Molecular Identification of Parasites Causing Cutaneous Leishmaniasis in Panama

ABSTRACT

Isolates from 475 cutaneous leishmaniasis (CL) patients from three endemic regions were studied by three typing techniques. The molecular analysis from lesion scrapings based on hsp70 PCR-restriction fragment length polymorphism (RFLP) showed that 78.1% (371/475) restriction patterns corresponded to Leishmania (Viannia) panamensis, 19% (90/475) to Leishmania (Viannia) guyanensis, and 3.0% (14/475) to Leishmania (Viannia) braziliensis. Promastigotes isolated by culture from lesions of 228 patients (48.0%, 228/475) were identified by multi-locus enzyme electrophoresis. Of them, 95.2% (217/228) were typified as L. (V.) panamensis, 1.3% (3/228) as L. (V.) guyanensis, 2.2% (5/228) as L. (V.) braziliensis, and 1.3% (3/228) as hybrids (L. [V.] braziliensis/L. [V.] panamensis). However, a partial sequencing analysis of the hsp70 gene from 77 selected samples showed 16.9% (13/77) typified as L. (V.) panamensis, 68.8% (53/77) as Leishmania (V.) sp., 1, 3.9% (3/77) as L. (V.) guyanensis, 1.3% (1/77) as L. (V.) braziliensis outlier, 2.6% (2/77) as Leishmania (Viannia) naiffi, 2.6% as (2/77) Leishmania (V.) sp., and 2 and 3.9% (3/77) hybrid isolates of L. (V.) braziliensis/L. (V.) guyanensis. These results confirm L. (V.) panamensis as the predominant species and cause of CL lesions in Panama and that L. (V.) guyanensis, L. (V.) braziliensis, and L. (V.) naiffi are circulating to a lower degree. Furthermore, the determination of parasite isolates belonging to atypical clusters and hybrid isolates suggests the circulation of genetic variants with important implications for the epidemiology and clinical follow-up of CL in Panama. No evidence of the existence of parasites of the Leishmania (Leishmania) mexicana complex in Panamanian territory was found in this study.

INTRODUCTION

Worldwide, more than 350 million people are considered at risk of contracting leishmaniasis, and some two million new cases occur yearly.¹ Clinical features of this disease can vary significantly, reflecting the wide range of parasites in the genus Leishmania capable of infecting humans,^2–5 the host immune response, and factors found in the saliva of the sand fly vector.^6,7

Cutaneous leishmaniasis (CL) is the most common clinical form causing ulcers on exposed parts of the body and leaving scars in most cases. Moreover, some Leishmania species can disseminate from the cutaneous lesions to the nasopharyngeal mucosa and cause mucocutaneous leishmaniasis that can partially or totally destroy the mucous membranes of the nose, mouth, and throat cavities and surrounding tissues, causing serious disability.⁸

Cutaneous leishmaniasis is a worldwide neglected disease that is an emerging public health problem in Panama. There are 1,000–3,000 cases per year, but the health authorities consider that it could be 50% more.⁹ By 2011, the Pan American Health Organization (PAHO) reported that Panama was the country with the higher incidence of CL per 100,000 inhabitants in America.¹⁰ Panama, Bocas del Toro, and Cocle are the western provinces in Panama with the higher reported number of CL cases¹¹ (Figure 1). Within these provinces, the transmission is concentrated in forested and rural areas, mainly among the marginalized population. In Panama, CL is a neglected disease, affecting the poorest, a situation similar to other endemic countries. This leads to an increase in morbidity, and inequity in a country that has already one of the highest levels of inequality in the region.¹²

Figure 1.

Map of Panama showing location of the study sites and the cutaneous leishmaniasis transmission intensity based on the median number of confirmed cases (2015–2017). Map was created using ArcGIS Desktop 10.6.1 software (ArcGIS Desktop^© 10.6.1 software, 2018, Redlands, CA). Data were obtained from information reported by the Panamanian Ministry of Health-National Leishmaniasis Surveillance Services to the PAHO as published in PAHO, leishmaniasis, epidemiological report in the Americas, Washington, PAHO, 2019. Available at: www.paho.org/leishmaniasis. PAHO = Pan American Health Organization. This figure appears in color at www.ajtmh.org.

