|Home | About | Journals | Submit | Contact Us | Français|
Conceived and designed the experiments: RF AW YM. Performed the experiments: RF AW YM IA AN LFS FP JPD. Analyzed the data: RF MT IA AN. Contributed reagents/materials/analysis tools: CJ AN IA FP JPD LFS. Wrote the paper: RF AW LFS.
In 2006/7, 18 cases of cutaneous leishmaniasis (CL) were reported for the first time from Sde Eliyahu (pop. 650), a village in the Beit She'an valley of Israel. Between 2007–2011, a further 88 CL cases were diagnosed bringing the total to 106 (16.3% of the population of Sde Eliyahu). The majority of cases resided in the south-western part of the village along the perimeter fence. The causative parasite was identified as Leishmania major Yakimoff & Schokhor, 1914 (Kinetoplastida: Trypanosomatidae). Phlebotomus papatasi (Scopoli), 1786 (Diptera: Psychodidae) was found to be the most abundant phlebotomine species comprising 97% of the sand flies trapped inside the village, and an average of 7.9% of the females were positive for Leishmania ITS1 DNA. Parasite isolates from CL cases and a sand fly were characterized using several methods and shown to be L. major. During a comprehensive survey of rodents 164 Levant voles Microtus guentheri Danford & Alston, 1880 (Rodentia: Cricetidae) were captured in alfalfa fields bordering the village. Of these 27 (16.5%) tested positive for Leishmania ITS1 DNA and shown to be L. major by reverse line blotting. A very high percentage (58.3% - 21/36) of Tristram's jirds Meriones tristrami Thomas, 1892 (Rodentia: Muridae), found further away from the village also tested positive for ITS1 by PCR. Isolates of L. major were successfully cultured from the ear of a wild jird found positive by ITS1 PCR. Although none of the wild PCR-positive voles exhibited external pathology, laboratory-reared voles that were infected by intradermal L. major inoculation, developed patent lesions and sand flies became infected by feeding on the ears of these laboratory-infected voles. This is the first report implicating M. guentheri and M. tristrami as reservoirs of Leishmania. The widespread co-distribution of M. guentheri and P. papatasi, suggests a significant threat from the spread of CL caused by L. major in the Middle East, central Asia and southern Europe.
This study dealt with the ecological and epidemiological characterization of a newly emerged dermal disease caused by Leishmania parasites transmitted by the bite of infected sand flies. In the first five years of its emergence, the human infection rate in Kibbutz Sde Eliyahu, Israel reached 16%. The majority of cases resided along the periphery of the Kibbutz. To identify the vector species, we performed systematic trapping of sand flies in different habitats inside and outside the inhabited areas. Sand flies were identified in the laboratory and shown to comprise mainly Phlebotomus papatasi. Since the causative agent of human cases was shown to be L. major, we examined sand flies for Leishmania infections by PCR and found a high infection rate amongst sand flies trapped near houses in late summer (11.8%). To confirm the Leishmania species, we isolated live parasites from the gut of a wild-caught sand fly female. The cultured isolate was identified as L. major by several methods. The Levant vole, abundant in the surrounding fields and Tristram's jird found further away from the Kibbutz were tested for infection rate and their capacity for hosting the parasite and were both incriminated as new possible hosts.
The leishmaniases are parasitic diseases caused by Leishmania parasites and transmitted by phlebotomine sand flies in tropical, subtropical and temperate regions of some 98 countries , . Three species of Leishmania cause leishmaniasis in Israel. Infections with L. major and L. tropica Wright, 1903 cause cutaneous leishmaniasis (CL) while L. infantum Nicolle, 1908 causes mainly canine but also human visceral leishmaniasis (VL) . Recent studies have clearly documented the rapid geographical expansion and steep increase in the number of CL cases caused by L. tropica in northern Israel and the Palestinian West Bank , , . Similarly, the incidence and geographic distribution of dogs with VL is clearly on the increase , . On the other hand, L. major foci remained geographically stable with case numbers being directly proportional to the number of persons encroaching upon the zoonotic foci in the deserts of southern Israel , .
