A Systematic Review (1990–2021) of Wild Animals Infected with Zoonotic Leishmania

Leishmaniasis are neglected diseases caused by several species of Leishmania that affect humans and many domestic and wild animals with a worldwide distribution. The objectives of this review are to identify wild animals naturally infected with zoonotic Leishmania species as well as the organs infected, methods employed for detection and percentage of infection. A literature search starting from 1990 was performed following the PRISMA methodology and 161 reports were included. One hundred and eighty-nine species from ten orders (i.e., Carnivora, Chiroptera, Cingulata, Didelphimorphia, Diprotodontia, Lagomorpha, Eulipotyphla, Pilosa, Primates and Rodentia) were reported to be infected, and a few animals were classified only at the genus level. An exhaustive list of species; diagnostic techniques, including PCR targets; infected organs; number of animals explored and percentage of positives are presented. L. infantum infection was described in 98 wild species and L. (Viania) spp. in 52 wild animals, while L. mexicana, L. amazonensis, L. major and L. tropica were described in fewer than 32 animals each. During the last decade, intense research revealed new hosts within Chiroptera and Lagomorpha. Carnivores and rodents were the most relevant hosts for L. infantum and L. (Viannia) spp., with some species showing lesions, although in most of the studies clinical signs were not reported.


Introduction
Included in the group known as neglected tropical diseases, the leishmaniases are a group of diseases caused by flagellated protozoan parasites from more than 20 species belonging to the genus Leishmania. The disease can appear with a variety of clinical pictures, depending on the species involved, the geographic region and the response of the host. Most people and animals infected by the parasite do not develop symptoms but, if present, the disease can follow three basic clinical forms in humans: cutaneous, mucocutaneous and visceral, while in animals, only cutaneous and/or visceral forms are observed. Leishmaniasis is a vector-borne disease transmitted by phlebotomine sand flies (order Diptera, family Psychodidae) with a worldwide distribution (Europe, Africa, the Americas, Asia, and one species recently described in Australia) and an endemic presence in more than 90 countries [1]. There are an estimated 350 million people at risk of Leishmania infection. The World Health Organization (WHO) estimates more than one million new cases per year in people around the world, making it one of the most relevant yet neglected parasitic diseases (https://www.who.int/leishmaniasis/en/, accesed on 8 January 2021). Leishmaniasis is one of the leading causes of morbidity in the world among infectious diseases and one of the leading causes of death among tropical diseases [2]. The disease is present in 98 of the 200 countries that collaborate with the WHO, and information is regularly updated by the organisation [3]. At least 39 described species of Leishmania can be found in the literature, in addition to a significant number of informal or incomplete All species of the genus follow a biological cycle with the same pattern, alternating amastigote forms that replicate intracellularly in the vertebrate host, and promastigote forms that reproduce in the digestive system of the insect vector ( Figure 1). Sand flies (mainly Phlebotomus and Lutzomya genera) become infected while feeding on a parasitised reservoir. Through the bite, they ingest macrophage-bearing blood and tissue with amastigotes. Natural vectors have been experimentally proven to be highly susceptible, and one or two parasites are sufficient to initiate an infection [7]. For a species of sand fly to be a vector of zoonotic Leishmania, it must meet five conditions: (1) be anthropophilic; (2) feed from reservoir hosts in cycles of zoonotic transmission; (3) be infected in nature with the same Leishmania species that infects humans; (4) harbour the complete development of the parasite until it becomes infective; (5) be capable of transmitting the parasite through the bite [8].
Other infectious routes, such as venereal and vertical transmission, have been proved for L. infantum in a canine host [9], and it is seriously considered in humans [10]. Biting is a route suspected for canids [11], and the oral route has been confirmed in hamsters [12], which has been suggested to be associated with the ingestion of micromammals by common genets and servals [13], or the ingestion of phlebotomines by insectivorous bats [14]. In another study, the presence of Leishmania amastigotes and promastigotes in the faeces of gorillas have been reported [15]. Indeed, translocation of bacteria from the gut to Table 2. Wild animals reported infected with zoonotic Leishmania (Viannia) species. Organs or tissues where the parasite was detected are indicated, as well as the techniques employed for detection. L (Viannia) species are as follows: Lb: L. braziliensis, Lg: L. guyanensis, LVsp: Leishmania (Viannia) spp., Lsp: Leishmania sp., Lpa: L. panamensis, Lpe: L. peruviana, Ln: L. naiffi, Ls: L. shawi and Ll: L. lainsoni.
3.3. Wild Animals Infected with Leishmania amazonensis L. amazonensis was described in countries from Central and South America, where data were available, including Costa Rica, Panama, Venezuela, Colombia, Ecuador, Peru, Argentina, Uruguay, French Guiana, Surinam, Brazil and Bolivia, the last two countries being the most widely studied [3,5]. This review includes updated information of wild animals infected with the parasite from surveys carried out in Brazil, Bolivia and Argentina.
Among carnivores, L. amazonensis was detected only in one of two skunks analysed from a focus of leishmaniasis in Bolivia [41]. More information was retrieved from three studies including several species of bats [14,45,66]. Most of the species analysed were insectivorous bats, but the parasite was also detected in hematophagous (Desmodus rotundus), nectarivorous and omnivorous species. Prevalence values varied from 1% to 25%, probably depending on the geographic area, the species and the organs or techniques employed for detection. Higher values were observed in Sturnira lilium and Eumops auripendulus from urban areas and remnants of primitive forest of Sao Paulo (Brazil), employing nested PCR from liver and spleen (Supplementary Materials File S2).
