated the presence of two catabolic branches, both of them having levels of several intermediates affected by CTA administration; one branch leading to the production of indole 3-acetate and related compounds, and the other one the kynurenine pathway. Indeed, previous research has shown that BSF T. brucei excrete high quantities of indole 3-acetate and indole 3-pyruvate resulting from tryptophan metabolism (el Sawalhy et al., 1995). The specific target(s) of CTA could not yet be identified. Levels of several metabolites in both branches of tryptophan catabolism were decreased, yet the only step shared by both branches is tryptophan import into the cell. But yet, the intracellular concentration of the amino acid was increased. It thus seems likely that early steps in both branches are targeted, with the aminotransferase in the pathway leading to the indole compounds and a not yet identified tryptophan-2,3-dioxygenase and/or indoleamine-2,3-dioxygenase at the start of the kynurenine pathway as candidates. Interestingly, the candidate tryptophan aminotransferase is most likely an enzyme that is actually structurally most related to aspartate aminotransferases (ASATs). The cytosolic ASAT of T. brucei appeared to possess a broad substrate specificity with its highest affinity and catalytic efficiency for tryptophan, in contrast to the mitochondrial ASAT that is highly specific towards aspartate (Marciano et al., 2008). Inhibition of this cytosolic ASAT may also affect the levels of other metabolites, like oxaloacetate, which acts as the preferred amino group acceptor in the reaction catalyzed by this enzyme with tryptophan as a donor and leading to the formation of aspartate and indole 3-pyruvate (McGettrick et al., 2016), or alternatively 2-oxoglutarate as the acceptor resulting in glutamate production. Supporting this notion was an observed decrease in aspartate level, suggesting that the ASAT might indeed be a CTA target, but the level of oxaloacetate was not found to be increased but decreased. However, this may result from its decreased production, that in BSF T. brucei occurs mainly by phosphoenolpyruvate carboxykinase in the succinate-producing branch of glycolysis (see Figure 1). Indeed, also levels of the other intermediates of this branch dropped after CTA treatment: PEP, malate, fumarate and succinate. This may result from either a decreased glucose consumption, or from a redirection of a major part of glucose 6-phosphate away from glycolysis to the pentose-phosphate pathway allowing the cell to cope with the stress caused by the toxic effects of CTA.

What exactly would make tryptophan metabolism essential for trypanosomes remains to be determined. Relevant, however, is that cytosolic ASAT has been shown, by RNA interference, to be essential (McGettrick et al., 2016). This was attributed to its role in maintaining a sufficient intracellular level of aspartate, which is essential in purine salvage and de novo pyrimidine biosynthesis but has a low uptake rate in T. brucei.

6.1.2.3 Halogenated tryptophan derivatives

Other evidence that the cytosolic ASAT of T. brucei plays an important role in tryptophan metabolism and is a likely drug target has been obtained in a study by Cockram et al. (2020). They developed several halogenated (Cl-, Br-, I-) analogs of tryptophan that showed a capacity to kill cultured BSF trypanosomes, with no detectable activity on human-derived HeLa cells. Particularly tryptophan derivatives with the substitution at position 7 of the indole ring were effective (EC₅₀ = 42.0 – 47.4 μM). While these modified compounds would, most likely, be taken up by the aromatic amino acid transporter (Marchese et al., 2018), an additional ∼15-fold increase of the trypanocidal activity was achieved by methyl esterification of the amino acid’s carboxyl group, supposedly to increase the permeation of the compounds through the plasma membrane, followed by their cleavage by cytosolic esterases to release the free acids. Supplementation of growth medium with tryptophan above its usual concentration significantly reduced the trypanocidal activity of these halogenated compounds, indicating that the analogs interfered with the amino acid’s natural role in the cell. But proteomic analysis of trypanosomes treated with 7-³⁵Cl-tryptophan at EC₁₀ showed that the substituted analog was not incorporated into proteins.

