Species diversity causing human cutaneous leishmaniasis in Rio Branco, state of Acre, Brazil

Tropical Medicine and International Health doi:10.1111/j.1365-3156.2006.01695.x volume 11 no 9 pp 1388 1398 september 2006 Species diversity causing human cutaneous leishmaniasis in Rio Branco, state of Acre, Brazil Anna Christina Tojal da Silva 1, Elisa Cupolillo 2,Ângela Cristina Volpini 2, Roque Almeida 3 and Gustavo Adolfo Sierra Romero 1 1 Núcleo de Medicina Tropical, Universidade de Brasília, Campus Darcy Ribeiro, Brasília, DF, Brazil 2 Laboratório de Pesquisas em Leishmanioses, Departamento de Imunologia, Instituto Oswaldo Cruz, Rio de Janeiro, Brazil 3 Serviço de Imunologia, Hospital Universitário Professor Edgar Santos, Universidade Federal da Bahia, Salvador, Brazil Summary objective Information on Leishmania species diversity in western Brazilian Amazon and the clinical picture of human cutaneous leishmaniasis it causes is scarce. We describe clinical findings, diagnostic procedures and identification of Leishmania species in patients from that region. methods The sample consisted of 50 patients, prospectively evaluated for epidemiological and clinical characteristics by means of a structured questionnaire. Conventional and molecular tools were applied to confirm the parasitological diagnosis and identify the species responsible for the disease. results Patients were predominantly male (76.5%) and living in rural areas. Median average age was 1 18 years and median average disease evolution was 8 weeks. For the diagnostic procedures of leishmanin skin test, direct visualization of amastigotes in dermal scrapings and parasite culture of aspirates of the ulcer border were positive for 98%, 52% and 34%, respectively. Molecular methods applied to DNA extracted from skin biopsies of the 50 patients yielded 100%, 82% and 44% positivity by PCR minicircle kdna, PCR-RFLP ITS1rDNA and PCR-glucose-6-phosphate (G6P), respectively. Fourteen samples from 13 patients were successfully isolated and identified. Multilocus enzyme electrophoresis, PCR-RFLP ITS1rDNA and PCR-G6P permitted identification of the Leishmania species responsible for the aetiology of American tegumentary leishmaniasis in 60% of the examined patients: 16 Leishmania (Viannia) braziliensis, 12Leishmania (Viannia) lainsoni, 1Leishmania (Viannia) guyanensis and 1 putative hybrid of Leishmania (Viannia) naiffi and L. (V.) lainsoni. conclusion The clinical and epidemiological behaviour of cutaneous leishmaniasis in Acre, Brazil, is similar to other Amazon scenarios previously described; however Acre s complex parasite diversity may be contributed to the concomitant circulation of at least three distinct Leishmania species. The implementation of control interventions in the studied area must take into consideration the possibility of various expected phlebotomine vectors and reservoirs. keywords cutaneous leishmaniasis, Leishmania (Viannia) braziliensis, Leishmania (Viannia) lainsoni, Leishmania (Viannia) naiffi, western Brazilian Amazon, genetic diversity Introduction American tegumentary leishmaniasis (ATL) is a zoonotic disease caused by parasites belonging to the Leishmania genus transmitted by sand fly bites. The widely disseminated disease afflicts individuals mainly in tropical and subtropical regions. Many clinical presentations have been described involving skin and the mucosal surface of the upper airway, and the most common syndrome is characterized by localized cutaneous disease. In Brazil, the genetic diversity among the Leishmania parasites is great; at least seven Leishmania species have been described as the aetiological agent of human cutaneous disease: Leishmania (Viannia) braziliensis, Leishmania (Viannia) guyanensis, Leishmania (Viannia) shawi, Leishmania (Viannia) lainsoni, Leishmania (Viannia) naiffi, Leishmania (Viannia) lindenbergi and Leishmania (Leishmania) amazonensis (Grimaldi et al. 1989; Shaw 1994; Silveira et al. 2002). From 1980 to 2003, 552 059 cases of ATL were officially reported in Brazil with an annual incidence from 3.83 to 22.94 cases per 100 000 inhabitants (Ministério da Saúde 2004). Knowledge on the geographical distribution of each species causing ATL is scarce for some regions, such as the state of Acre, in the western Brazilian Amazon, where the disease is endemic (Figure 1). The Amazon region is especially interesting because of its biological diversity of 1388 ª 2006 Blackwell Publishing Ltd

Figure 1 Political map of Brazil and neighbouring countries, detaching the state of Acre with municipalities where patients came from during the study period. potential vectors and reservoirs, which may quite possibly favour the sympatric circulation of various Leishmania species (Grimaldi et al. 1989; Lainson et al. 1994). From 1980 to 2003, reported cases from Acre increased from 41 to 1298, calling the attention of health officers to the problem that is considered a public health challenge because of its diagnostic, therapeutic and follow-up complexity. Limited clinical and demographic data on cases from Acre have been published from the information in notification files, and there are no prospective studies aiming at parasite species identification in that region (Silva et al. 1999). Species identification in ATL is important not only for taxonomic purposes but also for prognostic issues, because the response to treatment with pentavalent antimonial is dependent upon parasite species (Navin et al. 1992; Romero et al. 2001a). The gold standard for Leishmania species identification is the multilocus enzyme electrophoresis (MLEE) approach that differentiates all parasites causing ATL in Brazil (Rioux et al. 1990; Cupolillo et al. 1994). Monoclonal antibodies have been employed in identification procedures as well as discrimination among parasite species