Primates and opossum have been scarcely reported with L. amazonensis in the last 30 years, but the species was recently detected by PCR of the ITS region in the ear tissue of 2.8% of 209 black howler monkeys (Alouatta caraya) from Argentina [64]. In addition, a clinical case of a spider monkey (Ateles paniscus) from a zoo in Brazil, which showed weight loss and pale mucous membranes, was further confirmed by PCR and RFLP from blood [67]. L amazonensis DNA was also detected in 1.1% of the analysed woolly-mouse opossum (Marmosa paraguayanus) from Brazil [48] (Supplementary Materials File S2).
Three studies from Bolivia and Brazil reported the presence of L. amazonensis in blood or skin (tail or ear) by PCR followed by sequencing, in 7.1-33.3% of the analysed rodents (Hylaeamys, Oryzomys, Akodon, Necromys and Olygoryzomis genera) [35,41,68]. Some of these rodents displayed old lesions including scars on the tail or ear [68].
Animals found infected with L. amazonensis, as weel as the techniues employed, are summarized in Table 3.

Wild Animals Infected with Leishmania mexicana
In this review, data regarding infection with L. mexicana in wild animals were mainly from the USA and Mexico, two countries where the parasite is frequently reported, but data from rodents and carnivores from Brazil and Bolivia were also included (Supplementary Materials File S3). The species was present in other American territories such as Venezuela, Colombia, Ecuador and all countries in Central America [3,5].
According to the data from the systematic review, thirty-one species of wild animals from six orders were found to be infected with L. mexicana. In carnivores, only one out of two Molina's hog-nosed skunks (Conepatus chinga rex) were infected with L. mexicana in Bolivia, and the parasite was isolated by inoculation in hamster and subsequently analysed by isoenzyme analysis and hybridisation [41]. Samples from seven grey foxes (Urocyon cinereoargenteus) were analysed by ELISA, showing 100% prevalence in Mexico [69] (Supplementary Materials File S3). Both species should be considered as sentinel or even reservoirs, due to the parasite's isolation and high values of positivity.
Thirteen species of bats (order Chiroptera) were also found infected with the parasite in Mexico, employing skin, heart, liver and spleen in a PCR of kDNA and SSU [5]. The authors found infection rates ranging from 4-100% of the animals [70].
The order Primates was less explored, and only indirect evidence of the infection was reported by serology (ELISA, IFAT and Western blot). A prevalence of 5-37.5% was found in two species of howler monkeys (Alouatta palliate and A. pigra) in Mexico [80].
The base of the tail was the election site for detection or isolation of L. mexicana in rodents and marsupials, with 100% of infection in many studies in which animals with lesions were sampled [71][72][73]77], but the ear and foot were also included [76] (Supplementary Materials File S3). When other organs were investigated, such as liver, spleen, kidney or heart, they were also infected, but at lower percentages (11-66.7%) [72]. Heart, liver, spleen and skin were also employed to find infections in Chiroptera [70], while lymph nodes, lung, spleen, liver and kidney tissues were used in the northern anteater, with DNA detection by PCR only in spleen [79].
In general, wild animals showed mild clinical signs of leishmaniasis and no external signs were reported in the orders Carnivora, Chiroptera, Pilosa and Primates. On the contrary, rodents and marsupials were reported with cutaneous clinical signs in most of the surveys, including swollen skin, depigmentation, ulcers, alopecia and erythema, mainly at the base of the tail. This fact can be explained because the authors were searching for lesions to find reservoirs of the disease [71,72,77].
Wild animals infected with L. mexicana, along with techniques and organs or tissues positive to the parasite are shown in Table 4.  In previous reviews, several species from publications prior to 1990 were reported to be infected with L. mexicana (Agouti paca, Marmosa robinson, Nyctomys sumichrasti, Oryzomis capito, Proechymis spp., Reithrodontomys gracilis) [23,24].

Wild Animals Infected with Leishmania infantum (L. chagasi)
L. infantum is the most globally distributed of all species of zoonotic Leishmania. Australia is considered free of L. infantum, but the protozoan is present on almost all continents with available data, including Southern Europe, Africa, Asia and the Americas from north (excluding Alaska and Canada) to south. African countries and Brazil report more than 90% of the human VL cases around the world, but detailed characterised focusses are more frequently reported in Brazil and Mediterranean countries (North Africa and South Europe) [3,5].
Techniques employed to detect infection with L. infantum in different parts of the world are similar to those previously described for L. braziliensis. Serology was mainly employed in carnivores, primates and occasionally in marsupials or other species, such as rodents or Lagomorpha (Supplementary Materials Files S4 and S5), while the rest of the species were examined preferentially by PCR. Among the serological techniques, ELISA, IFAT, DAT or rapid test (rK39) were extensively employed. The most frequent target, again, was kDNA, but other targets, such as SSU and ITS1 and the repeat region, were also used in several studies and animal species. Less frequently used targets include cytochrome B (Cyt B), HSP70, ITS2, glyceraldehyde phosphate hydrogenase (GAPDH) and α-tubulin (for details, see Supplementary Materials File S4). Xenodiagnosis or culture were employed only in a few occasions.