Metabolomic analysis of extracts from BSF T. brucei treated with the 7-bromo-tryptophan ester at EC₁₀ and EC₃₀ provided indeed evidence that it disrupts tryptophan metabolism. Intracellular levels of its transamination product indole 3-pyruvate were 61% and 75% reduced, respectively, compared to untreated cells, and even 91% in spent medium from EC₅₀-treated cells. Also affected were levels of intracellular and excreted indole 3-lactate, a metabolite further downstream in this tryptophan-catabolism branch, and those of phenylpyruvate, 4-hydroxyphenylpyruvate and imidazole 4-pyruvate which are transamination products of phenylalanine, tyrosine and histidine, respectively. Importantly, the cytosolic ASAT of T. brucei has been shown to exert catalytic activity – better than to aspartate – not only to tryptophan, but also to these other aromatic amino acids (Marciano et al., 2008). Furthermore, the 7-bromo-tryptophan ester could not be detected in the cell extract, in contrast to small amounts of the brominated free acid as well as its transamination product 7-bromo-indolepyruvate and another, not identified brominated metabolite. But these brominated metabolites were not detected in the spent medium, suggesting they were not exported. Thus, the brominated tryptophan itself appeared to be a substrate for the transaminase. Together, these data suggest that the trypanocidal activity of the halogenated tryptophan analogs is due to their inhibitory activity, or that of their metabolic products, on the cytosolic ASAT that, as described above in the cited RNAi experiment by McGettrick et al. (2016), is an essential enzyme in BSF T. brucei.

6.1.2.4 Alkylresorcinol

Siheri et al. (2019) analyzed propolis samples from Libya for anti-trypanosomal activity. Propolis, or “bee glue,” is a resinous mixture produced by honeybees to seal their hives to protect them from external threats. Its composition is dependent on the resins of the plants visited by the bees. Propolis is known for its activity against bacteria and a wide variety of multicellular and protistan parasites, including trypanosomatids, and its anti-inflammatory, and antioxidant properties (Siheri et al., 2019). By fractionation of propolis samples selected for their antiparasitic activity, the authors obtained fractions containing highly enriched compounds with significant EC₅₀ values against BSF T. brucei, less activity against several other parasites analyzed, and no measurable effects on the HFF human cell line at up to 2 mg/mL. The compounds in the fractions were identified by NMR spectroscopy. The most potent fraction, with trypanocidal activity of EC₅₀ = 0.70 μg/mL, contained 60% cardol, a.k.a bilobol, an alkylresorcinol. To determine the MOA of this cardol-rich fraction, metabolomics was performed on BSF T. brucei, exposed to the compound at 0.5 x EC₅₀ for 10 h. Major changes were detected for a wide range of lipids; a significant decrease in the levels of several phosphocholine and phosphoserine lipids containing highly unsaturated long-chain fatty acids, with a corresponding increase of choline and CDP-choline, probably a reaction to counter the lipid decrease. Also, a drop in the levels of several phosphatidylglycerol and phosphatidylinositol (PI) lipids was seen. Many of the lipids that were lowered in the cell extract were found correspondingly elevated in the growth medium of the treated cells. Levels of some fatty acids were also increased in the medium. These data suggest the occurrence of cell lysis, although viability measurements did not indicate extensive cell death under the conditions used. The spent medium of treated cultures also contained traces of high-energy phosphates. Based on these results, the authors proposed that the cardol-rich fraction targeted the cell membrane, acting as a surfactant, extracting lipids from the membrane making it leaky. Such a MOA would be similar to that of miltefosine, a known anti-leishmanial drug that affects the fluidity of membranes. Moreover, this explanation is also in agreement with a previous report about electrolyte leakage from erythrocytes caused by alkylresorcinols (Kozubek and Tyman, 1999).

