超越利什曼原虫（L. infantum）：巴西东北部内脏利什曼病流行区家犬和家猫中利什曼原虫属的血清学和分子学调查。 机翻

Introduction

Leishmaniasis is a complex vector-borne disease caused by protozoa of the genus Leishmania, transmitted by phlebotomine sandflies, and remains a major public health concern in tropical and subtropical regions, affecting both humans and domestic animals. In the Americas, visceral leishmaniasis (VL) is primarily associated with Leishmania infantum [8, 11]. American tegumentary leishmaniasis is associated with several Leishmania species, with Leishmania (Viannia) braziliensis recognized as the predominant species in the Americas and particularly in Brazil, while Leishmania (Leishmania) amazonensis and Leishmania (Viannia) guyanensis are also important etiological agents [16, 44].

These transmission cycles involve interactions among Leishmania species, vertebrate hosts, and diverse phlebotomine sand fly vectors, whose ecological and geographic distributions contribute to distinct regional transmission patterns [34, 38]. In anthropized environments, domestic and peridomestic transmission may overlap with sylvatic cycles, as vertebrate hosts occupying different ecological habitats can contribute to connections between these transmission interfaces [2, 38].

However, the classical clinical–epidemiological distinction is being challenged by evidence of overlapping transmission cycles in peri-urban and anthropized environments. Studies suggest that ecological overlap, host permissiveness, and environmental changes may allow different species to circulate together, complicating the traditional dichotomous classification [25, 50].

In Brazil, visceral leishmaniasis has undergone a marked epidemiological transition over recent decades, shifting from predominantly rural transmission to sustained urban and peri-urban cycles. This expansion reflects the adaptability of Leishmania spp. and their vectors to human-modified environments, facilitating closer contact between vectors, domestic reservoirs, and human populations. Environmental degradation, migration of human and animal populations, and precarious housing are recognized as important factors contributing to parasite persistence and geographic expansion [26, 50].

Climate change may further influence leishmaniasis transmission by altering environmental conditions that affect vector distribution and disease occurrence, with potential changes in the geographic distribution of leishmaniasis in Brazil [23]. Despite the implementation of national surveillance and control strategies, transmission remains intense in several urban centers, particularly in northeastern Brazil, where endemicity persists under complex socio-environmental conditions [26, 50].

Domestic dogs are considered the main urban reservoir of L. infantum; however, growing attention has been given to other domestic species, particularly cats, whose epidemiological role remains incompletely understood [12]. L. infantum and Leishmania braziliensis are considered the most epidemiologically relevant species in canine leishmaniasis in South America. Notably, L. amazonensis has been isolated from dogs with visceral manifestations, highlighting atypical host–parasite interactions and the need for species-level identification. These findings reinforce the need for species-level molecular discrimination in endemic settings, as distinct Leishmania species may coexist in the same area [14]. Importantly, the occurrence of multiple Leishmania species is not restricted to domestic animals, as coinfection has also been documented in small mammals in northeastern Brazil, including the simultaneous detection of L. (Viannia) spp. and L. (L.) infantum in different host species [19].

Beyond visceral leishmaniasis, the increasing domestic transmission of American Tegumentary Leishmaniasis (ATL) has also raised concerns regarding the epidemiological role of dogs [18]. Likewise, there is no consistent evidence supporting the role of cats in the epidemiological chain of the disease, although infected animals are frequently detected in endemic areas [13]. These uncertainties highlight gaps in understanding host participation and transmission dynamics, particularly where multiple Leishmania species may coexist. Accurate diagnosis and species identification are essential, especially in endemic regions, in order to characterize the epidemiological profile of the disease and support effective control strategies [39, 43].

Reports of molecular detection of atypical species in domestic animals have raised questions regarding under-recognized transmission patterns in endemic areas. However, the frequency, distribution, and epidemiological significance of co-circulating Leishmania species in domestic environments remain poorly understood, particularly in urban endemic settings [33].