causing ATL, although less efficiently, as there are no species-specific monoclonals available for some of these species. Both techniques enable intraspecific variation identification as extensively reported (Miles et al. 1981; McMahon-Pratt et al. 1982; Cupolillo et al. 1994; Grimaldi & McMahon-Pratt 1996; Romero et al. 2002). Molecular tools for species identification are under evaluation, and some techniques such as PCR and PCR-RFLP are extremely promising (Castilho et al. 2003; Schönian et al. 2003; Volpini et al. 2004). The objective of our study is to contribute to the understanding of ATL epidemiology in the state of Acre located in the western Brazilian Amazon, by describing the clinical findings, the positivity of diagnostic procedures and the identification of Leishmania species responsible for cases of cutaneous disease in this area. Patients and methods Patients Our study was conducted in Rio Branco municipality where 25% of the state s ATL reported cases occurred in the last 3 years. A convenience sample composed of patients with suspected lesions was evaluated from March to September 2002. Patients were invited to participate during diagnostic consultation in the main State Laboratory or during home visits by the National Foundation of Health officers. Inclusion criteria were the presence of any number of cutaneous lesions with more than 2 weeks of evolution suggestive of leishmaniasis and a signed consent agreement form. Exclusion criteria were the presence of mucosal disease and children under 6 years old because of the difficulties involved in invasive diagnostic procedures in young children. This work was in accordance with resolution 196/96 of the National Health Council of the Ministry of Health of Brazil, which regulates research in human beings. The Ethics Research Committee of the ª 2006 Blackwell Publishing Ltd 1389

University of Brasilia approved the study protocol. All patients were treated following the recommendations of the Brazilian Ministry of Health for leishmaniasis treatment. ATL clinical cases were defined as patients with clinically suspected lesions of more than 2 weeks evolution plus positive leishmanin skin test or any positive parasitological diagnostic test such as: direct visualization of amastigotes, successful parasite isolation or amplification of conserved kdna region by the PCR described below. Skin test The Montenegro skin test was performed with an antigen prepared as described formerly (Reed et al. 1986). Briefly, 0.1 ml of soluble components of L. (L.) amazonensis promastigotes (MHOM/BR/86/BA 125 strain), with antigen concentration of 250 lg/ml, was applied by intradermic injection on the volar surface of the left forearm, and the reaction was measured after 48 72 h. Induration diameters 5 mm were considered positive. (E.C.2.7.1.1), glucose-6-phosphate dehydrogenase (G6PDH; E.C.1.1.1.49), isocitrate dehydrogenase (E.C.1.1.1.42), malate deshydrogenase (E.C.1.1.1.37), malic enzyme (E.C.1.1.1.40), acid phosphatase (E.C.3.1.3.2), glucose phosphate isomerase (E.C.5.3.1.9), nucleopeptidases 1 and 2 (E.C.3.2.2.1), proline peptidase (E.C.3.4.13.9), phosphoglucomutase (PGM; E.C.1.4.1.9) and 6-phosphogluconate dehydrogenase (6PGDH; E.C.1.1.1.43); sample preparation and visualization procedures followed the protocols described by Cupolillo et al. (1994). Seven reference strains of different Leishmania species were used as control for parasite identification (Table 1). Isolates were evaluated using numerical methods to determine their taxonomic positions and the relationship among them. Briefly, the table containing the electromorphs (alleles) positions was transformed in a binary matrix, and then a similarity matrix was constructed applying Jaccard s coefficient to finally obtain the phenogram using the UPGMA algorithm (NTSYS program, version 1.70). Parasite isolation and visualization Leishmania cultures of the ulcer border were collected using the technique described previously (Romero et al. 1999a). Some patients underwent satellite lymph node aspiration for the same purposes employing the procedures described already (Romero et al. 1999b). Scarification of the ulcer border was performed to obtain smears for direct examination sought for amastigotes. Smears were fixed with methanol, stained with Giemsa and observed via optical microscopy (1000 ). Multilocus enzyme electrophoresis Parasite identification was carried out by enzyme electrophoresis using 12 enzymatic loci as follows: hexokinase PCR and RFLP Biopsies of the ulcer border were obtained with disposable 4 mm sterile punches and stored at )20 C until the DNA extraction procedure. DNA was extracted from biopsies using the Wizard TM Genomic DNA Purification System (Promega, Madison, WI, USA) following manufacturer s instructions. In order to identify the species incriminated as the ATL aetiological agent in the studied region, mainly involving patients where parasite isolation was unsuccessful, different PCR-based methods were employed: (1) PCR to amplify Leishmania conserved region of minicircle kdna (mkdna) of 120 base pairs that are Leishmaniagenus-specific (PCR mkdna), adhering to exactly the protocol previously described (Volpini et al. 2004), (2) PCR-RFLP of ITS1 region from rdna (Schönian et al. International code IOC/Zà Species MHOM/BR/75/M4147 23 Leishmania (Viannia) guyanensis MHOM/BR/75/M2903 27 Leishmania (Viannia) braziliensis IFLA/BR/67/PH8 07 Leishmania (Leishmania) amazonensis MHOM/BR/74/PP75 01 Leishmania (Leishmania) chagasi MHOM/BR/81/M6426 15 Leishmania (Viannia) lainsoni MDAS/BR/79/M5533 36 Leishmania (Viannia) naiffi MCEB/BR/84/M8408 26 Leishmania (Viannia) shawi Table 1 International code and zymodemes assignment of the Leishmania reference strains employed in this study The strains are deposited at the CLIOC Coleção de Leishmania do Instituto Oswaldo Cruz (registration no. WDC 731, WFCC World Data Center on Microorganisms Directory for identification procedures). àioc/z: zymodeme of the Instituto Oswaldo Cruz Leishmania collection. 1390 ª 2006 Blackwell Publishing Ltd

