An Updated Geographical Distribution of Leishmania Species Involved in Human Leishmaniasis in Colombia

Leishmaniasis in an endemic disease in Colombia. This country ranks second in number of cases of leishmaniasis in Latin America and is one of the four countries with the largest number of Leishmania species. Although leishmaniasis is a notiable disease, the information about the geographic distribution of Leishmania species circulating in the country is fragmented. Since 1990, when the rst distribution map was published in Colombia, very few new reports have described the Leishmania species present in Colombia and their geographic distribution. In this report we included 1,833 natural isolates obtained from human cases of leishmaniasis. Their geographical origin together with those reported by other authors let us conrm the presence of Leishmania species in 30 of 32 Colombian departments, being Quindío and San Andres/Providencia free of transmission. In this work, L. braziliensis is described for rst time in the departments of Atlántico, Caldas and Vaupes and L. panamensis in the departments of Guajira and Vaupés. It is also conrmed that L. panamensis and L. braziliensis coexist in the same regions being the main causative agents of cutaneous leishmaniasis in Colombia; nonetheless, L. panamensis predominates in the north and western regions whereas L. braziliensis predominates in the southeast region.


Introduction
Leishmaniasis is a parasitic disease caused by species of the genus Leishmania. The parasites are maintained in reservoirs that may be wild or domestic animals, including humans, and are transmitted by the bite of numerous species belonging to the Psychodidae Family and Phlebotominae subfamily (Lutzomyia, Phlebotomus, Warileya, Sergentomyia and Brumptomyia genus). The disease is characterized by its large clinical polymorphisms and can be grouped into three major categories: cutaneous leishmaniasis (CL) with lesions circumscribed to the skin, mucosal leishmaniasis (ML) characterized by lesions affecting the mucous membranes of the nasooropharynx, and visceral leishmaniasis (VL), with lesions affecting organs and tissues, such as the spleen, liver, and bone marrow. Leishmaniasis in its different clinical forms is endemic in 98 countries located in tropical and subtropical regions worldwide [1,2].
The World Health Organization (WHO) estimates that leishmaniasis continues to be a major health problem in 4 regions of the world: The Americas, East and North Africa and West and Southeast Asia. There are 92 countries or territories endemics for CL and 83 countries or territories endemic for VL. More than 1 billion people are at risk of infection because they live in endemic areas. It is estimated more than 1 million new cases of CL and 30,000 VL cases occur every year. In 2018, 253,435 new CL and 17,223 new VL cases were reported to WHO, 18% of them in the Americas region (AMR) [3]. In December 2019, the Pan American of Health Organization (PAHO) declared that 17 of the 18 endemic countries in the Americas reported a total of 46,041 new cases of leishmaniasis in the Region, of which, 84% were reported by Brazil (16,432), Colombia (6,362), Peru (6,321), Nicaragua (3,722), Bolivia (3,127) and Venezuela (2,612) [4].
In AMR, where current reporting rates are high, there was a slight decreasing trend in CL cases, from a peak of 71,866 cases in 2005 to 46,265 cases in 2018 [3]. Nevertheless, Colombia has still the second highest number of CL cases in the Americas, after Brazil [2,3,5]. Because many cases of CL and VL occur in remote regions, which makes it di cult to quantify the exact burden of the disease, the incidence for Latin America is estimated considering an underreporting of 2.8 -4.6-fold [2]. To date, 22 species of Leishmania are described as human pathogens [1]. The identi cation has been based on the parasite´s isoenzymes content determined by a multilocus enzyme electrophoresis (MLEE) [6][7][8] or the detection of parasite membrane biomarkers with monoclonal antibodies (MoAb) [9][10][11].
Even though is well accepted that some species are correlated with some clinical forms, one species may cause different clinical forms. On the other hands, the clinical response to treatments also varies inside species, according to the susceptibility of Leishmania species strain to antileishmanial drugs [21]. Therefore, the WHO Expert Committee recommends treating cases with speci c therapeutic strategies for the parasite species causing the disease. Hence, identi cation of the species that cause leishmaniasis is important not only from an epidemiological perspective to clarify the distribution of species in a certain region and the prognosis of the disease but also to make decisions regarding the therapeutic options to treat patients [1]. Thus, elucidating the species distribution in a given geographical area and determining the species of the parasite that is affecting the patient are necessary for the adjustment of treatment [1]. This issue is particularly important in countries such as Colombia, which is one of the four countries worldwide with the highest biodiversity of Leishmania species with at least 10 of them circulating in the territory [L. braziliensis, L. panamensis, L. guyanensis, L. mexicana, L. amazonensis, L. infantum, L. colombiensis, L. equatorensis,L. lainsoni and L. nai ] [27][28][29][30][31][32][33], all of them isolated from patient samples. Other countries with a high number of Leishmania species include Brazil, Venezuela and Nicaragua [1].