In this context, São Luís, in northeastern Brazil, represents a relevant epidemiological setting, characterized by sustained visceral leishmaniasis transmission, peri-urban expansion, and environmental conditions that may favor the interaction between vectors, domestic animals, and wildlife. Phlebotomine fauna reported in the metropolitan region includes Lutzomyia longipalpis and Lutzomyia whitmani among the most abundant species, as well as Lutzomyia flaviscutellata, vectors associated with visceral and tegumentary leishmaniasis transmission cycles [30].

Therefore, this study aimed to investigate the serological and molecular occurrence of Leishmania spp. in domestic dogs and cats from an endemic area for visceral leishmaniasis in northeastern Brazil, with special emphasis on identifying potential co-circulation of L. infantum and L. amazonensis. This approach was intended to explore whether the epidemiological context of Leishmania infection in domestic animals in this area could extend beyond the classical urban cycle predominantly associated with L. infantum.

Methods

Ethical Approval

This study was approved by the Ethics Committee on Animal Use of the State University of Maranhão (CEEA/UEMA no. 09/2018) and by the Brazilian National Research Ethics Commission (Plataforma Brasil). Written informed consent was obtained from all animal owners prior to sample collection.

Study Area

The study was conducted in three peri-urban neighborhoods of São Luís (2° 31′ S, 44° 18′ W), Maranhão State, northeastern Brazil — Fumacê (2° 33′ 36.4″ S, 44° 20′ 23.9″ W), Anjo da Guarda (2° 33′ 45.1″ S, 44° 19′ 35.0″ W), and Mauro Fecury (2° 32′ 32.5″ S, 44° 19′ 54.2″ W) (Fig. 1). These neighborhoods are located in the Itaqui-Bacanga region, which comprises residential neighborhoods and major port and industrial activities [1]. The region also presents limitations in urban infrastructure and sanitation, including restricted sewage coverage [40]. The municipality has a tropical humid climate with marked seasonal rainfall and is historically endemic for visceral leishmaniasis [42]. According to the Brazilian Notifiable Diseases Information System (SINAN), 205 confirmed human cases of visceral leishmaniasis were reported among residents of São Luís between 2020 and 2025, indicating continued transmission in the municipality [7].

[图：Maranhão State map, highlighting the city of São Luís and the three studied locations]

Study Design and Sample Size

A cross-sectional study was conducted using convenience sampling of domiciled dogs and cats older than six months.

Sample size for dogs was estimated using the StatCalc module of Epi Info (CDC), considering a finite population. In the absence of previous prevalence data for the canine population in the study area, an expected prevalence of 50% was adopted as a conservative assumption for sample size estimation [49], with a 95% confidence level and 5% precision, resulting in an adjusted sample size of 323 dogs, which was rounded up to 330 animals. The estimated canine population in the study area was approximately 2,035 animals, based on a ratio of one dog per 10 human inhabitants, as previously used in a Brazilian epidemiological study of canine visceral leishmaniasis [3], and a human population of approximately 20,355 inhabitants in the sampled area, according to data provided by the Municipal Health Department.

No formal sample size calculation was performed for cats. Feline sampling (n = 21) was exploratory and resulted in a smaller sample because cats were less frequently available than dogs in the households visited. In addition, feline sampling was constrained by difficulties in restraint and blood collection, and some owners declined blood collection because of concerns about handling-related stress.

Sample Collection

Sampling was conducted from July to October 2020 in the three urban neighborhoods included in the study (Fumacê, Anjo da Guarda, and Mauro Fecury). Households were recruited door-to-door with the assistance of local Community Health Agents. All eligible domiciled dogs and cats older than six months whose owners agreed to participate were included, with no predefined limit on the number of animals sampled per household. Thus, when multiple eligible animals were present in a participating household, all could be sampled. A total of 110 dogs were sampled in each neighborhood, resulting in 330 dogs being included in the study.

Peripheral blood was collected by cephalic or jugular venipuncture. Serum was separated by centrifugation and stored at −20 °C until serological testing. Whole blood aliquots were preserved in ethanol at a 1:1 (v/v) blood-to-ethanol ratio immediately after collection in the field, following Pacheco et al. [29]. The samples were transported under refrigeration in insulated containers with reusable ice packs until arrival at the laboratory and subsequently stored at −20 °C for molecular analyses. Peripheral blood was selected due to its minimally invasive nature, which ensures higher owner compliance in large-scale epidemiological surveys, allowing for a more representative population sampling.