In 2007 physicians at the dermatology department of Ha'emek Hospital in Afula, northern Israel began to see increasing numbers of CL cases among residents of the Beit She'an Valley, north-eastern Israel. Skin scrapings from lesions were sent for diagnosis to the Hebrew University – Hadassah Medical School where the causative parasites were identified as L. major using ITS1 PCR and subsequent restriction fragment length polymorphism (RFLP) analysis , . Concurrently, residents of Kibbutz Sde Eliyahu (Beit She'an valley) approached us for assistance in controlling sand flies.
L. major in Israel was traditionally restricted to desert foci where the parasites infect Sand rats Psammomys obesus Cretzschmar, 1828 (Rodentia: Muridae) and Sundevall's jirds Meriones crassus Sundevall, 1842 (Rodentia: Muridae), the chief reservoir hosts in the region , , , , , . P. papatasi is the proven vector of L. major throughout much of its distribution including Israel , , . However, preliminary studies in Sde Eliyahu made it clear that this emerging L. major focus was ecologically distinct; the village of Sde Eliyahu is surrounded by agricultural fields not desert. Moreover, it is located north of the recognized geographic range of both P. obesus and M. crassus which inhabit wild desert regions characterized by sandy or loess soils and chenopod bushes ( and G.Shenbrot, pers. comm.).
Here we report on findings from a comprehensive study conducted over five years in Kibbutz Sde Eliyahu in which we decipher the transmission cycles, incriminate the vector and implicate two new reservoir hosts of L. major. Our findings suggest the likelihood of CL caused by L. major spreading to new areas in Israel, neighboring countries and perhaps northwards into south European countries and central Asia.
Medical records: Patients all sought medical treatment for CL and were not enrolled in a study. We received geographical data on where patients live and anonymous diagnostic data were collected in the lab to be correlated with the geographical data. Private medical records were not accessed in this study nor were names or other identifying aspects provided. The Helsinki Committee on Research Involving Human Subjects of The Hebrew University – Hadassah Medical School of Jerusalem, Israel approved the study on “Emergence of Cutaneous Leishmaniasis in the Middle East: An Investigation of Leishmania tropica” under which medical data were obtained (permit No. 362-7.09.07).
Trapping of wild sand flies and rodents and experimentation on rodents was approved by the Israel Nature and Parks Authority (INPA permit No. 2009/32202), under the wild life protection act of 1955, and the national parks protection and memorial sites act of 1998 (www.parks.org.il/).
The INPA permit allowed for trapping of 1000 voles and 300 jirds on the condition they would be sacrificed or otherwise kept in captivity when experimentation was over; anesthetization was to be carried out by intraperitoneal injection of ketamine and xylazine according to specimen weight. Euthanization would follow the ketamine/xylazine overdose route.
All field work was conducted in or near Sde Eliyahu (32°26′N, 35°30′E, Alt. −185 m. Pop. 650), a collective agricultural settlement (=Kibbutz) located 6 km south of the town of Beit She'an, in the northern Jordan Valley of Israel. The climate is hot and dry; the mean maximal daily temperature in summer is 35°–37°C and 18°–22°C in winter. Average annual rainfall is 280 mm. Farmers in Sde Eliyahu practice mostly organic farming, utilizing very few fertilizers or pesticides (www.seliyahu.org.il/eOrganic.htm).
The diagnostic data of all CL cases from Sde Eliyahu that were registered in either the local clinic or regional hospital (HaEmek Hospital, Afula), were used to associate between the season and locality of morbidity. This list included the year of birth, date of positive diagnosis, diagnostic test conducted and diagnosing facility. Personal details were withheld, and locations of their residences were mapped anonymously on a blank map of the site (Fig. 1).
Sand flies: Sand flies were trapped using CO2 (dry ice or compressed gas) - baited miniature CDC - type traps assembled in our lab . Sand fly trapping within the Kibbutz was conducted after approval of the local household inhabitants. Trapping in the fields was given consent by the Kibbutz farming secretariat. Sand flies were transported to the laboratory where they were counted, sexed and preserved in 100% ethanol, frozen (−20°C) for PCR or dissected fresh for parasite isolation. Methods for mounting and identification were described previously .