2003), employing in each reaction 0.20 lm of each primer, 1.5 U of Taq polymerase and 1.5 mm MgCl 2, followed by one cycle of 4 min at 94 C and 35 cycles of 94 C for 30 s, 53 C for 30 s and 72 C for 40 s then finally an extension cycle of 72 C for 8 min and (3) PCR of the glucose- 6-phosphate (G6P) gene with combinations of primers described as L. (Viannia) specific, non-l. (V.) braziliensis and specific to L. (V.) braziliensis (Castilho et al. 2003). The PCR reactions were conducted introducing some modifications in the previously described protocols. In the experiment directed towards discrimination between L. (Viannia) and non-l. (Viannia) parasites, 50 pmol/ll of the ISVC primer and 10 pmol/ll of the ISVA, 2 U of Taq polymerase and 2 mm MgCl 2 was utilized. An initial PCR cycle at 94 C for 4 min followed by 30 cycles of 94 C for 30 s, 60 C for 1 min and 72 C for 30 s followed by one final cycle at 72 C for 8 min was adopted. The PCR cycle used for the reactions, which employed primers to discriminate between L. (V.) braziliensis and non-l. (V.) braziliensis, consisted of one initial step for 4 min at 94 C and then 30 cycles of 94 C at30s,60or62 C (for ISVC + ISVB and ISVC + ISVNB, respectively) for 1 min and 72 C for 30 s with a final step of 8 min at 72 C. Positive PCR controls containing DNA from reference strains of L. (V.) braziliensis, L. (V.) lainsoni, L. (V.) guyanensis and L. (V.) naiffi (Table 1) together with a negative control with no DNA were included in each reaction set. In order to avoid the risk of PCR contamination, all reactions aiming at mkdna amplification were performed as described in Volpini et al. (2004). Briefly, (a) the extraction, preparation of the PCR mix and amplification were conducted in different rooms with separate reagents, plastics and equipment and (b) the use of Ready- To-Go TM sans PCR Beads (GE Healthcare, Piscataway, NJ, USA) instead of the hand-made PCR reagent mix. After PCR amplification, the amplicons corresponding to the mkdna were digested with restriction enzyme ApaLI (PCR-RFLP mkdna) and subjected to 6% polyacrylamide gel electrophoresis (Volpini et al. 2004). The PCR products of ITS1rDNA were digested with restriction enzyme Sau3AI, subsequently exposed to electrophoresis in the GenePhor apparatus (GE Healthcare) using a 12.5% polyacrylamide gel (GeneGel; GE Healthcare). Amplicons from the G6P gene were visualized in 6% polyacrylamide gel electrophoresis. All gels were silver stained (DNA silver staining kit; GE Healthcare). Results Of the 51 patients studied, 50 had localized cutaneous lesions and one had the disseminated disease. For the purpose of counting and lesions localization, we excluded the patient with the disseminated disease who had 30 lesions. There were 88 lesions observed in 50 patients: 53 lesions (60.2%) appeared in the lower limbs, 22 (25.0%) in the upper limbs, 10 (11.4%) in the trunk and 3 (3.4%) in the face and neck. Other clinical findings and demographic characteristics are described in Table 2. Fifty patients were submitted to an aspiration procedure of the ulcer border, and Leishmania cultures were positive in 17 patients (34%), four isolates were lost during transportation. Eight patients were submitted to lymph node aspiration, and parasite isolation was successful in one. The positivity of diagnostic procedures is described in Table 3. Fourteen isolates were obtained from 13 patients. In one patient, there was successful parasite isolation from the skin ulcer and its satellite lymph node. All isolates were identified by enzyme electrophoresis and belonged to 10 different zymodemes, nine of them new enzymatic profiles. Table 2 Demographic and clinical findings of 51 patients with cutaneous leishmaniasis attended in Rio Branco, Acre, Brazil from March to September 2002 Characteristics Number of patients or medians %of quartiles Male 39 76.5 Median age (years) 18 15 and 34 Living in rural areas 27 52.9 Median disease 8 4 and 12 evolution (weeks) Signs and symptoms Regional lymphadenopathy 10 19.6 at first examination Itching 30 58.8 Fever 8 15.7 Secretion 41 80.4 Table 3 Positivity of diagnostic procedures in patients with cutaneous leishmaniasis attended in Rio Branco, Acre, Brazil from March to September 2002 Diagnostic test No. of positives/tested Positivity (%) Leishmanin skin test 49/50 98 Direct visualization of amastigotes 26/50 52 in scrapings from ulcer border Ulcer skin culture 17/50 34 Lymph node culture 2/8 25 PCR mkdna 50/50 100 PCR-RFLP ITS1rDNA 41/50 82 PCR-G6P 22/50 44 ª 2006 Blackwell Publishing Ltd 1391

Eight isolates were new zymodemes of L. (V.) braziliensis, one isolate was identical to zymodeme IOC/Z23 of L. (V.) guyanensis, three isolates corresponded to a new zymodeme of L. (V.) lainsoni and one isolate named MHOM/ BR/2002/NMT-RBO-OO4P was identified as a putative hybrid between L. (V.) lainsoni and L. (V.) naiffi. This strain presented the same profile of L. (V.) lainsoni in 9 of the 12 enzymatic loci used, in one locus the pattern observed could not discriminate between the two species, and in 2 loci (6PGDH and PGM) the patterns were suggestive of heterozygotes for a dimeric and monomeric enzyme, respectively. A representative panel of electrophoretic patterns of these isolates is shown in Figure 2. The dendogram (Figure 3) demonstrates the relation among 14 isolates belonging to 10 different zymodemes. Biopsies of 50 patients were obtained and DNA was extracted to perform different PCR-based methods useful for the aetiological diagnosis of ATL. The results of PCR mkdna, PCR-RFLP ITS1 and G6P