Blood, skin, liver and spleen were the most employed tissues for PCR detection, but heart, lungs, lymph nodes, intestines, kidney and bone marrow were also used in several studies. Blood was more frequently employed in carnivores, marsupials and primates, because it is easier to obtain, while other tissues were accessible only during post-mortem examinations or after fatal clinical cases or euthanasia of the animals. This was the case for rodents, some bats, several clinical cases of carnivores, and road-killed mammals. Hair and eye swabs were also successfully tested in some studies (Supplementary Materials Files S4 and S5).

L. infantum in the Americas
According to the literature, eight orders of wild animals are infected with L. infantum in the Americas: Carnivores, Chiroptera, Cingulata, Didelphimorphia, Lagomorpha, Pilosa, Primates and Rodentia (Supplementary Materials File S4). Carnivores were the most widely studied, mainly because domestic and wild carnivores are considered reservoirs of the disease, but also because clinical cases are more frequently reported in them, both in nature and in zoological parks [24,25]. The crab-eating fox (Cerdocyon thous) is a widespread carnivore in South America that can act as a reservoir of leishmaniasis for humans, since it can be found in forest locations as well as in residential areas. It was found to be infected with L. infantum or exposed to the parasite (positive serology) in several publications, some of which were clinical cases, and thus were not considered in this study. The percentage of infection varied widely among the studies when including more than one animal (4-75%), and exposure to the parasite was demonstrated by serology (i.e., ELISA, IFAT), while culture, microscopy of smears, PCR followed by sequencing and inoculation of hamsters were employed to detect the parasite [81][82][83][84][85][86][87][88][89]. Several organs and tissues tested positive via PCR: bone marrow, heart, lymph nodes, liver, lungs, skin and spleen. Mainly serological test were employed in the maned wolf (Chrysocyon brachyurus) with prevalence values from 10% to 75% depending on the study [82,84,85,[88][89][90], while the parasite was found only in bone marrow and skin by PCR [84,85]. In the bush dog (Speothos venaticus), several techniques were employed including serology, culture of isolates, PCR, smears and histopathology. While most studies reported results from only one or two animals, only three studies analysed a higher number (4-6) and found 33.3% positives using PCR (blood) and 60-100% of the animals positive by serology [84,85,[89][90][91] (Supplementary Materials File S4). The potential transmission to the vector was demonstrated in manned wolves and bush dogs [89], which reinforces their role as reservoirs.
The DNA of L. infantum was found in at least 17 species of bats in nine studies, including one hematophagous species and several frugivorous, omnivorous or insectivorous ones [14,32,44,45,66,[95][96][97][98]. The feeding habits of the animals were relevant, since the oral route was suggested for transmission in animals feeding on insects, including the vector of leishmaniasis [14]. They also shared the same ecological niche in bat caves and probably other locations. Values of infection varied widely, from less than 1% to 100% of the analysed bats, being infected mainly in the blood but also in the liver, skin, spleen and even in oral swab samples [32]. PCR followed by RFLP or sequencing was employed in the studies and, when sensitive primers were employed, a prevalence higher than 30% was usually obtained [95].
A small number of species of the orders Cingulata and Pilosa were found to be infected with L. infantum in Brazil. The lesser anteater (Tamandua tetradactyla) was reported to test positive by PCR (kDNA) in blood and bone marrow in 2013 [99] and again in 2014, together with giant anteaters (Myrmecophaga tridactyla) and one seven-banded armadillo (Dasypus septemcinctus) found dead on the roads in Brazil, employing PCR from several tissues [86].
Marsupials were studied in several surveys by PCR or serology (Supplementary Materials File S4). The white-eared opossum (Didelphis albiventris) was analysed by six groups in Brazil, who found the parasite in blood, bone marrow, lungs, kidney, skin and spleen by culture or PCR (kDNA, ITS1 or SSU) and sequencing or RFLP [38,39,50,54,86,100,101], with percentages of infection between 6.3% and 22.2%. The big-eared opossum (Didelphis aurita) was positive at a low percentage in Brazil by PCR, spleen imprints and serology (rK39), and one of the animals displayed spleen enlargement, but no other clinical signs were recorded from the rest [102]. In Brazil as well as in Colombia and Venezuela, the common opossum (Didelphis marsupialis) was widely analysed [102][103][104][105][106][107]. Two studies in Colombia demonstrated the transmission of isolates from common opossums to hamster, highlighting their role as reservoirs of L. infantum [105,106]. The parasite was found in several tissues employing PCR (kDNA, SSU and ITS1) followed by hybridisation or RFLP [38,103,107]. Two studies in Brazil employed serology and PCR simultaneously. In the first one, the authors found 9-21.6% of the animals positive using serology, and only 5% positive by PCR-RFLP [38], while the other study analysed 112 individuals of two species (i.e., whiteeared and big-eared opossums), and found high percentages of positivity (71-91.6%) with both techniques (see Supplementary Materials File S4 for details).
Lagomorphs were scarcely reported as exposed to L. infantum in the Americas, with one European hare (Lepus europaeus) found positive in Brazil by DAT and with a low antibody titre (1:320) [92].
