- Email: [email protected]
Susceptibility and natural resistance of Trypanosoma cruzi strains to drugs used clinically in Chagas disease
TRANSACTIONS OF THE ROYAL SOCIETY OF TROPICAL MEDICIXE
AND HYGIENE (1987) 81, 755-759
755
Susceptibility and natural resistance of Trypanosoma cruzi strains to drugs used clinically in Chagas disease L. S. FILARDI AND Z. BRENER Centro de Pesquisas Rend Rachou, FIOCRUZ, Belo Horizonte, Brazil, C.P. 1743* and Department of Zoology, University of Minas Gerais, Belo Horizonte, Brazil
Abstract
The susceptibility and natural resistance to two nitroheterocyclic derivatives used clinically in Chagas disease (nifurtimox and benznidazole) were investigated in 47 Trypanosoma cruzi strains isolated from human patients, domestic vectors and sylvatic reservoirs or vectors. A large gradient of drug efficacy from 0% to 100% was detected. Drug susceptibility apparently related to geographical distribution of some T. cruzi strains was also observed. Drug resistancewas identified among T. cruzi populations isolated from sylvatic vectors from an area where autochthonous human Chagasdisease does not exist. Thus, natural drug-resistance of sylvatic strains might be a way of introducing this character into a T. cruai domestic cycle. Most of the 47 studied strains were either sensitive or resistant to both compounds, an intriguing finding considering that nifurtimox and benznidazole apparently have different mechanisms of action against T. cruzi. Introduction
Chagas disease is endemic in large areas of the American continent where lo-20 million inhabitants are infected by Trypanosoma cruzi. Morbidity is relatively high; 30 to 40% of chronic chagasicpatients display variable degrees of a myocardiopathy and 8-10% suffer from a digestive form characterized by pathological dilations of the oesophagus and colon. The remaining patients are in the indeterminate form in which symptoms are not detected by routine clinical examination although post-mortem examination may disclose slight heart histopathological lesions. Specific chemotherapy with the two available standard drugs (nifurtimox and benznidazole) has been indicated for the treatment of the short-lasting acute phase and, under strict medical supervision and prolonged follow-up, for patients in the indeterminate form. Both drugs are rather toxic nitroheterocyclic derivatives which are given in long-term schedules and cure only a very low percentage of the chronic cases (BRENER, 1984). Natural resistance of T. cruzi to nitro derivatives has been currently suggestedto be an important factor in explaining the low rates of cure detected in treated chagasicpatients. Differences in the susceptibility of a small number of T. cruzi strains to nifurtimox and benznidazole have been described (BRENER et al., 1976). More recentlv. ANDRADEet al. (1985) reDorted a correlation between resistance of T. cruzi’st&ns to both drugs and morphobiological characteristics of the parasite populations. They also suggested that certain strain types, with their particular drug susceptibilities, may predominate in a particular geographical area, and that this may explain contradictory results reported after specific treatment of Chagasdisease.In the present pa-&r we studied the susceptibility to nifurtimox and benznidazole of 47 different T. cruzi strains isolated from different hosts, including sylvatic vectors and reservoirs, in order to determine the *Address for correspondence.
distribution of the natural resistance to these drugs among those populations. Materials and Methods T. cruzi strains The 47 strains used in this work are deposited in the crvobank of T. cruai (World Health Organization Collaborating Centre for the Cjopreservation of-American Typanosoma) maintained at the Laboratory of Chagas Disease in the Centro de PesquisasRenCRachou, FundacHoOswald0 Cruz in Belo Horizonte, as infected blood samplesfrozen in liquid nitrogen (FILARDI& BRENER, 1975). The T. cm.& strains from patients or sylvatic reservoirs were isolated by xenodiagnosis and then inoculated into mice, whose blood was stored in the cryobank; blood from one acute human casewith extremely high parasitaemia was collected directly and cryopreserved in liquid nitrogen; parasitesfrom naturally infected vectors were inoculated into mice and then cryopreserved. Parasitesof human origin were isolated only from patients who had not received specific treatment. Table 1 lists all strains studied: 26 isolated from humans, 7 from the strictly intradomiciliary vector Triatoma infesrans,8 from the sylvatic vector Pansrrongvlusmegistus,1 from a domestic cat and 5 from wild reservoirs. Inoculation and treatment of animals Groups of 35 male albino mice, 18-20 g, were inoculated intraperitoneally with 10 000 T. crwi blood forms. The number of parasites in the inoculum was determined according to BRENER (1962). The inoculated animals were divided into two groups of 15 and then treated orally from the 4th day for 20 consecutive days, with lOOmg/kgof either the nitrofuran derivative nifurtimox 13-methvl-4(5’-nitrofurfurylidene-amino)-tetrahydro-4H-1:4-thiazine- i, 1-dioxide] or the 2-nitroimidazole derivative benznidazole [N-benZy!-2nitro-1-imidazolacetamide]. Both compounds were suspended in water and each mousereceived 0.25 ml of the drug suspension daily, by gavage. The remaining 5 mice were similarly inoculated and used as untreated controls. Blood examination
Blood collected from the tail of mice was examined microsconicallv for living flagellates. With treated animals the examination started about 10 days after treatment and was performed 2-5 times over a period of 15 days. In the control mice blood was examined for a variable number of
756
SUSCEPTIBILITY
AND
RESISTANCE
$gctiFner inoculation, to ensure the presence of patent Haemoculture 30 to 45 days after the end of treatment, treated and surviving untreated animals were bled from the orbital venous sinus and 0.4 to 0.6 ml of blood inoculated into 2 tubes containing 5 ml LIT (liver-infusion tryptose) medium (CAMARGO,1964). The tubes were incubated at 26-28°Cfor 30-60 days and examined microscopically for living flagellates. Indirect immunofluorescence test (IFT) This test was performed using formalin-fixed amastigote
OF
T.
