A Critical Review on Chagas Disease Chemotherapy

BACKGROUND

Chagas disease is endemic in Latin America, affecting 16-18 million people, with more than 100 million exposed to the risk of infection (WHO 1997). Its etiological agent is Trypanosoma cruzi, an hemoflagellate protozoan (family Trypanosomatidae, order Kinetoplastida) (Hoare & Wallace 1966), whose life cycle involves obligatory passage through vertebrate (mammals, including man) and invertebrate (hematophagous triatomine bugs) hosts, in a series of stages. The trypomastigote ingested by the insect differentiates into the proliferative epimastigote form that, on reaching the posterior intestine, evolves to metacyclic trypomastigotes. This latter form, following invasion of vertebrate host cells, undergoes differentiation into amastigotes, which after several reproductive cycles transform to trypomastigotes, the form responsible for the dissemination of the infection. The transmission of the disease occurs mainly by the vector (80 a 90%), blood transfusion (5 a 20%) and congenital routes (0.5 a 8%) (Dias 2000).

In humans, after infection and a subsequent incubation period, the acute phase of Chagas disease begins, and in the absence of specific treatment, the symptoms persist for about two months, with a mortality rate of 2 to 8%, especially among children. T. cruzi is able to invade and multiply within different host cells, including macrophages, smooth and striated muscles, fibroblasts and even neurons. The first reaction to T. cruzi is focal mononuclear inflammation due to rupture of parasitized cells. Within days to two weeks can be detected in the serum the presence of immune complexes, decrease in C3 level, besides necrosis in the inflammatory foci. Severe inflammation is usually accompanied by necrosis of parasitized and non-parasitized cells, especially in the heart. Platelet aggregation, eosinophils degranulation, microvascular pathology, edema, thrombosis, blood stasis and ischaemia have also been demonstrated (Andrade & Andrade 1999).

After the acute infection, the patient presents strong evidence of immunity, but has a tendency to remain infected. Some parasites evading the immune response and focal inflammatory lesions are seen in several organs. Amastigote forms can be detected by conventional histology, by immunofluorescence and genomic markers by in situ hybridization. The combination of rising immunity against the parasite with specific immunological suppression of hypersensibility and reduction of inflammatory reaction seem to be the main pathways to the indeterminate phase of Chagas disease (Andrade 1999).

In the chronic phase that follows, most patients remain asymptomatic, with about 20 to 50% of the cases, accordingly to the endemic area analyzed, developing the characteristic symptoms of this phase, namely cardiac, digestive or neurological disturbances (reviwed in Brener et al. 2000). Chronic, active, fibrosing myocarditis have been attributed to hypersensibility to parasite antigens, neoantigens or autoimmunity. The presence of cross-reactive antigens between heart muscle and T. cruzi has been demonstrated, but the autoimmunity does not entirely explain Chagas heart disease. The high positivity of xenodiagnosis and hemoculture, and reactivation of chronic disease by immunosuppression demonstrate the presence of the parasite in chronic cases. High frequency of parasites and/or antigens associated with myocardial inflammation is an important guide to the therapeutic procedures in the chronic phase.

The pathogenesis of Chagas disease is not yet completely defined and understood, with two basic inflammatory lesions, one focal and the other diffuse. The focal lesion is associated with the parasite, and occurs when parasitized cells are disrupted. The diffusion lesion is out of proportion in relation to the presence of parasites. During the acute infection there are diffuse lesions in the heart and focal lesions in several other organs. In the chronic heart disease, severe fibrosis and inflammatory lesions seem not to be related only to the presence of T. cruzi, but also to a strong delayed hypersensibility host response and ischaemic lesions (Andrade 1999, Andrade & Andrade 1999, Higuchi 1999). Two mechanisms are proposed for pathogenesis in chronic chagasic infections: the persistence of the parasite results in chronic inflammatory reactivity and it induces immune responses targeted at self-tissues (reviewed in Tarleton 2001). Several clinical reports reinforces the first mechanism (Higuchi et al. 1993, 1997, Jones et al. 1993, Anez et al. 1999), while the main support of the second one is that signs of the disease are evident in tissues in the apparent absence of parasites.

Several reviews about clinical and/or experimental Chagas disease treatment have been published as articles (Coura & Silva 1961, Prata 1963, Brener 1975, 1979, 1984, De Castro 1993, Coura 1996, Urbina 1999, Stoppani 1999) and as book chapters (Brener 1968, 2000, Cançado & Brener 1979, Cançado 1968, 1985, 1997, 1999, 2000, Rassi & Luquetti 1992, Storino et al. 1994). We will detail here just some of the publications and refer readers to the reviews cited above.

The aim of this "Critical Review" is to analyse drugs employed in the clinics since the 70s bringing attention to the recommendations about treatment, and evaluation of cure, and the studies about the development of new drugs, considering potential targets in the parasite and summarizing the experimental studies performed with new compounds assayed against T. cruzi after 1992/1993. For a more complete coverage of experimental in vitro and in vivo studies we suggest the reviews of Brener (1975, 1979) and De Castro (1993).

EXPERIMENTAL AND CLINICAL TREATMENT

Drugs assayed up to the decade of 70

The first compounds assayed experimentally for the treatment of Chagas disease, soon after its discovery by Carlos Chagas in 1909, were atoxyl (arsenical), fucsin (rosanilin dye), tartar emetic (or antimony potassium tartarate, a pentavalent antimonial) and mercury chloride, employed experimentally by Mayer and Rocha Lima (1912, 1914) and all of them without effective results. Until the publication of the "Manual de Doenças Tropicaes e Infectuosas" by Carlos Chagas and Evandro Chagas (1935) "there was no specific treatment for American trypanosomiasis. Drugs with trypanocidal activity have been assayed by a great number of researchers, but without success", affirm the authors (p. 144).

Among the chemotherapeutic agents employed until 1962 stand out quinolein derivatives, several other antimalarials, arsenobenzoles, and other arsenicals, phenantridines, salts of gold, bismuth, copper and tin, sodium iodide, gentian violet, aminopterin, para-amino salicylic acid, nicotinic acid hydrazide, antihistaminics, sulfonamides, ACTH, cortisone, stylomycin derivatives, amphotericin B, more than 30 antibiotics and some nitrofurans (reviewed in Coura & Silva 1961, Brener 1968, Cançado 1968).

Brener (1968) made a meticulous evaluation of the experimental drugs assayed in vitro and in vivo against T. cruzi, registering 27 compounds and more than 30 antibiotics, assayed between 1912 and 1962, that were inactive. He also considered that the following compounds had a suppressive effect on the parasitemia but were not curative: the bisquinaldine Bayer 7602 (Cruzon, Imperial Chemical Industry, UK), the phenatridine carbidium sulfate (74C48), aminoquinolines (pentaquine, isopentaquine and primaquine), trivalent arsenicals (Bayer 9736 and Bayer 10557 also named spirotrypan), aminoglycoside of stylomycin, nitrofurans and antibiotics.