Infection with L. infantum in primates were studied in five surveys, and eleven species were reported with DNA of the parasite. Several species of captive primates showed high prevalence values when employing PCR (kDNA) in an endemic area of Brazil including brown howler monkeys (Alouatta guariba), black-headed night monkeys (Aotus nigriceps), black-fronted titi (Callicebus nigrifons), golden-bellied capuchin (Cebus xanthosternos), golden-headed lion tamarin (Leontopithecus chrysomelas), bald-faced saki (Pithecia irrorata) and emperor tamarin (Saguinus imperator). Among them, one black-fronted titi was found dead with clinical signs compatible with leishmaniasis, but the rest did not show clinical signs [108]. On the other hand, free-ranging howler monkeys (Alouatta caraya) sampled at the marginal area of an endemic region from Argentina displayed low values of prevalence (6.3%) [64]. Two other studies found indirect evidence of infection with the parasite using DAT in one white-tufted-ear marmoset (Callithrix jacchus) [109] and 26.9% of the black-tufted marmosets (Callithrix penicillata) [109]; the last study also employed PCR of the skin. Positive serology was detected in twenty-two percent of red howler monkeys (Alouatta seniculus) in French Guiana, and data were further confirmed by PCR (110).
Rodents occupied most of the attention of researchers investigating L. infantum in the Americas, and twelve studies fulfilled the inclusion criteria of this systematic review [21,36,39,50,53,58,86,92,101,106,107,110]. Most of the studies employed different PCR approaches in several tissues, although in one study the authors detected antibodies. Brazilian guinea pigs were reported to be infected in Brazil by PCR in heart tissue [86]. The infection was also found in two species of porcupines from Brazil: the prehensile tailed porcupine (Coendu-Sphiggurus villosus) by serology (DAT) [92], which is indirect evidence of the parasitism, and the Paraguayan hairy dwarf porcupine (Coendou-Sphiggurus spinosus) by PCR and sequencing from several tissues Supplementary Materials File S4) [86]. Agouties were also reported to test positive for L. infantum in the spleen (16.7%), skin and blood by PCR [36,110]. The giant rodent capybara (Hydrochoerus hydrochaeris) was positive in the lungs by PCR and sequencing [86]. Several species from wild mice, rats and cricetidae of the genera Cerradomys, Clyomis, Holochilus, Hylaeamys, Nectomys, Oryzomys, Proechymis, Rhipidomys and Trichomys were reported to be infected with L. infantum in several surveys, and in previous studies the potential role as reservoir of some of them was indicated [24]. The authors employed distinct approaches of PCR followed by RFLP, hybridisation or sequencing [21,36,39,50,58,106,110].
Synanthropic rodents, such as the house mouse (Mus musculus), the black rat (Rattus rattus) and the brown rat (Rattus norvegiccus) were investigated in Brazil and Venezuela. Researchers found 20% of house mice to be infected [50], while the prevalence in black rats varied widely, with values from 0.1% to 100% using several approaches of PCR followed by hybridisation, RFLP or sequencing [21,39,50,53,58,107]. Almost 17% of brown rats were positive by nPCR and sequencing [50]. These synanthropic species of rodents could act as relevant reservoirs of leishmaniasis, since they were infected at high percentages and share habitats with humans.
Additional species were analysed in other surveys or in some of the previous studies in which the authors could not characterise the parasite at the species level (Supplementary Materials File S4) [44,91,[111][112][113][114][115][116][117][118]. Indirect evidence of Leishmania spp. was found in the USA using rapid tests (rk30 antigen) in several wild carnivores such as coyotes (Canis latrans, 1.6%), American red foxes (Vulpes fulvus, 9.1%) and grey foxes (Urocyon cinereoargenteus, 2%) [111,112]. On the other hand, DNA of Leishmania spp. was detected in several species, including carnivores, such as South American grey foxes (Lycalopex-Pseudalopex griseus) [113] and Neotropical otters (Lontra longicaudis), and some species of primates, such as black-headed night monkeys (Aotus nigriceps), black-bearded sakis (Chiropotes satanas) and grey-woolly monkeys (Lagothrix cana) in Brazil [114]. Among rodents, Leishmania spp. DNA was reported in the blood of a red-tailed squirrel (Sciurus granatensis) in Venezuela [103]. All this information is summarized in Table 5.     The following species have been reported to be infected with L. infantum in the Americas in previous reviews: the rodent Proechymis spinosus, the Brazilian porcupine (Coendu prehensilis) and the fennec fox (Vulpes zerda). More information can be retrieved from the abovementioned reviews [23][24][25]27] L. infantum infection was reported in a large list of wild carnivores including 18 species. The golden jackal (Canis aureus) was positive in four studies at low percentages. The lowest values were found in Georgia using a rapid test (2.6%) [120] and in Romania employing PCR and sequencing of the ITS1 region from bone marrow samples (3%) [121]. In Iran, 11.6% of the jackals were found positive by serological test (DAT), and subsequent smears and culture from skin lesions, lymph nodes, spleen and liver were furthered characterised by PCR and sequencing [122]. Only 7.8% of the animals were found positive by PCR using blood samples [123]. The target or the technique used in each study could influence the results obtained by the different authors.