CtWi
STRAINS
TO DRUGS
and trypomastigote stages from “Vero” tissue culture cells infected with T. cruzi and fluorescein conjugated anti-mouse Ig (Institut Pasteur, France). The preparations were examined under ultra-violet light in a Wild-Leitz “Ortholux” microscope equipped with an HBO 500 w/CAC lamp. Sera were collected from treated animals and controls maintained in the laboratory for 9-12 months after the end of treatment.
Results Table 1 lists all the strains studied, their origin, and the percentages of cure induced by nifurtimox or benznidazole in the different groups of mice. Fig. 1A shows the distribution of the 47 strains in relation to
Table I-Percentages of cure in mice inoculated with Ttypanosoma cruzi strains from different origins and geographical areas, submitted to specific treatment with nifurtimox and benznidazole Geographical Benznidazole Nifurtimox T. cruzi strain Origin % cure % cure area Noel Gilmar
J
Generoso
JAT !I%
Basileu Silvio MVC VLlO PNM ABC 7557
Berenice kane
SMJ
Y DM
knFaelipe
Ibarra CA1 Roman0 Colombiana E MR Buriti Guaraizinho Barra Seca YUYU SC-16 SC-17 g:; SC-20 SC-25 SC-7
SC-1 A-88 Chaninha EXR 426/l EXR 44217 EXR 42413 Mlo Pelada *Ind.=indeterminate
Minas Gerais I, ,, ,, ,, ,I ,, I, ,, ,, ,, ,, I, ,, ,! Goi& ,r ,I Slo Paul0 Rio Grande do Sul Bahia Argentina I, I, I, Colombja Rio Grande do Sul ,, I, N I, BAa Santa Catarina I! ,I I, I, I, I, ,, Santa Catarina Minas Gerais SHo Paul0 I, !I Minas Gerais
Acute,, case !, II ,I ,I ,, ,, I, Chronii (Ind.)* Chronic (Dig.)* Chronic (Card.-Dig.)* Chronic (Card.)* Chronic (Ind.) Chronic (Card.-Dig.) Chronic (Card.-Dig.) Chronic (Card.) Acute case Chroni;A:art.-Dig.) Acute case Chronic Chronic (Card.) Acute case chronic
0
1000 100 ik ifi
100
2
81 100 ii 100
69
12 15 100 47 100 ii 100 100 12 92 100 100 1x: 0
Tria’oma,, infestans ,I ,, I, I, I, Panstrongyf~s megistus !, !I !I r, ,, ,, Didelphis azarae azarae Felis domesticus Didelphis marsupialis D . marsupialis Philander opossum Procyon cancrivorus nigri$es
1;: 1:; 93 100 0 60 :; 71 100 -86 60 100 100 100 100 100
form, Dig. =digestive form, Card.-Dig.=Cardiac-digestive form, Card.=cardiac form.
L.
100
S.
FILARDI
AND
Z.
BRENER
-
is2 anma
OOJ
. a
0 COX
80
. ”
-
. .
al
0
.
:
OCW
. . 0 z
0
.
aa
60
0
”
l
3
.
.
” s
4c I
0
.
co
--
.
0 0
0
.
20
. 00
CI
.
0
en
0 0
.
8
A l
C
Nifurtimox
”
E
D
o Benrnidazole
Fig. 1. Percentage of cure in groups of mice inoculated with different T. of drug sensitivity of all 47 strains tested; 1B: T. cmzi strains isolated humans or domestic vectors in Rio Grande do Sol (Brazil) and Argentina; T. cruzi strains isolated from wild animal reservoirs and vectors in
cnui strains and treated with nifunimox or benznidazole. IA: Distribution from humans in Minas Gerais, Brazil; 1C: T. cruzi strains isolated from 1D: T. cruai strains isolated from Goias, S&o Paulo and Bahia (Brazil); 1E: Minas Gerais, Santa Catarina and SBO Paula (Brazil)
100
80
-
Nifurtimox
-
Benznidazole
60 t?! (3 is?