Packchanian (1952, 1957) opened a new and promising line of potential drugs with the nitrofurans that led to nitrofurazone (5-nitro-2-furaldehyde-semicarbazone). This derivative administered by oral route for 53 days consecutively in the dose of 100 mg/kg/day mice experimentally infected with T. cruzi cured 95.4% of the animals (62/65) (Brener 1961)

Ferreira (1961, 1962) and Ferreira et al. (1963) treated the first ten cases of acute Chagas disease with this nitrofuran, obtaining "good clinical results" with few collateral effects but the xenodiagnosis became positive in five cases after treatment Coura et al. (1961, 1962) treated 14 chronic cases with this drug in long-term schemes, observing in the first four patients, that received progressive doses of 10 to 30 mg/kg/day, important side effects that led to suspension of the treatment due to severe sensitive polyneuropathy, that began at the third week of nitrofurazone administration. With reduction of the dose to 10 mg/kg/day and association with complex B, administered by parenteral route, five patients tolerated the treatment for 60 days, in spite of the side effects (anorexia, weight loss, paresthesia, and sensitive polyneuropathy). Another patient was treated with 20 mg/kg/day and presented paresthesics manifestations only at the 53th day of treatment, evolving to a severe sensitive polyneuropathy, with termination of the treatment by his own decision (informed consent). From the six patients submitted to long-term treatment, two of them were considered cured, based on xenodiagnosis and serology (complement fixation) persistently negative. Cançado et al. (1964) also treated five chronic patients with 10 mg nitrofurazone/kg/day during 10 to 34 days when the treatment had to be suspended due to polyneuritis, being all the five patients considered as therapeutic failures.

In a critical analysis of the literature about the clinical experiences, Cançado (1968) emphasized the lack of methods in the execution, preferential selection of acute cases, based on the remission of the symptoms and signs that could also be spontaneous. He cited, as example, the reports of Mazza et al. (1937, 1942) and Pifano (1941) about the results with Bayer 7602 and Bayer 9736 that were considered "excellent results" only based on the reduction of the symptoms and signals. Both compounds were in fact ineffective, since the xenodiagnosis after treatment remained positive and untreated cases had also reduction of the symptomatology. In subsequent works (Cançado et al. 1973, Cançado 1981) reviewed the results of therapeutic trials in the period of 1936 to 1965 and proposed basic criteria for the evaluation of the specific treatment, that were later updated by 15 experts from Latin America (OPS/OMS 1998).

Following the requirements of the World Health Organization (WHO) the ideal drug for the treatment of Chagas disease should fulfill the following requirements: (i) parasitological cure of acute and chronic cases; (ii) effective in single or few doses; (iii) accessible to the patients, in other words, of low cost; (iv) no collateral or teratogenic effects; (v) no need of hospitalization for the treatment; (vi) no induction of resistance.

As we will see bellow this ideal drug does not exist and possibly it will take a long period of time to be obtained. Since the end of 1960 beginning of the 70s two drugs have been used for the treatment of Chagas disease: nifurtimox and benznidazole.

Nifurtimox and benznidazole

The drugs and Chagas disease treatment

Nifurtimox (Nif) is a 5-nitrofuran (3-methyl-4-(5'-nitrofurfurylideneamine) tetrahydro-4H-1,4-tiazine-1,1-dioxide (Bayer 2502) and benznidazole (Bz) is a 2-nitroimidazole (N-benzyl-2-nitroimidazole acetamide (RO 7-1051), commercialized, respectively, with the names Lampit and Rochagan in Brazil (Radanil in Argentina) (Fig. 1a,b). Nif that was the most active 5-nitrofurfurilidene derivative experimentally assayed (Bock et al. 1969) and Bz showed a high in vitro and in vivo activity against T. cruzi (Richle 1973). Since the 80s Nif had its commercialization discontinued, first in Brazil, and then in Argentina, Chile and Uruguay. The mode of action of Nif involves generation of nitroanion radical by nitroreductases that, in the presence of oxygen, led to reactive intermediates and being T. cruzi is partially deficient in free radical detoxification mechanisms, it is susceptible to such intermediates (reviewed in DoCampo & Moreno 1986). On the other hand, this oxidative damage was not the key action of Bz, the detection of corresponding nitroanion radical occurred only at concentrations much higher than those that killed the parasite. The action of Bz could involve covalent bond or other interactions of nitroreduction intermediates with parasite components (Polak & Richle 1978), or binding to DNA, lipids and proteins (Diaz de Toranzo et al. 1988).

Nif and Bz have been employed by different authors, especially in Brazil, Chile and Argentina (Cançado et al. 1969, 1973, 1975, Cançado & Brener 1979, Bocca-Tourres 1969, Rubio & Donoso 1969, Schenone et al. 1969, 1972, 1975, 1981, Rassi & Ferreira, 1971, Rassi & Luquetti 1992, Cerisola et al. 1972, 1977, Prata et al. 1975, Ferreira 1976, 1990, Coura et al. 1978, 1997, Macêdo & Silveira 1987, Viotti et al. 1994, Andrade et al. 1996, Sosa Estani et al. 1998).

The results obtained with both drugs varied according to the phase of Chagas disease, the period of treatment and the dose, the age and geographical origin of the patients. Good results have been achieved in the acute phase, in recent chronic infection (children under 12 years old), congenital infection and laboratory accidents. For the acute phase treatment and congenital cases it is recommended 8 to 10 mg/kg/day of Nif or 5 to 7.5 mg/kg/day of Bz during 30 to 60 days consecutively, and divided in two or three daily doses. Patients with less than 40 kg can take up to 12 mg/kg/day of Nif and up to 7.5 mg/kg/day for Bz during 30 to 60 days (OPAS/OMS 1998). For recent chronic infection (children under 12 years old) or individuals infected in the last 10 years, the treatment should be made with 8 mg/kg/day of Nif or 5 mg/kg/day of Bz during 30 to 60 days. In the case of accidental infection the treatment must begin immediately and last for only 10 to 15 consecutive days. Cases of late chronic infections without clinical manifestation or with mild cardiac or digestive manifestations should be treated during 60 to 90 days, in accordance with the tolerance to the drugs, aiming to prevent or reduce the evolution of Chagas disease to more severe forms, a fact that is not yet definitely proved.

Side effects and contraindications

The more frequent collateral effects with Nif treatment are anorexia, loss of weight, psychic alterations, excitability or sleepiness and digestive manifestations, such as nausea, vomit and occasionally intestinal colic and diarrhea. The adverse reactions with Bz could be classified in three groups: (i) symptoms of hypersensibility, dermatitis with cutaneous eruptions (usually appearing between the 7th and 10th day of treatment), generalized edema fever, lymphadenopathy, articular and muscular pain; (ii) depression of bone marrow, thrombocytopenic purpura and agranulocytosis, the most severe manifestation; (iii) polyneuropathy, paresthesia and polyneuritis of peripheric nerves.

