Abstract
Human leishmaniasis is an emerging problem in Italy and is on the increase in the Emilia–Romagna region, northeastern part of the country. Nevertheless, studies dealing with the molecular characterization of Leishmania spp. circulating in these areas are limited. In the present work, we explored the genetic polymorphism of Leishmania isolates from 28 cases of canine leishmaniasis and three cases of human visceral leishmaniasis (VL), which occurred in 2013–2014 in the Emilia–Romagna region. The characterization was carried out in comparison with nine human isolates of Leishmania from other VL endemic Italian regions and two reference strains. Nucleic acid from 31 Leishmania-positive phlebotomine sandfly pools, sampled in 2012–2013 in the Emilia–Romagna region, were also evaluated. DNA amplification and sequencing of the ribosomal internal transcribed spacer-1 and of a repetitive nuclear region on chromosome 31 were carried out for genotyping. Two size polymorphic targets were also analyzed by PCR, the cpb E/F-gene and the k26-gene. Altogether, the analysis showed the circulation of different Leishmania infantum genotypes in the Emilia–Romagna region: two genotypes found in dogs from public kennels were similar to VL isolates from other Italian regions, whereas a third genotype was detected in VL cases of the Emilia–Romagna region and in all but one of the sandfly pools. The combined molecular tools applied in this study can constitute a helpful support for parasite tracking (e.g., in outbreak investigations) and for a better understanding of the epidemiological evolution of leishmaniasis in northeastern Italy.
Introduction
T
In southern Europe, VL is endemic, and is considered a zoonotic infection as humans may enter the zoonotic cycle and contract the infection. Dogs represent the main domestic reservoir hosts for human infection, which is caused by Leishmania infantum. Nevertheless, the circulation of Leishmania species other than L. infantum cannot be excluded, taking into account the increased human migration (Di Muccio et al. 2015).
In Italy, classical endemic zones for VL are the Tyrrhenian littoral, the southern peninsular regions, and the islands. Since the 1990s, the incidence of VL has increased in humans and dogs, with new foci not only within traditional endemic areas, but also in northern regions previously regarded as nonendemic (Maroli et al. 2008, Biglino et al. 2010, Gramiccia et al. 2013). In the Emilia–Romagna region (northeastern Italy), leishmaniasis has been repeatedly described in both dogs and humans (Pampiglione et al. 1974, Baldelli et al. 2001, Mollicone et al. 2003, Santi et al. 2014). From November 2012 to May 2013, a human outbreak occurred in the Bologna province within the Emilia–Romagna region (Varani et al. 2013); 14 autochthonous cases of VL were detected compared with a previous average number of 2.6 cases per year (Emilia–Romagna Public Health Department 2015). In the same period, an increasing number of VL cases was reported in the neighboring Modena province (Franceschini et al. 2016).
To monitor the evolving situation, surveillance activities have been implemented, involving humans, dogs, and vectors (Santi et al. 2014).
Comparison of Leishmania strains circulating in a geographical area could be useful for epidemiological studies. Multilocus Enzyme Electrophoresis (MLEE) is still regarded as the gold standard for the identification and classification of Leishmania isolates. However, MLEE has several disadvantages because it is a laborious, time-consuming, and expensive method and requires a high quantity of parasites in culture (Schönian et al. 2008). Several DNA-based methods have been developed for Leishmania characterization at species and strain level, including polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) or sequencing of antigen coding (i.e., gp63, hsp70, cpb) or noncoding genes (i.e., internal transcribed spacer-1, repetitive DNA sequences), multilocus typing such as Multilocus Sequence Typing (MLST), or microsatellite analysis (Van der Auwera and Dujardin 2015).
The aim of this study was to genotype canine and human Leishmania isolates from the Emilia–Romagna region and to compare the results with those obtained with human Leishmania isolates from other endemic Italian areas. Leishmania DNA from sandflies collected during 2012–2013 in the Emilia–Romagna region was also included for typing. Two DNA targets were amplified by PCR and sequenced: the ribosomal internal transcribed spacer-1 (ITS-1), and a repetitive nuclear region of 250 bp on chromosome 31 (rnr_chr.31). For other two targets, the cysteine protease B (cpb) gene and a repeat region of the hydrophilic acylated surface protein B (HASPB) gene, also known as k26, the size polymorphisms of the amplicons were determined by PCR.