40
20
0
T cruzi Fig. 2. Individual
sensitivity
strains
of 47 T. cmzi strains to nifurtimox
and benznidazole determined
in experimentally
infected mice
758
SUSCEPTIBILITY AND RESISTANCE OF T. CtW.2;STRAINS TO DRUGS
their sensitivity to both standard drugs; a gradient of drug efficacy was detected with percentagesof cure from 0% to 100%. Figs lB, C, D, E show the results with the strains grouped according to their different origins. The data suggest a higher sensitivity of the southern strains (Fig. 1C) to both drugs. Strains from animal reservoirs (Fig. 1E) also showed a wide range of drug sensitivity. Fie. 2 shows individual sensitivities of the 47 T. cruzi&ains to either drug. Most strains responded to chemotherapy similarly, being either sensitive or resistant to both compounds. However, with at least one strain isolated from an onossum (Did&his mmupialis), a clear difference was observed between the percentages of cure obtained with nifurtimox (100%) and benznidazole (50%). Bv usina an arbitrarv level of 50% cure as a criterion~for &scriminating naturally drug-resistant and drug-sensitive T. crwi populations, 13 strains (27.6%) might be considered as resistant to both or either drug. In our experience haemoculture was a very sensitive and reliable method to ascertain cure in treated mice, as shown by the comparative results between this method and IFT. In a group of 425 mice (26.3% of all studied animals) thi agreement between both methods was 92.9%. whereas onlv 2.6% of the treated mice gave negative’haemoculture and positive IFT. Moreover, of 203 untreated control mice inoculated with the different T. cr& strains, only 2 failed to give positive haemocultures. This rules out the possibility that some of the negative results reported herein might be due to spontaneouscure rather than to drug effect. Discussion Differences in the susceptibility of T. cti strains to active drugs were first demonstrated using a bisauinaldine derivative (HAUSCHKA, 1949). Following the demonstration that long-term treatment with nitrofuran and nitroimidazole derivatives can induce parasitological cure in mice experimentally infected with T. crwi (BRENER, 1962), and the further use of these drugs in the human disease,drug-resistance of T. cruzi toboth compounds has been demonstrated in a limited number of strains (HABERKORN & GUNNERT, 1972; BRENER et al., 1976). More recently, ANDRADE et al. (1985) reported great variability in the susceptibility of a number of strains from different origins to the standard drugs nifurtimox and benznidazole . The practical implication of drug resistance in T. crun’ was difficult to assessa few years ago when, becauseof the lack of a dependable criterion of cure, Chagas disease was considered an incurable disease and T. cti an “unbeaten adversary” (BRENER, 1973). Recent study of the immune response to Chagas disease, as well as prolonged follow-up of treated patients. permitted the conclusion that some acute cases,and even a few chronic patients, can be cured (CERISOLA et al.. 1970: KRETTLI et al., 1982, 1984).It is now worth determining the importance of natural drug-resistancein the refractoriness of Chagas disease to treatment with the available drugs. We observed a wide range of drug susceptibility, and almost 30% of the parasite populations induced infections resistant to both or either drug-a figure that could be higher if a more stringent criterion were
used. This is significant, considering that the doses administered to mice (100 mg/kg) were rather high compared with those given to patients (5-8 mg/kg). Interestingly, drug resistance was detected in a number of strains isolated from wild reservoirs and sylvatic vectors in which previous contact with the drugs can be surely excluded; among those strains 3 had been isolated from strictly wild vectors in the Santa Catarina State, Brazil, where there is no autochthonous human Chagas disease. Differences in drug susceptibility of T. crwi strains related to their geographical origin were reported by ANDRADE et al. (1985), who detected extremely high resistanceto nifurtimox and benznidazole in 9 strains isolated from patients living in Montalvania, Minas Gerais, an area of recent introduction of Chagas disease. In our material, 7 strains from south Brazil (Rio Grande do Sul State) and 4 from Argentina had very high cure rates (93.3 to loo%), rather different from those observed with strains from Minas Gerais State which displayed variable degrees of drug sensitivity. This geographical variation is interesting considering the existence of reports suggesting higher cure rates of acute patients in Argentina than in central or southwest Brazil. Thus, whereas in Argentina 81% cures have been reported, using as a criterion persistently negative parasitological and serological tests (CERISOLA et al., 1970), Brazilian authors using similar methods reported at best 30-50% cures (FERREIRA, 1976; RASSI & FERREIRA, 1971).