Rodents: Voles that do not usually enter rodent traps were captured by hand after being expelled from their burrows by flooding with water or by plowing. Several active burrow systems in alfalfa fields located 30–300 m from the peripheral houses of the village were selected for flooding (Fig. 1: marked with W in blue). A 1000 L mobile water tank with a 20 m long hose was towed by tractor between the burrows in alfalfa fields. Each tank-full enabled the flooding of 3–4 burrow systems (5–8 exit holes). Voles escaping from flooded burrows were captured by hand wearing protective gloves. In advance of scheduled plowing of alfalfa fields we were notified and arrived a day in advance. We walked behind the tractor-drawn plow and captured voles that emerged from their destroyed burrows. Plowing was conducted in three alfalfa fields up to 400 m away from the village fence (Fig. 1; marked with P in red stars).
Other rodents (e.g. jirds and mice) were trapped with live rodent traps (Sherman, H.B. Sherman Traps. Tallahassee, FL, USA). Traps were dispersed 15–20 m apart, baited with a peanut snack (Bamba, Osem Ltd. Holon, Israel) attractive to many rodents .
Human cases of CL: The following procedure was performed routinely by a certified physician or nurse on suspected CL patients referred by their Health Maintenance Organizations (HMO) for diagnosis at HaEmek or Hadassah hospitals. The skin was sterilized and the exudates from superficial skin scraping of the margins of the lesion were blotted on to sterile filter paper . In three cases, Leishmania isolates were obtained by seeding lesion exudates mixed with sterile PBS into NNN slants overlaid with M199 medium supplemented with 15% fetal calf serum (FCS) containing penicillin (100 IU/mL), streptomycin (100 µg/mL), pH 7.4 . Identification of the causative agent of the disease was performed at the department of microbiology and molecular genetics in The Hebrew University of Jerusalem, using molecular tools (ITS1 PCR followed by RFLP) described below.
Sand flies: For parasite isolation, sand fly females were anesthetized, placed in a tea strainer and washed in 5% detergent solution. They were rinsed in sterile water, dipped briefly in 70% ethanol and transferred to sterile PBS supplemented with penicillin (100 IU/mL), streptomycin (100 µg/mL). Their guts were dissected on sterilized glass slides, using flame-disinfected forceps and examined for the presence of promastigotes under a phase-contrast microscope with a 40× objective. The guts of promastigote-infected sand flies were seeded into NNN slants (see above) further supplemented with 5-fluorocytosine (1,500 µg/ml, Sigma) to suppress fungal growth and 2% sterile human urine to enhance promastigote growth .
Rodents: Field-caught rodents were anaesthetized using a ketamine/xylazine mixture. Their ears were scrubbed with detergent solution (10%) followed by ethanol (70%). Small skin snips from the ear pinnae were mashed onto filter paper (Whatman, 3MM) for PCR and/or seeded into NNN slants (see above) for parasite isolation. For monitoring infection in lab-reared voles and jirds superficial skin scraping was performed using sterile blood lancets and the blood/exudate droplets were blotted onto pre-marked filter papers and microscope slides. Filter papers and glass slides were allowed to air-dry and kept at room temperature until use. Procedures were performed under permit of the Israel Nature and Parks Authority (INPA, permit No. 2009/32202).
Molecular biological (ITS1/RFLP)  and serological (EF sero-typing)  characterization of Leishmania isolates were performed at the Department of Microbiology and Molecular genetics, The Hebrew University-Hadassah Medical School, Jerusalem. Biochemical characterization using isozyme electrophoreses was done at the Centre National de référence des Leishmania, Université Montpellier 1, Montpellier, France .
Samples were excised from the filter papers using a scalpel or a paper punch. For samples from humans and rodents, DNA extraction was performed using the phenol-chloroform procedure described in detail previously . Sand fly DNA extraction was performed using potassium acetate as described elsewhere .