exhibited 100%, 82% 575 579 1545 005 025 037 011 1365 004 027 027g 1023 036 013 566 018 035 040 029 1365 Figure 2 One percent agarose gel showing some alleles observed for the 6PGDH loci. Reference strains: 575 L. (L.) amazonensis, 579 L. (L.) chagasi, 1545 L. (V.) shawi, 1365 L. (V.) naiffi, 1023 L. (V.) lainsoni and 566 L. (V.) braziliensis. The other strains correspond to isolates from Rio Branco, Acre, Brazil. Arrow indicates the profile corresponding to the putative hybrid (004) between L. (V.) naiffi and L. (V.) lainsoni. and 44% positivity, respectively. After the digestion of the 120 bp fragment, representing the amplification of the mkdna, it was possible to determine that all the samples analysed represent species belonging to the L. (Viannia) subgenus. The patterns obtained through PCR-RFLP ITS1 from 41 samples displayed 13 compatible patterns with the reference strains of L. (V.) lainsoni and 28 with either L. (V.) braziliensis or L. (V.) guyanensis, as this methodology has not enabled discrimination between L. (V) braziliensis and L. (V.) guyanensis yet. The results of 22 positives for G6P demonstrate that 16 were L. (V.) braziliensis and 6 were non-l. (V.) braziliensis. These results were compared and in agreement with those obtained by MLEE. The three methods together (MLEE, PCR-RFLP ITS1rDNA and PCR-G6P) enabled identification of the Leishmania species responsible for the aetiology in 60.8% (31/50) of the patients who had tissue biopsies for molecular diagnosis. In the sample studied, 12 L. (V.) lainsoni, 16 L. (V.) braziliensis, 1 L. (V.) guyanensis and 1 putative hybrid between L. (V.) naiffi and L. (V.) lainsoni were detected. In 11 samples, it was not possible to determine the Leishmania species involved in the infection, but it is certain that the patients were infected with L (V.) braziliensis or L. (V.) guyanensis, according to the PCR- RFLP ITS1rDNA results. In 18% of the samples submitted for molecular assays (9/50), it was only possible to determine that the ATL was being caused by one of the L. (Viannia) species, considering that only the PCR-RFLP of the mkdna yielded positive results. The putative hybrid presented a profile identical to L. (V.) lainsoni by the PCR- RFLP ITS1rDNA. Table 4 demonstrates the clinical and diagnostic findings of the 31 patients whose Leishmania 100 0.00 83 90 100 0.25 0.50 0.75 1.00 L. shawi Z26 L. guyanensis Z23 013P Z23 L. naiffi Z36 L. lainsoni Z15 027P 044P Z86 027G 036P 004P Z87 L. braziliensis Z27 005P Z78 025P 037P Z79 018P Z80 035P Z81 040P Z82 011P Z83 029P Z84 Figure 3 Dendogram obtained through the UPGMA algorithm and Jaccard s similarity coefficient. 1392 ª 2006 Blackwell Publishing Ltd

Table 4 Clinical, demographic and diagnostic findings of 31 patients attended in Rio Branco, state of Acre, Brazil, who had successful parasite identification through multilocus enzyme electrophoresis and/or molecular methods Isolate/international code Species IOC/Zà Age (years) Sex No. of lesions Lesion site Disease duration (weeks) Lesion appearance Lymph node involvement MHOM/BR/2002/NMT-RBO-005 L. (V.) b. 78 42 M 2 Lower limb 8 Round ulcer Not MHOM/BR/2002/NMT-RBO-008 L. (V.) b. ND 23 M 3 Lower limb/upper limb/thorax 4 Round ulcer Not MHOM/BR/2002/NMT-RBO-009 L. (V.) b. ND 7 M 1 Abdomen 2 Round ulcer Yes MHOM/BR/2002/NMT-RBO-011 L. (V.) b. 83 17 M 2 Upper limb 20 Round ulcer Not MHOM/BR/2002/NMT-RBO-012 L. (V.) b. ND 14 M 1 Lower limb 4 Round ulcer Yes MHOM/BR/2002/NMT-RBO-015 L. (V.) b. ND 6 M 1 Lower limb 10 Round ulcer Yes MHOM/BR/2002/NMT-RBO-017 L. (V.) b. ND 18 M 1 Head 9 Round ulcer Yes MHOM/BR/2002/NMT-RBO-018 L. (V.) b. 80 14 M 1 Face 4 Round ulcer Yes MHOM/BR/2002/NMT-RBO-022 L. (V.) b. ND 29 F 2 Lower limb 16 Round ulcer Not MHOM/BR/2002/NMT-RBO-023 L. (V.) b. ND 38 M 2 Lower limb 12 Round ulcer Yes MHOM/BR/2002/NMT-RBO-025 L. (V.) b. 78 14 M 3 Upper limb 24 Round ulcer Yes MHOM/BR/2002/NMT-RBO-028 L. (V.) b. ND 21 M 1 Lower limb 8 Round ulcer Yes MHOM/BR/2002/NMT-RBO-029 L. (V.) b. 84 32 M 2 Upper limb 8 Round ulcer Not MHOM/BR/2002/NMT-RBO-035 L. (V.) b. 81 16 M 1 Lower limb 4 Round ulcer Yes MHOM/BR/2002/NMT-RBO-037 L. (V.) b. 79 16 M 1 Thorax 3 Irregular ulcer Not MHOM/BR/2002/NMT-RBO-040 L. (V.) b. 82 16 M 1 Lower limb 7 Verrucous Not MHOM/BR/2002/NMT-RBO-045 L. (V.) b. ND 25 M 3 Lower limb 4 Round ulcer Yes MHOM/BR/2002/NMT-RBO-010 L. (V.) l. ND 25 M 3 Lower limb/thorax/abdomen 3 Round ulcer Yes MHOM/BR/2002/NMT-RBO-014 L. (V.) l. ND 43 M 1 Upper limb 3 Irregular ulcer Yes MHOM/BR/2002/NMT-RBO-016 L. (V.) l. ND 41 F 1 Head 6 Round ulcer Yes MHOM/BR/2002/NMT-RBO-020 L. (V.) l. ND 10 F 5 Lower limb 2 Round ulcer Yes MHOM/BR/2002/NMT-RBO-021 L. (V.) l. ND 12 F 1 Lower limb 2 Round ulcer Yes MHOM/BR/2002/NMT-RBO-027 L. (V.) l. 86 17 M 3 Lower limb 8 Round ulcer Yes MHOM/BR/2002/NMT-RBO-031 L. (V.) l. ND 14 M 1 Lower limb 4 Round ulcer Yes MHOM/BR/2002/NMT-RBO-032 L. (V.) l. ND 30 F 5 Lower limb/upper limb/abdomen 4 Round ulcer Not MHOM/BR/2002/NMT-RBO-036 L. (V.) l. 86 34 M 1 Upper limb 10 Round ulcer Yes MHOM/BR/2002/NMT-RBO-043 L. (V.) l. ND 15 F 1 Upper limb 13 Round ulcer Yes MHOM/BR/2002/NMT-RBO-044 L. (V.) l. 86 10 M 1 Thorax 9 Irregular ulcer Not MHOM/BR/2002/NMT-RBO-048 L. (V.) l. ND 18 M 1 Thorax 7 Round ulcer Yes MHOM/BR/2002/NMT-RBO-013 L. (V.) g. 23 6 M 5 Upper limb 5 Round ulcer Yes MHOM/BR/2002/NMT-RBO-004 L. (V.) l./l. 87 17 F 1 Lower limb 2 Round ulcer Not (V.) n. L. (V.) b., Leishmania (Viannia) braziliensis; L. (V.) l., Leishmania (Viannia) lainsoni; L. (V.) n., Leishmania (Viannia) naiffi; L. (V.) g., L. (V.) guyanensis; ND, not determined. History of lymph node involvement at any time. àzymodeme code number of the Instituto Oswaldo Cruz collection. ª 2006 Blackwell Publishing Ltd 1393