Leishmania (Viannia) panamensis, which has been considered a local variant of Leishmania (Viannia) guyanensis,¹³ is referred as the most widespread and main etiologic agent for most of the human cases in the country.^14–16 Only a few sporadic human cases of CL caused by Leishmania (Viannia) braziliensis, Leishmania (Leishmania) amazonensis, and Leishmania (Leishmania) mexicana were reported decades ago using isoenzymic methods for species characterization.^15,17–19 Other species such as Porcisia hertigi, originally identified as Leishmania hertigi,²⁰ and Leishmania (Leishmania) aristidesi were detected in few wild mammals in 1965 and 1968, respectively.^21–23 Also, Endotrypanum colombiensis, originally identified as Leishmania colombiensis,²⁰ was isolated from four sand flies and a sloth between 1980 and 1986.²⁴ More recently, Leishmania (Viannia) naiffi was identified in sand flies collected in Barro Colorado Island located in Gatún Lake Panama Canal and also in sand flies in a hyperendemic CL community in the district of Capira, province of Panama Oeste.^25,26 Although the gold standard for diagnosis is still parasite visualization (smears/culture), identification of the species causing the infection is critical in this era of species-guided treatment recommendations from the WHO/PAHO.²⁷ Likewise, characterization of the species can provide important complementary information about disease progression. In this regard, a significant association between the infecting Leishmania species and treatment outcome has been described.^28–30 Furthermore, identification of the exact parasite species infecting humans and animal host, as well as sand fly vectors, also gains increasing importance for epidemiological surveillance and CL control.^20,31–33

The Panama isthmus is the geographical link between Central and South America, providing a biologic corridor between Colombia and Costa Rica, that increases the risk for the entrance of pathogens, vectors, and reservoirs from Central and South America. Studies conducted in neighboring countries have reported human leishmaniasis caused by species not detected previously or of uncertain endemicity in Panama, as L. (L.) infantum, L. (V.) guyanensis, E. colombiensis, and L. (V.) braziliensis.^24,34,35 However, for a better understanding of the identity and genetic characteristics of the causative agents of CL in Panama, thorough studies are necessary, especially in regions of high prevalence and using an adequate number of biological samples. To investigate the Leishmania species diversity causing human leishmaniasis in Panama, we typed parasites found in 475 human CL cases from different endemic areas, by hsp70 PCR-RFLP, multi-locus enzyme electrophoresis (MLEE), and hsp70 sequencing analysis. The results suggest that although L. (V.) panamensis is present in most of the evaluated cases, other Leishmania (Viannia) species and genetic variants causing localized CL are also circulating in Panama.

MATERIALS AND METHODS

Study sites and samples.

The species of Leishmania present in cutaneous lesions from CL Panamanian patients were investigated. Study patients were part of a phase 3 clinical trial of a topical cream for the treatment of CL conducted in Panama, sponsored by the Surgeon General, Department of the Army, USA.³⁶ The analyzed samples corresponded to 475 positive DNA lesion samples (kinetoplast DNA [KDNA] PCR/hsp70 PCR) and 228 promastigote-positive cultures from these patients. During the clinical trial, all lesion samples were initially evaluated by parasitological diagnostic procedures (smear/culture) and a PCR test that amplifies a specific sequence of Leishmania (Viannia) sp. KDNA³⁷ (Supplemental Information 1). Lesion samples were collected between April 2013 and August 2015 from patients who attended one of the three study sites: Clinical Investigation Unit at the Gorgas Memorial Institute (ICGES) in Panama City, Panama Province (8°58′14″N, 79°32′1″W); Hospital Aquilino Tejeira in Penonome, Cocle Province (8°30′48″N, 80°21′3″W); and Health Center Materno Infantil Sandra Hernández in Changuinola (9°22′60″N, 82°31′60″W), Bocas del Toro Province (Figure 1).