Although leishmaniasis is endemic in most of the Colombian territory and it is a mandatory noti able disease since the end of the 1980's, information about the geographic distribution of the circulating Leishmania species in the country is very fragmented. To date, very few publications have described the Leishmania species present in Colombia and their geographical distributions. The present work describes and complement the current distribution of Leishmania species in Colombia, isolated from patients with different clinical forms of leishmaniasis attended in our laboratory and in endemic transmission foci studied in the last 35 years with the objective to update the information about the geographic distribution of Leishmania species in Colombia. The species were classi ed by extrinsic characteristics in 2 subgenera, Viannia and Leishmania. Of the 4,539 patients positive for leishmaniasis, the Leishmania species was identi ed in 1,833 (40%) patients. In 1,771 isolates the identi ed strains corresponded to 7 different species: L. panamensis, L. braziliensis, L. amazonensis, L. guyanensis, L. mexicana, L. infantum and L. nai . In the remaining 39 isolates the identi cation was only achieved for the genus Leishmania spp or subgenus L. (Viannia) spp. The number of the isolates by species and department/region is given in Table 1. L. panamensis and L. braziliensis were the species identi ed in the highest percentages (78% and 20%, respectively). The remaining 2% of the species corresponded to L. mexicana, L. infantum, L. guyanensis, and L. amazonesnsis (0.1% for each) and mixed L. braziliensis/L. panamensis (0.2%). Although an attempt was made to identify all the species by the PCR-RFLP method, in the remained isolates (0.4%) it was not possible to nd that they belonged to any known species.

Results
Both L. panamensis and L. braziliensis were found in 17 of the 29 departments with leishmaniasis cases diagnosed in our laboratory ( Figure 1). Mapping analysis by region did show that L. panamensis predominated in the northwest paci c coast and Andean regions, and was involved in 92% of cases diagnosed by us in these regions, whereas L. braziliensis predominated in the Southeast, Orinoquian and Amazonian regions towards the border with Brazil and being responsible for 74% of the leishmaniasis cases diagnosed by us in those regions.
Antioquia, Tolima, Meta and Cundinamarca departments have the highest diversity with six Leishmania species followed by Guaviare and Caquetá with ve species of Leishmania affecting humans. Santander, Norte de Santander, Casanare and Cauca have four Leishmania species, while the rest of departments have 3, 2 and 1 species.
In Arauca, Cauca, Sucre, Risaralda and Valle only L. panamensis was found, whereas in Atlántico and Magdalena L. braziliensis was only found (Figure 1). L. mexicana and L. amazonensis were found in Antioquia. L. mexicana and L. guyanensis were only found in the department of Meta, and L. infantum was found only in the department of Córdoba. In Antioquia and Meta departments, four different species were identi ed: 3 species were present in both departments (L. braziliensis, L. panamensis, L. mexicana) and L. amazonensis was found only in the Antioquia department, while L. guyanensis was in Meta department).
The comparison of the Leishmania species distribution found in the present study with the distribution previously reported by other authors, allowed the update of the information available for Leishmania species and their distribution by departments and regions. The table 2 summarizes the reported Leishmania species diversity for Colombia. It was observed that L. panamensis and L. braziliensis are the most frequently reported species. From seven authors that have reported the distribution of Leishmania species in Colombia, four of them have reported L. panamensis in the departments of Bolivar, Córdoba, Antioquia, Chocó, Caldas, Cundinamarca, Risaralda, Huila, Tolima, Boyacá, Santander and Meta. In turn, L. braziliensis have been reported in the departments of Cundinamarca, Santander, Meta and Vichada for four or more authors. Apparently, the departments of Quindío and San Andrés and Providencia archipelago remains free of autochthonous cases of leishmaniasis in Colombia, up to date.

Discussion
To date, there are about a dozen studies that have reported the presence of Leishmania species in different geographic regions in Colombia, most of them with a small number of isolates: 340 [27], 511 [28], 76 [29], 137 [30], 16 [31], 35 [21], 81 [33], 327 [32], 93 [12], 273 [34] and 96 [35]. Here we report the geographical distribution of 1,833 isolates of Leishmania from patients we have diagnosed from 29 departments in all geographic regions of the country. In this work no isolates from Vichada department were reported.
