To begin examining diversity at the interspecific level, the DNA segment containing ITS 1 and 2, as well as the 5.8 S RNA gene, was amplified by PCR using as template DNA from 17 L. amazonensis strains isolated from patients with different clinical manifestations of leishmaniasis, DNA from a reference L. amazonensis strain (M2269) and DNA from a L. major strain (MHOM/81/IL/Friedlin), which was used as an out-group (Table 1). Except for strain BA115 which was, therefore, omitted from subsequent analyses, PCR amplification yielded a product of ~ 1.2 kb for all strains studied (data not shown). Sequencing of the amplification product revealed 1131 bp orthologous sites and saturation analyses showed no severe saturation, all positions were informative (data not shown). The data set was then subjected to the probability ratio test (LRT) approach. The model that best fit the pattern of nucleotide substitution for our dataset was the HKY85 model 20 plus gamma distribution of α = 0.3641 (indicating heterogeneity of the sites' substitution rates over time). Considering these findings, we performed a phylogenetic signal analysis and found 25.6% of unresolved trees in the likelihood mapping method. As seen in Fig. 1, the tree topology obtained using the HKY model plus gamma distribution revealed two internal nodes, with 100% bootstrap support, in which strains BA32, BA112 and BA75 are clustered. We can only speculate that the clade containing BA32, BA112 and BA75 could be a sister group of the clade containing the BA69, BA73, BA88, and BA137 strains since the boostrap value (58%) was not meaningful.

All VL isolates (BA137, BA32 and BA112) were clustered in these two clades, with the exception of BA109. The remaining tree branches, however, were found to be unresolved and some strains, such as BA109, BA125 and BA276, were even found to be identical at this locus such. Interestingly, BA106, a DCL case documented outside Maranhão State, where this disease is endemic, remained in a separate node with bootstrap support of 72%. This strain was isolated in Bahia State, in 1986, whereas the other two DCL strains studied (BA199 and BA276) were isolated from patients from Maranhão state, in 1989.

We then compared the topology of phylogenetic tree obtained by ITS sequencing with the dendrograms generated by RAPD and SSR-PCR. Branches observed in the resulting RAPD or SSR-PCR dendrograms were not strongly supported by bootstrap analysis (data not shown). A combined analysis of RAPD and SSR-PCR results is shown in Fig. 2. The grouping of BA109 and BA112 (both obtained from VL patients) as well as the grouping of strains BA113 and BA114 (both from ML patients) were supported by significant bootstrap values (91 and 98%, respectively). In this combined analysis, BA106 (obtained from a DCL strain) also remained in a separate branch, as observed in the individuals RAPD and SSR-PCR dendrograms (data not shown). Overall, either by ITS, RAPD or SSR-PCR, we did not detect any association between genetic polymorphism and geographic origin.

In order to investigate the degree of chromosome size polymorphism, chrosomomes from L. amazonensis BA106, BA109, BA113, BA125, BA276 which are representative of the leishmaniasis clinical manifestation spectrum, L. amazonensis M2269, L. braziliensis strain (MHOM/BR/94/H3227) and L. major Friedlin were separated by PFGE and hybridized to specific chromosome markers. Fig. 3 shows the hybridization patterns observed for each marker used. The hybridization pattern indicates a wide variation in terms of chromosome size for the L. amazonensis strains studied. Hybridization with a gene present in both chromosomes 1 and 2 (Fig. 3A) shows variation in terms of chromosome size for the strains tested and, importantly, we observed size polymorphisms between chromosome homologues. A similar result was observed by hybridization with the spliced leader RNA gene present in chromosome 2 (Fig. 3B). In contrast, size polymorphism of chromosome 6, identified by hybridization with the dihydrofolate reductase-thymidulate-synthase (DHFR-TS) gene (Fig. 3C) was less evident among the strains tested. The genomic clone used to identify chromosome 14 (Fig. 3D) has reiterated elements, which are present in other three larger chromosomes of L. major, therefore, it allowed the comparison of 4 chromosomes at a time. The results indicate a narrow size variation of these chromosomes within the L. amazonensis strains tested.

A total of 18 L. amazonensis strains were used in this study (Table 1). Strains were obtained between 1985 and 1990 by aspiration of lesions from cutaneous, mucosal, diffuse and post-Kalazar dermal leishmaniasis or by bone marrow aspiration of visceral leishmaniasis patients, each isolate originated from a different human subject 1. Strains have been maintained in cryopreservation tanks since the time of isolation in culture. For the purpose of this study, isolates were defrosted and cultured until a parasite density of 10⁷/ml was obtained. Species identification was performed by PCR 40 and by ELISA using monoclonal antibodies 41. For genomic DNA extraction, stocks were grown in 199 medium supplemented with 10% heat inactivated fetal bovine serum (both from Invitrogen), at 26°C. Parasites (10⁷/ml) were harvested by centrifugation and genomic DNA was purified using Wizard^® Genomic DNA Purification Kit (Promega). DNA was ressuspedend in 50 μl TE buffer (10 mM Tris, 1 mM EDTA, pH 8.0) and stored at -20°C until used. DNA concentration was estimated by spectrophometry by reading the absorbance at 260 nm. Sample purity was checked by agarose gel electrophoresis and ethidium bromide staining.