parasites, including the putative hybrid, were identified by any of the diagnostic methods. Discussion The increasing number of ATL cases in the state of Acre may be ascribed to a number of factors such as the improvement of the official surveillance system, migratory movements for agricultural purposes of susceptible individuals to active transmission areas, progressive deforestation with negative impact on parasite, vector and reservoir natural equilibrium and exploitation activities of forest products such as fruits, rubber and lumber. The clinical presentation of cutaneous leishmaniasis in this study was similar to descriptions from other geographical regions of Brazil where the species belonging to the Leishmania (Viannia) subgenus are responsible for the majority of ATL cases (Grimaldi et al. 1989; Grimaldi & 2 Tesh 1993). The predominance of L. (V.) braziliensis or L. (V.) guyanensis is also in agreement with the epidemiology of ATL in Brazil where these species have been associated with most of the cases described in the different endemic regions (Grimaldi et al. 1989, Romero et al. 2001b). However, it is important to mention that the relative proportion of each observed species in our study might be inaccurate, as this work was based on a small, non-random sample that may have excluded rare cases caused by other species. In fact, some other species are quite abundant in neighbouring countries (e.g. Bolivia and Peru presenting several ATL cases associated with L. lainsoni and Leishmania peruviana), which may quite possibly affect the epidemiology of ATL in our studied region. No clinical signs were specifically associated to any of the species identified in our study reinforcing the fact that it is almost impossible to identify the species causing the disease based on clinical findings. The performance of isolation procedures was lower than expected, if we consider the experience with the same technique in other regions where at least in 50% of the patients the parasite was successfully isolated. There were no exclusion criteria based on disease duration, which may be the determinant factor to explain the results. In fact, patients with longer disease more frequently had negative isolation procedures. Despite our low unsuccessful Leishmania isolation, 13/50 3(26%), which has also been reported by others in literature (Cuba et al. 1984; Weigle et al. 1987; Rodriguez et al. 1994), our data suggest a wide diversity of Leishmania, with at least three different species plus a putative hybrid circulating in the studied region. The diversity of genotypes found was impressive with nine newly described zymodemes, seven of them belonging to the L. (V.) braziliensis complex, confirming the great diversity of this species, as already observed in other regions (Ishikawa et al. 2002; Cupolillo et al. 2003). The identification of one isolate as L. (V.) guyanensis, belonging to the same zymodeme as the reference strain (Cupolillo et al. 1994), which circulates in the north of the Amazon River (Grimaldi et al. 1989; Romero et al. 2001a), suggests the possibility of this species spreading to the western Brazilian Amazon. This patient had 5 weeks of lesion evolution and had moved from Bolivia to Acre 7 months before contracting the disease. We did not consider the possibility of the infection being contracted in Bolivia; as far as we knew there were no reports of L. (V.) 4 guyanensis infection there. New World parasites considered as hybrids have been identified with monoclonal antibodies, MLEE, molecular karyotyping and RAPD because of shared biochemical, antigenic or genomic characteristics from two species termed parental. These findings were previously described with Old World parasites (Evans et al. 1987), however, recently reported in the Americas (Bonfante-Garrido et al. 1992; Belli et al. 1994; Dujardin et al. 1995; Delgado et al. 1997). Of note is the fact that in this study, the first description of Leishmania isolates exhibiting heterozygotic profiles suggesting a hybrid pattern between L. (V.) lainsoni and L. (V.) naiffi is introduced. Our results imply that the hybrid parasite is genetically closer to L. (V.) lainsoni than to L. (V.) naiffi. It is also important to mention that we did not characterize any L. (V.) naiffi in the sample examined by any of the adopted methodologies. However, L. (V.) naiffi have been encountered infecting humans and sylvatic animals in several areas of the Brazilian Amazon region near the state of Acre (Naiff et al. 1991; Gil et al. 2003), and the dissemination of this species throughout the South American countries has already been demonstrated (Pratlong et al. 2002). The genetic mechanisms involved in the formation of Leishmania hybrid parasites remain unclear. Whether Leishmania parasites presenting hybrid profiles are contributing to or are a result of the genetic diversity detected in the area is another point that needs to be addressed. The fact that all biopsies were positive for PCR mkdna was not surprising as all were obtained from the lesions of suspected ATL patients, and the sensitivity of the technique is recognized. Although it has recently been announced that human DNA can be amplified when some primers targeted to Leishmania kinetoplast are applied (Vergel et al. 2005), we excluded the possibility of falsepositive results. Previous studies utilizing the same primers employed herein displayed negative results when human DNA obtained from patients presenting nonleishmaniasis diseases was tested (Pirmez et al. 1999; Marques et al. 2001). Furthermore, the PCR-RFLP 1394 ª 2006 Blackwell Publishing Ltd