The two most serious complications induced by Bz are agranulocytosis, initiated by neutropenia, sore throat, fever and septicemia, and thrombocytopenic purpura, characterized by reduction of platelets, petechiae, hemorrhagic blister and even mucosal bleeding. At the first signs of such manifestations, medication must be immediately suspended and should be started a treatment with antibiotics in the case of septicemia plus corticosteroids for the control of the agranulocytosis and of the thrombocytopenic purpura. Other manifestations of intolerance and hypersensibility could be circumvented with the reduction of the dose or suspension of the drug, depending on their intensity, introduction of symptomatic medication and eventually of anti-histaminics and corticosteroides.

Teixeira et al. (1990, 1994) have been alerting for the appearance of lymphomas and mutagenic and carcinogenic activities in experimental animals (rabbits and mice) treated with Bz. However, a broad review of thousands of patients treated with these drugs by several authors has not demonstrated such effects.

Bz and Nif should not be indicated for pregnant patients, in cases of severe diseases associated with Chagas disease, such as systemic infections, cardiac, respiratory, renal or hepatic insufficiency, hemopathies and neoplasies without the possibility of treatment, old-aged and very debilitated persons.

Effect of the treatment on the evolution of Chagas disease

Since 1969 several therapeutic experiences have been performed in acute and chronic cases of Chagas disease using Nif (Bocca-Tourres 1969, Rubio & Donoso 1969, Schenone et al. 1969, 1972, Rassi & Ferreira 1971, Cerisola et al. 1972, 1977, Silva et al. 1974, Prata et al. 1975, Cançado et al. 1975, 1976, Cançado & Brener 1979), Bz (Schenone et al. 1975, Ferreira 1976, Coura et al. 1978, Viotti et al. 1994, Andrade et al. 1996, Sosa Estani et al. 1998) and also comparing the efficacy and tolerance of both drugs in different groups of patients, therapeutic schemes and periods of follow-up and cure evaluation criteria (Schenone et al. 1981, Ferreira 1990, Coura et al. 1997, Lazzari & Freilig 1998).

The results of such experiences showed a great variability, according to the authors, and the type of casuistic and of cure control employed. In general, results obtained were good for acute phase and recent infection cases, especially among children, who, besides tolerating a long-term treatment much better, presented high indexes of cure, as demonstrated in a randomized field study with Bz in Brazil (Andrade et al. 1996) and in Argentina (Sosa Estani et al. 1998). For acute cases and recent infections an average index of parasitological cure around 60% is estimated. In relation to chronic infection cases, results have been poor, for the Brazilian experience, while in the South Cone, they were much better, possibly due to the type of parasite strain (Silva et al. 1974, Cerisola et al. 1977).

Studies on the clinical evolution of Chagas disease after specific treatment are controversial and results are not convincing, due to differences in casuistics, methods of evaluation, time of follow-up and interpretation of the data. Macêdo and Silveira (1987) studying 171 adults with chronic disease treated with Nif or Bz with a follow-up of 7 years, observed electrocardiographic (ECG) evolution of cardiopathy in 6.7% of the cases, against 8.8% for untreated patients, indicating no significant differences between the two groups. Ianni et al. (1993) monitored 33 adults in the indeterminate phase for 8 years, reported the evolution in 13.3% of the cases treated with Bz (n = 15) and 0% of the cases that received placebo (n = 18), not allowing a definitive conclusion, due to the small casuistic. Miranda et al. (1994 apud OPAS/OMS 1998), in 120 patients (adults and children), observed ECG evolution in 10.5% of those treated with Bz and 63.6% of the placebo group. Although these authors monitored the patients for 10 to 16 years, the combination of the data obtained with adults and children and the ECG interpretation made the analysis of the results obtained difficult.

Viotti et al. (1994) in a well-designed study with 201 adult patients monitored for 8 years, observed ECG evolution in 7/131 cases (5.3%) treated with Bz (5 mg/kg/day for 30 days) and 16/70 (22.8%) in the control group. In patients with more than 50 years old, the ECG alterations occurred in 3/36 (8.3%) of the treated cases and 7/40 (17.5%) of the untreated ones; differences were not statistically significant. For those under 50 years old, such alteration occurred in 4/95 (4.2%) and 9/30 (30%), respectively, for treated and untreated patients, and were significantly different. Two patients died during the follow-up, one treated and one untreated. Although the study has been well conducted, after 8 years of follow-up, 68% of the untreated patients presented positive serology against 48.2% of the Bz-treated group. The low number of ECG alterations and their frequent mutability in chronic cases make the interpretation of the data difficult.

On the other hand, Fragata Filho et al. (1995) reported, in a study with 120 chronic patients with follow-up for 7-8 years, ECG evolution in 7% for Bz-treated cases (n = 71) and 14.3% for the untreated group (n = 49).

Andrade et al. (1996) performed a randomized field study in the State of Goiás (Brazil) treating children between 7 and 12 years old with positive serology for Chagas disease. Sixty-four patients received 7.5 mg/kg/day of Bz for 60 days and 65 received placebo. From these 129 children, 88.7% (58 treated and 54 control) were monitored for three years. The authors considered the treatment effective in 55.8% of the treated cases (most of them with a significant decrease in serological titers). However, no significant results were observed when the ECG abnormalities were compared. In a similar way, in Salta (Argentina), Sosa Estani et al. (1998) treated 55 children from 6 to 12 years old with 5 mg Bz/mg/day for 60 days and 51 children with placebo, and monitored the study for four years. They observed negativation or significant decrease of the specific serology in 62% of the treated group and none for the control group. In relation to xenodiagnosis the positivity was 4.7% and 51.2%, respectively for treated and untreated children, indicating an important suppressive activity of the treatment, however, the ECG alterations were similar for both groups: 2.5% (1/40) and 2.4% (1/41).

Rules and recommendations for the clinical treatment

A meeting of 13 specialists promoted in Brasília by the Ministry of Health of Brazil summarized by Luquetti (1997) and another one by the Pan American Health Organization and World Health Organization (OPAS/OMS 1998) that took place in Instituto Oswaldo Cruz (Fiocruz) in April 23-25 of 1998, established some rules and recommendations for the etiologic treatment of Chagas disease, with the present available drugs, that are summarized bellow:

Acute phase

In this phase the parasite is detected by direct examination of peripheral blood by weat smear or in stained cover slides or by concentration methods (centrifugation of the blood and microscopic examination of leukocyte cream or stained quantitative buffy coat techniques). With or without clinical manifestations, the detection of the parasite by direct methods or determination of IgM levels allows the diagnostic of the acute phase. Independently of the mechanism of transmission (vectorial, transfusional, by oral route, or laboratory accident) the patients must be treated, as indicated previously, since about 60% of them could be cured in the acute phase.