Whether the experimental data obtained in mice with T. cruai strains correspond to the response to drugs in human hosts has not yet been established, partly because stricter criteria of cure in Chagas disease now demand prolonged follow-up involving conventional serology (IFT, complement-fixation reaction), tests to detect protective or “lytic” antibodies which are associatedwith active infection, and parasitological tests (haemoculture, xenodiagnosis) (KRETTLI et al., 1982, 1984). Using such criteria we established a correlatton between experimental and clinical data in 4 acute casesfrom which strains had been isolated before treatment and submitted, after inoculation in mice, to our experimental protocol (strains Basileu. MVC. Silvio and AAS in Table 1). The first 2 patients were considered as cured according to the criteria described above and their strains were highly susceptible to the drugs; the patients Silvio and AAS (whose strains were resistant) still have positive xenodiagnoses after treatment ’ with benznidazole. No correlation could be established between the conventional characteristics of strains, such as virulence or pathogenicity in experimental animals, and their drug sensitivity. A wide range of virulence was detected in the investigated strains, but no correlation whatsoever could be established between drug susceptibility and the severity of infection evaluated by pre-patent period, curves of parasitaemiaor mortality rates. Acknowledgements
This work receivedfinancial support from the World
Health Organization and the National ResearchCouncil and FINEP in Brazil. References Andrade, S. G., Magalhles,J. B. & Pontes, A. L. (1985). Evaluation of chemotherapy with benznidazole and
L.
S.
FILARDI
nifurtimox in mice infected with Tiypanosoma crusi strains of different types. Bulletin of the World Health Organization,
63, 721-726.
Brener, Z. (1962). Therapeutic activity and criterion of cure on mice experimentally infected with Ttypanosoma cruzi. f8ymistst6doInstituto de Medicina Tropical de Sao Paula, 4, Brener, Z.‘(1973). Biology of Trypanosoma cruzi. Annual Review of Microbiology,
27, 347-383.
Brener, Z., Costa, C. A. G. & Chiari, C. A. (1976). Differences in the susceptibility of Ttypanosoma cruzi strains to active chemotherapeutic agents. Rev&a do Institute de Medicina Tropical de Sdo Paulo, l&450-455. Camargo, E. P. (1964). Growth and differentiation in Trypanosoma cruzi. I. Origin of metacyclic trypomastigotes in liquid media. Revista do Znstituto de Medicina Trotkal de Srio Paulo. 6. 93-100. Cerisoh, J. A., Alvarez, hI’& De Rissio, A. M. (1970). Imunodiagnbstico da doencade Chagas.Evolucao serologica de pacientes corn doenca de Chagas. Revista do Instituto de Medicina Tropical de Sao Pa&. 12.403-411. Ferreira. H. 0. (1976). Ensaio teratiutico-clh&o corn o do Znstituto de benz’mdazolna doenca de Chagas.~&vista Medicina Tropical de Sao Paulo, 18, 357-364. Filardi, L. S. & Brener, Z. (1975). Cryopreservation of
AND
2.
BRENER
759
Ttypanosoma cruzi bloodstream forms. 7ournal cf Protozoology, 22, 398401.
Haberkorn, A. & Gonnert, R. (1972). Animal experimental investigations in the activity of nifurtimox against Ttypanosoma cnm’. Amneimittel-Forschung,
22, 1570-
1581. Hauschka, T. S. (1949). Persistence of strain-specific behaviour in two strains of Trypanosoma cruzi after prolonged transfer through inbred mice. Journal of Parasitology,
35, 593-599.
Krettli, A. U., Cancado, J. R. & Brener, Z. (1982). Effect of specific chemotherapy on the levels of lytic antibodies in Chagas’ disease. Transactions of the Royal Society of Tropical Medicine and Hygiene, 76, 493-496.
Krettli, A. U., Cancado, J. R. & Brener, Z. (1984). Criterion of cure of human Chagas’ diseaseafter specific chemotheram: recent advances. Memorias do Znstituto Oswald0 C&? 79 (Suppl.), 157-164. Rassi, A. & Ferreua, H. 0. (1971). Tentativas de tratamento especffico da fase aguda da doenca de Chagas corn nitrofuranos em esquemasde duracao nrolongada. Revisto da Sociedade Rrasileira 235-262. Accepted for publication
CORPORATE MEMBERSHIP WITH THE ROYAL COMMONWEALTH SOCIETY 18 NORTHUMBERLAND AVENUE, LONDON, WC2 SBJ
The Society has taken up Corporate Membership with the above Societv. which has txemises iust off Trafalgar Square. Ove-keas Fellows’ can now use the facilities of this Society, when they come to London, for a period of 21 days in any three months provided that all bookina is done throuph the Secretarv at Manson Ho&. Fellows who-wish to use these
facilities must give the Secretary adequate notice of their requirements particularly during the summer when at least two months’ notice will be required. Details of the facilities available will be sent to Fellows on request.
de hedkina
10 December
Trovical.