The wolf was the focus of research of many zoonoses including L. infantum infections. In this review, nine studies reported the animal to be positive for the parasite, although three of them included a low number of animals (three or less). In Croatia, only one wolf was reported positive by PCR and sequencing of the cysteine protease B in lymph nodes [124]. The remaining authors employed PCR of the kDNA region to detect the infection and, in some cases, RFLP and sequencing were later applied. Prevalence values from 33% to 50% were found in Spain when using spleen, skin or lymph nodes as samples [125][126][127][128][129] including one study conducted in a non-endemic region [127]. The percentage of infected animals was lower when blood (9%) [130] or hair (4.1%) [131] were used to detect the parasite. In Italy, 25% of the animals were infected when samples from spleen were analysed [132]. Skin lesions were reported only in one study [127].
A smaller number of individuals from wild cats (Felis silvestris) [127,128,133] and genets (Genetta genetta) [125,[127][128][129]133,134] tested positive to L. infantum in samples from skin, liver or spleen employing PCR of the kDNA. The percentage of infection in both species reached 100% of the sampled animals, although wild cats showed 25% as the minimum value [128,133], while genets displayed a 10% prevalence in blood or spleen samples [134]. In two studies, additional PCR and sequencing of the ITS2 region were carried out [129,133].
Several studies were conducted employing similar approaches to detect L. infantum in various species of carnivores. The parasite was found in the spleen or blood of 28.6% of Egyptian mongoose (Herpestes ichneumon) from Spain [125] and 4.7% (only spleen) from Portugal [135]. Seventy percent of otters' (Lutra lutra) spleens [136] and 25% of Iberian lynxes' (Lynx pardinus) samples (spleen and blood) [135] were found to be infected in two studies. Six surveys reported the presence of L. infantum in hair, liver, spleen or lymph nodes of 29-100% of sampled beech martens (Martes foina) in Spain [127][128][129]131,133,137]. Values between 30% and 62% were found in pine martens (Martes martes) using the same techniques [127,133,134]. Badgers (Meles meles) were found infected in the liver or spleen in Italy (53%) [132] and Spain (26%) [133]. European minks (Mustela lutreola) were found to be infected with values of 50% in Spain using a similar methodology [133] but at lower percentages (e.g., 2.1%) when ELISA or PCR of the ITS1 were employed in Greece [136]. Moreover, 20% of pole cats (Mustela putorius), 45% of tigers (Panthera tigris) in a zoo, 20% of red squirrels (Sciurus vulgaris) and one individual of each species of American mink (Mustela vison) and brown bear (Ursus arctos) were reported to be infected with the protozoa [129,133,137,138].
The DNA of L. infantum was found in at least 17 species of bats in nine studies, including one hematophagous species and several frugivorous, omnivorous or insectivorous ones [14,32,44,45,66,[95][96][97][98]. The feeding habits of the animals were relevant, since the oral route was suggested for transmission in animals feeding on insects, including the vector of leishmaniasis [14]. They also shared the same ecological niche in bat caves and probably other locations. Values of infection varied widely, from less than 1% to 100% of the analysed bats, being infected mainly in the blood but also in the liver, skin, spleen and even in oral swab samples [32]. PCR followed by RFLP or sequencing was employed in the studies and, when sensitive primers were employed, a prevalence higher than 30% was usually obtained [95].
A small number of species of the orders Cingulata and Pilosa were found to be infected with L. infantum in Brazil (Supplementary Materials File S4). The lesser anteater (Tamandua tetradactyla) was reported to test positive by PCR (kDNA) in blood and bone marrow in 2013 [99] and again in 2014, together with giant anteaters (Myrmecophaga tridactyla) and one seven-banded armadillo (Dasypus septemcinctus) found dead on the roads in Brazil, employing PCR from several tissues [86].
Marsupials were studied in several surveys by PCR or serology (Supplementary Materials File S4). The white-eared opossum (Didelphis albiventris) was analysed by six groups in Brazil, who found the parasite in blood, bone marrow, lungs, kidney, skin and spleen by culture or PCR (kDNA, ITS1 or SSU) and sequencing or RFLP [38,39,50,54,86,100,101], with percentages of infection between 6.3% and 22.2%. The big-eared opossum (Didelphis aurita) was positive at a low percentage in Brazil by PCR, spleen imprints and serology (rK39), and one of the animals displayed spleen enlargement, but no other clinical signs were recorded from the rest [102]. In Brazil as well as in Colombia and Venezuela, the common opossum (Didelphis marsupialis) was widely analysed [102][103][104][105][106][107]. Two studies in Colombia demonstrated the transmission of isolates from common opossums to hamster, highlighting their role as reservoirs of L. infantum [105,106]. The parasite was found in several tissues employing PCR (kDNA, SSU and ITS1) followed by hybridisation or RFLP [38,103,107]. Two studies in Brazil employed serology and PCR simultaneously. In the first one, the authors found 9-21.6% of the animals positive using serology, and only 5% positive by PCR-RFLP [38], while the other study analysed 112 individuals of two species (i.e., whiteeared and big-eared opossums), and found high percentages of positivity (71-91.6%) with both techniques (see Supplementary Materials File S4 for details).
Lagomorphs were scarcely reported as exposed to L. infantum in the Americas, with one European hare (Lepus europaeus) found positive in Brazil by DAT and with a low antibody titre (1:320) [92].