Congenital infection

The diagnostic of congenital infection is based in cases of children from infected mothers, serologically positive, who presented T. cruzi in the blood of umbilical cord, specific IgM in the serum soon after birth, or IgG after 6 months, when the possibility of vectorial, transfusional and oral transmission are absent. The treatment is similar to that of the acute phase patients.

Accidental infection

The person, technician or researcher, that working with T. cruzi, was accidentally punctured by infected needle, ingested or had any contact with infected materials in lesions, wounds, mucosal, or any other form indicative of the possibility of the parasite penetration is considered infected. In these cases, blood is collected for serological test and treatment immediately begins, during 10 to 15 days, repeating the serology after 15, 30, and 60 days after the accident. It is recommended that all the laboratories that work with T. cruzi have the drug available.

Chronic phase

Recent chronic phase - Recent chronic phase is considered when the infection was acquired in the last 10 years, including children up to 12 years old or adults that have been infected occasionally in endemic areas of by blood transfusion in a known interval of time. The published work indicates that the results obtained by the etiologic treatment in this phase are much better than that performed in late chronic patients.

Late chronic phase - Patients with more than 10 years of infection are considered late chronic cases. There is no agreement about the clinical evolution of such cases, and parasitological cure is obtained in 10 to 20% of the patients, according to different experiences. The treatment must be elective, with priority to cases in the indeterminate form or with minor pathology that may be monitored by long periods of time, after treatment.

Transplant of organs - In cases of transplants it is always necessary to perform specific serological reaction both in the donor and the receptor. The transplant of organs from patients with infection by T. cruzi could transmit the parasite to the receptor, especially during the immunosuppression phase. On the other hand, in the infected receptor a reactivation of the disease could occur through the immunosuppression, compromising even the transplanted organ, mainly in cardiac cases. Jatene et al. (1997) refer to reactivation of Chagas disease in 40% of cases derived from immunosuppression in heart transplants. Both the donor and the receptor must be treated with Bz in the dose of 5 mg/kg/day for at least 60 days. Tominori-Yamashita et al. (1997) considered that "allopurinol seems a safe and effective treatment for reactivated Chagas disease after heart transplantation, although it is not recommended as a post-transplantation prophylaxis because reactivation of the disease is unpredictable".

Reactivation of Chagas infection

Reactivation of the chronic disease can occur due to immunosuppression by several diseases, such as leukemia, lymphoma and other neoplasies, infection by HIV/AIDS and in the cases of transplants with immunos-supression. Meningoencephalitis and acute myocarditis are the most frequent manifestations (reviewed in Ferreira et al. 1997a). In chronic patients parasites were detected in cutaneous lesions after transplant with immunosuppressive treatment and in the smooth muscle after cancer chemotherapy. Ferreira et al. (1997a,b) made an extensive review about this topic, recommending as first choice the treatment with Nif or Bz, indicating as alternatives triazole derivatives and allopurinol. Long-term secondary prophylaxis should be recommended for patients who respond to therapy, although it is uncertain which drug to use for this purpose (Ferreira et al. 1997b). Anyway all HIV-positive cases, patients with neoplasies or candidates to transplants must be thoroughly investigated for the possibility of concomitance with chagasic infection. There is no consensus about the prophylatic indication of the etiologic treatment of the infection, in cases without clinical reactivation, but with xenodiagnosis or hemoculture positive for T. cruzi.

Patients in the chronic phase and receiving corticoid because of concomitant diseases were treated with Bz since the beginning of the use of corticoid (n=12) or 15 days afterwards (n = 6) (Rassi et al. 1999). The authors observed that Bz prevented an increase of parasitemia, and suggested that in immunocompromized patients with chronic Chagas disease this drug could be useful.

Where and who should treat the patient

The patients in severe acute phase and the congenital symptomatic cases with diagnostic at birth, must be hospitalized for treatment. The acute oligosymptomatic or chronic cases can be treated in basic health care units under the supervision of an experienced physician. Bz and Nif must be considered drugs of high complexity, and recommended only by professionals with solid information about the side effects and the disease in itself. The acute phase and the accidental infection are emergency situations, and treatment must start immediately even with a professional without experience that must search a colleague or a qualified institution for orientation.

Evaluation of cure

The evaluation of cure of Chagas disease is certainly the more complex aspect of its treatment, leading several times to diverse and controversial results, in relation to both parasitological cure and clinic cure. The term parasitological cure itself is of difficult interpretation and the evaluation is almost impossible, since it would mean the total elimination of the parasite not only from the blood but also from all tissues. So, in humans it is not viable to be confirmed. On the other hand, clinic cure is the long-term evaluation and several times uncertain due to the pathogenesis of the disease, which involves the action of parasite and the immune and autoimmune response of the patient and in antigenic complexes deposition, generation of antibodies, inflammatory reactions, tissular lesions with cellular degeneration, ischaemia, fibrosis and their consequent clinical manifestations, sometimes for long periods of time.

For the evaluation of experimental animals, mostly in drug assays, the situation is less complex. The in vitro tests (tissue culture) are not necessarily reproduced in vivo. On the other hand, as occurs in humans, a suppressive effect on the parasitaemia does not correspond exactly to the effect of a drug in tissues. In his pioneer work Brener (1961) demonstrated parasitological cure in 94.5% of the mice treated with nitrofurazone. Later, Brener et al. (1969) demonstrated by electron microscopy that 13.5% of the amastigotes of the Y strain were intact in heart cells of the treated animals. A question remained: were these findings due to populations of the parasite with primary resistance or the drug did not reach all the infected cells? The group of Andrade reported development of resistance to both nitroheterocycles and the influence of T. cruzi strain in the cure rate, for example, Bz cured 87% of the mice infected with the Peru strain, and only 16.7% in the case of Colombian strain (Andrade et al. 1975, 1977, Andrade & Figueira 1977). Resistance to both drugs, including cross resistance, was also observed in animals infected with the Y strain (Costa Silva et al. 1990). Could this be due to a mechanism similar to that questioned above?

Parasitological evaluation

The suppressive activity on the parasitaemia is almost immediate after the beginning of the treatment when the strain (population) of T. cruzi is susceptible to the drug employed. In acute cases this fact could be verified by the direct examination of blood in fresh or stained preparations or by concentration methods. In chronic cases, the usual methods are xenodiagnosis standardized with 40 nymphs of 3rd/4th stage of Triatoma infestans or by 20 nymphs of this species and another 20 from Panstrongylus megistus or of other species that could give similar yield. The nymphs are distributed in four boxes (10 per box); two are placed in the internal face of each arm for about 30 min (Coura et al. 1991). In this study we observed a positivity of 50.7% in 570 xenodiagnosis performed in 246 patients. Nowadays, due to ethical questions and comfort for the patients, the preferential method is the artificial xenodiagnosis that consists in collection of 10 ml of the blood that are placed in a condom type membrane, with external heating, and then the nymphs are added and the reading done after 30, 45 and 60 days. The yield obtained is similar to that of the natural test (Pineda et al. 1998).