Ethics.

As previously reported,³⁶ the study and use of clinical samples were reviewed and approved by the National Review Board (Comité de Bioética de la Investigación, ICGES Panama City, Panama, assigned code PEI 50098, S-12-21), and by the Human Research Protections Office, U.S. Army Medical Research and Material Command. Written informed consent was obtained from all study participants and/or guardians before enrollment. Minors also provided assent to participate. All subjects gave permission to type and further use Leishmania parasites associated with their lesions.

Reference strains.

Leishmania (Viannia) reference strains used in this study as controls for hsp70-RFLP and phylogenetic analysis were as follows: L. (V.) panamensis (MHOM/PA/1998/WR2306), L. (V.) guyanensis (MHOM/BR/1975/M4147), L. (V.) braziliensis (MHOM/BR/1975/M2903), and Leishmania (Viannia) peruviana (MHOM/PE/2005/WR2771).

Molecular characterization methods.

DNA extraction.

DNA extraction of the skin scrapings in TE buffer was performed using the Qiagen QIAamp^® DNA Blood Mini Kit according to the manufacturer’s instructions (Qiagen, Hilden, Germany). Leishmania DNA from culture-isolated parasites was purified using Wizard™ Genomic DNA Purification Kit (Promega, Madison, WI) according to the manufacturer’s instructions.

PCR-hsp70 detection of Leishmania genus.

PCR was performed with oligonucleotides F25 and R1310 that amplify a 1,286-bp product from the repeated gene heat shock protein 70 (hsp70) as previously described.^38,39 Amplification reactions were performed in a final volume of 50 μL containing 25 μL of Go Taq Green Master Mix 2X (Promega), 0.6 μmol/L of each primer, 5 μL of DNA of clinical samples, and 1 ng of reference strains. Thermal cycling was performed in an Applied Biosystems^® (Foster City, CA) 2,720 Thermal Cycler. When negative results or results with faint agarose gel band of amplification products were obtained, re-amplification was performed using the same conditions and primers mentioned earlier with 1–5 μL of amplified product to improve the sensitivity of the PCR or to have enough product for endonuclease digestion.

Hsp70-RFLP analysis.

Leishmania hsp70-RFLP characterization was performed as described elsewhere.³⁸ In brief, an initial amplification was performed with the aforementioned hsp70 primers. The resulted amplicons were digested with Hae III and subsequently with BccI or RsaI endonucleases. Obtained restriction patterns were analyzed by electrophoresis and compared with reference strain patterns.

Multi-locus enzyme electrophoresis (MLEE).

Biochemical characterization by isoenzyme electrophoresis was performed in the Department of Leishmania Diagnosis at the Walter Reed Institute of Research, USA. In brief, parasite lysates were prepared from logarithmic-phase bulk cultures of isolated promastigotes, and their soluble enzymes were extracted. The isoenzymes were separated after electrophoresis on cellulose acetate as previously described.^40,41 For each patient sample, the electrophoretic mobility banding pattern was compared with standard patterns of known Leishmania species (WHO reference strains). The following Leishmania enzymes were assessed: phosphogluconate dehydrogenase, mannose phosphate isomerase, glucose phosphate isomerase, and peptidase D. When necessary, additional enzymes (glucose-6-phosphate dehydrogenase, malic enzyme, aspartate aminotransferase, alanine aminotransferase, and phosphoglucomutase) were used to confirm the identifications made with initial enzymes. In addition, isolates that could not be identified with the reference strains were further analyzed using other enzymes, including acid phosphatase, fumarate hydratase, glutamic oxaloacetic transaminase, hexokinase, isocitrate dehydrogenase, lactate dehydrogenase, malate dehydrogenase, glutathione reductase 1 (GSR1), glutathione reductase 2 (GSR2), 6-phosphofructokinase, and pyruvate kinase.

Hsp70 sequencing and phylogenetic analysis.