1986
5.
AND HYGIENE (1987) 81, 755-759
755
Susceptibility and natural resistance of Trypanosoma cruzi strains to drugs used clinically in Chagas disease L. S. FILARDI AND Z. BRENER Centro de Pesquisas Rend Rachou, FIOCRUZ, Belo Horizonte, Brazil, C.P. 1743* and Department of Zoology, University of Minas Gerais, Belo Horizonte, Brazil
Abstract
The susceptibility and natural resistance to two nitroheterocyclic derivatives used clinically in Chagas disease (nifurtimox and benznidazole) were investigated in 47 Trypanosoma cruzi strains isolated from human patients, domestic vectors and sylvatic reservoirs or vectors. A large gradient of drug efficacy from 0% to 100% was detected. Drug susceptibility apparently related to geographical distribution of some T. cruzi strains was also observed. Drug resistancewas identified among T. cruzi populations isolated from sylvatic vectors from an area where autochthonous human Chagasdisease does not exist. Thus, natural drug-resistance of sylvatic strains might be a way of introducing this character into a T. cruai domestic cycle. Most of the 47 studied strains were either sensitive or resistant to both compounds, an intriguing finding considering that nifurtimox and benznidazole apparently have different mechanisms of action against T. cruzi. Introduction
Chagas disease is endemic in large areas of the American continent where lo-20 million inhabitants are infected by Trypanosoma cruzi. Morbidity is relatively high; 30 to 40% of chronic chagasicpatients display variable degrees of a myocardiopathy and 8-10% suffer from a digestive form characterized by pathological dilations of the oesophagus and colon. The remaining patients are in the indeterminate form in which symptoms are not detected by routine clinical examination although post-mortem examination may disclose slight heart histopathological lesions. Specific chemotherapy with the two available standard drugs (nifurtimox and benznidazole) has been indicated for the treatment of the short-lasting acute phase and, under strict medical supervision and prolonged follow-up, for patients in the indeterminate form. Both drugs are rather toxic nitroheterocyclic derivatives which are given in long-term schedules and cure only a very low percentage of the chronic cases (BRENER, 1984). Natural resistance of T. cruzi to nitro derivatives has been currently suggestedto be an important factor in explaining the low rates of cure detected in treated chagasicpatients. Differences in the susceptibility of a small number of T. cruzi strains to nifurtimox and benznidazole have been described (BRENER et al., 1976). More recentlv. ANDRADEet al. (1985) reDorted a correlation between resistance of T. cruzi’st&ns to both drugs and morphobiological characteristics of the parasite populations. They also suggested that certain strain types, with their particular drug susceptibilities, may predominate in a particular geographical area, and that this may explain contradictory results reported after specific treatment of Chagasdisease.In the present pa-&r we studied the susceptibility to nifurtimox and benznidazole of 47 different T. cruzi strains isolated from different hosts, including sylvatic vectors and reservoirs, in order to determine the *Address for correspondence.
distribution of the natural resistance to these drugs among those populations. Materials and Methods T. cruzi strains The 47 strains used in this work are deposited in the crvobank of T. cruai (World Health Organization Collaborating Centre for the Cjopreservation of-American Typanosoma) maintained at the Laboratory of Chagas Disease in the Centro de PesquisasRenCRachou, FundacHoOswald0 Cruz in Belo Horizonte, as infected blood samplesfrozen in liquid nitrogen (FILARDI& BRENER, 1975). The T. cm.& strains from patients or sylvatic reservoirs were isolated by xenodiagnosis and then inoculated into mice, whose blood was stored in the cryobank; blood from one acute human casewith extremely high parasitaemia was collected directly and cryopreserved in liquid nitrogen; parasitesfrom naturally infected vectors were inoculated into mice and then cryopreserved. Parasitesof human origin were isolated only from patients who had not received specific treatment. Table 1 lists all strains studied: 26 isolated from humans, 7 from the strictly intradomiciliary vector Triatoma infesrans,8 from the sylvatic vector Pansrrongvlusmegistus,1 from a domestic cat and 5 from wild reservoirs. Inoculation and treatment of animals Groups of 35 male albino mice, 18-20 g, were inoculated intraperitoneally with 10 000 T. crwi blood forms. The number of parasites in the inoculum was determined according to BRENER (1962). The inoculated animals were divided into two groups of 15 and then treated orally from the 4th day for 20 consecutive days, with lOOmg/kgof either the nitrofuran derivative nifurtimox 13-methvl-4(5’-nitrofurfurylidene-amino)-tetrahydro-4H-1:4-thiazine- i, 1-dioxide] or the 2-nitroimidazole derivative benznidazole [N-benZy!-2nitro-1-imidazolacetamide]. Both compounds were suspended in water and each mousereceived 0.25 ml of the drug suspension daily, by gavage. The remaining 5 mice were similarly inoculated and used as untreated controls. Blood examination
Blood collected from the tail of mice was examined microsconicallv for living flagellates. With treated animals the examination started about 10 days after treatment and was performed 2-5 times over a period of 15 days. In the control mice blood was examined for a variable number of
756
SUSCEPTIBILITY
AND
RESISTANCE
$gctiFner inoculation, to ensure the presence of patent Haemoculture 30 to 45 days after the end of treatment, treated and surviving untreated animals were bled from the orbital venous sinus and 0.4 to 0.6 ml of blood inoculated into 2 tubes containing 5 ml LIT (liver-infusion tryptose) medium (CAMARGO,1964). The tubes were incubated at 26-28°Cfor 30-60 days and examined microscopically for living flagellates. Indirect immunofluorescence test (IFT) This test was performed using formalin-fixed amastigote
OF
T.