Infection with L. infantum in primates were studied in five surveys, and eleven species were reported with DNA of the parasite. Several species of captive primates showed high prevalence values when employing PCR (kDNA) in an endemic area of Brazil including brown howler monkeys (Alouatta guariba), black-headed night monkeys (Aotus nigriceps), black-fronted titi (Callicebus nigrifons), golden-bellied capuchin (Cebus xanthosternos), golden-headed lion tamarin (Leontopithecus chrysomelas), bald-faced saki (Pithecia irrorata) and emperor tamarin (Saguinus imperator). Among them, one black-fronted titi was found dead with clinical signs compatible with leishmaniasis, but the rest did not show clinical signs [108]. On the other hand, free-ranging howler monkeys (Alouatta caraya) sampled at the marginal area of an endemic region from Argentina displayed low values of prevalence (6.3%) [64]. Two other studies found indirect evidence of infection with the parasite using DAT in one white-tufted-ear marmoset (Callithrix jacchus) [109] and 26.9% of the black-tufted marmosets (Callithrix penicillata) [109]; the last study also employed PCR of the skin. Positive serology was detected in twenty-two percent of red howler monkeys (Alouatta seniculus) in French Guiana, and data were further confirmed by PCR (110).
Rodents occupied most of the attention of researchers investigating L. infantum in the Americas, and twelve studies fulfilled the inclusion criteria of this systematic review [21,36,39,50,53,58,86,92,101,106,107,110]. Most of the studies employed different PCR approaches in several tissues, although in one study the authors detected antibodies. Brazilian guinea pigs were reported to be infected in Brazil by PCR in heart tissue [86]. The infection was also found in two species of porcupines from Brazil: the prehensile tailed porcupine (Coendu-Sphiggurus villosus) by serology (DAT) [92], which is indirect evidence of the parasitism, and the Paraguayan hairy dwarf porcupine (Coendou-Sphiggurus spinosus) by PCR and sequencing from several tissues (Supplementary Materials File S4) [86]. Agouties were also reported to test positive for L. infantum in the spleen (16.7%), skin and blood by PCR [36,110]. The giant rodent capybara (Hydrochoerus hydrochaeris) was positive in the lungs by PCR and sequencing [86]. Several species from wild mice, rats and cricetidae of the genera Cerradomys, Clyomis, Holochilus, Hylaeamys, Nectomys, Oryzomys, Proechymis, Rhipidomys and Trichomys were reported to be infected with L. infantum in several surveys, and in previous studies the potential role as reservoir of some of them was indicated [24]. The authors employed distinct approaches of PCR followed by RFLP, hybridisation or sequencing [21,36,39,50,58,106,110].
Synanthropic rodents, such as the house mouse (Mus musculus), the black rat (Rattus rattus) and the brown rat (Rattus norvegiccus) were investigated in Brazil and Venezuela. Researchers found 20% of house mice to be infected [50], while the prevalence in black rats varied widely, with values from 0.1% to 100% using several approaches of PCR followed by hybridisation, RFLP or sequencing [21,39,50,53,58,107]. Almost 17% of brown rats were positive by nPCR and sequencing [50]. These synanthropic species of rodents could act as relevant reservoirs of leishmaniasis, since they were infected at high percentages and share habitats with humans.
The infection was demonstrated also in Bennett's wallabies in a zoo in Madrid (Spain) using serology (rk39) and PCR followed by sequencing of the ITS1 and ITS2 regions. Thirtythree percent of the animals were infected in several tissues including blood, kidneys, lymph nodes, liver, lungs, skin and spleen [147]. In addition, two North West Bornean orangutan (Pongo pygmaeus) with clinical signs of visceral leishmaniasis were found to be infected using PCR (ITS1) of the bone marrow and serology [148].
Over the last decade, new reservoirs of leishmaniasis in Europe have been reinforced through investigation with bats, lagomorpha and hedgehog species, enlarging the list of wild animals infected with the parasite. Only one study in Europe demonstrated the presence of L. infantum in spleen, hair and blood of 51.9% common urban bats (Pipistrellus pipistrellus) in Spain, using PCR and sequencing of the repeat region [149]. One hundred percent of Algerian hedgehogs were found infected in two studies in Tunisia [150,151]. The authors employed smears, PCR and RFLP and sequencing of several targets. Spleen, liver, kidney, heart, lymph nodes, blood and eye swabs were positive to all of the techniques employed. In Spain, the European hedgehog was found to be infected using ELISA and qPCR of the kDNA region, finding higher values of infection in the spleen than in skin samples [137]. The parasite was also detected in a hair sample from one animal [131].
The DNA of L. infantum was first detected in 2013 in the spleen of European hares (Lepus europaeus) and Iberian hares (Lepus granatensis) from Spain, with 43.6% of the animals infected [152], since the outbreak of human leishmaniasis in Madrid motivated research on wild reservoirs. Since then, other studies were conducted finding the parasite in spleen, hair and skin with molecular (PCR of several targets), histological (direct antibody fluorescence assay) and serological analyses (IFAT) in Spain [153,154], Italy [155,156] and Greece [136]. Similar approaches were used for the European rabbit (Oryctolagus cuniculus), which was investigated in six studies from Spain [128,129,153,154,157,158]. The authors found positive values ranging from 0.6% to 59%, depending on the time of the year, the sampling area or the techniques employed (i.e., smears, ELISA, IFAT, PCR). In general, serology displayed lower percentages of infection than PCR. In Italy and Greece, lower values of infected animals were obtained, but as previously pointed out, this might be due to the techniques employed (serology or PCR of ITS) [136,145].