Epidemiological Data

Owners completed a structured questionnaire addressing animal characteristics (age, sex, clinical signs, lifestyle), environmental conditions (presence of organic matter, other animals, previous leishmaniasis cases), and preventive measures. Specifically, the epidemiological variables included age, sex, access to the street, mode of acquisition of the animal, accumulation of organic matter, presence of other animals in the household, presence of another animal with suspected or diagnosed leishmaniasis in the household, history of Leishmania-positive animals or human cases in the household, presence of wild animals around the household, peridomiciliary characteristics, and adoption of preventive and control measures.

The presence of wild animals around the household was recorded based on information reported by the owners. These animals were not necessarily classified as synanthropic. Access to the street was defined as whether the animal was allowed to leave the household premises. Preventive and control measures included the use of insecticide-impregnated collars for dogs, environmental management and removal of organic matter from the peridomicile, insecticide use, installation of screens on doors and windows, use of repellents or mosquito nets, and routine veterinary monitoring of companion animals. Clinical evaluation included the recording of signs compatible with canine leishmaniasis, including cutaneous ulcers, diffuse alopecia, dermatitis, periocular scaling, onychogryphosis, lymphadenomegaly, hepatosplenomegaly, mucosal pallor, keratoconjunctivitis sicca, cachexia and other clinical abnormalities observed during physical examinations. These clinical findings were recorded as epidemiological variables and were not considered diagnostic confirmation of canine leishmaniasis.

Serological Analysis

Indirect immunofluorescence antibody tests (IFAT) were performed for L. infantum and L. amazonensis as previously described [27]. Promastigote forms cultured in vitro were used as antigens. Serum samples were screened at a cutoff dilution of 1:80, as previously adopted for IFAT-based serological detection of anti-Leishmania antibodies in dogs and cats [28, 32]. The same cutoff was applied to both L. infantum and L. amazonensis antigens to standardize the serological comparison between the two antigens. IFAT was selected in accordance with the Brazilian Ministry of Health guidelines available at the time the study was conducted [6]. Positive samples were serially diluted two-fold to determine endpoint titers. Samples were considered positive when clear peripheral fluorescence of promastigotes was observed.

DNA Extraction and PCR Amplification

Genomic DNA was extracted using the PureLink™ Genomic DNA Mini Kit (Thermo Fisher Scientific) following the manufacturer’s instructions. The ITS1 region was amplified using primers described by Schönian et al. [41]. PCR reactions were performed in a final volume of 20 µL containing 12.7 µL of ultrapure water, 2 µL of 10× PCR buffer, 0.8 µL of MgCl₂ (50 mM), 2 µL of dNTPs (2 mM), 0.1 µL of Taq DNA polymerase (5 U/µL; Invitrogen®, Carlsbad, USA), 0.2 µL of each primer (10 µM), and 2 µL of template DNA. The thermal cycling conditions consisted of an initial denaturation at 95 °C for 3 min, followed by 35 cycles of denaturation at 95 °C for 30 s, annealing at 53 °C for 30 s, and extension at 72 °C for 1 min, with a final extension at 72 °C for 5 min. Reference DNA from L. infantum and L. amazonensis, obtained from the Brazilian Trypanosomatid Collection, was included as positive-control material, whereas ultrapure water was used as the negative control to monitor contamination. Samples showing an amplification product of approximately 300–350 bp, corresponding to the expected ITS1 fragment, were considered PCR-positive. Amplified products were visualized on 1.5% agarose gels stained with SYBR Safe.

Digestion of Polymerase Chain Reaction Fragments

To differentiate Leishmania species, all PCR-positive products were subjected to restriction digestion with HaeIII according to Schönian et al. [41]. Each digestion reaction contained 12.5 µL of purified PCR product, 2 µL of 10× reaction buffer, 1 µL of HaeIII restriction enzyme, and ultrapure water to complete the final reaction volume. PCR amplification and restriction digestion were repeated when necessary. Only restriction profiles that were sufficiently clear and interpretable by comparison with the reference controls were used for species characterization. Samples yielding absent, weak, or poorly defined restriction profiles were not assigned to a Leishmania species.