Leishmania-specific ribosomal Internal Transcribed Spacer 1 (ITS1) gene was amplified using the primers LITSR (5-CTG GAT CAT TTT CCG ATG-3) and L5.8S (5-TGA TAC CAC TTA TCG CAC TT-3) . L. major MHOM/IL/1967/Jericho II was used as a positive control, after DNA was extracted from cultured parasites.
PCR products (8 to 20 µl) were digested with BsuRI (MBI Fermentas), a Hae III prototype, according to the manufacturer's instructions, and the restriction fragments were analyzed by gel electrophoresis at 120 V in 1 X Tris-acetate-EDTA buffer in 2.5% agarose gels (FMC BioProducts, Rockland, ME). The fragments were visualized by UV light , .
Leishmania species specific ITS1 oligonucleotide probes were designed and hybridization procedures followed those previously described . Briefly, 5′-amino modified oligonucleotides were immobilized on Biodyn C membrane strips (Pall Biomedical, USA). Strips were incubated in pre-hybridization solution (2× SSC, 0.1% SDS) for 30 minutes at 46°C with gentle shaking followed by hybridization with Biotinylated Leishmania ITS PCR product at 46°C for 1 hour. The membrane strips were washed with 0.7× SSC, 0.1% SDS for 20 minutes. The Hybridized biotinylated amplicon was detected by incubating the strips in strepavidin-HRP (diluted in 2× SSC, 0.1% SDS) for 30 minutes at room temperature and color development after the addition of 0.1 mg/ml of 3,3′,5,5′ tetramethylbenzidine (TMB) (Sigma), dissolved in 0.003% H2O2 in 0.1 M sodium citrate (pH 5.0).
Phlebotomus papatasi sand flies originating in Sde Eliyahu were reared according to the method of Modi and Tesh (1983) . Five-day-old females (F1) were fed through chick-skin membranes on heparinized rabbit blood containing 5×106/ml stationary-phase L. major promastigotes (IPAP/IL/2010/LRC-1475) isolated from a sand fly caught in Sde Eliyahu. Blood-fed flies were starved for 24 h and thereafter, maintained on a 50% aqueous honey (wild flowers) solution, at 26°C and 65% relative humidity. Blood-engorged females were sacrificed for determining infection 6–8 days post blood feeding. Parasite density was graded according to accepted criteria .
In an attempt to perform xenodiagnosis on voles and jirds, 5-day-old P. papatasi females were starved for 24 h and allowed to feed on anesthetized wild-caught rodents that were found positive for Leishmania by ITS1 PCR or infected artificially. The infected skin was exposed to blood-questing females inside a 20 ml glass vial covered by a tight cotton-mesh. Feeding was allowed for the duration of the anesthesia (approximately 45–60 min) in a darkened room at 24°C.
The experimental procedures were approved by the Israel Nature and Parks Authority (INPA permit No. 2009/32202). The voles used in this study were F2 progeny of wild-caught voles from Sde Eliyahu that were reared at the Israel Ministry of Agriculture rodent quarantine facility (PPIS) at Beit Dagan. Prior to all experimental procedures, voles were anesthetized intraperitoneally with ketamine (10%, 50 µl) and xylazine (2%, 10 µl). Animals were inoculated intradermally in the ear pinnae, above the snout, and in hind foot pads, using 106/ml stationary-phase L. major promastigotes (Sde Eliyahu sand fly derived strain IPAP/IL/2010/LRC-1475), in 20 µl M-199 culture medium using a 29G needle. Blood-spot and smear biopsies for PCR diagnosis were taken weekly from the inoculation sites, beginning one week post inoculation (P.I).
From 2007 to 2011, 106 cases of CL from Sde Eliyahu were recorded. The majority (84.0%) were persons residing in peripheral houses close to the southern and south-western borders of the kibbutz (Fig. 1, Table 1). The first three cases were diagnosed in 2006 followed by 15 in 2007, and the peak of the outbreak that occurred in 2008 with 50 (reported) CL cases in Sde Eliyahu. In 2009, 2010 and 2011 case incidence dropped to 11, 9 and 18, respectively. The case incidence in Sde Eliyahu included almost all the residents of the peripheral houses (Fig. 1) and circa 16% of the entire Kibbutz population (106/650).