Wild and synanthropic rodents were always the focus of researchers interested in L. infantum epidemiology, similar to what happened with other species of Leishmania. In Spain, the wood mouse (Apodemus sylvaticus) was positive with values of 20-50%, depending on the study, in several tissues [128,129,159]. The authors employed culture and smears [159], but also PCR followed by RFLP or sequencing of ITS1, ITS2, kDNA and SSU regions. Blood and spleen from shrews (Crocidura russula) were found to be infected at a low prevalence [160] as well as Algerian mice (Mus spretus), in percetanges from 4.3% to 42.9% using PCR in several tissues as well as serology [137,160]. In Iran, 39% of the sampled shor-tailed bandicoot rat (Nesokia indica) were positive in smears of skin and spleen, and later characterised as L. infantum by nested PCR of the kDNA when compared with reference isolates [161].
Synanthropic rodents were studied in eight surveys displaying high prevalence values for Leishmania infection. In most of them, L. infantum was found in the skin, liver and spleen of house mice (Mus musculus) and brown rats (Rattus norvegicus) in Portugal and Morocco [162,163], with animals displaying skin lesions in both studies. Smears were employed in Portugal while PCR and sequencing were employed in both studies, although different targets were selected (kDNA, ITS1, SSU). Prevalence values ranged from 22% to 33.3% in mice and 33% in rats. In Spain, 50% of house mice were found to test positive using PCR and sequencing of blood, bone-marrow and skin samples [159]. Brown rats were found infected using similar methologies, with 33-100% of animals positive in Spain [129,131] and 5.9% in Greece [164]. Black rats (Rattus rattus) were found infected in 18.2% of sampled animals in Saudi Arabia [163], 15.5% in Italy [165], 33.3% in Spain [157] and 7.5% in Morocco [161].
3.6. Wild Animals Infected with L. major L. major infections extended through Asian and African countries [3,5], and nineteen studies were carried out from 1990 in wild animals in Algeria, Tunisia, Iran, Israel, Ethiopia, Kenya, Cameroon and Morocco (Supplementary Materials File S5).
In the included studies, the most employed techniques to detect L. major were PCR of the kDNA (nine studies) and ITS (nine studies) regions, although SSU and the repeat regions were also employed, mainly followed by RFLP and/or sequencing (four and two studies, respectively). Only seven studies exclusively employed the skin to search for the parasite, but the rest of the studies employed also other anatomical sites such as liver, spleen, heart, blood, kidney, lymph nodes, eye swabs or even feces. Noteworthy were the higher prevalence values observed in smears compared with the PCR of the kDNA in some studies, probably due to the methodology employed, since the DNA was extracted from fixed smears [161,[167][168][169][170][171] (for details, see Supplementary Materials File S5).
The infection was demonstrated in the orders Chiroptera, Eulipotyphla, Primates, and Rodentia, the last group, again, being the most widely studied. Only one species of bat (Nycteris hispida) was reported in Ethiopia to be infected with L. major in the spleen by qPCR and sequencing of the kDNA and ITS regions [172]. The DNA of the parasite was found in several organs and tissues in three species of hedgehogs in Algeria, Iran and Tunisia including spleen, skin, heart, kidney, liver, blood and eye swab. A hundred percent prevalence was reported in two studies carried out on the Algerian hedgehog (Atelerix algirus) in Tunisia [150,151], while 36.8% was reported in Algeria by serology and PCR-RFLP of the kDNA region employing the spleen and skin [173] (Supplementary Materials File S5). Two studies, including the long-eared hedgehog (Hemiechinus auritus), reported prevalence rates ranging from 33%, using nPCR of the ITS and smears of the skin [173], to 53.3%, employing nPCR of the kDNA from the skin and smears from skin, liver and spleen [174]. The desert hedgehog (Paraechinus aethiopicus) was found to be infected with L. major in two studies. The first study employed qPCR of the kDNA region from spleen and skin as well as serology [175], and the other reported the infection in the kidney, blood, liver, eye swab and lymph node of one animal by qPCR of the kDNA, SSU and repeat regions [151]. The high values of infection found in these animals suggest their role as reservoirs, and these animals should be monitored in endemic areas.
One study investigated the immune response to L. major of three species of primates in Kenya (Cercopithecus mitis, Chlorocebus aethiops and Papio cynocephalus anubis) including humoral (ELISA and Western blot) and cellular responses (lymphoproliferative assay) [176]. The authors include 57-213 individuals per technique and found that 60-77% of the animals were previously exposed to the parasite. Surprinsingly, one study found parasites (amastigote and promastigotes forms) and DNA of L. major in the feces of gorillas (Gorilla gorilla) in Cameroon, and the authors pointed to the ingestion of phlebotomines by the animals [15]. However, no other method to measure exposure (serology) or presence of the parasite in organs was employed.