The restriction profiles obtained from the samples were compared with those of L. infantum and L. amazonensis reference controls. Digested fragments were separated on 2% agarose gels and visualized under UV light to determine species-specific banding patterns.

Statistical Analysis

Data were analyzed using EpiInfo version 6.0 (CDC, Atlanta, USA). Associations between categorical variables were evaluated using Chi-square or Fisher’s exact test, as appropriate. Crude odds ratios (ORs) and their respective 95% confidence intervals (95% CIs) were calculated to quantify the magnitude and precision of the associations identified in the univariable analysis. The epidemiological analyses were considered exploratory, and no multivariable adjustment was performed. Statistical significance was set at p < 0.05.

Results

Study Population

A total of 330 domiciled dogs and 21 cats were included in this study. Among dogs, 165 (50%) were male and 165 (50%) were female, and the majority were older than one year (254/330; 76.96%). Among cats, 11 (52.38%) were male, and most were older than one year (16/21; 76.19%).

Serological Findings in Dogs and Cats

Seroreactivity against L. amazonensis antigen was detected in 125/330 (37.9%) dogs and against L. infantum antigen in 76/330 (23.0%). These totals include animals seroreactive to both antigens. Accordingly, among the 330 dogs, 75 (22.7%) were seroreactive only to L. amazonensis antigen, 26 (7.9%) only to L. infantum antigen, and 50 (15.2%) to both antigens (Fig. 2A). Antibody titers among the 125 dogs seroreactive to L. amazonensis antigen and the 76 dogs seroreactive to L. infantum antigen ranged from 1:80 to 1:5,120 (Table 1). Among dogs seroreactive against L. amazonensis antigen, 74/125 (59.2%) were male and 107/125 (85.6%) were older than one year.

[图：Venn diagrams showing seroreactivity to Leishmania infantum and Leishmania amazonensis antigens in dogs ( A ) and cats (]

Clinical signs compatible with canine leishmaniasis were recorded in 16/76 (21.1%) dogs seroreactive to L. infantum antigen, including cutaneous ulcers, diffuse alopecia, dermatitis, onychogryphosis, cachexia, hepatosplenomegaly and lymphadenomegaly.

In cats, seroreactivity against L. amazonensis antigen was detected in 6/21 (28.6%) animals and against L. infantum antigen in 3/21 (14.3%). These totals include animals seroreactive to both antigens. Accordingly, among the 21 cats, 4 (19.0%) were seroreactive only to L. amazonensis antigen, 1 (4.8%) only to L. infantum antigen, and 2 (9.5%) to both antigens (Fig. 2B). Antibody titers among the six cats seroreactive to L. amazonensis antigen and the three cats seroreactive to L. infantum antigen ranged from 1:80 to 1:160 (Table 1). One seroreactive cat showed cutaneous lesions.

Molecular Detection and Species Identification

PCR detected Leishmania DNA in 58/330 (17.6%) dogs and 2/21 (9.5%) cats. Molecular identification via PCR–RFLP targeting the ITS1 region allowed for the differentiation of Leishmania species based on well-established restriction profiles. Species identification was achieved in 9/58 PCR-positive canine samples. The remaining 49 PCR-positive canine samples did not yield restriction profiles of sufficient quality for reliable species assignment and therefore remained uncharacterized at the species level. Eight samples showed restriction fragments characteristic of L. infantum, while one sample demonstrated a mixed restriction pattern, suggesting the presence of genomic fragments from both L. infantum and L. amazonensis.

Among the eight canine samples showing an L. infantum restriction profile, four were seroreactive only to L. infantum antigen and four were seroreactive to both L. infantum and L. amazonensis antigens. The canine sample showing the mixed restriction profile was also seroreactive to both antigens. Both PCR-positive feline samples showed patterns compatible with L. infantum. Both cats were seronegative by IFAT for L. infantum and L. amazonensis.