CtWi
STRAINS
TO DRUGS
and trypomastigote stages from “Vero” tissue culture cells infected with T. cruzi and fluorescein conjugated anti-mouse Ig (Institut Pasteur, France). The preparations were examined under ultra-violet light in a Wild-Leitz “Ortholux” microscope equipped with an HBO 500 w/CAC lamp. Sera were collected from treated animals and controls maintained in the laboratory for 9-12 months after the end of treatment.
Results Table 1 lists all the strains studied, their origin, and the percentages of cure induced by nifurtimox or benznidazole in the different groups of mice. Fig. 1A shows the distribution of the 47 strains in relation to
Table I-Percentages of cure in mice inoculated with Ttypanosoma cruzi strains from different origins and geographical areas, submitted to specific treatment with nifurtimox and benznidazole Geographical Benznidazole Nifurtimox T. cruzi strain Origin % cure % cure area Noel Gilmar
J
Generoso
JAT !I%
Basileu Silvio MVC VLlO PNM ABC 7557
Berenice kane
SMJ
Y DM
knFaelipe
Ibarra CA1 Roman0 Colombiana E MR Buriti Guaraizinho Barra Seca YUYU SC-16 SC-17 g:; SC-20 SC-25 SC-7
SC-1 A-88 Chaninha EXR 426/l EXR 44217 EXR 42413 Mlo Pelada *Ind.=indeterminate
Minas Gerais I, ,, ,, ,, ,I ,, I, ,, ,, ,, ,, I, ,, ,! Goi& ,r ,I Slo Paul0 Rio Grande do Sul Bahia Argentina I, I, I, Colombja Rio Grande do Sul ,, I, N I, BAa Santa Catarina I! ,I I, I, I, I, ,, Santa Catarina Minas Gerais SHo Paul0 I, !I Minas Gerais
Acute,, case !, II ,I ,I ,, ,, I, Chronii (Ind.)* Chronic (Dig.)* Chronic (Card.-Dig.)* Chronic (Card.)* Chronic (Ind.) Chronic (Card.-Dig.) Chronic (Card.-Dig.) Chronic (Card.) Acute case Chroni;A:art.-Dig.) Acute case Chronic Chronic (Card.) Acute case chronic
0
1000 100 ik ifi
100
2
81 100 ii 100
69
12 15 100 47 100 ii 100 100 12 92 100 100 1x: 0
Tria’oma,, infestans ,I ,, I, I, I, Panstrongyf~s megistus !, !I !I r, ,, ,, Didelphis azarae azarae Felis domesticus Didelphis marsupialis D . marsupialis Philander opossum Procyon cancrivorus nigri$es
1;: 1:; 93 100 0 60 :; 71 100 -86 60 100 100 100 100 100
form, Dig. =digestive form, Card.-Dig.=Cardiac-digestive form, Card.=cardiac form.
L.
100
S.
FILARDI
AND
Z.
BRENER
-
is2 anma
OOJ
. a
0 COX
80
. ”
-
. .
al
0
.
:
OCW
. . 0 z
0
.
aa
60
0
”
l
3
.
.
” s
4c I
0
.
co
--
.
0 0
0
.
20
. 00
CI
.
0
en
0 0
.
8
A l
C
Nifurtimox
”
E
D
o Benrnidazole
Fig. 1. Percentage of cure in groups of mice inoculated with different T. of drug sensitivity of all 47 strains tested; 1B: T. cmzi strains isolated humans or domestic vectors in Rio Grande do Sol (Brazil) and Argentina; T. cruzi strains isolated from wild animal reservoirs and vectors in
cnui strains and treated with nifunimox or benznidazole. IA: Distribution from humans in Minas Gerais, Brazil; 1C: T. cruzi strains isolated from 1D: T. cruai strains isolated from Goias, S&o Paulo and Bahia (Brazil); 1E: Minas Gerais, Santa Catarina and SBO Paula (Brazil)
100
80
-
Nifurtimox
-
Benznidazole
60 t?! (3 is?