6.1.2.5 Monoenone curcumin

Curcumin, obtained from roots and rhizome of the plant Curcuma longa, native to South and South-East Asia, has a long history in traditional medicine. It is mainly being considered for its anti-inflammatory, antioxidant and anticarcinogenic properties (Liu et al., 2022). But curcuminoids have also been shown to be effective in some infectious diseases, and have modest effect on the viability of some in vitro cultured pathogenic protists such as Plasmodium and Leishmania. But Changtam et al. (2010) reported that a series curcuminoid analogs carrying a C7 linker bearing a C4-C5 olefinic linker with a single keto group at C3 (enone linker) displayed a highly promising activity against BSF T. brucei: EC₅₀ 0.023 – 0.082 μM against different strains for the most active compound with a SI of 453 over the human HEK293 cell line. Subsequently, the MOA of the enone linker lead compound (1,7-bis(4-hydroxy-3-methoxyphenyl)hept-4-en-3-one, called AS-HK014) was assessed on BSF T. brucei by Alkhaldi et al. (2015). First, an AS-HK014-resistant (28-fold) cell line (TA014) was developed by long-term exposure to the compound. Growth rates of wild-type and resistant cell lines were identical and the resistant cells were not cross-resistant to curcumin. Previously, it had been reported that the curcumin and other common phytochemicals affect the integrity of biological membranes rather than acting on specific targets. However, the authors did not detect any effect of AS-HK014 on the calcium conductance, plasma membrane potential of wild-type T. brucei, or its levels of the intracellular signaling molecule cAMP upon treatment with curcumin or the lead compound at concentrations much higher than the EC₅₀, in contrast to their controls (calcium ionophore A23187, gramicidin and the phosphodiesterase inhibitor CpdA). Next, they performed metabolomics to assess the effect of AS-HK014 on cultured BSF trypanosomes incubated without or with AS-HK014 at 200 nM for 30 min, a period without apparent effect on cell survival or plasma membrane integrity. A striking effect on intracellular thiol levels was observed: >99% reduction in trypanothione level, while those of glutathione and S-glutathionyl-L-cysteine were reduced by >90% compared to non-treated cells. In a separate experiment, intracellular sulfhydryls were derivatized using the monobromobimane method to prevent oxidation in order to distinguish oxidized from reduced thiols. This allowed to establish that AS-HK014 treatment leads to a dramatic decrease in levels of both trypanothione redox forms and similar changes in derivatized glutathione, S-glutathionyl-L-cysteine and glutathionylspermidine (Figure 6). Together, these observations strongly suggest that the inhibition of trypanothione synthesis is due to the depletion of glutathione. A small decrease in the level of derivatized cysteine was also observed, but levels of the two other amino acids required for glutathione synthesis, glycine and glutamate, were not affected. No changes were observed either for metabolites involved in the synthesis of spermidine and putrescine: ornithine and AdoMet. Thiol levels in the resistant TA014 cells were similar to those in untreated wild-type cells, i.e., not affected by the compound. The effect of AS-HK010 could be attributed to the formation of an adduct with glutathione and trypanothione (Figure 6). However, the accumulation of the adducts in sensitive parasites was much lower than in resistant ones, indicating that not the adducts are toxic to the cell, but that the uptake of the compound is driven by adduct formation. The observation that thiol depletion is minimal in TA014 cells indicates their resistance is due to an increased capacity of maintaining their glutathione and trypanothione levels in the presence of the AS-HK014. Curcumin, at its corresponding toxic level of 3 μM, has a different mode of action than the monoenone curcumoids, because no profound changes were detected on thiol levels in wild-type and TA014 cells in a similar metabolomic analysis. Moreover, whereas addition of glutathione significantly increased the EC₅₀ of AS-HK014, it had no effect on the toxicity of curcumin, neither did it affect the EC₅₀ for the TA014 cell line, indicating that the residual inhibitory action of AS-HK014 in these resistant cells involves a mechanism different from that in the wild-type.

6.1.2.6 Fluoropyrimidines

Purines and pyrimidines are essential for the synthesis of nucleic acids and, therefore, their metabolism is being considered for development of drugs to treat infectious diseases and cancer. A well-known example is the pyrimidine nucleobase analog 5-fluorouracil, widely used in cancer therapy, whose principal MOA is inhibition of thymidylate synthase thus blocking DNA synthesis (Longley et al., 2003).

African trypanosomes are incapable of synthesizing the purine heterocycle de novo, but have transporters and enzymes to salvage them from the host. In contrast, these parasites are capable of both synthesis and salvage of pyrimidines (de Koning et al., 2005). Ali et al. (2013) tested the effect of a large number of pyrimidine nucleoside and nucleobase analogs on BSF T. brucei. Among the many compounds tested, 5-fluorouracil (5-FU) and 5-fluoroorotic acid (5-FOA) were highly effective with EC₅₀ values of 35.9 and 14.1 μM, respectively. Growth arrest was almost immediately induced, but killing of trypanosomes was delayed, even at 500 μM. Interestingly, the trypanocidal effect of 5-FOA was 2.5 times higher than that of 5-FU, whereas the uptake of orotate and its fluorinated analogs was <1% of the transport of uracil and its analogs which are excellent substrates for the high-affinity transporter TU3 of BSF trypanosomes. Furthermore, resistant parasites (131-fold to 5-FU and 83-fold to 5-FOA) were developed by exposure to stepwise increasing concentrations. Each of these cell lines displayed cross resistance to the other pyrimidine analog. The transport of 5-FU into resistant parasites was 76% reduced, although that of uridine was not affected. To determine the MOA of these fluorinated pyrimidines, metabolomic analysis was performed on wild-type and resistant trypanosomes after 12 h exposure to 100 μM of compound. In both cell types, 5-FU was found incorporated into several metabolites, such as 5F-UMP, 5F-UDP and 5F-UTP (Figure 7), although to a lower extent in the resistant parasites, in line with their reduced 5-FU transport capacity.