Materials and Methods
Study area
Emilia–Romagna region is located in northeastern Italy (Fig. 1), between 9°11′–12°50′ E longitude and 43°46′–45°05′ N latitude. The region covers an area of 22,452 km2 and approximately half of it (48%), in its northern part, consists of Po Valley plains, whereas the remaining part of the region is covered by hills (27%) and mountains (25%) up to an elevation of 2165 m above sea level. The climate condition is typically subcontinental, with a mean temperature of 16.5°C and mean annual precipitations of 815 mm, in the last 10 years (Ministero delle Politiche Agricole 2016).

Geographical distribution of human (squares) and canine (dots) isolates and Leishmania-positive sandfly pools (triangles), 2013–2014, Emilia–Romagna region (northeastern Italy). The k26 genotype is indicated by lines within symbols: vertical (626 bp), diagonal (878 bp), horizontal (680 bp), grid (626/680 bp), no lines (not-determined). The acronyms of the provinces PC (Piacenza), PR (Parma), RE (Reggio Emilia), MO (Modena), BO (Bologna), RA (Ravenna), FC (Forlì-Cesena), RN (Rimini) are indicated. Municipalities where canine Leishmania infantum isolates were collected: Piacenza (within PC); Parma (within PR); Reggio Emilia, Scandiano, Bagnolo in Piano, Novellara (within RE); Trebbo di Reno (within BO); Lugo (within RA); Forlì, Cesena (within FC); Montescudo, Riccione, San Clemente (within RN).
Previous entomological surveys have established the presence of two proven vectors of L. infantum in the Emilia–Romagna region: Phlebotomus perfiliewi and Ph. perniciosus (Pampiglione et al. 1974, Maroli et al. 2008). Sandflies are mainly distributed in the hilly areas of the region. Ph. perfiliewi represents the most abundant species (Santi et al. 2014); it is considered a less efficient vector of L. infantum in comparison to Ph. perniciosus (Maroli et al. 2008), however, it is able to support the transmission of infection (Baldelli et al. 2001).
Samples
A total of 40 Leishmania isolates were analyzed. Thirty-one isolates were collected in different locations across the Emilia–Romagna region in 2013 and 2014; 28 isolates were obtained from dogs and three isolates were from HIV-negative VL cases. Dogs were sampled at 17 different areas (Fig. 1), after seroconversion was detected in the frame of a surveillance program for canine leishmaniasis (canL) in public kennels of the Emilia–Romagna region (Santi et al. 2014). The human cases were diagnosed in patients involved in a VL outbreak reported in the Modena province (Fig. 1) (Franceschini et al. 2016).
In addition, nine human isolates were obtained from HIV-negative VL cases diagnosed between 2007 and 2012 in patients residing in VL endemic Italian regions, such as Lombardy, Tuscany, Liguria, Calabria, and Sicily, and henceforth mentioned as extra-region human isolates. Nucleic acid from 31 Leishmania spp.-positive sandfly pools (comprising a total of 6168 specimens) was also included in the study. The pools were previously collected in 11 sites from three areas of the Emilia–Romagna region in 2012 and 2013 (Fig. 1). A subsample of these specimens were morphologically identified, as reported by Calzolari et al. (2014), highlighting a great abundance of Ph. perfiliewi (2057 specimens) and a limited presence of Ph. perniciosus (6 specimens).
Two reference strains, L. infantum MHOM/TN/80/IPT1 (MON-1) and MHOM/IT/86/ISS218 (MON-72), were included in the study as the most common zymodemes responsible for VL in Italy (Gramiccia et al. 2013).
DNA extraction, PCR, and sequence analysis
Parasites were cultivated in Evans' modified Tobie's medium supplemented with 15% rabbit blood. The replicating parasites were harvested, washed with NaCl 0.3%, and centrifuged. The total DNA was extracted as follows: the pellet was lysed by heating at 96°C for 20 min with a 400 μL of a mixture containing 1% Tween 20 (Sigma, St. Louis, MO), 1% Nonidet P-40 (Roche, Germany), and 20% Chelex resin (Bio-Rad, Hercules, CA), in sterile distilled DNA-free water. The mixture was centrifuged at 14,000 g for 10 min at 4°C and the upper phase containing the DNA was collected and stored at −20°C (Reale et al. 2010).