patterns observed after digestion with ApaLI as well as other restriction enzymes (Volpini A.C. & Cupolillo E. 52006, unpublished data) were identical to those witnessed when promastigote DNA from Leishmania species was used. The positivity of PCR-ITS1rDNA (82%) and PCR- G6P (44%) markers were low compared with mkdna, but altogether MLEE, PCR-RFLP ITS1rDNA and PCR- G6PDH were able to identify the species involved in the infection of 30 samples submitted to molecular diagnostic methods. The presence of L. (V.) lainsoni in the study region should be explained by the dissemination of this species, as a number of cases were reported in Peru and Bolivia, both countries bordering the state of Acre (Lucas et al. 1998; Martinez et al. 2001). The four isolates (two from the same patient) identified as L. (V.) lansoni by MLEE belonged to the same zymodeme (IOC/Z86); however they differed from the reference strain of L. (V.) lainsoni (IOC/Z15) that was isolated in the Eastern Brazilian Amazon region. A comparison of L. (V.) lainsoni isolates from Acre, Peru and Bolivia by MLEE could give more information about the relationship and dispersion of these parasites. Disease acquired outside the borders of the state of Acre was considered during the interview prior to the diagnostic procedure. In all the patients examined, only three declared exposure outside the state of Acre. One patient had travel history to Bolivia and the second one travelled to the state of Roraima, Brazil. Both patients did not have positive species identification of their Leishmania parasites and it was possible only to determine that they were infected by L. (Viannia) species. The third was infected with L. (V.) guyanensis, as already commented. The genetic diversity of Leishmania parasites from patients presenting similar disease in the state of Acre is important to reinforce the need of species identification for individual purposes. The isolation of L. (V.) braziliensis, a recognized cause of mucosal disease (Marsden 1985), could be relevant for prognostic counselling after healing of cutaneous disease. The external validity of our results is limited by the non-random sampling and the exclusion of suspected mucosal disease; however, the observed parasite diversity indicates that the scenario for implementation of interventions could be complex in spite of the actual relative proportion of cases that might be attributed to each Leishmania species. Because of this relevant parasite diversity, it would be reasonable to expect heterogeneity of transmission cycles including various phlebotomine vectors and reservoirs. Control measures based on vector behaviour will be extremely complex considering the peculiarities of each species, and human exposure prevention could be difficult in a region where public policies stimulate the permanence of people in forested areas practising activities with low impact on natural environment. Acknowledgements The research was funded by the following Brazilian research agencies: the National Council for Scientific and Technological Development of the Ministry of Science and Technology (CNPq/MCT process 475106/-1-6; Elisa Cupolillo is a CNPq fellow researcher) and the Carlos Chagas Filho Research Foundation of the state of Rio de Janeiro. The English version of the article was reviewed and revised by Mitchell Raymond Lishon. Table Available on Request: Electromorphs (alleles) presented in each Leishmania (Viannia) zymodeme detected infecting patients in Acre State, Brazil. References Belli AA, Miles MA & Kelly JM (1994) A putative Leishmania panamensis/leishmania braziliensis hybrid is a causative agent of human cutaneous leishmaniasis in Nicaragua. Parasitology 109, 435 442. Bonfante-Garrido R, Méndez E, Barroeta S et al. (1992) Cutaneous leishmaniasis in western Venezuela caused by infection with Leishmania venezuelensis and L. braziliensis variants. Transactions of the Royal Society of Tropical Medicine and Hygiene 86, 141 148. Castilho TM, Shaw JJ & Floeter-Winter LM (2003) New PCR assay using glucose-6-phosphate dehydrogenase for the identification of Leishmania species. Journal of Clinical Microbiology 41, 540 546. Cuba CC, Llanos-Cuentas EA, Barreto AC et al. (1984) Human mucocutaneous leishmaniasis in Três Braços, Bahia-Brazil. An area of Leishmania braziliensis braziliensis transmission. I. Laboratory diagnosis. Revista da Sociedade Brasileira de Medicina Tropical 17, 161 167. Cupolillo E, Grimaldi Jr G & Momen H (1994) A general classification of New World Leishmania using numerical zymotaxonomy. The American Journal of Tropical Medicine and Hygiene 50, 296 311. Cupolillo E, Brahim LR, Toaldo CP et al. (2003) Genetic polymorphism and molecular epidemiology of Leishmania (Viannia) braziliensis in different host and geographic areas in Brazil. Journal of Clinical Microbiology 41, 3126 3132. Delgado O, Cupolillo E, Bonfante-Garrido R et al. (1997) Cutaneous leishmaniasis in Venezuela caused by infection with a new hybrid between Leishmania (Viannia) braziliensis and L. (V.) guyanensis. Memórias do Instituto Oswaldo Cruz 92, 581 582. Dujardin J-C, Bañuls A-L, Llanos-Cuentas A et al. (1995) Putative Leishmania hybrids in the Eastern Andean Valley of Huanuco, Peru. Acta Tropica 59, 293 307. ª 2006 Blackwell Publishing Ltd 1395