For 5-FOA, it was concluded that, in contrast to 5-FU, transport was not the main mechanism of resistance, because very similar levels of it were detected in wild-type and resistant cells. And despite the very large difference in transport of both compounds, they gave rise to very similar levels of active metabolites, converging on 5F-UMP, indicating that the production of the latter from 5-FOA is more efficient than that from 5-FU.

Interesting was also the significant amounts of 5F-UDP-glucose and 5F-UDP-N-acetyl-glucosamine produced upon addition of 5-FU and 5-FOA, suggesting the possibility of interference with protein glycosylation and/or the synthesis of glycosylphosphatidylinositol anchors. However, no changes in variant surface proteins were detected upon treatment with the fluorinated analogs. An alternative explanation for their toxicity, i.e., incorporation into RNA, was considered because of the detection of 5F-UTP. Indeed, upon purifying and analyzing RNA, significant amounts of 5F-UMP were detected. In contrast, no incorporation of fluorinated analogs into DNA was found, in agreement with the absence of 5F-dUDP and 5F-dUTP, different from what happens in human cells via the formation of 5F-2′deoxyuridine and 5F-dUMP, the latter being the substrate for thymidylate synthesis. Another possibility is the incorporation of the fluorinated compounds in precursors for lipid biosynthesis. Indeed, levels of several of its intermediates were changed. Thus, the trypanocidal effect of these compounds is most likely multifactorial.

6.1.2.7 Aroyl guanidine derivatives

In a high-throughput screening campaign, an aroyl guanidine derivative was identified exhibiting an EC₅₀ = 2.6 μM against BSF T. brucei with a SI = 29 over mammalian HEK293 cells. Subsequent SAR-guided optimisation yielded a number of compounds with increased potency and selectivity (Varghese et al., 2024). Metabolomics analyses were performed on two batches of trypanosomes exposed for 5 h to 1 μM of improved derivatives, respectively compound 18 (EC₅₀ 4.4 μM, SI > 19) and 38 (EC₅₀ 2.5 μM, SI > 33). Remarkably, they caused opposing effects with most significant changes in levels of intermediates of nucleotide and carbohydrate metabolism. Eight additional derivatives were studied, with four – all having substitutions of the amide ring of the initial hit acyl guanidine, like compound 18 – exhibiting an 18-like metabolic profile, and four others with – like compound 38 substituted with halogens in the aniline ring – causing similar metabolome changes as no 38. The ‘class 38’ compounds induced a significant decrease in the levels of all nucleotide triphosphates (NTPs) and their deoxynucleotide forms, and an increase of those of the corresponding diphosphate (NDPs), monophosphate (NMPs), nucleoside and purine/pyrimidine base forms. Inverse results were observed when trypanosomes were treated with ‘class 18’ compounds. Thus, the two classes have opposite effects on the energy balance of the parasites. Additionally, contrasting results were seen in glycolytic metabolism, where the ‘class 38’ compounds caused an increase of the glycolytic flux as inferred from the increased secretion of the end products of glucose metabolism – pyruvate, glycerol and succinate – and a significant depletion of the PPP intermediate 6-phosphogluconate. Inverse results were found after treatment with the ‘class 18’ compounds. The fact that an increased glycolytic flux, thus increased ATP production, is accompanied with decreased cellular energy balance, suggests an increased ATP consumption. Further studies are required to determine how the two classes of aroyl guanidines cause these and the reciprocal changes, and how both classes have a detrimental effect on trypanosome viability. An unknown regulatory metabolic mechanism is probably involved in causing the opposite changes observed after treatment by the two classes of compounds.