Integrated Analysis of Serological and Molecular Findings and Associated Factors

Among PCR-positive dogs (n = 58), 14 (24.14%) were also seroreactive to L. infantum antigen, 20 (34.48%) to L. amazonensis antigen, including eight dogs that were seroreactive to both antigens, and 32 (55.17%) were seronegative by IFAT for both antigens.

For L. infantum, univariable analysis showed significant associations between seroreactivity and age greater than one year (OR = 2.27; 95% CI 1.10–4.67; p = 0.023) (Table 2), male sex (OR = 1.81; 95% CI 1.07–3.07; p = 0.025), and the presence of clinical signs compatible with canine leishmaniasis (OR = 2.87; 95% CI 1.42–5.81; p = 0.002).

For L. amazonensis, significant associations were observed with age greater than one year (p = 0.003), male sex (p = 0.009), street access (p = 0.002), and presence of wild animals around the household (p = 0.003) (Table 3).

Discussion

This study investigated the occurrence of Leishmania spp. in domestic dogs and cats from a visceral leishmaniasis–endemic area in northeastern Brazil, focusing on the potential co-circulation of L. infantum and L. amazonensis. In dogs, seroreactivity to both L. infantum and L. amazonensis antigens were detected, while molecular characterization predominantly identified L. infantum, with one sample showing a mixed restriction profile suggestive of the presence of both species.

Serological analysis showed greater reactivity to L. amazonensis than to L. infantum in dogs. Previous studies in Brazil have reported variable seroprevalence rates for L. amazonensis, ranging from 3.5 to 74.8%, depending on the region and diagnostic method employed [20, 39]. In the present study, the higher seroreactivity to L. amazonensis antigen is noteworthy in an area traditionally recognized as endemic for visceral leishmaniasis, although this finding alone cannot be interpreted as evidence of species-specific infection.

The occurrence of distinct serological profiles, including animals with higher titers for L. amazonensis than for L. infantum, is noteworthy. However, serological reactivity alone does not necessarily indicate active or species-specific infection [37], particularly in areas where different Leishmania species may co-exist and serological cross-reactivity may occur.

Molecular detection identified Leishmania DNA in 58/330 (17.6%) canine samples, although interpretable PCR–RFLP profiles were obtained for only 9/58 PCR-positive samples. The remaining samples yielded absent, weak, or poorly defined restriction profiles and were therefore not assigned to a species. Although the use of peripheral blood may pose challenges for molecular detection because of low parasite loads [17, 51], it cannot alone explain the limited PCR–RFLP characterization observed here. Among the successfully characterized samples, L. infantum predominated, while one mixed restriction profile suggested the presence of both L. infantum and L. amazonensis. This finding is noteworthy in the context of seroreactivity to both antigens but should be interpreted cautiously given the limited number of samples characterized at the species level.

The discrepancy between serological and molecular results, including PCR-positive/seronegative animals, may reflect differences in infection stage, immune response, or sample type [15]. Similar discordant findings have been reported in endemic regions [26]. The limited molecular characterization, despite the higher seroreactivity to L. amazonensis antigen, may reflect differences in parasite load, sample type, or other methodological factors and should therefore be interpreted cautiously.

Factors associated with seroreactivity to L. infantum antigen included age greater than one year, male sex, and clinical signs, whereas seroreactivity to L. amazonensis antigen was associated with age greater than one year, male sex, street access, and the reported presence of wild animals around the household. These findings are consistent with previous reports suggesting that greater environmental exposure may be associated with increased contact with infected sand flies [5]. Adult animals may accumulate vector exposure over time, and roaming behavior may increase opportunities for vector contact.

Clinical signs compatible with canine leishmaniasis were observed in a subset of seroreactive dogs, consistent with the variable clinical presentation reported for canine leishmaniasis. Previous research has shown that infection does not necessarily lead to clinical symptoms, and a significant proportion of infected dogs remain asymptomatic [4, 45]. The wide range of clinical manifestations, extending from asymptomatic infections to severe disease, is influenced by the host's immune response and parasite load [46]. When observed in this study, these clinical signs were among the most common reported in canine leishmaniasis.