Evans DA, Kennedy WPK, Elbihari S, Chapman CJ, Smith V & Peters W (1987) Hybrid formation within the genus Leishmania? Parassitologia 29, 165 173. Gil LHS, Basano SA, Souza AA et al. (2003) Recent observations on the sand fly (Diptera: Psychodidae) fauna of the State of Rondônia, Western Amazonia, Brazil: the importance of Psychodopygus davisi as a vector of zoonotic cutaneous leishmaniasis. Memórias do Instituto Oswaldo Cruz 98, 751 755. Grimaldi G Jr & McMahon-Pratt D (1996) Monoclonal antibodies for the identification of New World Leishmania species. Memórias do Instituto Oswaldo Cruz 91, 37 42. Grimaldi G Jr, Tesh RB & McMahon-Pratt D (1989) A review of the geographic distribution and epidemiology of leishmaniasis in the New World. The American Journal of Tropical Medicine 6 and Hygiene 41, 687 725. Grimaldi G Jr & Tesh RB (1993) Leishmaniases of the New World: current concepts and implications for future research. Clinical Microbiology Reviews 6, 230 250. Ishikawa EA, Silveira FT, Magalhães AL et al. (2002) Genetic variation in population of Leishmania species in Brazil. Transactions of the Royal Society of Tropical Medicine and Hygiene 96 (Suppl. 1), s111 s121. Lainson R, Shaw JJ, Silveira FT, Sousa AAA, Braga RR & Ishikawa EAY (1994) The dermal leishmaniases of Brazil, with special reference to the eco-epidemiology of the disease in Amazonia. Memórias do Instituto Oswaldo Cruz 89, 435 443. Lucas CM, Franke ED, Cachay MI et al. (1998) Geographic distribution and clinical description of leishmaniasis cases in Peru. The American Journal of Tropical Medicine and Hygiene 59, 312 317. Marques MJ, Volpini AC, Genaro O, Mayrink W & Romanha AJ (2001) Simple form of clinical sample preservation and Leishmania DNA extraction from human lesions for diagnosis of American cutaneous leishmaniasis via polymerase chain reaction. The American Journal of Tropical Medicine and Hygiene 65, 902 906. Marsden PD (1985) Clinical presentations of Leishmania braziliensis braziliensis. Parasitology Today 1, 129 133. Martinez E, Le Pont F, Mollinedo S & Cupolillo E (2001) First case of cutaneous leishmaniasis due to Leishmania (Viannia) lainsoni in Bolivia. Transactions of the Royal Society of Tropical Medicine and Hygiene 95, 375 377. McMahon-Pratt D, Bennett E & David JR (1982) Monoclonal antibodies that distinguish subspecies of Leishmania braziliensis. Journal of Immunology 12, 926 927. Miles MA, Lainson R, Shaw JJ, Povoa M & de Souza AA (1981) Leishmaniasis in Brazil: XV. Biochemical distinction of Leishmania mexicana amazonensis, L. braziliensis braziliensis and L. braziliensis guyanensis aetiological agents of cutaneous leishmaniasis in the Amazon Basin of Brazil. Transactions of the Royal Society of Tropical Medicine and Hygiene 75, 524 529. 7 Ministério da Saúde (Brazil) (2004) Leishmaniose Tegumentar Americana. Distribuição de casos confirmados por unidade federada, Brasil 1980 2003. Available at: http://portal.saude. gov.br/portal/arquivos/pdf/lta.pdf (accessed in 29 November 2005). Naiff RD, Freitas RA, Naiff MF et al. (1991) Epidemiological and nosological aspects of Leishmania naiffi Lainson & Shaw, 1989. Memórias do Instituto Oswaldo Cruz 86, 317 321. Navin TR, Arana BA, Arana FE, Berman JD & Cajón JF (1992) Placebo-controlled clinical trial of sodium stibogluconate (Pentostam) versus ketoconazol for treating cutaneous leishmaniasis in Guatemala. Journal of Infectious Diseases 165, 528 534. Pirmez C, da Silva Trajano V, Paes-Oliveira Neto M et al. (1999) Use of PCR in diagnosis of human American tegumentary leishmaniasis in Rio de Janeiro, Brazil. Journal of Clinical Microbiology 37, 1819 1823. Pratlong F, Deniau M, Darie H et al. (2002) Human cutaneous leishmaniasis caused by Leishmania naiffi is wide-spread in South America. Annals of Tropical Medicine and Parasitology 96, 781 785. Reed SG, Badaró R, Masur H et al. (1986) Selection of a specific skin test antigen for American visceral leishmaniasis. The American Journal of Tropical Medicine and Hygiene 35, 78 85. Rioux JA, Lanotte G, Serres E, Pratlong F, Bastien P & Perieres J (1990) Taxonomy of Leishmania. Use of isoenzymes. Suggestions for a new classification. Annales de Parasitologie Humaine et Comparée 65, 111 125. Rodriguez N, Guzman B, Rodas A, Takiff H, Bloom BR & Convit J (1994) Diagnosis of cutaneous leishmaniasis and species discrimination of parasite by PCR and hybridization. Journal of Clinical Microbiology 32, 2246 2252. Romero GAS, Sampaio RNR, Macêdo VO & Marsden PD (1999a) Sensitivity of vacuum aspiratory culture technique for diagnosis of localized cutaneous leishmaniasis in an endemic area of Leishmania (Viannia) braziliensis transmission. Memórias do Instituto Oswaldo Cruz 94, 505 508. Romero GAS, Sampaio RNR, Macêdo VO & Marsden PD (1999b) Sensitivity of lymph node aspiration in localized cutaneous leishmaniasis due to Leishmania (Viannia) braziliensis. Memórias do Instituto Oswaldo Cruz 94, 509 511. Romero GAS, Guerra MVF, Paes MG & Macêdo VO (2001a) Comparison of cutaneous leishmaniasis due to Leishmania (Viannia) braziliensis and L. (V.) guyanensis in Brazil: therapeutic response to meglumine antimoniate. The American Journal of Tropical Medicine and Hygiene 65, 456 465. Romero GAS, Guerra MVF, Paes MG & Macêdo VO (2001b) Comparison of cutaneous leishmaniasis due to Leishmania (Viannia) braziliensis and L. (V.) guyanensis in Brazil: clinical findings and diagnostic approach. Clinical Infectious Diseases 32, 1304 1312. Romero GAS, Ishikawa E, Cupolillo E et al. (2002) Identification of antigenically distinct populations of Leishmania (Viannia) guyanensis from Manaus, Brazil, using monoclonal antibodies. Acta Tropica 82, 25 29. Schönian G, Nasereddin A, Dinse N et al. (2003) PCR diagnosis and characterization of Leishmania in local and imported clinical samples. Diagnostic Microbiology and Infectious Diseases 47, 349 358. 1396 ª 2006 Blackwell Publishing Ltd