In a multicenter study that involved researchers from 10 Brazilian institutions and 312 Bz-treated patients monitored by a media of 12 xenodiagnosis performed after treatment, suppression of parasitaemia was demonstrated in 78% of the cases (Coura et al. 1978). In another comparative controlled study with 77 chronic patients treated with Bz (n = 26), Nif (n = 27) or placebo (n = 24), suppression of the parasitaemia monitored xenodiagnosis by one year after treatment was achieved in 98.1% (2/110) of the Bz group, and in 90.4% (75/83) of the Nif group (Coura et al. 1997). However, this result does not imply that cases with xenodiagnosis negative are cured since only 34.3% of the control group was positive.

The hemoculture is the second parasitological method of choice for the control of cure in chronic Chagas disease, being equivalent in sensibility to xenodiagnosis (Chiari & Brener 1966). Both methods have a tendency to increase the positivity with the number of tests performed, amount of blood employed, cultivation medium, interval of time between blood collection and cultivation and other factors emphasized by Chiari et al. (1989) aiming standardization of the assay. Using 30 ml of blood seeded in six test tubes with LIT medium and readings up to 60 days, Chiari et al. (1989) obtained a positivity up to 50%, while with small modifications of the technique such as direct seeding soon after blood collection, refrigerated centrifugation for a short time, gentle homogenization and up to 120 days led Luz et al. (1994) to a positivity of 94%, not achieved by any other author, for assays with chronic cases of Chagas disease.

Polymerase chain reaction (PCR) was a major advance for the parasitological control of the cure of Chagas disease, with positivity 2 to 3 times higher for chronic cases when compared to routine xenodiagnosis and hemoculture. By this technique it is possible to detect one parasite or a fragment of T. cruzi DNA in 20 ml of blood (Ávila et al. 1991).

Sturm et al. (1989) amplifying minicircles of DNA of T. cruzi obtained fragments of 83 and 22 pairs of base (bp) from variable regions that were employed for detection of the parasite (Moser et al. 1989). Ávila et al. (1991) using a solution of 6 M guanidine plus EDTA and equal amount of blood of chronic patients, promoted lysis of proteins and maintained the integrity of the DNA sample at room temperature. The treatment of this lysate with phenantroline-copper (OP-Cu2+) led to the cleavage of DNA and liberation of minicircles, allowing the identification of a single parasite in 20 ml of blood when three initiators for the fragments of 83, 122 and 330 bp from the variable and constant regions of the minicircles. By this technique, the authors identified T. cruzi in samples of 10 ml of blood of five chronic patients, four of them with negative xenodiagnosis.

Several authors (Wincker et al. 1994, Britto et al. 1995, 2001, Junqueira et al. 1996) have demonstrated the efficiency of PCR for the diagnosis of chronic disease and for the control of cure after treatment. However, Junqueira et al. (1996), in a comparative study among PCR, xenodiagnosis and hemoculture in 101 chronic cases, observed positivity of, respectively, 59.4%, 35.6% and 25.7%, but in five cases with positive xenodiagnosis and/or hemoculture, the result of PCR was negative. Recently, Britto et al. (2001) comparing PCR and xenodiagnosis in the control of cure of 85 chagasic patients submitted to specific treatment and 15 chronic assymptomatic cases that received placebo, reported that in all the cases of positive xenodiagnosis, positivity was obtained also by PCR. On the other hand when xenodiagnosis was negative, PCR was positive in 18.5% of the acute phase group (n = 37), 29% of the chronic phase group (n = 45) and 57.1% of the control group. These results demonstrate the advantage of PCR over conventional techniques to demonstrate persistent infections in patients that underwent chemotherapy.

Serological evaluation

This evaluation is certainly the most simple, more broad and reliable for the control of cure of chagasic infection after treatment, especially in the chronic phase when the serology is positive in almost 100% of the untreated cases. Whereas the positivity of parasitological methods depends on the random presence of the parasite in the blood sample, the presence of antibodies is almost warranted in all the samples. On the other hand, the serology since the Guerreiro and Machado reaction (1913) until the qualitative and quantitative reaction of complement fixation (Kelser 1936, Freitas & Almeida 1949), the indirect immunofluorescence assay (Fife & Muschel 1959, Camargo 1966), the hemagglutination assay (Neal & Miles 1970), the conventional ELISA (Voller et al. 1975) and with recombinant antigens (reviewed in Silveira 1992, Silveira et al. 2001), the lysis mediated by complement (Krettli & Brener 1982, Kretlli et al. 1984) and finally techniques using immunoblots have been improved as confirmatory tests due to their sensibility and specificity (Umezawa et a1. 1996, 2001).

The three basic serological reactions for the diagnosis of Chagas disease are indirect immunofluorescence, hemagglutination and ELISA. During the years, the great polemics is their negativation in cases of parasitological cure. Some authors as Cançado (1963, 1997) consider cure as "definitive post-therapeutic reversion to negativity of parasitological and serological tests", whereas others, like Rassi and Luquetti (1992), Andrade et al. (1996) and Sosa Estani et al. (1998) admit a long period of negativation of the reactions and even low serological titers as criteria of cure. Andrade et al. (1991) demonstrated that, in mice infected with T. cruzi and parasitologically cured by chemotherapy, parasite antigens persist in interstitial dendritic cells in the spleen and the animals present positive serology. Recently Andrade et al. (2000) reported the importance of the presence of parasite antigens in the same type of cells in the heart of infected dogs and the pathogenesis of the chagasic myocarditis probably by presentation ofT. cruzi antigens to immune-competent cells, and, as consequence, maintenance of the response to the infection.

Clinical evaluation

This type of the cure evaluation after chemotherapy is perhaps the most difficult and long topic to be addressed. In this review, we have already discussed some aspects of the clinical evaluation when analyzing the evolution of Chagas disease after treatment with Nif or Bz, so, in this topic we will discuss only some essential clinical tests for monitoring the disease, before, during and after treatment.

In the clinical evaluation of the treatment we must consider, besides the anamnesis and clinical examination, the electrocardiographic and radiologic aspects together with other non-invasive tests with high sensibility, such as dynamic electrocardiography (Holter) for the study of arrhythmias and echocardiography for the anatomophysi-ological evaluation of the cardiac function, the endoscopy and manometry for the anatomofunctional study of the digestive system and some other tests for evaluation of the autonomous nervous system and neuronal lesion, besides biopsies for histological and histochemical studies, that will not be evaluated here in depth.