6.1.2.8 3-(Oxazolo [4,5-b]pyridine-2-yl)anilide

Phenotypic library screening with T. brucei yielded a compound with an oxalopyrimidine core from which, after SAR-directed optimisation, 3-(oxazolo [4,5-b]pyridine-2-yl)anilide (OXPA) was obtained with an EC₅₀ of 0.07–0.17 μM and minimal toxicity to human cells (Ferrins et al., 2013). Metabolomics analysis of BSF T. brucei exposed for 5 h to 0.85 μM OXPA showed significant accumulation of ceramides as the most prominent changes (Stoessel et al., 2016). A lipidomics analysis confirmed that 85% of the impacted lipids were ceramides of which 11 out of 28 had significantly increased levels. Imaging of trypanosomes with fluorescent labeled ceramides showed that the accumulation occurred most likely in the Golgi apparatus where the complex lipids containing ceramides, sphingomyelins and ethanolamine phosphorylceramides, are synthesized. However, future research will be required to identify the OXPA target that, when affected, leads to the observed phenotype. Possibilities considered by the authors are the synthesis, processing and intracellular transport of these ceramide-containing complex lipids.

6.2 Identification of targets in Trypanosoma cruzi

Compared with T. brucei, a smaller number of metabolomics studies have been performed on T. cruzi. The available literature encompasses both approved drugs and experimental compounds and has revealed a broad range of (potential) drug-induced metabolic alterations affecting several metabolic pathways and cellular processes (Figure 8).

6.2.1 Existing drug

6.2.1.1 Benznidazole

Trochine et al. (2014) exposed T. cruzi epimastigotes to 20 μM benznidazole (Bnz), for 6 h, during which ∼80% of the trypanosomes remained viable, and compared their metabolome with that of untreated control cells. Relatively few changes were detected. Among those were significant decreases in the levels of the redox-active thiols trypanothione, homotrypanothione (i.e., trypanothione in which the two glutathione moieties are connected by aminopropylcadaverine instead of spermidine, only detected in T. cruzi; Hunter et al., 1994) and cysteine and accumulation of three glutamate containing di-peptides, probably representing the γ-glutamyl dipeptides involved in glutathione recycling (Figure 8A). In addition, several Bzn-derived metabolites were detected, as well as covalent adducts of reduced Bzn with cellular metabolites, such as the thiols trypanothione, glutathione, spermidine, and ovothiol A, and reduction products and fragments. The thiol depletion observed is probably the result of the adduct formation, rather than inhibition of enzymes involved in the synthesis of thiols. When epimastigotes were exposed to 50 μM for 6 h, with ∼50% of the trypanosomes remaining viable, even more Bzn-derived metabolites were detected. In addition, changes in the amounts of more endogenous metabolites were observed: an increase in those of two long-chain acyl-carnitines and the dipeptide γ-glutamylcysteine and a decrease in the levels of three sterols. In this respect, it is relevant that sterol metabolism is considered a validated drug target for T. cruzi, and that Bzn and the ergosterol inhibitor posaconazole have been shown to act synergistically in a mouse CD model (Diniz et al., 2013). The metabolomics analysis did not reveal any indication for DNA damage, in contrast to a previous study in which a higher Bzn concentration and longer exposure time were used (Rajão et al., 2014). Low molecular weight adducts of glyoxal, a proposed toxic end-product of the Bzn metabolism, were neither detected.

6.2.2 Compounds in research

6.2.2.1 Bestatin

Peptidases play an important role in the physiology and virulence of parasitic protists and are thus considered potential drug targets. Genome sequence analysis of T. cruzi has revealed 28 genes encoding putative aminopeptidases, amongst which there are three leucine aminopeptidases of the M17 family of metallopeptidases. One of these leucine aminopeptidases, LASPTc, has been shown to mediate a major leucyl aminopeptidase activity in T. cruzi epimastigotes and, in vitro studies, was inhibited by bestatin (Cadavid-Restrepo et al., 2011). Bestatin is a peptidase inhibitor produced by actinomycetes (Umezawa et al., 1976). Trochine et al. (2015) exposed T. cruzi epimastigotes to 300 μM bestatin for 6 h, a treatment without any detectable toxic effect on the parasites, to perform a metabolomics analysis. The most striking finding was the significant accumulation of many different (∼35) dipeptides, due to bestatin’s inhibition of one or more peptidases catalyzing the hydrolysis of dipeptides into single amino acids. Almost 90% of these upregulated dipeptides contained leucine and/or isoleucine, valine, threonine, and/or histidine residues (Figure 8B). Although the authors could not identify the specific in vivo target of bestatin, the results are consistent with the inhibition of LAPTc. But future studies will be required to determine whether bestatin exerts its antiparasitic effects by the inhibition of also other aminopeptidases.