The hsp70-amplified products were submitted to electrophoresis in 1.5% agarose gels in 1X TBE (89 mM Tris borate, 2 mM ethylenediaminetetraacetic acid [EDTA], pH 8.3). Product bands were excised from agarose gel and purified using the Qiaquick Gel Extraction Kit (Qiagen) following the manufacturer’s instructions. DNA sequencing of both strands was carried out using primers F25, R1310, 6F, and R617^38,39; and BigDye Terminator 3.1 Cycle Sequencing Kit (Applied Biosystems). Primers and deoxynucleotide triphosphates were removed using Xterminator Kit (Applied Biosystems). The clean sequencing reaction was run through an ABI 3130x sequencer (Foster City, CA). The chromatograms were edited by the assembling-to-reference tool of UGENE tool kit⁴² (Novosibirsk, Russia) using a trimming quality valor of 35. Sequences were then multiple-aligned using MAFFT software MAFFT (Multiple Alignment using Fast Fourier Transform, Philadelphia, PA) also included in the bioinformatic UGENE tool kit with a maximum number of iterative refinement of three and a gap penalty of 1.53. The generalised time-reversible model [GTR] (GTR + I + G) model was found as the best DNA evolution model by the program JModelTest 2 (Coruña, Spain).⁴³ A phylogenetic tree reconstruction of Leishmania was implemented applying Bayesian inference with the Mr. Bayes v. 3.2 program (Bayesian inference of phylogeny, Rochester, NY). Ten Markov chains were proceeded for eight millions of generations, and trees were sampled for every 1,000 generations. Twenty-five percent of the sample tree were discarded, and the remaining trees were used to build up a consensus tree and calculation of posterior probabilities of clades. The results of Bayesian analyses were visualized using Figtree v. 1.4.2 (Edinburgh, Scotland). The hsp70 gene sequence Trypanozoma cruzi was used as root. Transformation of the leaves and schematic representation of the root were applied for visualization purposes.

The p distance among group of sequences was assessed using the software MegaX⁴⁴ (Philadelphia, PA) set up to perform a bootstrap analysis of 1,000 replicates to estimate variation, including a Gamma correction of inter-site variation using four categories as parameter and also set up to delete gaps as missing treatment. Both transitions and transversions were included in the calculation of P-distances.

RESULTS

Hsp70-RFLP typing.

Four hundred seventy-five PCR products of hsp70 gene were digested with Hae III and subsequently with BccI or RsaI endonucleases. After comparing the digestion patterns with the ones observed in reference strains, 371 (78.1%) showed a restriction pattern coincident with L. (V.) panamensis, 90 (18.9%) with L. (V.) guyanensis, and 14 (3.0%) with L. (V.) braziliensis. Table 1 shows the hsp70-RFLP typing results obtained in each study site.

Table 1.

Characterization of Leishmania species by hsp70-RFLP analysis of skin scrapings from Panamanian cutaneous leishmaniasis patients

Site of study	Typed samples	hsp70 PCR-RFLP*
		BccI pattern		RsaI pattern
		Leishmania panamensis	L. guyanensis	L. braziliensis
Panamá (site 1)	266	210 (78.9%)	46 (17.3%)	10 (3.8%)
Coclé (site 2)	161	115 (71.4%)	42 (26.1%)	4 (2.5%)
Bocas del Toro (site 3)	48	46 (95.8%)	2 (4.2%)	–
Total	475	371 (78.1%)	90 (18.9%)	14 (3.0%)

L. braziliensis = Leishmania braziliensis; L. guyanensis = Leishmania guyanensis.

^*After hsp 70 gene sequencing (1,245 bp), most samples typed as L. guyanensis or L. braziliensis regrouped into minor clusters near to reference strains.

Multi-locus enzyme electrophoresis (MLEE).

Two hundred twenty-eight positive cultures were typed by MLEE and compared with hsp70-RFLP typing results (Table 2). The isolates characterized as L. (V.) panamensis by MLEE (n = 202) showed a perfect agreement with hsp70-RFLP typing results. Three isolates characterized as L. (V.) guyanensis by hsp70-RFLP coincided with the MLEE typing result. However, 15 isolates considered L. (V.) guyanensis by hsp70-RFLP were characterized by MLEE as L. (V.) panamensis. Five isolates characterized by hsp70-RFLP as L. (V.) braziliensis were also confirmed by MLEE. In addition, three samples identified as L. (V.) braziliensis by hsp70-RFLP were assigned as L. (V.) braziliensis/L. (V.) panamensis hybrid isolates by MLEE.