Shaw JJ (1994) Taxonomy of the genus Leishmania: present and future trends and their implications. Memórias do Instituto Oswaldo Cruz 89, 471 478. Silva NS, Viana AB, Cordeiro JA & Cavasini CE (1999) Leishmaniose tegumentar Americana no Estado do Acre, Brasil. Revista de Saúde Pública 33, 554 559. Silveira FT, Ishikawa EAY, De Souza AAA & Lainson R (2002) An outbreak of cutaneous leishmaniasis among soldiers in Belém, Pará State, Brazil, caused by Leishmania (Viannia) lindenbergi n. sp. A new leishmanial parasite of man in the Amazon region. Parasite 9, 43 50. Vergel C, Walker J & Saravia NG (2005) Amplification of human DNA by primers targeted Leishmania kinetoplast DNA and post-genome considerations in the detection of parasites by a polymerase chain reaction. The American Journal of Tropical Medicine and Hygiene 72, 423 429. Volpini AC, Passos VMA, Oliveira GC & Romanha A (2004) PCR-RFLP to identify Leishmania (Leishmania) amazonensis causing American cutaneous leishmaniasis. Acta Tropica 90, 31 37. Weigle KA, Dávalos M, Heredia P, Molineros R, Saravia NG & D Alessandro A (1987) Diagnosis of cutaneous and mucocutaneous leishmaniasis in Colombia: a comparison of seven methods. The American Journal of Tropical Medicine and Hygiene 36, 489 496. Corresponding Author Gustavo Adolfo Sierra Romero, Núcleo de Medicina Tropical, Universidade de Brasília, Campus Darcy Ribeiro, Asa Norte, CP 04517, Brasília, DF 70904-970, Brazil. Tel.: +55-61-3273-5008; Fax: +55-61-3273-2811; E-mail: gromero@unb.br La diversité des espèces responsables chez l homme de leishmanioses cutanées à Rio Branco, Brésil objectif Il y a peu d informations sur la diversité d espèces des Leishmania en Amazonie brésilienne de l ouest et sur la diversité des manifestations cliniques de la leishmaniose cutanée humaine. Nous décrivons dans cet article les résultats cliniques, les procédures diagnostiques et d identification des espèces de Leishmania rencontrées chez des patients de cette région. méthodes Notre échantillon de patients était composé de 50 patients, prospectivement évalués sur leurs caractéristiques épidémiologiques et cliniques au moyen d un questionnaire pré-établi. Des outils conventionnels et moléculaires ont été utilisés pour confirmer le diagnostic parasitologique et pour identifier les espèces responsables de la maladie. résultats Les patients étaient principalement de sexe masculin (76.5%) et vivaient dans une zone rurale. L âge moyen était 18 ans. L évolution médiane de la maladie était de 8 semaines. Les résultats des tests cutanés à la leishmanine, des examens directs de visualisation des formes amastigotes dans les sucs dermiques obtenus après scarification et la mise en culture des sérosités prélevées en bordure d ulcération étaient respectivement positives dans 98%, 52% et 34% des cas. Les tests moléculaires sur ADN extrait à partir de biopsies de peau étaient respectivement positifs chez les 50 patients à 100%, 82% et 44% par les techniques PCRmkDNA, PCR-RFLP ITS1rDNA et PCR G6P. 14 échantillons provenant de 13 patients ont été isolés et identifiés avec succès. Grâce aux techniques par MLEE, PCR-RFLP-ITS1rDNA et PCR-G6P, nous avons identifié les espèces de Leishmania responsables des manifestations cutanées pour 60% des patients examinés. Il s agissait de: 16 cas de L. (V.) braziliensis, 12 cas de L. (Viannia) lainsoni, 1 cas de L. (V.) guyanensis et 1 cas parent hybride de L. (V.) naiffi et de L. (V.) lainsoni. conclusion Les caractéristiques épidemiologiques et les manifestations cliniques des leishmanioses cutanées dans l état de l Acre au Brésil, sont semblables à celles des régions d Amazonie précédemment décrites. Toutefois au moins 3 espèces différentes circulent dans l état de l Acre. Dans cette région, la mise en œuvre de mesure de contrôle doit prendre en compte la diversité des vecteurs et des réservoirs impliqués. mots clés Leishmaniose cutanée, Leishmania (Viannia) braziliensis, Leishmania (Viannia) lainsoni, Leishmania (Viannia) naiffi, Amazone brésilienne occidentale, diversité génétique ª 2006 Blackwell Publishing Ltd 1397

Diversidad de especies causantes de leishmaniasis cutánea en Rio Branco, Brasil objetivo La información sobre la diversidad de especies de Leishmania en el Amazonas occidental de Brasil y del panorama clínico que la leishmaniasis cutánea humana causa es escasa. Describimos los hallazgos clínicos, los procedimientos de diagnóstico y la identificación de las especies de Leishmania en pacientes de esa región. métodos La muestra consistió en 50 pacientes, prospectivamente evaluados por características epidemiológicas y clínicas, por medio de un cuestionario estructurado. Se aplicaron herramientas moleculares y convencionales para confirmar el diagnóstico parasitológico e identificar las especies responsables de la enfermedad. resultados Los pacientes fueron predominantemente del sexo masculino (76.5%), y habitantes de zonas rurales. La mediana de edad fue 18 años y la mediana de evolución de la enfermedad fue de 8 semanas. En el procedimiento de diagnóstico para la prueba cutánea de leishmaniasis, la directa visualización de amastigotes en raspados dérmicos y cultivo de parásitos de aspiraciones del borde de las úlceras fueron positivos para 98%, 52% y 34% respectivamente. Los métodos moleculares aplicados para el ADN extraído de las biopsias de piel de los 50 pacientes produjeron 100%, 82% y 44% de positividad para PCR mkdna, PCR-RFLP ITS1rDNA y PCR G6P, respectivamente. 14 muestras de 13 pacientes fueron exitosamente aisladas e identificadas. MLEE, PCR-RFLP-ITS1rDNA y PCR-G6P permitió la identificación de las especies de Leishmania responsables por la etiología de ATL en 60% de los pacientes examinados: 16 L. (V.) braziliensis, 12 L. (Viannia) lainsoni, un L. (V.) guyanensis y un híbrido putativo de L. (V.) naiffi y L. (V.) lainsoni. conclusiones El comportamiento clínico y epidemiológico de la leishmaniasis cutánea en Acre, Brasil, es similar a otros escenarios amazónicos previamente descriptos. No obstante, la complejidad de la diversidad parasitaria en Acre puede contribuir a la circulación concomitante de por lo menos tres especies distintivas de Leishmania. La implementación de intervenciones de control en las áreas estudiadas debe tomar en consideración la posibilidad de varios flebotominos vectores y reservorios. palabras clave leishmaniasis cutánea, Leishmania (Viannia) braziliensis, Leishmania (viannia) lainsoni, Leishmania (Viannia) naiffi, Amazonas occidental brasilero, diversidad genética 1398 ª 2006 Blackwell Publishing Ltd