6.2.2.2 Cruzioseptins

Antimicrobial peptides (AMPs) are small peptides synthesized by many prokaryotes and eukaryotes, where they perform functions in the innate immune response (reviewed by Rojas-Pirela et al., 2023). AMPs receive much interest because of their activity against pathogenic microorganisms, including parasitic protists. They act primarily by inserting into the lipid core of the membrane, affecting its integrity, thereby causing lysis while also inhibiting intracellular processes. Several AMPs, called cruzioseptins (CZS), have been isolated from the skin of the frog Cruziohyla calcarifer, native to the tropical forests of Central-America (Proaño-Bolaños et al., 2016). Some of these CZS have been shown to possess activity against both the extracellular and intracellular forms of T. cruzi (Proaño-Bolaños et al., 2024). Osorio-Méndez et al. (2025) then explored the MOA of the synthetic version of the cruzioseptin that exerted the most potent and selective action on this parasite, CZS-5, a 32-residues long peptide with a net positive charge of +1.25. This AMP displayed an EC₅₀ = 4.7 μM in epimastigote viability assays, with an SI = 50.3 versus human erythrocytes as assessed by hemolysis. Molecular dynamics based on the sequence indicated that the peptide has the propensity to form toroidal pores in the membrane. Scanning electron microscopy revealed that addition of CZS-5 to the parasite results in membrane blebbing, cell rupture and flagellar detachment from the cell body, while a nucleic acid release assay showed a time- and dose-dependent release of these macromolecules from the cell, results that all are consistent with membrane disruption as the initial effect. To determine specific metabolic effects, a metabolomic analysis was performed on cells treated with CZS-5 at 5 μM for 2 and 4 h. Lipid-related molecules comprised 56.8% of the 118 metabolites that were changed compared to untreated control cells. Among them were glycerophospholipids (40.7%), fatty acids (13.6%) and glycerolipids (2.5%). Other changes were observed for amino acids, peptides and derivatives (14.4%), carbohydrates and conjugates (11.9%), carboxylic acids (5.9%), nucleosides and nucleotides (3.4%) and other metabolites (7.6%). Enrichment was found for lipids such as lysophospholipids, phosphatidylinositols and the fatty acids oleic acid and hydroxy isovaleric acid, the amino acids serine, threonine leucine and aspartate, as well as glutathione, and several intermediates of glycolysis and the PPP. Decreased levels were found for phosphatidylethanolamines, phosphatidylserines, and amino acids such as tryptophan, alanine and proline. Trends in the changes of the lipid-type metabolites were consistent between cells treated for 2 and 4 h. In contrast, for metabolites such as amino acids, peptides, and their derivatives the proportion decreased in the initial hours was greater.

These data indicate that the metabolic response to the CZS-5 treatment is complex, with the observed alterations primarily in energy metabolism and membrane composition (Figure 8C). Further studies will be required to determine which of the metabolic changes observed are a direct result of AMP interference and which ones are secondary effects.

6.2.2.3 Clethra fimbriata hexanic extract

An hexanic extract from the Colombian plant Clethra fimbriata has been shown to possess activity against all stages of T. cruzi (Pardo-Rodriguez et al., 2022), with EC₅₀ values of 153.9 ± 29.5 (epimastigotes), 39.3 ± 7.2 (trypomastigotes), and 45.6 ± 10.5 (amastigotes) μg/mL, and exhibiting a selectivity over monkey Vero kidney cells with a SI ranging from 6.49 to 25.4. With the aim of identifying the inhibited target(s), Pardo-Rodriguez et al. (2023) performed a metabolomics analysis on epimastigotes exposed to the extract at 690 μg/mL for 36 h. When compared to non-treated cells, 154 metabolites showed changes in level: 57.4% an increase and 42.6% a decrease. Most of the alterations (119 or 77.4%) were found for lipid-related molecules (glycerophospholipids, fatty acids, sphingolipids, glycerolipids and steroids), while organic acids (21), nucleosides (5), oxygenated organic compounds (4) and others compounds (5) were among the other 22.6% molecules that had changed in level. Upward trends were mainly observed for the group of glycerolipids, fatty acids (fatty acid amides and carnitine esters) and sphingolipids, whereas downward trends occurred predominantly among nucleosides, amino acids and peptides, carboxylic acids (pyruvate, PEP, 3-phosphoglycerate – all metabolites of the lower part of the glycolytic pathway – as well as glyceric acid and D-allose), and other compounds such as trypanothione, xanthine, hypoxanthine and carnitine. To estimate which metabolic pathways were most affected by the treatment with the hexanic extract, the authors analyzed the data with pathway enrichment analysis software. This approach highlighted parts of amino-acid metabolism, glycolysis and the metabolism of sphingolipids and glycerophospholipids. The changes in the metabolome caused by the treatment were then interpreted as a predominantly inhibitory effect on the energy metabolism involving carbohydrates and some amino acids as substrates, compensated by an increased energy supply through increased β-oxidation of fatty acids from lipids and TCA cycle activity (Figure 8D).