Table 2.

Discrimination of Leishmania Viannia species from Panamanian cutaneous leishmaniasis patients by hsp70PRC-RFLP, MLEE, and sequencing

Leishmania Viannia species	hsp70 PCR-RFLP (n = 228)	MLEE (n = 228)	Sequencing (n = 77)
L. panamensis	202	202	13
L. guyanensis	15	15 (L. panamensis)	15 + 38* (Leishmania sp. 1)
L. guyanensis	3	3	3
L. braziliensis	5	5	1 (Leishmania braziliensis outlier)
			2 (Leishmania naiffi)
			2 (Leishmania sp. 2)
L. braziliensis	3	3 (L. braziliensis/L. panamensis)	3 (L. braziliensis/L. guyanensis)

L. (V.) braziliensis = Leishmania (Viannia) braziliensis; L. (V.) guyanensis = Leishmania (Viannia) guyanensis; L. (V.) panamensis = Leishmania (Viannia) panamensis; MLEE = Multi-locus enzyme electrophoresis. Hsp70PCR-RFLP: PCR products were digested with Hae III and subsequently with BccI or RsaI endonucleases. Sequencing: hsp70 PCR-amplified product sequencing of both strands using primers F25, R1310, 6F and R617, followed by phylogenetic analysis. Results that do not match hsp70 RFLP analysis are shown in parentheses.

^*Besides 15 cultured samples, 38 DNA samples from skin scraping found as L. guyanensis by hsp70 PCR-RFLP were sequenced.

Hsp70 sequencing and phylogenetic analysis.

To confirm the hsp70-RFLP and MLEE typing results, we further sequenced 1,245 bp of the hsp70 gene from 39 isolated Leishmania parasites (positive cultures): 13 samples typed as L. (V.) panamensis by hsp70-RFLP and 26 samples typed as non-panamensis Leishmania by hsp70-RFLP (18 typed as L. [V.] guyanensis and eight as L. [V.] braziliensis). In addition, we sequenced the following reference strains: L. (V.) panamensis (WR 2306), L. (V.) guyanensis (M4147), L. (V.) braziliensis (M2903), and L. (V.) peruviana (WR 2771).

The sequences from the 13 samples that were typed as L. (V.) panamensis by hsp70-RFLP clearly clustered with reference sequences belonging to L. (V.) panamensis (Figure 2). Of the 18 samples that were typed as L. (V.) guyanensis by hsp70-RFLP, three grouped in the cluster of L. (V.) guyanensis reference strains. The remainder subset of 15 samples (here named as Leishmania [V.] sp. 1) clustered together in a distinct group, but close to L. (V.) guyanensis and Leishmania (Viannia) shawi reference strains (Figure 2). In addition, 38 lesion scraping samples identified as L. (V.) guyanensis by hsp70-RFLP were also sequenced. All of them clustered with the previously found Leishmania (V.) sp. 1 group.

Figure 2.

Hsp70-based Bayesian phylogenetic tree of Leishmania species obtained constructed under the GTR (Nst = 6) model with a gamma rate of four. The Bayesian consensus tree was searched by the program Mr. Bayes after 8,000,000 cycles of the Markov Chain Monte Carlo algorithm with a subsampling frequency of 1,000. Clade credibility values are shown as values at each clade node. Reference sequence codes appear in black color. All sequence codes obtained in this study are highlighted in red color. Reference sequences are abbreviated as follows: bra: Leishmania braziliensis; braO: Leishmania braziliensis outlier; guy: Leishmania guyanensis; lai: Leishmania lainsoni; nai: Leishmania naiffi; pan: Leishmania panamensis; per: Leishmania peruviana. This figure appears in color at www.ajtmh.org.