Of the 71 CL cases from the Beit She'an Valley diagnosed in the Hebrew University lab during the years 2006–2011, 56 were positive for Leishmania DNA by ITS1 PCR. Of these 51 were characterized by RFLP as L. major species (Fig. 2, Table 2). Five L. major isolates; three from human CL cases, one from a naturally infected jird and one from a naturally infected P. papatasi female sand fly captured near Sde Eliyahu, were also characterized using isoenzyme electrophoresis and excreted factor (EF) serotyping (Table 3). All isolates' ITS1 gene were sequenced and shown to be L. major (99.9–100%).
Routine sand fly monitoring was conducted from early July to late October 2009 (21 nights X 35 traps=735 trap/nights). In total 3,087 females were trapped along the Kibbutz' peripheral fence and near houses and in an alfalfa field south-east of the Kibbutz (Fig. 1). A sample of 385 sand flies captured near houses and 250 sand flies from the field were identified showing that P. papatasi was by far the predominant species both near houses (97.5%) and in the fields (78%). P.(Larrousius) tobbi Adler & Theodor, 1930 and P.(L.) perfiliewi Parrot, 1930, neither of which is known to transmit L. major, made up the balance (Fig. 3). A sample of 1,047 (34%) P. papatasi females, trapped throughout the 2009 sand fly season next to the houses, were used to screen for Leishmania infections by ITS1 PCR. An average of 7.9% of the females was positive for Leishmania ITS1 DNA. The infection rate increased from 5.2% in August, peaking at 11.76% by late October (Table 4). Sand fly numbers decreased significantly in November as the sand fly season ended. A sample of ten ITS1 PCR products from wild caught sand flies was analyzed by reverse line blotting (RLB) and DNA sequencing to identify the Leishmania species. All samples were shown to be L. major (data not shown).
In total 164 voles (M. guentheri) were captured by flooding burrows or by plowing of alfalfa fields (approximate locations marked with blue circled ‘W’ in Fig. 1). Thirty six (36) Tristram's Jirds (M. tristrami), were trapped using Sherman traps. Although the infection rates in jirds were high, their overall numbers were lower than voles and they were not found in proximity to the houses. Twenty-five of 36 were trapped in the olive grove (800–1000 m W of the nearest house), 9 in an alfalfa field (1 km E of the houses) and two were trapped in the date palm plantation (700 m NW of the houses) (Fig. 1). Conversely, voles were plentiful near the fence and the highest infection rates were recorded in the two fields closest to the Kibbutz (marked with red stars in Fig. 1).
Animals were anesthetized and examined for skin lesions. There were no obvious signs of disease in any of the animals of either species. Small skin biopsies from ears and snout were tested for Leishmania infection by ITS1 PCR. Twenty-seven (27/164 or 16.5%) voles were positive for Leishmania DNA, as were 58.3% (21/36) of the jirds (Table 5). Samples that yielded sufficient PCR products, were shown to be L. major using reverse-line blotting (Fig. 4). One of the infected jirds that was kept in the animal facility for six weeks, developed patent lesions in its ear pinnae. Biopsies from the lesions were seeded into NNN culture tubes. The promastigotes that grew in the culture were identified as L. major by sequencing the ITS1 PCR product and excreted factor serotyping (Table 3).
In addition, 44 house mice Mus musculus Linnaeus, 1758 (Rodentia: Muridae) were trapped in different locations in or near Sde Eliyahu. Leishmania DNA was detected by PCR in a single mouse trapped in an olive grove, some 1,500 m NW of the Kibbutz (Table 5). A very faint band corresponding to Leishmania ITS1 PCR product was observed on the gel. There was no discernible external pathology (data not shown).