Our study con rms that leishmaniasis is present in almost the entire country with only two departments with apparently no reports of autochthonous cases of leishmaniasis: San Andres and Providencia and Quindío departments, despite some foci study conducted by us in those departments where no cases were not detected nor infected phlebotomine or reservoirs. The highest percentage of patients were mainly from the department of Antioquia followed by Meta, Chocó, Caquetá, and Caldas. This abundance of cases are because our laboratory is located in Medellín, the capital city of Antioquia department and close foci patients come to our consulting room and also because in these regions we have carried out the largest number of foci studies and clinical trials of the "leishvacin" vaccine between 1998 -2008 [36] or some therapeutic alternatives such as miltefosine [37], thermotherapy [38] and Anfoleish® [39], or the implementation of the Leishmaniasis Control Program in military forces conducted by us from 2005-2008.
Results showed here agree with results previously reported by other authors, in which the department of Antioquia contributes the largest number of tegumentary leishmaniasis cases in Colombia but also with the highest number of Leishmania species (all identi ed species except L. guyanensis and L. infantum). Other departments with high numbers of cases in this work were Chocó and Meta followed by Córdoba, Caquetá and Caldas. Several factors explain why Antioquia is the department that reports the highest number of CL cases and they are related to human activities such as deforestation, mining, illicit crops and also as a consequence of the war against illicit crops and groups on the margins of the law.
We also con rm that L. panamensis and L. braziliensis coexist in the same regions in Colombia but in different proportions according to regions. Thus, L. panamensis predominates in the Paci c and Andean regions, whereas L. braziliensis predominates in the Orinoquian and Amazonian regions. The presence of L. panamensis in the Northwestern Colombia and, L. braziliensis in Southwestern Colombia, is understandable because it can be assumed that these species have gradually spread from Panama and Brazil to Colombia, respectively. It can be also assumed that the expansion of both species has found the barrier of the Andean and therefore the expansion has not been perhaps faster to the rest of Colombian territory.
However, there are zones with mixture presence of L. panamensis and L. braziliensis in the central area of Colombia. How the presence of both species of parasite was achieved in the same territory is something that needs to be determined; it could be likely due to migration in both routes of infected domestic reservoirs, through gallery forests that cross valleys and mountains. On the other hand, the higher prevalence of a species in a given area may be due to the presence and greater abundance of the main vector species in each geographic area or to the presence of new vectors of these Leishmania species in these areas. Nonetheless, these hypotheses remain to be validated.
The third most reported species is L. infantum which is present in 14 departments where cases of VL has been documented, with predominance in the Caribbean and Andean region but also in the departments of Meta (Orinoquian region) and Putumayo (Amazonian region) ( Table 2). Recently, a rst case of urban canine VL caused by L. infantum was described in Cali, Colombia (Valle department, Paci c region) [40].
The presence of L. mexicana and L. amazonensis are reported 14 departments of Colombia (Table 2). Both species cocirculating in almost the same departments, with the exception of Nariño, Caldas and Risaralda where only L. mexicana circulates and Chocó, Huila and Boyacá where only L. amazonensis has been reported [31,32,35]. There are no reports of L. mexicana nor L. amazonensis in the Caribbean region. This work reports L. amazonensis in Antioquia, with a strain identi ed in a patient from Anorí in 1991, maybe representing the rst report for Antioquia. L. mexicana was reported in Antioquia in 2006 in a patient [41]. Notably, L. mexicana has been reported as causative agents of CL, but also as responsible of one case of diffuse CL [42] and other case of transfusion-transmitted VL in a immunocompromised patient [43].
L. guyanensis is reported in 12 departments in Colombia, especially those located in the Andean and Amazonian region but also in the departments of Cordoba and Sucre in the Caribbean region (Table 2). It is notorious that in the Amazon, only L. guyanensis has been reported [27] and there are no new reports since then. A recent publication that compiled the scienti c literature available for Leishmania species identi ed in Colombia between 1985 and 2017 declares the presence of L. guyanensis in Valle, Guajira and Antioquia Departments [44]; nevertheless, it is not clear which authors are responsible for those reports. Few cases of L. guyanensis have been reported in Colombia; this could be due to the di culties in taxonomic differentiation between L. panamensis and L. guyanensis [29].
Although L. colombiensis was identi ed early in the 80`s in samples from patients from the municipalities of Apartadó and San Carlos in the department of Antioquia [45], and in the sand y P. shannoni from Santander in three individuals belonging to the sample bank of the Colombian National Institute of Health [32], in the present study this species of Leishmania was not found in any of the 138 isolates from San Carlos or the 181 isolates from Apartadó. Because no new reports of this species in Colombia have occur since long time ago, it is suggested that the taxonomic classi cation of this species is controversial.