As an initial step, DNA samples obtained from Leishmania isolates were amplified for the ITS-1 region according to Kuhls et al. (2005). Furthermore, a PCR amplifying a repetitive region of 250 bp on chromosome 31 (rnr_chr.31) was performed (Piarroux et al. 1995). All amplicons were purified with the AMPure XP PCR Purification Kit (AgenCourt) and sequenced in both directions by CEQ 8000 sequencer (Beckman Coulter), using the GenomeLab DTCS Quick Start Kit (Beckman Coulter). The results were analyzed and assembled using the “Sequencing” and “Investigator” packages of CEQ 8000 version 8.0 software. Sequences were aligned using the ClustalW application of BioEdit version 7.0.8.0. (Hall 1999) and compared with the public sequences available in GenBank using the BLAST server from the National Center for Biotechnology Information (
In addition, the short cpb E/F-PCR (Zackay et al. 2013) and the k26-PCR (Haralambous et al. 2008) were performed on all the DNA samples. The accurate estimation of the k26-PCR product size was determined by capillary electrophoresis on a CEQ 8000 sequencer (Beckman Coulter), in the presence of size-standard MAP MARKER D1 LADDER 50–1000 bp (BioVentures, Inc., USA). The k26 genotypes were assigned according to the size of the PCR products and adjusted considering the gene size variability, due to the number of 39/42 repeated nucleotide motifs (Haralambous et al. 2008, Bhattacharyya et al. 2013, Zackay et al. 2013).
In the sandfly pools, the Leishmania spp. positivity was detected by a Real-Time PCR targeting minicircle kinetoplast DNA (Galletti et al. 2011). Owing to weak parasitic DNA in sandfly specimens, only the rnr_chr.31 target was amplified and sequenced as described above. For the same reason, only two sandfly pools were tested for the k26 gene.
Phylogenetic analysis
One rnr_chr.31 sequence per type (type A or type C according to Minodier et al. 1997), representative of human and canine isolates from the Emilia–Romagna region, extra-region human isolates, Leishmania-positive sandfly pools and reference strains, was aligned with 17 sequences available in GenBank using ClustalW, as implemented in BioEdit.
The evolutionary histories of the Leishmania rnr_chr.31 sequences were inferred using maximum likelihood (ML) and Bayesian inference (BI) algorithms. The most appropriate nucleotide substitution model, based on corrected Akaike Information Criterion and Bayesian Information Criterion, were evaluated by jModeltest version 2.1.10 (Darriba et al. 2012).
The ML inference was performed using the software MEGA6 (Tamura et al. 2013) for the Hasegawa-–Kishino–Yano model (Hasegawa et al. 1985), with 1000 bootstrap replicates. BI was conducted using MrBayes 3.2.6 (Huelsenbeck and Ronquist 2001) for the same substitution model, by allowing four incrementally heated Markov chains to proceed for 10,000,000 generations, with samples taken every 200 generations. The first 25% was discarded as burn-in and the following data set was being sampled with a frequency of every 200 generations. Stationarity was deemed to have been reached when the average standard deviation of split frequencies was less than 0.01
Results
The DNA from Leishmania isolates was first examined by the ITS-1 sequencing (GenBank accession nos. KX096612–KX096651). The analysis showed that all the 40 isolates belonged to the L. infantum species (Table 1). The isolates clustered in the same ITS group, including several L. infantum strains from the Mediterranean basin that were reported by Kuhls et al. (2005) (data not shown).
na, not available; i, isolates; s, DNA; ^, Cluster 1; °, Cluster 2; *, 361 bp; **, 400 bp.