Sequencing results from the eight samples that were typed as L. (V.) braziliensis by hsp70-RFLP were as follows: one was grouped with L. (V.) braziliensis outlier reference strains, two clustered in the L. (V.) naiffi reference strains, and two samples were grouped together in a separate group (here named as Leishmania [V.] sp. 2), independent of the L. (V.) braziliensis/L. (V.) peruviana and L. (V.) braziliensis outlier reference sequences (Figure 2). Finally, the other three isolates identified as L. (V.) braziliensis by hsp70-RFLP and as hybrids by MLEE showed ambiguous nucleotides in their hsp70 sequences that suggest a possible event of genetic exchange between parental strains L. (V.) braziliensis and L. (V.) guyanensis. The sequences of these hybrid isolates were not included in the phylogenetic analysis presented in Figure 2. The different parasites found in the three regions, based on hsp70 sequence analysis, are presented in Table 3.

Table 3.

Characterization of Leishmania (Viannia) species by hsp70 gene sequencing from Panamanian cutaneous leishmaniasis patients

	Typed samples	hsp 70 gene sequencing (1,245 bp)*
Site of study		Leishmania (Viannia) panamensis	L. (V.) guyanensis	L. (V.) braziliensis outlier	Leishmania (Viannia) naiffi	L. (V.) braziliensis/guyanensis	L. (V.) sp. 1	L. (V.) sp. 2
Panamá (site 1)	51	13	2†	1	1	2	31	1
Coclé (site 2)	23	–	1	–	1	1	20	1
Bocas del Toro (site 3)	2	–	–	–	–	–	2	–
Total	77	13	3	1	2	3	53	2

L. (V.) braziliensis = Leishmania (Viannia) braziliensis; L. (V.) guyanensis = Leishmania (Viannia) guyanensis.

^*Hsp70 PCR-amplified products sequencing of both strands using primers F25, R1310, 6F, and R617, followed by phylogenetic analysis.

^†One of these two cases was found to be imported from French Guyana.

Leishmania hsp70 sequences obtained in this work (N = 77) were deposited in GenBank-National Center for Biotechnology Information (accession numbers can be found in Supplementary Information 2). For the phylogenetic analysis, we also re-sequenced hsp70 gene of cultured reference strains, and in addition, we retrieved 92 Leishmania sp. hsp70 reference sequences from GenBank. The final alignment contained 168 sequences including 77 sequences obtained in this study. The Bayesian phylogenetic tree is shown in Figure 2.

DISCUSSION

In this study, we characterized the Leishmania parasites present in 475 CL Panamanian patients from different geographical areas. The findings confirm that L. (V.) panamensis is the species most often associated (78.1%) with CL cases from the evaluated endemic areas in Panama. However, we also described, based on hsp70 gene partial sequencing and MLEE, a sympatry of L. (V.) panamensis with L. (V.) braziliensis, L. (V.) naiffi, L. (V.) guyanensis, hybrid isolates, and two genetic variants (named here as Leishmania (V.) sp. 1 and Leishmania (V.) sp. 2) (Supplemental Information 2). No evidence of the existence of parasites of the L. (L.) mexicana complex was found.

The Bayesian phylogenetic analysis based on hsp70 gene revealed that 53 samples analyzed belonged to Leishmania (V.) sp. 1, a cluster close to L. (V.) guyanensis and L. (V.) shawi and appeared to be closely related to the last species (P = 0.033 and P = 0.028, respectively). On the other hand, two samples analyzed in this study grouped together in a new cluster denominated Leishmania (V.) sp. 2 near the group of L. (V.) braziliensis complex in the cladogram (Figure 2). It seems that these two samples belong to the L. (V.) braziliensis complex as suggested by the P distance between groups (P = 0.0012). Studies using other markers are ongoing to reveal the status of Leishmania (V.) sp. 1 and Leishmania (V.) sp. 2 groups in the L. (V.) guyanensis and L. (V.) braziliensis complexes, respectively.