40
20
0
T cruzi Fig. 2. Individual
sensitivity
strains
of 47 T. cmzi strains to nifurtimox
and benznidazole determined
in experimentally
infected mice
758
SUSCEPTIBILITY AND RESISTANCE OF T. CtW.2;STRAINS TO DRUGS
their sensitivity to both standard drugs; a gradient of drug efficacy was detected with percentagesof cure from 0% to 100%. Figs lB, C, D, E show the results with the strains grouped according to their different origins. The data suggest a higher sensitivity of the southern strains (Fig. 1C) to both drugs. Strains from animal reservoirs (Fig. 1E) also showed a wide range of drug sensitivity. Fie. 2 shows individual sensitivities of the 47 T. cruzi&ains to either drug. Most strains responded to chemotherapy similarly, being either sensitive or resistant to both compounds. However, with at least one strain isolated from an onossum (Did&his mmupialis), a clear difference was observed between the percentages of cure obtained with nifurtimox (100%) and benznidazole (50%). Bv usina an arbitrarv level of 50% cure as a criterion~for &scriminating naturally drug-resistant and drug-sensitive T. crwi populations, 13 strains (27.6%) might be considered as resistant to both or either drug. In our experience haemoculture was a very sensitive and reliable method to ascertain cure in treated mice, as shown by the comparative results between this method and IFT. In a group of 425 mice (26.3% of all studied animals) thi agreement between both methods was 92.9%. whereas onlv 2.6% of the treated mice gave negative’haemoculture and positive IFT. Moreover, of 203 untreated control mice inoculated with the different T. cr& strains, only 2 failed to give positive haemocultures. This rules out the possibility that some of the negative results reported herein might be due to spontaneouscure rather than to drug effect. Discussion Differences in the susceptibility of T. cti strains to active drugs were first demonstrated using a bisauinaldine derivative (HAUSCHKA, 1949). Following the demonstration that long-term treatment with nitrofuran and nitroimidazole derivatives can induce parasitological cure in mice experimentally infected with T. crwi (BRENER, 1962), and the further use of these drugs in the human disease,drug-resistance of T. cruzi toboth compounds has been demonstrated in a limited number of strains (HABERKORN & GUNNERT, 1972; BRENER et al., 1976). More recently, ANDRADE et al. (1985) reported great variability in the susceptibility of a number of strains from different origins to the standard drugs nifurtimox and benznidazole . The practical implication of drug resistance in T. crun’ was difficult to assessa few years ago when, becauseof the lack of a dependable criterion of cure, Chagas disease was considered an incurable disease and T. cti an “unbeaten adversary” (BRENER, 1973). Recent study of the immune response to Chagas disease, as well as prolonged follow-up of treated patients. permitted the conclusion that some acute cases,and even a few chronic patients, can be cured (CERISOLA et al.. 1970: KRETTLI et al., 1982, 1984).It is now worth determining the importance of natural drug-resistancein the refractoriness of Chagas disease to treatment with the available drugs. We observed a wide range of drug susceptibility, and almost 30% of the parasite populations induced infections resistant to both or either drug-a figure that could be higher if a more stringent criterion were
used. This is significant, considering that the doses administered to mice (100 mg/kg) were rather high compared with those given to patients (5-8 mg/kg). Interestingly, drug resistance was detected in a number of strains isolated from wild reservoirs and sylvatic vectors in which previous contact with the drugs can be surely excluded; among those strains 3 had been isolated from strictly wild vectors in the Santa Catarina State, Brazil, where there is no autochthonous human Chagas disease. Differences in drug susceptibility of T. crwi strains related to their geographical origin were reported by ANDRADE et al. (1985), who detected extremely high resistanceto nifurtimox and benznidazole in 9 strains isolated from patients living in Montalvania, Minas Gerais, an area of recent introduction of Chagas disease. In our material, 7 strains from south Brazil (Rio Grande do Sul State) and 4 from Argentina had very high cure rates (93.3 to loo%), rather different from those observed with strains from Minas Gerais State which displayed variable degrees of drug sensitivity. This geographical variation is interesting considering the existence of reports suggesting higher cure rates of acute patients in Argentina than in central or southwest Brazil. Thus, whereas in Argentina 81% cures have been reported, using as a criterion persistently negative parasitological and serological tests (CERISOLA et al., 1970), Brazilian authors using similar methods reported at best 30-50% cures (FERREIRA, 1976; RASSI & FERREIRA, 1971).