In cats, seroreactivity was lower than in dogs, and molecular detection was limited to two samples identified as L. infantum. Reports of feline infection by L. amazonensis remain rare in Brazil [9, 47, 48]. The molecular detection of L. infantum in two cats confirms the occurrence of feline infection in the study area, consistent with previous reports indicating that infection in cats is generally subclinical or associated with dermatological symptoms [31]. The small feline sample size in this study limits any inference regarding their epidemiological role, but seroreactivity and molecular detection were observed, and their contribution to transmission dynamics remains a subject for further investigation.

Environmental characteristics reported in the study areas—including limitations in sanitation and sewage infrastructure and the reported presence of animals such as Mus musculus and Didelphis marsupialis around households—may favor vector maintenance [22]. Although entomological surveys were not performed, previous reports have documented the presence of competent sandfly vectors in São Luís [21, 35], supporting the plausibility of active transmission in these peri-urban settings.

The vector component is also relevant when considering the possible occurrence of different Leishmania species in this setting. Lutzomyia longipalpis, the principal vector of L. infantum in the Americas, is considered a permissive sand fly vector because it can support the development of different Leishmania species, whereas selective vectors exhibit more restricted parasite–vector compatibility [36, 52]. Although L. amazonensis is classically associated with Lutzomyia flaviscutellata, its DNA has also been detected in naturally infected Lu. longipalpis in northeastern Brazil [10]. These observations illustrate that vector–parasite associations may be more complex than a strict correspondence between a single sand fly and a single Leishmania species. However, permissiveness under experimental conditions or detection of parasite DNA in field-collected sand flies does not establish their role in natural transmission. Because no entomological survey was performed in the present study, the serological and molecular findings in domestic animals cannot be attributed to any particular vector species.

The animal findings observed in this study should also be considered within the local epidemiological context of human visceral leishmaniasis. The Itaqui-Bacanga region was involved in the urban expansion of visceral leishmaniasis in São Luís, and Anjo da Guarda, one of the neighborhoods investigated here, was among the areas affected by human cases during the urbanization and geographic expansion of the disease [24]. More recent surveillance data indicate that human VL continues to occur in the municipality, with 205 confirmed cases reported between 2020 and 2025 [7]. In this context, the detection of Leishmania DNA in dogs and the predominance of L. infantum among the successfully characterized samples are consistent with the endemic context of VL in São Luís. However, because human infection was not investigated and the available surveillance data refer to the municipality as a whole, no direct relationship between the animal findings and the occurrence of human cases in the three neighborhoods can be established.

Importantly, seroreactivity to L. amazonensis antigen, together with the detection of a mixed PCR–RFLP profile in one dog, suggests the possibility of involvement of different Leishmania species in this epidemiological setting, although the mixed profile alone does not confirm coinfection.

This study has limitations, including the exclusive use of peripheral blood for molecular detection, the low number of feline samples, and the reduced success rate of species identification by PCR–RFLP. Additionally, the absence of tissue sampling and entomological investigation may have limited the detection of parasite diversity and transmission dynamics in the study area.

Because the epidemiological associations were based on univariable analyses, potential confounding among the evaluated factors cannot be excluded. Therefore, these findings should be interpreted as exploratory associations rather than independent risk factors.

In conclusion, seroreactivity to both L. infantum and L. amazonensis antigens was detected in domestic dogs and cats from a visceral leishmaniasis–endemic area, while molecular characterization predominantly identified L. infantum. The higher seroreactivity to L. amazonensis antigen in dogs, together with the mixed PCR–RFLP profile observed in one animal, raises the possibility that Leishmania occurrence in domestic animals in this area may not be restricted to L. infantum. However, these findings should be interpreted cautiously considering potential serological cross-reactivity and the limited species-level molecular characterization. Overall, the results highlight the value of considering different Leishmania species in epidemiological investigations and of combining serological and molecular approaches to better understand Leishmania occurrence in urban endemic settings.