Thirty-nine Leishmania isolates were also sequenced for rnr_chr.31 and two different sequence types (ST) were obtained (GenBank accession nos. KU680884–KU680922); 28 out of 28 canine isolates, eight out of eight extra-region human isolates, as well as the two reference strains (GenBank accession nos. KU680924–KU680925) showed the same nucleotide sequence, corresponding to ST-A (Minodier et al. 1997), except for one nucleotide substitution (C → G) at position 187 (Table 1). Conversely, the sequences from the three VL cases residing in the Emilia–Romagna region were identical to each other and showed a 100% identity to the ST-C described by Minodier et al. (1997). Sequencing failed for one extra-region human isolate. Concerning the sandfly pool Leishmania DNA, sequencing of amplicons showed a ST-C in 30 out of 31 sandfly pools and a ST-A in the remaining pool (GenBank accession nos. of 26 out of 31 sequences: KU680926–KU680951, five sequences were not deposited because less than 200 nucleotides were obtained).
Regardless of the strategy used for phylogenetic reconstruction (ML or BI), rnr-chr.31 trees (Fig. 2) revealed overall congruent topologies and a consistent placement of the sequences obtained in this study in two clusters. Sequences from dogs, extra-region humans, and reference strains (ST-A) were found exclusively in Cluster 1; on the other hand, sequences from all but one sandfly pools and from VL cases residing in Emilia–Romagna (ST-C) gathered in Cluster 2. Strains in this cluster shared five nucleotide substitutions and one insertion in the rnr-chr.31 sequence, not found in any of the other strains of this study.

Phylogenetic tree based on the partial sequences (190 bp) of a repetitive nuclear region on chromosome 31 (rnr_chr.31) of Leishmania spp. isolates/DNA. The tree was constructed by the alignment of seven sequences obtained in the present study (marked with filled triangles). Seventeen available GenBank sequences were also added. GenBank accession numbers are indicated, as well as the species of Leishmania and the acronym of the isolates. The evolutionary history was inferred by using the ML method based on the Hasegawa–Kishino–Yano model. The tree with the highest log likelihood (−456.8183) is shown. The tree is drawn to scale, with branch lengths measured in the number of substitutions per site. Branch values represent statistical support for 1000 bootstrap repetition for ML (first value) and posterior probability in Bayesian inference using MrBayes software (second value). ML, maximum likelihood.
Furthermore, all canine and extra-region human isolates, as well as the two reference strains, showed the cpb type E (361 bp) amplicon by the cpb E/F-PCR, whereas the three human isolates from patients residing in the Emilia–Romagna region exhibited the cpb type F (400 bp) amplicon (Table 1).
Finally, k26-PCR was carried out on DNA from all isolates and from two sandfly pools. Among the 40 isolates, capillary electrophoresis detected three different k26-PCR product sizes, including 626 bp (27/40), 680 bp (3/40), and 878 bp (10/40) (Table 1). In particular, the product size of 626 bp was detected in 21 canine isolates, in 6 extra-region human isolates and in L. infantum reference strains; the product size of 680 bp was observed only in the three human isolates from patients residing in the Emilia–Romagna region, whereas the product size of 878 bp was detected in seven canine isolates sharing the same geographical origin within Emilia–Romagna (Rimini province) and in three extra-region human isolates. Concerning the two sandfly pools tested, a product size of 680 bp and two product sizes of 680 and 626 bp, respectively, were detected.
Detailed data on typing for human and canine isolates and reference strains are shown in Supplementary Table S1 (Supplementary Data are available online at
Discussion
Studies on the molecular epidemiology of leishmaniasis are scarce (Ferroglio et al. 2006, Salvatore et al. 2016), since identification of parasite species and strain typing are complex to perform; for this reason, the diagnosis of leishmaniasis is often limited to genus recognition. Nevertheless, a surveillance system based on typing at species and subspecies level would be helpful to discriminate between autochthonous and imported cases of VL, to identify the introduction of new species into a nonendemic area, and to provide data on distribution of parasite variants among distinct foci of infection.
In this study, four molecular markers were investigated to explore polymorphism in Leishmania parasites circulating in northeastern Italy. Overall, different resolving power in discriminating groups within L. donovani complex was achieved by different typing methods.