Three anesthetized voles were inoculated intradermally with 105 late log-phase L. major promastigotes (Sde Eliyahu sand fly derived strain IPAP/IL/2010/LRC-1475). Leishmania DNA was detected by ITS1 PCR in all three voles beginning one week after inoculation and weekly henceforth (Fig. 5A). Infections flared up in the hind foot pads first (Fig. 6D) followed by the ears (Fig. 6C). No lesions developed in the snouts of any of these voles. The ear lesions exhibited pronounced disfiguring, swelling and hair loss (Fig. 6C). The hind foot pad lesions were distended and presented a dry yellowish ulceration (Fig. 6D). Giemsa-stained smears taken from infected ears and hind foot pads exhibited large numbers of amastigote-laden macrophages 4–5 weeks P.I onward (Fig. 5B).
Five-day-old laboratory-reared P. papatasi females (Sde Eliyahu colony) were fed through a chick-skin membrane on heparinized rabbit blood containing 2×106 late-log phase L. major promastigotes/ml (strain IPAP/IL/2010/LRC-1475). Four females dissected three days after feeding had heavy infections in the thoracic midgut. The remaining ten females were left for seven more day to digest their blood meals and then to re-feed on anesthetized, healthy voles. This trial was repeated twice, however, females were reluctant to take a second blood meal and thus, artificial infections of voles by sand fly bite were not achieved.
To determine infectiousness of voles to sand flies, five day-old P. papatasi females that were starved for 24 hours were allowed to feed on the ear or footpad of an anesthetized vole that had been infected with L. major six weeks earlier. Fed females were separated and maintained in the insectary at 26°C on 50% aqueous honey solution. Females were sacrificed 6–11 days after feeding and their gut examined under phase-contrast illumination. In three repeats, a total of 20 P. papatasi females had fed and ten of these were infected. Most infected flies had heavy infections with promastigotes forming a plug in the thoracic midgut, stomodaeal valve area.
The outbreak in Sde-Eliyahu followed a pattern characteristic of CL caused by L. major resulting in a very high percentage (16.3%) of the village population, most of them residing along the perimeter, contracted the disease during the initial five years (Table 1). P. papatasi, the vector of L. major is highly anthropophilic (attracted to humans) and endophilic (enters and feeds inside houses) , , . In contrast, P. (Paraphlebotomus) sergenti Parrot, 1917 that transmits CL caused by L. tropica in the region prefers animals and is much less endophilic , , . Therefore, the efficiency of transmission is greater and the case incidence of CL in L. major foci is usually significantly higher than in L. tropica foci .
Concurrent with the studies reported here, we conducted a sand fly control trial by erecting a tall insecticide-treated barrier along the southern portion of the fence of Sde Eliyahu. The intervention markedly reduced the number of sand flies captured near houses and a significant relief in bite burden, was reported by the residents . Moreover, many of the persons living on the periphery of Sde Eliyahu had already become infected by 2009 consequentially reducing the pool of susceptible individuals available for infection. Therefore, the causes for the decrease in CL cases noted in 2009 should be interpreted with caution.
The Sde Eliyahu focus is ecologically very different from the typical L. major foci throughout the wide geographical distribution of this parasite. Zoonotic L. major foci are normally found in desert or semi arid habitats where chenopod plants and other wild bushes flourish. P. obesus, the most important reservoir host, is highly fastidious, feeding on wild plants and burrowing under their roots , . This rodent does not adapt well to highly modified agricultural environments , , , , . Sde Eliyahu receives at least twice as much rain-fall annually than most other L. major foci and is surrounded by agricultural fields (Fig. 1). Significantly, both P. obesus and M. crassus, the known reservoir hosts of L. major in the Middle East were absent from Sde Eliyahu. On the other hand, innumerable burrows of voles were readily encountered in alfalfa fields, as well as orchards, vegetable gardens, date-palm plantations and olive groves bordering alfalfa fields surrounding the village . Tristram's Jirds were not trapped in the immediate vicinity of the village only in orchards and olive groves some 1–2 km away , , . Leishmania specific ITS1-PCR followed by RFLP showed that a high percentage of both rodent species (16.5% of the voles and 58.3% of the jirds) were exposed to L. major infections. In a PCR survey of wildlife conducted recently M. tristrami from south-western Israel were found infected with L. major in (8/59; 14%) . Eight additional Meriones spp. have been positively incriminated as reservoir hosts of L. major in different parts of this parasite's distribution , , , , . Therefore, finding M. tristrami naturally infected with L. major was not surprising.