Overall, many metabolic changes were observed after the 32 h-long exposure to the plant extract. A time-dependent analysis would be required to determine if the changes can be differentiated into early, primary effects and secondary ones. Moreover, the complexity of the data may also be due to the chemical composition of the hexanic extract that was shown to contain several compounds, all pentacyclic triterpenes known to be present in C. fimbriata (Pardo-Rodriguez et al., 2022). These compounds may have caused different effects in the parasite.

6.2.2.4 Cytochrome b inhibitors

An often-used strategy for the identification of new drugs against trypanosomiases involves, as the first step, the phenotypic screening of compound libraries followed by target deconvolution. However, many such screenings result in the identification of hits for a small number of promiscuous targets. For the discovery of new drugs, interfering with activity of alternative targets through a novel MOA, it is important to implement already at an early-stage counter-screening methods for compounds acting on such common targets. MacLeod et al. (2026) developed such a methodology, based on untargeted metabolomics. In the case of drug discovery for CD, cytochrome b, a catalytic component of complex III of the mitochondrial respiratory system is a frequently identified target. As the authors showed, obvious changes in the metabolome, when inhibiting electron transfer at cytochrome b, are the rapid accumulation of succinate and depletion of fumarate, the substrate and product of complex II, a decrease of the ATP/ADP ratio due to incapacity to generate the transmembrane proton gradient necessary for the F₁F_o-ATP synthase activity, and a more gradual increase in the level of lactate resulting from a switch to anaerobic glycolysis and decrease in that of proline which is used as alternative energy source (Figure 8E). However, such changes are not sufficiently specific; for example, they will also occur when other components of the OxPhos system are inhibited. To differentiate early in the drug discovery workflow between inhibition of cytochrome b and that of other known – and supposedly unknown – targets , the authors developed a method to compare changes in levels of metabolites in epimastigotes treated with four different cytochrome b inhibitors and parasites treated with three inhibitors of other known targets (lysyl-tRNA synthetase, N-myristoyltransferase, proteasome). This involved small cultures in 12-well culture plates, 4 h incubation times that would allow detectable inhibition effects without resulting in death at 5 x EC₅₀ doses, and mass-spectrometry based metabolomics analysis of the epimastigotes. This allowed them to create a signature list of 79 metabolites showing a ≥ 2-fold differential abundance, due to either an increase or a decrease in the level of cytochrome b versus the non-cytochrome b target. Subsequently, this cytochrome b signature list was validated with 17 different compounds: 6 specific inhibitors of cytochrome b, 2 of other OxPhos components and 9 of other known cellular targets.

It should be relatively easy to implement the method developed by MacLeod and coworkers in other laboratories performing compound library screening, and using the signature list to aid in confidently identifying hits acting on cytochrome b and thus deprioritizing them from further medicinal chemistry research in favor of compounds inhibiting non-cytochrome b targets.

7 Concluding remarks

In this review, we summarized all (to our knowledge by May 2026) published intracellular metabolomics profiling studies performed on axenically cultured BSF T. brucei and epimastigote forms of T. cruzi treated with drugs or promising drug-like compounds. The aim of these studies was to identify the targets of the compounds, their MOA, and how their action possibly leads to growth arrest and death of the parasites. The disequilibrium in the amount of information presented about the respective Trypanosoma species reflects the current state of knowledge. All available studies on T. brucei were performed using the relatively easily cultivatable mammalian-pathogenic form of the parasite, whereas for T. cruzi the insect-specific form was used. It would be technically much more difficult to perform such experiments with the amastigote form that proliferates intracellularly in mammalian cells and is responsible for the pathogenicity. Relevant information about the target and MOA of a drug can undoubtedly be obtained by analyzing epimastigotes, because enzymes of core metabolic processes are expressed in all life-cycle stages. However, epimastigotes and amastigotes reside in environments which differ in available nutrients, and transcriptomics and genome-scale metabolic modelling predicted significant quantitative differences in the metabolic fluxes of both forms (Shiratsubaki et al., 2020). Thus, the flux control exerted by a target enzyme on a metabolic process and consequently also the trypanosome’s susceptibility for a drug targeted to such an enzyme are likely to differ.