First, analysis of ITS1 indicated that all tested samples and reference strains shared a monomorphic sequence, 100% consistent with L. infantum. This target proved to be highly species-specific, as all samples clustered in the same ITS group of several L. infantum strains from the Mediterranean basin. Second, phylogenetic inference based on rnr_chr.31 sequences separated isolates in two distinct clusters: Cluster 1, including L. infantum reference strains (MON-1 and MON-72), canL and extra-regional VL isolates, as well as 1 sandfly pool, and Cluster 2, including 3 regional VL and 30 sandfly pools. Third, when focusing on the cpb-PCR, previously designed to discriminate among species in the L. donovani complex (Hide and Bañuls 2006, Zackay et al. 2013), the strains belonging to Cluster 2 by rnr_chr.31 sequencing showed the cpbF amplicon, which is typical of L. donovani. The multicopy nature of the cpb gene might give rise to variations within the L. donovani complex, as reported for Tunisian L. infantum MON-24 strains (Chaouch et al. 2013); in the current study, the cpb target confirmed the intraspecies variability of the Emilia–Romagna strains. Fourth, the highest discrimination power was achieved by the k26 typing tool, with isolates being differentiated in three genotypes. Most of the isolates exhibited a 626 bp amplicon that is typical of L. infantum MON-1 strains from Western Europe (Haralambous et al. 2008, Chicharro et al. 2013); this zymodeme is highly prevalent in humans and dogs within Italy (Gramiccia et al. 2013). Other so far unreported amplicons were also detected, including a 680 and a 878 bp; similar product sizes have been previously described for L. infantum strains from Spain, Greece, and France (870 bp) and for Indian and Cypriotic L. donovani strains (660 and 700 bp, respectively) (Haralambous et al. 2008, Gouzelou et al. 2013, Zackay et al. 2013).
The multitarget approach of this study allowed us to identify intraspecies polymorphism within L. infantum population in northeastern Italy, also providing first insights into spatial distribution of parasite variants. The combined analysis made it possible to identify three distinct L. infantum genotypes sustaining active foci of transmission in 2013–2014; two genotypes, commonly causing VL in endemic Italian regions, were widespread in dogs from the Emilia–Romagna region, but not in human samples. Despite the limited number of regional human isolates included in this study, worth noting is the detection of a novel L. infantum genotype causing VL. Such genotype (ST-C/cpbF/k26-680) was clearly distinct from those found in dogs of the same area and commonly observed in the rest of Italy (ST-A/cpbE/k26-626 and ST-A/cpbE/k26-878).
Finally, analysis of Leishmania DNA from sandflies showed circulation of two sympatric genotypes in northeastern Italy; that is, ST-C/k26-680 bp and ST-A/k26-626 bp. While the cocirculation of both genotypes was detected in a limited area (Fig. 1), the strains with ST-C were present since 2012 in the territory of three provinces of the Emilia–Romagna region, overlapping with areas characterized by increasing VL cases (Varani et al. 2013, Franceschini et al. 2016). Notably, the genotype ST-C/cpbF/k26-680 was recently detected by typing DNA samples from two VL cases of the human outbreak that occurred in the neighboring Bologna province (Varani et al., 2013 and unpublished data), confirming the geographical spreading of this novel genotype. Together with the historical presence of Leishmania in the Emilia–Romagna region (Pampiglione et al. 1974, Zerbini et al. 1975), this study highlights the need to intensify molecular surveillance of Leishmania in northeastern Italy.
The unique ratio between sandfly species of this area, with an overwhelming presence of Ph. perfiliewi in comparison to Ph. perniciosus—the latter being the primary vector of the leishmanial parasite in Italy (Maroli et al. 2008)—may play a role in the selection of distinct Leishmania genotypes at a local level. However, this hypothesis should be tested on additional strains from different hosts, with methods developed for Leishmania population genetics, such as microsatellite analysis (Kuhls et al. 2007).
Conclusions
This study provides first insights into the genetic polymorphism of Leishmania population in Emilia–Romagna, northeastern Italy. Remarkably, a distinct L. infantum strain affecting humans seems to be adapted to the local sandfly fauna and involved in recent cases of VL in the same area. The combined molecular tools applied in this study can constitute a helpful support for parasite tracking (e.g., in outbreak investigations) and for a better understanding of the epidemiological evolution of leishmaniasis.
Footnotes
Acknowledgments
This work was partly supported by a research grant from the Italian Ministry of Health (IZS LER 08/11 RC) and by Lab P3 funds from the Emilia–Romagna Region. The authors wish to thank all the colleagues of the Official Veterinary Services who contributed to sampling activities.
Author Disclosure Statement
No competing financial interests exist.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