It is not clear how other new species such as L. lainsoni or L. equatorensis arrived in some areas of the interior of the country. This does not seem to be due to geographic expansion of these species, as occurs with L. panamensis and L. braziliensis. Most probably the arrival could be occurred through a human or a reservoir. There is very few information about L. lainsoni. Since the description of this species has been recorded in three states of brazilian amazon basin [46], Perú [47], Bolivia [48,49] and Ecuador [49].
It was expected that this species should be found in the Colombian Amazonian, as Araracuara, Caquetá where Lu. ubiquitalis, the insect vector of L. lainsoni [50] has been described [51]. Nevertheless, the presence of this species in Antioquia that is intriguing because no incrimination or occurrence of L.ubiquitalis has been recorded.
Leishmania and phlebotominae classi cation remain controversial and currently there is an increasing number of species described and phylogenetic analyses show di culties with the current classi cation [52]. For example, the taxonomic classi cation of L. archibaldi as a species following its initial identi cation as a causative species of VL in Sudan remains controversial. Similarly, even though many authors have accepted that L. chagasi is synonymous with L. infantum based on its isoenzyme classi cation, other authors consider the parasite a different species or perhaps a subspecies of L. infantum.
In conclusion, this work reports the presence of Leishmania species in 30 departments and con rm that San Andres and Providencia and Quindio apparently remained free of transmission. Our nding also concludes that at least ten Leishmania species circulate in Colombia thus validating the great biodiversity of Leishmania species in this country. In addition, this work complements previous reports by other researchers and increase the area of distribution of Leishmania species in Colombia.

Methods
A retrospective study was performed with patients with a con rmed diagnosis of leishmaniasis (cutaneous, mucosal or visceral) treated during the active search for cases performed in multiple transmission foci of Colombia and during consultation at the medical facility of the Program for Study and Control of Tropical Diseases, Universidad de Antioquia, Medellín-Colombia. All experimental protocols were approved by the Human Bioethics Committee of the Sede de Investigación Universitaria, Universidad de Antioquia, Medellín-Colombia.
After sign the Inform Consent Form for use of samples and clinical data in research activities, patients were sampled for parasitological diagnosis of leishmaniasis and species identi cation. In the case of CL and ML samples were taken by scraping and/or aspirated material from the lesion (bottom and/or borders). After that, scraping material was stained with Giemsa for direct examination under the microscope and aspirated material was cultured in NNN medium (Novy, Nicolle and McNeal) and incubated at 26°C [1]. For patients with ML, samples were taken from aspirated material or scraped of mucosal lesions whiles in patients with VL samples were taken from spleen or bone marrow aspirates [53]. Conventional PCR was also used for diagnosis. All methods were carried out in accordance with WHO guidelines and regulations.
The clinical records of each patient were reviewed, and any information related to the clinical form was extracted, Leishmania species and, the most probable place of infection (municipality, county -"vereda" or both), according to the epidemiological information of the patient, displacements in weeks prior to the beginning of the injury, etc. All data analyzed were anonymized.
Over 30 years the identi cation of Leishmania species were carried out using the standardized methodologies recommended by the WHO, using MLEE and MoAb in the 80´s and 90´s and the PCR-RFLP in the last decade. For MLEE and MoAb, isolation of parasite and cultivation in laboratory conditions was mandatory. Thus, between 1984 and 1995, Leishmania species were identi ed according to the zymodeme patterns following the procedure described by Roiux (1990) [7]. From 1995 to 2008, strains were identi ed using MoAb [9,11,54] and since 2009, the identi cation has been conducted mainly using the PCR-RFLP technique and conducted with DNA isolated both, from parasites in culture or directly from DNA puri ed from the clinical samples with sets of primers targeting the cysteine B or the Hsp70 genes as described by others [12,33,55].
The restriction pattern generated allowed inferring the identity of the Leishmania species.
To address the geographical distribution of leishmaniasis cases and Leishmania species reported in the present work, a database was constructed with a list of patients with positive diagnosis of leishmaniasis and the and the most probable place of infection Although we have the location at the municipality and village level of most of the patients, the analysis was presented in a digital map at the department level for easier understanding.
Declarations IDV conceptualized the work. SMR and CM wrote the main manuscript text and prepare gures and tables. All authors reviewed the manuscript.
Data availability.
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.