Drug action in trypanosomes is often difficult to interpret, because most metabolomics studies revealed changes in levels of many intermediates, often also belonging to different parts of the metabolic network, or even molecules probably derived from structural components of the cell such as membranes. It is possible that a compound, at the concentration used, exerts pleiotropic effects due to simultaneously perturbing multiple metabolic processes. Moreover, for the identification of the target and its MOA it is essential to distinguish the primary effect from secondary ones, thus to analyze the metabolome in parasites exposed to the compound for different periods. Often, when a primary target is part of a multi-enzyme system, changes in levels of the metabolites immediately affected (the target’s substrate and product) will be transient. Due to the combined kinetic properties of its constituting enzymes, the system may, after its disturbance, reach a new steady state with levels very similar to the initial ones, despite a considerably reduced flux through the pathway with lowered product levels causing downstream toxic effects. For example, multiple secondary metabolome changes with drastic effects may occur when inhibition of the target affects the formation of multipurpose metabolites such ATP, redox factors like NAD(P)H and trypanothione, essential components for nucleic acid or protein synthesis. Resulting stress may then lead to damage of the cells and ultimately to uncontrolled or programmed cell death (Duszenko et al., 2011; Menna-Barreto, 2019). Additionally, these parasites display remarkable transcriptomic flexibility, enabling adaptive responses to drug-induced stress (e.g., Ospina et al., 2025) which may also become evident in the metabolome of samples taken exposed for longer time, as exemplified in the study of suramin-treated T. brucei by Zoltner et al. (2020).

For some drugs, the primary action will be very fast, when the compound is readily taken up, and affects an essential process with immediate effect such a decreased glycolytic flux by the inhibition of PFK by the pyrrolopyridine CTCB405, resulting in a rapid drop of the ATP level in BSF T. brucei soon followed by death (McNae et al., 2021; Nare et al., 2023). Interestingly, the above cited interpretation of the metabolomics data for the trypanocidal effect of suramin by Fall et al. (2024b) is that this drug similarly acts by inhibition of T. brucei glycolysis and ATP production, but exhibits a delayed metabolomic effect eventually resulting in parasite death due to the long time necessary for its accumulation to inhibitory concentrations within glycosomes where the target enzyme is located. Delayed trypanocidal effect may also be expected for a drug that acts by slowly increasing the level of a metabolite to toxic levels and possibly too for drugs like fexinidazole, benznidazole and nifurtimox that, after having been converted into reactive molecules, bind to, and modify many cell components in a rather non-selective manner.

Metabolomics profiling has its limitations, like genome-wide expression- or cytology-based profiling; interpretation of the data can be difficult. Therefore, data collected by these methodologies have often to be combined, metabolic pathway fluxes should be measured and/or still other approaches need to be applied as well. Additional methods considered for unambiguously determining the target of a drug and its MOA are those in which the susceptibility of a parasite for a drug is assessed by changing the expression of a candidate target by reverse genetics or using a reactive analog of the drug to bind it covalently to the target, in so-called chemoproteomics.

In some cases, metabolomics profiling may also provide information about the mechanism of drug resistance. Such resistance can be due to several different mechanisms: diminished uptake, increased degradation or increased export of the drug, mutation of the target to affect drug binding, or increased expression of the target (Alsford et al., 2012). Metabolomics will only be able to indicate the involvement of a mechanism that causes a decreased intracellular concentration of the drug. An example is eflornithine resistance in BSF T. brucei due to the inability to accumulate the drug that could be attributed to the loss of the amino acid transporter TbAAT6 (Vincent et al., 2010).

In summary, the studies reviewed in this paper show that metabolomic profiling of trypanosomes treated with drugs or drug-like compounds is a valuable experimental approach to collect information about the molecular target(s) and mode of action of the compounds, and may also provide insight in secondary processes resulting from the initial drug activity that eventually result in compromising the viability of the parasites.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Leticia Pérez Díaz, University of the Republic, Uruguay

Reviewed by: Fernanda Ramos Gadelha, State University of Campinas, Brazil

Carlos Sanz-Rodriguez, Universidad de la República, Uruguay