A careful clinical examination after a detailed anamnesis, especially analyzing the cardiovascular, digestive and neurologic systems, before, during and after treatment, monthly in the first year and at least once a year subsequently is fundamental importance for the evolutive study of the patients. The ideal condition, when ethics allows, would be the monitoring of a control group, of the same age and sex, untreated or receiving a placebo, with aspect similar to the drug, for evaluation of collateral effects and a comparative study between the treated and the control groups.

The standard electrocardiogram, with the limb leads (D1, D2, D3, aVR, aVL and aVF) and chest leads from V1 to V6 is the most simple and most important examination in the clinical evaluation. This test must be associated with the anamnesis and the physical examination in each consultation during all the follow-up period.

The radiological examination is a less sensible method and more expensive than ECG, so it should be performed once before treatment, 6 and 12 months later and then once per year of monitoring. This examination should consist of a chest RX with postero-anterior and lateral views with esophagus contrasted immediately after ingestion of barium and also after 1 min for the evaluation of the time necessary to drain the contrast (Rezende et al. 1960). The barium enema with previous preparation and radiography of colon is the only test capable of evaluating an established megacolon (Rezende 1959).

The dynamic electrocardiography and the echocardi-ography are suitable techniques for evaluation of arrhythmias and anatomophysiology of the heart and must be performed by a cardiologist. In the same way, endoscopic and manometric tests need an experienced gastroenterologist. The examination of the autonomous and peripheric nervous systems, with or without stimulation with cholinergic drugs, such as pilocarpine, can be done by a physician (Macêdo 1997).

The histopathology study of biopsies fragments of the digestive system or endomyocardic with conventional microscopy or analysis by immunoperoxidase or immunofluorescence are indicated only in some research cases and could only be performed following strict protocols from the ethics point of view.

New drugs in clinical tests

For several years in Brazil, and more recently in Argentina, Chile and Uruguay, only Bz is commercialy available as the development of drugs for tropical diseases is of little interest for the pharmaceutical industry (Fairlamb 1999). After the introduction of Nif and Bz, few compounds were assayed in chagasic patients.

Allopurinol

The results obtained with allopurinol (4-hydro-xypyrazolo (3,4-d) pyrimidine HPP, Fig. 1c) in experimental animals and the knowledge about its mode of action led to its clinical assays for the treatment of Chagas disease. This compound is a hypoxanthine analog that acts as an alternative substrate of hypoxanthine-guanine phospho-ribosyltransferase (HGPRT) and is incorporated into the RNA. This incorporation leads to formation of non-physiological nucleotides and to blockade of the synthesisde novo of purine nucleotides (reviewed in Marr 1991).

In the treatment of six acute phase patients, allopurinol was ineffective, with maintenance of positive xenodiagnosis and serology (Lauria-Pires et al. 1988). In a study, with chronic patients, Galleano et al. (1990) treated two groups of patients with 600 and 900 mg/kg/day of allopurinol for 60 days, compared with other two treated with Nif and Bz. In the four groups the percentage of negativation of xenodiagnosis was in the range of 75-92%, and those treated with allopurinol presented less collateral effects. Allopurinol (600 mg/day for 2 months) was administered to two cases of reactivation of Chagas disease due to cardiac transplant. Erythematous lesions on the superior and/or inferior members characterized this reactivation. In one patient the lesions disappeared in 3 weeks, and in the second one, after 2 weeks there was a clinical improvement of the lesions. In both cases after treatment, xenodiagnosis and hemoculture tests were negative and in the follow-up of 38 and 17 months, respectively, no reactivation of Chagas disease occurred, even with continued immunosuppression (Tomimori-Yamashita et al. 1997). Apt et al. (1998) treated 104 chronic patients with allopurinol (8.5 mg/kg/day for 60 days) that were monitored by clinical examination, serology, xenodiagnosis, hemoculture and electrocadiogram. Parasitological cure was achieved in 44% of the allopurinol-treated patients. The criteria for parasitological cure were maintenance of negative xenodiagnosis and/or complement-mediated lysis for at least four years. A double blind randomized longitudinal study must be performed to reevaluate the efficacy of this drug for the treatment of Chagas disease.

Ketoconazole

In eight chronic patients, ketoconazole (cis-(dl)-1-acetyl-4-[4-[[(2-2,4-dichlorophenyl)-2-(1H-imidazol-1-yl -methyl)-1,3-ioxalan-4-yl]methoxy]-phenyl]-piperazine, Fig. 1d) was administered in doses between 3.1 and 8.7 mg/kg by oral route during 51 to 96 days and cure evaluation was performed by hemoculture, conventional serology and complement-mediated lysis. The patients were monitored up to 60 months and it was observed that the drug was unable to erradicate the parasites, from 6 out of 8 patients with positive hemoculture and two others with positive serology (Brener et al. 1993).

In a case of reactivation of Chagas disease in a patient in the inderteminate phase due to infection by HIV, ketonazole (400 mg/day) was administered for 70 days leading to a negative xenodiagnosis. The treatment was suspended by the patient´s own decision and after one month occurred signs of reactivation of the disease, including development of myocarditis. Bz was introduced (200 mg/day) and after four days, although negativation of parasitemia, the patient presented signs of neurological deterioration. This drug was maintained for 45 days and then replaced by ketoconazole (400 mg/day) as a suppressive treatment. However, after a general clinical improvement, there was a new neurologic deterioration and the patient died (Galhardo et al. 1999).

Ketoconazole was one of the first imidazoles that showed in vitro activity against T. cruzi, with accumulation of metabolites of sterol metabolism in epimastigotes. In vivo ketoconazole led to parasitological cure in experimental animal in the acute phase, but was ineffective in the chronic phase (reviewed in De Castro 1993). A synergic effect of ketoconazole and Bz was observed in mice infected with the CL or Y strain, what did not occur in the case of the Bz-resistant Colombian strain (Araújo et al. 2000).

Fluconazole and itraconazole

An haemophylic patient was infected by blood transfusion with HIV and T. cruzi and brain biopsy revealed the presence of amastigotes inside glial, macrophages and endothelial cells. Initially he was treated with Bz (400 mg/day) but due to general worsening of his clinical condition, the medication was changed to itraconazole (200 mg/day), and latter, aiming a better CNS penetration, to fluconazole (400 mg/day). The azoles were administered for 11 weeks, and during this period the fever resolved and neurological symptoms stabilized. No significant collateral reaction was observed and three months after treatment the xenodiagnosis was negative and the titer of indirect hemagglutination test was 1:16 (Solari et al. 1993). Following the same methodology described for the treatment with allopurinol, Apt et al. (1998) treated 135 chronic patients with itraconazole (6 mg/kg/day for 120 days) observing parasitological cure and normalization of ECG in 36.5% of the treated patients but new abnormalities of the ECG appeared in 48.2% after treatment. As in the case of allopurinol these azoles must be further investigated in a well-designed protocol for treatment of chagasic patients.