The Leishmania (V.) sp. 1 genetic variant was detected in the three study sites, whereas samples of Leishmania (V.) sp. 2 were detected in sites 1 and 2 (Table 3). This finding, however, may be a sampling bias as the number of samples in site 3 was comparatively smaller. In general, it seems that the distribution of Leishmania (V.) sp. 1 and Leishmania (V.) sp. 2 is not restricted to particular endemic regions of Panama. However, the potential link between the eco-epidemiological characteristics of an endemic area and the frequency of CL cases induced by these genetic variants needs to be more adequately defined.

On the other hand, one sample characterized by hsp70-RFLP as L. (V.) braziliensis and by hsp70 sequencing as L. (V.) braziliensis “outlier” was collected in site 1 of the study. Epidemiological data suggest that this patient came from the region of Panama Oeste Province (near Panama City), confirming that currently L. (V.) braziliensis is infecting humans in Panamanian territory. In a previous 1990 study, L. (V.) braziliensis was reported near the border with Colombia, but the real origin of this sample was not totally defined.⁴⁵ The L. (V.) braziliensis “outlier” cluster was originally named because in 2012, some isolates were identified as L. (V.) braziliensis by MLEE, but using molecular analyses, they clustered separately from the main L. (V.) braziliensis–L. (V.) peruviana clade.⁴⁶ Also, it is known that parasites belonging to L. (V.) braziliensis “outlier” group can produce mucocutaneous lesions.³⁹ The genetic diversity of L. (V.) braziliensis has been reported in isolates even from very close geographical areas.⁴⁷ Thus, determining the degree of genetic variability presented by this species in Panamanian endemic regions will improve our knowledge about the epidemiology of CL in this country.

Another important typing result from this study were two samples identified initially by hsp70-RFLP as L. (V.) braziliensis, which after hsp70 sequencing were finally confirmed as L. (V.) naiffi. In this sense, it has been reported that the Rsa I enzyme used in our hsp70-RFLP approach is unable to distinguish between L. (V.) braziliensis and L. (V.) naiffi.³⁸ However, the nucleotide sequence of the 1,245-bp hsp70 amplified product performed during the present study ruled out the possibility of misidentification of this species. It is also important to mention that a recent study using hsp70-RFLP describes that L. (V.) naiffi isolates can present polymorphisms that must be taken into consideration.⁴⁸ The epidemiological data revealed that one of these two patients was infected in Darién Province near the Colombian border, whereas the other came from Cocle Province (Table 3). It should be emphasized that L. (V.) naiffi has been previously detected in sand flies in Barro Colorado Island located in Gatún Lake Panama Canal and in a hyperendemic CL community in the district of Capira, province of Panama Oeste.^25,26 Cutaneous leishmaniasis cases caused by L. (V.) naiffi evolves with a benign clinical course, and there is currently no association observed between L. (V.) naiffi and mucosal leishmaniasis.⁴⁹ However, therapeutic failure of leishmaniasis caused by L. (V.) naiffi has been reported.⁴⁹ This species had not been reported as a CL etiological agent previously in Panama; nevertheless, it has been frequently described in Brazil, French Guyana, Surinam, and Ecuador.^50–54

We also identified by hsp70-RFLP and sequencing three cases of CL caused by L. (V.) guyanensis in Panamanian patients. According to the epidemiological data, one of these patients acquired the infection in French Guyana where the infection with L. (V.) guyanensis is naturally endemic. This is the first report of an imported case of CL caused by L. (V.) guyanensis in Panama. The other two patients acquired the infection in Panama and Cocle regions (Table 3). In South America, L. (V.) guyanensis has been reported as responsible of CL in Colombia, Brazil, French Guyana, Surinam, Ecuador, Peru, and Venezuela^52,55,56 transmitted by Nyssomyia umbratilis and Nyssomyia anduzei.⁵⁴ The first of these vector species has not been found yet in Panama, but the second one was reported since 1972.⁵⁷ Nevertheless, it is possible that other phlebomine sand fly species, from the about 76 species described in Panama,⁵⁸ are able to transmit L. (V.) guyanensis. Early reports mention that L. (V.) guyanensis is eliminated with difficulty by pentavalent antimonials, with patients frequently needing many courses of treatment. By contrast, L. (V.) panamensis is in most cases susceptible to these drugs.⁵⁹