Whether the experimental data obtained in mice with T. cruai strains correspond to the response to drugs in human hosts has not yet been established, partly because stricter criteria of cure in Chagas disease now demand prolonged follow-up involving conventional serology (IFT, complement-fixation reaction), tests to detect protective or “lytic” antibodies which are associatedwith active infection, and parasitological tests (haemoculture, xenodiagnosis) (KRETTLI et al., 1982, 1984). Using such criteria we established a correlatton between experimental and clinical data in 4 acute casesfrom which strains had been isolated before treatment and submitted, after inoculation in mice, to our experimental protocol (strains Basileu. MVC. Silvio and AAS in Table 1). The first 2 patients were considered as cured according to the criteria described above and their strains were highly susceptible to the drugs; the patients Silvio and AAS (whose strains were resistant) still have positive xenodiagnoses after treatment ’ with benznidazole. No correlation could be established between the conventional characteristics of strains, such as virulence or pathogenicity in experimental animals, and their drug sensitivity. A wide range of virulence was detected in the investigated strains, but no correlation whatsoever could be established between drug susceptibility and the severity of infection evaluated by pre-patent period, curves of parasitaemiaor mortality rates. Acknowledgements
This work receivedfinancial support from the World
Health Organization and the National ResearchCouncil and FINEP in Brazil. References Andrade, S. G., Magalhles,J. B. & Pontes, A. L. (1985). Evaluation of chemotherapy with benznidazole and
L.
S.
FILARDI
nifurtimox in mice infected with Tiypanosoma crusi strains of different types. Bulletin of the World Health Organization,
63, 721-726.
Brener, Z. (1962). Therapeutic activity and criterion of cure on mice experimentally infected with Ttypanosoma cruzi. f8ymistst6doInstituto de Medicina Tropical de Sao Paula, 4, Brener, Z.‘(1973). Biology of Trypanosoma cruzi. Annual Review of Microbiology,
27, 347-383.
Brener, Z., Costa, C. A. G. & Chiari, C. A. (1976). Differences in the susceptibility of Ttypanosoma cruzi strains to active chemotherapeutic agents. Rev&a do Institute de Medicina Tropical de Sdo Paulo, l&450-455. Camargo, E. P. (1964). Growth and differentiation in Trypanosoma cruzi. I. Origin of metacyclic trypomastigotes in liquid media. Revista do Znstituto de Medicina Trotkal de Srio Paulo. 6. 93-100. Cerisoh, J. A., Alvarez, hI’& De Rissio, A. M. (1970). Imunodiagnbstico da doencade Chagas.Evolucao serologica de pacientes corn doenca de Chagas. Revista do Instituto de Medicina Tropical de Sao Pa&. 12.403-411. Ferreira. H. 0. (1976). Ensaio teratiutico-clh&o corn o do Znstituto de benz’mdazolna doenca de Chagas.~&vista Medicina Tropical de Sao Paulo, 18, 357-364. Filardi, L. S. & Brener, Z. (1975). Cryopreservation of
AND
2.
BRENER
759
Ttypanosoma cruzi bloodstream forms. 7ournal cf Protozoology, 22, 398401.
Haberkorn, A. & Gonnert, R. (1972). Animal experimental investigations in the activity of nifurtimox against Ttypanosoma cnm’. Amneimittel-Forschung,
22, 1570-
1581. Hauschka, T. S. (1949). Persistence of strain-specific behaviour in two strains of Trypanosoma cruzi after prolonged transfer through inbred mice. Journal of Parasitology,
35, 593-599.
Krettli, A. U., Cancado, J. R. & Brener, Z. (1982). Effect of specific chemotherapy on the levels of lytic antibodies in Chagas’ disease. Transactions of the Royal Society of Tropical Medicine and Hygiene, 76, 493-496.
Krettli, A. U., Cancado, J. R. & Brener, Z. (1984). Criterion of cure of human Chagas’ diseaseafter specific chemotheram: recent advances. Memorias do Znstituto Oswald0 C&? 79 (Suppl.), 157-164. Rassi, A. & Ferreua, H. 0. (1971). Tentativas de tratamento especffico da fase aguda da doenca de Chagas corn nitrofuranos em esquemasde duracao nrolongada. Revisto da Sociedade Rrasileira 235-262. Accepted for publication
CORPORATE MEMBERSHIP WITH THE ROYAL COMMONWEALTH SOCIETY 18 NORTHUMBERLAND AVENUE, LONDON, WC2 SBJ
The Society has taken up Corporate Membership with the above Societv. which has txemises iust off Trafalgar Square. Ove-keas Fellows’ can now use the facilities of this Society, when they come to London, for a period of 21 days in any three months provided that all bookina is done throuph the Secretarv at Manson Ho&. Fellows who-wish to use these
facilities must give the Secretary adequate notice of their requirements particularly during the summer when at least two months’ notice will be required. Details of the facilities available will be sent to Fellows on request.
de hedkina
10 December
Trovical.
1986
5.