The segment containing the ITS 1, 2 and the 5.8 S region, within the rRNA gene locus, were amplified by PCR using primers IR1 (5'-GCTGTAGGTGAACCTGCAGCAGCTGGATCATT-3'), IR2 (5'-GCGGGTAGTCCTGCCAAACACTCAGGTCTG-3') 15 and an internal primer (5'-CGGCGCATGGGAGAAGCT3-3'). PCR products were purified with the CONCERT rapid PCR purification system (Gibco BRL), following the manufacturer's instructions. Amplified PCR products were sequenced in both strands in an ABI377^® system using the BigDye Terminator Cycle Sequencing Kit^® (Applied Biosystems). Quality analysis and assembling of the sequences were performed using Phred/Phrap/Consed^® package 42. Consensus sequences with Phred quality > 15 were locally aligned with sequences submitted to GenBank using BlastN program 43. Consensus sequences were submitted to GenBank and are available under the accession numbers DQ300179–DQ300195. Sequence alignment was performed manually using the BIOEDIT software (v5.0.9) 44 and for all sequences, alignment was unambiguous. The mutational saturation was examined using PAUP* (v4.0b10) 45. The nucleotide substitution model and the gamma distribution were obtained by the likelihood ratio test (LRT) 46, employing the MODELTEST software (v3.0b6) 47. The phylogenetic signal was tested employing the TREE-PUZZLE (v 5.3) software 48, using the likelihood mapping method 49. Maximum likelihood (ML) method was then performed in PAUP* (v4.0b10), with a starting tree obtained via Neighbor-Joining algorithm. The reliability of internal branches was assessed using the bootstrap method with 1000 pseudo-replicates. Tree topology was visualized using the TREEVIEW program (v 1.6.2) 50. The homologous sequence of L. major Friedlin strain (MHOM/IL/81/FRIEDLIN) was adopted as the outgroup.

RAPD amplification was done as previously described 13, employing one the following primers: 3302 (5'-CTGATGCTAC-3'); 3303 (5'-TCACGATGCA-3') and 3304 (5'-GCACTGTCA-3') (all from Invitrogen). Amplification was performed under the following conditions: an initial denaturation step at 95°C for 5 min, two cycles with annealing at 30°C for 2 min, extension at 72°C for 1 min and denaturation at 95°C for 30 s, followed by 33 cycles in which the annealing temperature was increased to 40°C. The final extension was carried for 5 min at 72°C. SSR-PCR amplification was performed using two primers: [5'-(CA)₈RY-3'], which amplifies inter-repeat segments 10 and [5'-CAA(CT)₆ -3'], which amplifies inter-repeat segments as well as the repeat segment 51. Amplification reactions were performed under the following conditions: 26 cycles of denaturation at 94°C for 30 s, annealing at 50°C for 45 s for the [5'-CAA(CT)_6-3'] primer or at 57°C for 45 s for the [5'-(CA)₈RY-3'] primer and extension at 72°C for 7 min. All reactions were performed in triplicate. The amplification patterns were visualized after electrophoresis in a 1.5% ethidium bromide-stained agarose gel.

Amplicons were scored by eye and well-defined, reproducible bands (generated in at least three amplification reactions) were used to construct a binary table. Each amplification band was numbered and scored as present (1) or absent (0) and genetic similarity among strains was calculated using Jaccard's similarity coefficient. The relationships between strains were estimated through dendrograms representing RAPD and SSR-PCR data. Dendrograms were constructed based on a matrix of genetic distance through UPGMA 52 using FreeTree (v. 0.9.1.50) 53 and the bootstrap option was used to run 1000 replicates to obtain confidence estimates for the groupings.

Chromosomal DNA was prepared in low-melting temperature agarose plugs 54. Chromosomes were separated by PFGE run on a contour-clamped homogeneous electric field apparatus (CHEF DR II, BioRad). Gels were run at 14°C, at 6.0 V cm^-1 for 28 h with a 35.4- to 73.6-s pulse ramp time. Electrophoresed DNAs were transferred to nylon membranes (GeneScreen Plus).

Four probes were used in this study in order to explore size-polymorphisms of some chromosomes (1, 2, 6 and 14). These were (i) a L. major oligopeptidase gene, present in the extremities of chromosomes 1, 2 and 27 of L. major; (ii) a L. major spliced leader RNA gene, present in chromosome 2; (iii) a L. major dihydrofolate reductase-thymidulate-synthase (DHRF-TS) present in chromosome 6 and (iv) a L. major genomic fragment from chromosome 14 (cosmid 10E07), which has internal elements repeated in three other larger chromosomes of L. major (A.K. Cruz, unpublished results). All probes were labeled with [α-³²P]dCTP by random priming. Hybridization and washing conditions were performed as described 37 and membranes were exposed at -70°C with an intensifying screen.