The azoles fluconazole (a-(2,4-difluorophenyl)-a-(1H-1,2,4,-triazol-1-ylmethyl)-1H-1,2,4-triazol-1-ethanol, Fig. 1e) and itraconazole (cis-4[4-4-4[[2-(2-4-dichlorophenyl)-2-(1H-1,2,4,triazol-1-methyl)-1,3-dioxolan-4-yl]-1 -piperazinyl]phenyl]-2,4-dihydro-2-(1-methyl-propyl)-3H -1,2,4-triazol-3-one, Fig. 1f) have been previously assayed in experimental animals, and their mechanism of action against T. cruzi involve interference on ergosterol synthesis (reviewed in De Castro 1993). A more recent study showed potent effect of the D(+) isomer of fluconazole _ compound D0870 _ in both acute and chronic mice models, with 30-50 times higher activity than ketoconazole and Nif and leading to 60-70% of parasitological cure (Urbina et al. 1996). A formulation of D0870 as loaded nanospheres administered by intravenous route to mice, showed also a significant cure rate (Molina et al. 2001). It is important to stress that this compound, based on several cure parameters was also active in a chronic phase model. We believe that the pharmacokinetics of D0870 is now being investigated.

DEVELOPMENT OF NEW DRUGS

Development of anti-parasite chemotherapy could emerge from screening of synthetic or natural libraries, of compounds with structural similarities, with a drug with recognized activity, of assays with agents already approved for other diseases or through the determination of a specific target, identified in a key metabolic pathway. Although several putative targets have been presented, there is a need for their validation. The criteria for such validation was discussed by Wang (1997), who suggested that preliminary verifications can be indicated by in vitro activity of an inhibitor of the putative target, but before a major effort is directed to the design of specific inhibitors, three approaches should be used: (i) correlation of the target inhibition and anti-parasite activity among a series of drug derivatives; (ii) the comparison of the target between drug-sensitive and drug-resistant parasites; (iii) the knock out of the gene encoding such target in the parasite. In this paper Wang (1997) also pointed out the inherent difficulties of such approaches of target validation.

Promising targets

Recent developments in the study of the basic biochemistry of T. cruzi have allowed the identification of novel targets for chemotherapy that include sterol metabolism, enzymes such as trypanothione reductase, cystein proteinase, hypoxanthine-guanine phosphoribosyltrans-ferase, glyceraldehyde-3-phosphate dehydrogenase, DNA topoisomerases, dihydrofolate reductase and far-nesylpyrophosphate synthase (reviewed in DoCampo 2001 and in Rodriguez 2001).

Sterol synthesis

The knowledge about sterol synthesis on fungi opened the possibility of interference in this pathway, leading several pharmaceutical companies to develop drugs for the treatment of different types of superficial mycosis and systemic fungal infections. Since the main sterol of T. cruzi is ergosterol, an intensive and fruitful investigation about the potential effect of inhibitors of this sterol, especially by the group of Urbina in Venezuela. Parallel to clinical studies with azole derivatives, the trypanocidal activity and mechanism of action of new compounds is under intensive investigation.

The triazole posaconazole (SCH56592, Schering-Plouch), inhibited epimastigote proliferation and ergosterol synthesis at levels 30 to 100 times higher than ketoconazole and D0870. In experimental infections, this compound led to a cure rate of 50% in animals infected with strains resistant to Nif, Bz and ketoconazole (Molina et al. 2000). Another triazole derivative UR-9825 was very active against epimastigotes and intracellular amastigotes. At the minimum inhibitory concentration for epimastigotes occurred also depletion of 4,14-desmethyl endogenous sterols, such as ergosterol, and their replacement by methylated sterols, indicating inhibition of C14-alpha demethylase, as previously reported for other azoles. This drug induced also alteration in the phospholipid profile of the parasite (Urbina et al. 2000).

The induction of resistance of T. cruzi to azoles, such as fluconazole, and also the cross resistance between ketoconazole, miconazole and itraconazole, observed in in vitro experiments point to difficulties in the use of such compounds as chemotherapeutic agents (Buckner et al. 1998). In a subsequent work, Buckner et al. (2001) reported the development of inhibitors of a key enzyme in sterol biosynthesis, oxidosqualene cyclase, which converts 2,3-oxidosqualene to lanosterol. The lead compound, N-(4E,8E)-5,9, 13-trimethyl-4,8, 12-tetradecatrien-1-ylpyridinium, was shown to cause an accumulation of oxidosqualene and decreased production of lanosterol and ergosterol in T. cruzi. This compound and 27 related derivatives were tested against T. cruzi, and 12 of them were highly active against trypomastigotes.

Trypanothione reductase

Trypanosomatids present trypanothione (N1,N8-bis(glutationyl)spermidine) and of specific enzymes for this cofactor, trypanothione reductase (TR) and trypanothione oxidase (reviewed in Fairlamb & Cerami 1992). TR is an NADPH-dependent flavoprotein that maintains trypanothione in its reduced form and able to be oxidized by trypanothione oxidase, leading to reduction of free radicals levels and contributing to the maintenance of an intracellular reducing environment. TR has been used as a target for rational drug design against trypanosomiasis and leishmaniasis in a number of laboratories, since this enzyme and the mammalian counterpart (gluthatione peroxidase/glutathione reductase system) differ on the substrate specificity (reviewed in Augustyns et al. 2001). The determination of the structure of the active center of TR (Krauth-Siegel et al. 1987) allowed the search of inhibitors of this enzyme, being assayed different classes of compounds. In most cases the studies analyzed the effect of a putative inhibitor on the purified enzyme, and depending on the results obtained, new compounds based on molecular modeling were developed. A first group of inhibitors reported were the so-called "subversive substrates", due to the futile-cycling of TR induced by redox-damaging drugs, such as nitrofurans, and naphtoquinones (Henderson et al. 1988, Salmon-Chemin et al. 2001). Subsequently the structure of tricyclic neuroleptic showed to be a promising backbone class of TR inhibitors, and based on computational design techniques several tricylcic compounds were investigated (Chan et al. 1998, Gutierrez-Correa et al. 2001). Some compounds of the series of 2-amino diphenylsulfides, that have lower neuroleptic activity than phenothiazines, were potent inhibitors of TR (Girault et al. 1998). Polyamine derivatives (Bonnet et al. 1997, Li et al. 2001), bisbenzylisoquinoline alkaloids (Fournet et al. 1998) and platinum II complexes (Bonse et al. 2000) were also studied in their capacity of inhibiting TR of T. cruzi.

Cystein protease

Cruzipain, also known as cruzain or GP57/51, is a member of the papain C1 family of cystein proteinases (CPs). The T. cruzienzyme consists of a catalytic moiety with high homology to cathepsins S and L, and is absent in all other C1 families described so far (reviewed in Cazzulo et al. 2001). Irreversible inhibitors of cruzipain, such as several peptidyl diazomethylketones, peptidyl fluoro-methylketones and peptidyl vinyl sulphones interfered with the in vitro intracellular cycle of T. cruzi, killing the parasite (reviewed in McKerrow 1999).