The vole M. guentheri is abundant in agricultural areas  and geographically widespread, ranging from Israel's north and west through Syria and Lebanon to Turkey, Greece, Bulgaria and former Yugoslavia . The species is also found eastwards all the way to Iran . Abundant as the species may be, it had not been documented to harbor Leishmania parasites – possibly because of its abstention from entering traps . In our studies we utilized alternative methods for capturing voles by plowing and flooding burrows. These techniques were highly effective but adult-biased since the adult voles were the ones captured escaping from disturbed burrows. In this respect, these approaches may have increased the chances of collecting L. major infected animals.
The ears of laboratory-infected voles were shown to be infectious to P. papatasi females. Ears are the preferred site for development of L. major in infected P. obesus, M. tristrami, M. crassus and apparently in voles as well. The sparse, short hair make ears accessible to probing sand flies, thus increasing chances of infection by bite. The inverse experiments, attempting to infect naïve voles by bite of artificially infected sand flies failed due to reluctance of the infected female P. papatasi to re-feed on the anesthetized voles. We were unsuccessful in obtaining live L. major cultures from wild-caught PCR-positive voles. Despite these somewhat disappointing results, we are reasonably confident in incriminating voles as the main L. major reservoir host in the Sde Eliyahu focus because, M. guentheri were abundant close to houses where CL patients lived and high proportions were PCR-positive for L. major. Moreover, lab experiments confirmed their susceptibility to local L. major (Figs. 5, ,6)6) and infected voles were infectious to local P. papatasi. It is possible that M. tristrami served as bridge-hosts, facilitating the spread of L. major from P. obesus, naturally found in the Central Jordan Valley some 50 km to the south . However, the sparse populations of M. tristrami found only in distant locations (>500 m) from houses' preclude this rodent from being regarded as the main direct reservoir host in the Sde Eliyahu focus.
Incriminating P. papatasi as the vector of L. major in Sde Eliyahu came as no surprise since this phlebotomine species is by far the most important vector of L. major throughout most of the very wide geographic distribution of this parasite , . Moreover, the distribution of P. papatasi far exceeds that of L. major and extends northwards into Turkey, Greece and the Adriatic region, former Yugoslavia, the Balkans, Portugal and other south European countries and the Mediterranean littoral , , , , . The absence of CL caused by L. major from these countries has hitherto been explained by the lack of suitable reservoir hosts. Our findings point to the possibility that L. major, having adapted to voles, will spread northwards into Turkey, central Asia and southern Europe where reservoir hosts and vector exist sympatrically.
Sde Eliyahu farmers practice mostly organic farming. Little-to-no rodenticides or insecticides are used and voles are controlled by encouraging nesting of barn-owls Tyto alba (Scopoli, 1769) (Strigiformes: Tytonidae) and Kestrels Falco tinnunculus Linnaeus, 1758 (Falconiformes: Falconidae) . Sand flies are normally highly susceptible to insecticides . Perhaps because insecticides are only used sparingly in Sde Eliyahu, sand flies were very abundant in agricultural fields and plantations (Fig. 3). It is possible that local organic farming practices combined with biological rodent control provided the setting for the intensive interaction of voles and sand flies, thereby, facilitating the adaptation of L. major to voles, and the emergence of CL in Sde Eliyahu. Notably, human CL cases have now spread to near-by agricultural communities, none of which practice organic farming.
We thank E. Klement for help with statistical analyses, M. Freund, O. Kirstein, H. Guetta and A. Moncaz for assisting with the field work and manuscript preparation, Alex Shtern for the sand fly colony and Svetlana Vaisman who reared the wild-caught rodents. We also wish to thank residents of Kibbutz Sde Eliyahu, and Kibbutz Tirat Tzvi. C.L.J. holds the Michael and Penny Feiwel Chair in Dermatology.