Eleven species of rodents were reported as infected with the parasite in nine studies from Iran, and another one from Israel (Supplementary Materials File S5), following the same tendency of other zoonotic species of Leishmania included in the systematic review. The higher prevalence was found in Meriones libycus by PCR of kDNA in the skin [177], which is also the species most studied regarding Leishmania infections in Iran [167][168][169][170][171]173,177,178]. Other species of Meriones (M. hurrianae, M. persicus and M tristrami) and Microtus (M. guentheri and M. socialis) were reported to be infected with parasites with values of 5.7-58.3% [161,168,171,179]. Mus musculus was found to be infected with low percentages of infection (2.3-33%) in three studies carried out in Iran and Israel. However, PCR of the ITS region from skin samples, and smears from skin, liver and spleen, were used instead of PCR of kDNA, [161,179]. Nesokia indica was found to be infected in three studies from Iran, ranging from 8% in skin by smears and PCR of the ITS region [168] to values higher than 61% employing smears and PCR of skin, liver and spleen tissues [161,167]. The same techniques (PCR of the ITS and kDNA regions) were employed in three studies in Iran to detect L. major infections in Tatera indica from skin, liver or spleen, with values of prevalence ranging from 3,7 to 50% [168,171,178]. The great gerbil (Rhombomys opimus) was reported infected with the parasite in Iran [168,178] using smears, PCRs and inoculation of hamsters (see Supplementary Materials File S5 for details). The high prevalence found in many of these rodent species points to their reservoir role.
Most of the studies did not find clinical signs in infected animals, or the authors did not look for them; however, skin lesions were recorded in Meriones libycus from Iran [177].
All the information concerning L. major infection in wild animals is summarised in Table 7.

Wild Animals Infected with Leishmania tropica
Four studies conducted in Ethiopia [175,180], Kenya [181] and Egypt [182] demonstrated the presence of L. tropica DNA in the spleen of rodents and bats. One hundred and sixty-three bats (Cardioderma cor) were analyzed in Ethiopia by qPCR of the kDNA and ITS1 regions and 4.9% were found to be infected [175]. Using the same techniques, the authors found prevalences from 9.9% to 20% in the rodents Acomys sp. Arvicanthis niloticus and Gerbillus nanus [180] (Supplementary Materials File S5). In Egypt, L. tropica was found in 14.3% of the analyzed Anderson's gerbils (Gerbillus andersoni) with clinical signs by smears and PCR of the ITS1 region [182]. Finally, employing nested PCR and sequencing of the SSU and the ITS1 regions, 22% of the sampled house mice (skin) were found infected with L. tropica in Morocco [163].
The species Procavia capensis and Arvicanthis niloticus were reported in previous reviews infected with L. tropica [23].

Wild Animals Infected with Leishmania donovani
In Africa, rodents were infected with L. donovani in three studies. One employed serology and found 5.5% of African grass rats (Arvicanthis niloticus) to be positive by ELISA in Sudan [183], while another study found 18.2% of black rats (Rattus rattus) with clinical signs carrying the parasite in Saudi Arabia [164], employing culture, smear and inoculation of hamster. The other study found DNA of the L. donovani complex in the spleen in 15.3% of Mastomys erythroleucus, 7.7% of Gerbilliscus nigricaudus and 17.4% of Arvicanthis niloticus from Ethiopia using PCR of the kDNA and ITS regions [180].
Finally, one study obtained a prevalence of 23.5% of the L. donovani complex in European hares from Greece by nested PCR of the ITS region, employing the spleen of 166 animals [184] (Supplementary Materials File S5).
Details of infection with L. donovani of Felis serval can be found in a previous review [23].
Some of the previously mentioned studies obtained positive results for Leishmania spp., but they could not further determine the species (see Supplementary Materials File S5). The spleen of other four species of bats (Glauconycteris variegate, Miniopterus arenarius, Neoromicia somalica and Scotophilus colias) were found to be infected with Leishmania spp. in Ethiopia by PCR [175]. In a similar approach, 40% of the analyzed rodents of Aethomys spp. were also positive [180]. Smears of liver and spleen and indirect haemagglutination test were employed to detect the parasite in 40% of the sand cat (Felis margarita) in Saudi Arabia [185]. The same techniques were employed to detect Leishmania spp. infections in Gerbillus pyramidum and Rattus norvegicus in Egypt [186].
In previous reviews, other species of mammals, such as Crycetomys gambianus, Heterohyrax brucei and Dendrohyrax arboreus, were reported to be infected with L. aethiopica [23].

Conclusions
Knowledge of the role of wild animals as suitable hosts or reservoirs of Leishmania zoonotic species is essential in order to apply control measues or monitoring programmes. In this review, a systematic search of wild animals infected with zoonotic species of Leishmania was conducted, starting from 1990 and following PRISMA methodology. One hundred and eighty-nine species of wild animlas from ten orders (i.e., Carnivora, Chiroptera, Cingulata, Didelphimorphia, Diprotodontia, Lagomorpha, Eulipotyphla, Pilosa, Primates and Rodentia) were included in the review. Rodents and carnivores were the orders more widely explored, being the most probable main reservoirs, and also the ones presenting more clinical signs. L. infantum was the most widely distributed species, both geographically and in the range of species, followed by L. (Viannia) braziliensis, but this fact could be due to the more exhaustive investigation on these species.
More studies on the role of infected wild animals are necessary in order to implement specific measures when an outbreak of the disease appears.