The treatment of acutely infected mice with the vinyl sulphone N-piperazine-Phe-hPhe-vinyl sulphone phenyl led to the absence of myocardial lesions, lymphocyte infiltration and intracellular amastigote clusters. This drug kills T. cruzi by inducing an accumulation of unprocessed cruzipain in the Golgi cisternae, interfering with the secretory pathway (Engel et al. 1998a,b). Cruzipain exposed to biotin-labelled peptidyl diazomethane inhibitors with a spacer arm showed a stronger reaction than the counterparts without such spacer, probably due to differences in the topologies of the binding sites of proteinases, differences that could be exploited to improve specificity against trypanosomal CP (Lalmanach et al. 1996). Roush et al. (2000) substituting the L-leucine residue of the natural peptidylepoxysuccinate E-64, a selective irreversible inhibitor of CP, by a D-threonine obtained a derivative with much higher activity against cruzipain than against bovine cathepsin B. Yong et al. (2000) commented that a possible limitation of CP as a target would be the emergence of parasite populations developing resistance to inhibitors. These authors reported a phenotypically stable T. cruzi cell line (R-Dm28) that displays increased resistance to Z-(SBz)Cys-Phe-CHN2, an irreversible cysteine proteinase inhibitor, which preferentially inactivates cathepsin L-like enzymes.

Hypoxanthine-guanine phosphoribosyltransferase

Trypanosomatids must rely upon the salvage of exogenous purines for nucleotide synthesis, while in mammals these nucleotides are synthesized both de novo and salvaged from recycled purine bases. These protozoa convert purine bases to ribonucleotides, by the single enzyme HGPRT. This enzyme can also initiate in these parasites the metabolism of certain cytotoxic purine base analogs, such as allopurinol. This implies that either inhibitors or substrates of HGPRT have the potential of being effective and selective chemotherapeutic agents. The hgprt genes from T. cruzi and other pathogenic trypanosomatids have been cloned, sequenced and overexpressed in Escherichia coli, and the recombinant proteins have all been purified and characterized (reviewed in Ullman & Carter 1997).

The purine analogs 3'-deoxyinosine, 3'-deoxyadenosine and allopurinol inhibited the proliferation of amastigotes in HeLa cells, being the latter the most active. Among the pyrimidine analogs, 3'-azido-3'-deoxythymidine showed high activity against T. cruzi(Nakajima-Shimada et al. 1996). Purine analogs were assayed for their interaction with the HGPRTs from T. cruzi and man and some of them showed affinity for the trypanosomal enzyme (Eakin et al. 1997). A structure-based docking method identified 22 potential inhibitors of the enzyme. Three compounds (2,4,7-trinitro-9-fluorenyl-idenemalononitrite, 3-(2-fluorophenyl)-5-(phenoxy)-1,2,4-triazolo (4,3-C)-quinazoline and 3,5-diphenyl-4´-methyl-2-nitrobiphenyl) were effective against intracellular amastigotes and one [6-(2,2-dichloro-aceta-mido)chrysene] was a potent inhibitor of the trypanosomal HPRT (Freymann et al. 2000).

DNA topoisomerases

DNA topoisomerases II are enzymes that alter the topology of DNA and in kinetoplastids have been the focus of considerable study in the areas of molecular and cellular biology and also experimental chemotherapy. The gene encoding T. cruzi type II topoisomerase was isolated and the comparison with the amino acid sequence of the corresponding enzymes of T. brucei andCrithidia fasciculata showed a high degree of conservation (Fragoso & Goldenberg 1992). The enzyme is expressed in epimastigotes but not in trypomastigotes, although both forms of the parasite present the mRNA encoding the enzyme and is localized exclusively in the nucleus of the parasite (Fragoso et al. 1998).

Several inhibitors of bacterial DNA topoisomerase II showed activity against T. cruzi, inhibiting both proliferation and differentiation processes, and causing damage to kinetoplast and/or the nucleus of epimastigotes (Kerschmann et al. 1989, Gonzales-Perdomo et al. 1990), suggesting that both organelles could be the targets of the drugs. Camptothecin, inhibitor of eukaryotic DNA topoisomerase I, induced cleavage of nuclear and mitochondrial DNA in T. cruzi (Bodley & Shapiro 1995).

Dihydrofolate reductase

Dihydrofolate reductase (DHFR) and thymidylate synthetase exist as a bifunctional protein in different species of protozoa. This enzyme has successfully been used as a drug target in chemotherapy of cancer, malaria and bacterial infections. Thegene coding for the DHFR domain from T. cruzi was cloned and expressed (Reche et al. 1996). Zucotto et al. (1998) described the modelling of T. cruzi's DHFR based on the crystal structure of Leishmania major enzyme. From methotrexate, inhibitor of the human enzyme, among several derivatives synthesized, some of them showed a greater selectivity for the parasite enzyme than for the human counterpart (Zuccotto et al. 1999). In the same line, Chowdhury et al. (2001) designed and synthesized novel inhibitors of DHFR of trypanosomatids, however the compounds showed weak activity against both the enzyme and intracellular amastigotes of T. cruzi.

Glyceraldehyde-3-phosphate dehydrogenase

Since intracellular amastigotes possibly derive its energy from glycolysis, inhibition of glycolytic enzymes such as glyceraldehyde-3-phosphate dehydrogenase (GAPDH) may be a novel approach for the development of anti-T. cruzi drugs. The structure of GAPDH from glycosomes was reported and comparison with that of the mammalian counterpart led to the group of Oliva to consider the possibility of development of specific inhibitors of the parasite enzyme (Souza et al. 1998). In a subsequent work the isolation of flavonoids from the fruits of Neoraptua magnifica led to the compound 3',4',5',5,7-pentamethoxy-flavone that showed the highest activity over flavones and pyrano chalcones against the GAPDH of the parasite (Tomazela et al. 2000). Crystal structure of trypanosomatids and human GAPDHs provided details about the interaction of adenosyl moiety of NAD+ with proteins. Although adenosine is a very poor inhibitor, addition of substituents to the 2' position of ribose and the N6-position of adenosine led to a series of disubstituted nucleosides, and [N6-(1-naphthalenemethyl)-2'-(3-chlorobenzamido) adenosine] inhibited the proliferation of amastigotes without effect on the corresponding human enzyme (Bressi et al. 2001).

Farnesylpyrophosphate synthase

In pathogenic protozoa the pathway responsible for the synthesis of a variety of sterols and polyisoprenoids involves the enzyme farnesylpyrophosphate synthase, leading to the formation of farnesylpyrophosphate that marks the branching point of these synthet