Finally, three isolates were typified as hybrids (L. [V.] braziliensis/L. [V.] panamensis) after analysis by MLEE. Two of these patients came from site 1 and one from site 2 (Table 3), and were originally characterized by hsp70-RFLP as L. (V.) braziliensis. However, sequencing analysis of an hsp70 region showed a large number of ambiguities that were consistent with the hybrid condition of these isolates. Nevertheless, the sequences of analysis suggest that the most likely parental isolates were L. (V.) braziliensis and L. (V.) guyanensis. Although genetic exchange in Leishmania parasites is considered to be an uncommon event in South America,⁶⁰ CL caused by hybrid Leishmania species has been reported in some countries of this region.^61–65 Moreover, hybrid Leishmania species has been reported infecting sand fly vectors.⁶⁶ Interestingly, these Leishmania hybrids are reported for the first time in Panama; however, we are unaware of the epidemiological and clinical consequences associated with them. In this sense, it is important to consider that Leishmania hybrids can have a strong selective advantage, capable of enhancing its virulence, modifying its tissue tropism, and conferring them drug resistance.⁶⁶ Furthermore, hybrids can infect and be transmitted by new vectors, which could change the geographical distribution of the disease.⁶⁶

Leishmania (V.) braziliensis, L. (V.) guyanensis, and L. (V.) panamensis are responsible for causing different clinical manifestations as localized CL (the most benign form), disseminated leishmaniasis, and mucosal leishmaniasis.⁶⁷ Although L. (V.) guyanensis and L. (V.) panamensis are able to invade mucosal tissues,^14,45,68 their involvement in mucosal disease is less destructive, differing from the chronic and severe forms characteristically caused by L. (V.) braziliensis.⁶⁹ In Panama, mucosal involvement has been observed in 4.2% of positive patients from where L. (V.) panamensis has been demonstrated.²² However, based on the findings presented in this study, it is important to highlight that L. (V.) braziliensis and L. (V.) braziliensis outlier should be considered as potential etiological agents in mucosal leishmaniasis cases reported in Panama.

Relapses, therapeutic failure, and treatment resistance have been reported in human leishmaniasis caused by species of Leishmania subgenus Viannia.^69–74 In this sense, the biology, clinical relevance, and response to treatment of the non–L. (V.) panamensis parasites (including genetic variants “Leishmania [V.] sp. 1, Leishmania [V.] sp. 2” and hybrids isolates) described in this study are uncertain, and therefore need further evaluation.

Although the hsp70 methodology used in this study has been recommended for typing Leishmania spp. from regions where many circulating species are endemic,⁷⁵ its discriminatory power is not sufficient to clearly determine the phylogenetic position of some Leishmania species/genetic variants.^48,76 Probably, the analyzed region of the hsp70 gene (1,245 pb) presents a low polymorphism to readily distinguish between closely related Leishmania species. In this regard, complementary studies using more powerful molecular analysis tools, like multi-locus sequence typing and complete genome sequencing,^{22–24,77,78} are necessary to clearly define the phylogenetic position of the Leishmania isolates (here referred as Leishmania (V.) sp. 1, Leishmania (V.) sp. 2) and hybrids.

The genetic variability of the Leishmania (Viannia) parasites has been already reported in South America.^{47,77,79–81} The presence and frequency of these species/species variants are linked to eco-biological conditions not clearly defined, or only partially in some studies.⁴⁷ Nevertheless, like other pathogens, the genetic diversity could be linked to particular eco-epidemiological patterns, virulence, pathogenicity, and drug responses. In conclusion, the results here presented not only improve our knowledge about the genetic diversity of the parasites that cause localized CL in Panama but also highlight the need to perform additional studies with others molecular methodologies that confirm and extend these findings.

Note: Supplemental information and table appear at www.ajtmh.org.