Serological study of feline leishmaniasis and molecular detection of Leishmania infantum and Leishmania braziliensis in cats ( Felis catus )

adriane@ufmg.br Abstract Blood samples and swabs from ocular conjunctiva and mouth were obtained from 64 cats. Of 64 serum samples, 19 were positive for Leishmania antibodies by ELISA (29.80%). Eight cats were positive by PCR (12.5%) in swab samples from mouth and/or ocular mucosa. Poor kappa agreement between serological and molecular results (k = 0.16) was obtained. From five positive PCR samples one was L. braziliensis and four were L. infantum. Phylogenetic analysis performed with the five isolates of Leishmania , showed that samples of L. infantum isolated from the cats were phylogenetically close to those isolated from domestic dogs in Brazil, while the L. braziliensis is very similar to the one described in humans in Venezuela. The study demonstrated that, despite high seropositivity for Leishmania in cats living in the study region, poor agreement between serological and molecular results indicate that positive serology is not indicative of Leishmania infection in cats. Parasite DNA can be detected in ocular conjunctiva and oral swabs from cats, indicating that such samples could be used for diagnosis. Results of phylogenetic analyzes show that L. infantum circulating in Brazil is capable of infecting different hosts, demonstrating the parasite’s ability to overcome the interspecies follows: initial denaturation at 94ºC for 5 min, followed by 35 cycles at 94ºC for 30 sec, 61ºC for 1 min, and 72ºC for 3 min, with a final extension at 72ºC for 10 min. Negative (no DNA) and positive (0.01 and 1.0 pg of Leishmania DNA) PCR controls were included in each set of reactions. Positive control for the PCR reactions included DNA extracted from promastigote forms of the L. infantum 1 (ITS1) Further analyses from PCR positive samples were performed to identify the Leishmania species. DNA extracted from cat samples was used as a template for detection of Leishmania via amplification of a 300-350 bp fragment of an intergenic region in the Leishmania genome (ITS1), using the primers LITSR: 5′-CTGGATCATTTTCCGATG-3′ and L5.8S: 5′-TGATACCACTTATCGCACTT-3′ (El Tai et al., 2000; Schönian et al., 2003). Positive controls for the PCR reactions included DNA extracted from promastigote forms of the following Leishmania strains: L. amazonensis (IFLA/BR/67/PH8), L. braziliensis (MHOM/BR/75/M2903), L. infantum (MHOM/BR/74/PP75), and L. PCR RFLP ITS1 detected Leishmania DNA in five of eight kDNA positive samples. It was not possible to perform PCR analysis on DNA of three cats. At the species level, four samples were identified as positive for L. infantum and one for L. braziliensis, the latter detected in an oral swab. The five positive samples were subjected to sequencing


Introduction
Visceral Leishmaniasis (VL) is a zoonosis caused by Leishmania infantum (Baneth et al., 2008). Dogs are considered the main reservoir of the parasite in domestic and peridomestic areas; however, recent studies have reported cats infected with the protozoan and capable of acting as infectious reservoirs (Maroli et al., 2007;Maia & Campino 2011).
Leishmania infantum is transmitted by the bite of phlebotomine sandflies, primarily by Lutzomyia longipalpis and secondarily by Lutzomyia cruzi in Brazil (Andrade-Filho et al., 2017). The most frequent clinical signs of leishmaniasis in cats include skin changes, such as papules, nodules, ulcers, and alopecia (Silveira et al., 2015); ocular signs have also been described (Pimenta et al. 2016). Leishmania species identified in cats include L. mexicana, L. venezuelensis, L. braziliensis, L. amazonensis, and L. infantum  . Leishmania infection shows different clinical presentations in the affected hosts as a result of the Leishmania species and the host immune response. L. braziliensis is associated with cutaneous and mucosal leishmaniasis and L. infantum with visceral and cutaneous (unusual presentation) (Pace, 2014). Notably, it was demonstrated that cats showing dermatological lesions were highly associated with visceral leishmaniasis (Vides et al., 2011).
The use of noninvasive sampling techniques is of interest, especially in feline medicine (Athanasiou et al., 2018). Recently Leishmania DNA was detected from conjunctival (Benassi et al., 2017) and oral swabs of cats .
The Leishmania genus is a slowly evolving monophyletic group and includes important human pathogens (Marcili et al., 2014). Despite the widespread distribution of leishmaniasis among humans and animals in Brazil, little is known about the genetic characteristics of the causative agents. Moreover, phylogenetic studies have focused on human or canine strains (Cortes et al., 2014), while very few of them involve genetic analysis of Leishmania obtained from cats (Ceccarelli et al., 2018;Pereira et al., 2020).
Since cats are potential reservoirs of Leishmania and no study has identified the presence of Leishmania DNA in oral swabs from cats in Brazil, the objectives of the present study were to investigate the occurrence of Leishmania antibodies in cats from a region endemic for canine and human leishmaniasis by ELISA and the presence of Leishmania DNA using PCR analysis of ocular conjunctiva and oral swab samples. From the genomic sequences of the Leishmania isolates, a characterization and evaluation of the phylogenetic relationships between Leishmania strains based on the Leishmania Internal transcribed spacer 1 (ITS1) gene was made.

Animals
The study included 64 cats (Felis catus) living in Belo Horizonte and Confins in Minas Gerais State, Brazil. These municipalities are endemic for canine leishmaniasis (CanL) and have also registered human cases of leishmaniasis (Silva, 2015;Belo Horizonte, 2019). The collection of the biological samples was carried out with the permission of the owners and the experimental procedures were approved by the Ethics Committee on Animal Experimentation, UFMG (CETEA 242/2014). All procedures performed were in accordance with the ethical standards of the institution or practice at which the studies were conducted Cats were clinically evaluated before sample collection and were clinically healthy upon examination. Cats of either sex and of any breed and age were included in this study.

Sample collection
Using disposable syringes and needles, 5 mL blood samples were collected from the jugular or cephalic veins of the cats and transferred into tubes without anticoagulant. Blood samples were centrifuged at 1500 x g for 6 min, and 1 mL of the separated serum was transferred into plastic microtubes and stored at -20°C for serological analysis.
Sterile swabs manufactured for use in bacteriological isolation were used for the collection of specimens for Leishmania DNA analysis. Swabs were rubbed on the lower portion of both ocular conjunctiva and on the right of the oral mucosal of the lips of each cat. The ends of the swabs were separated and stored in 1.5 mL microtubes at -20°C until processing for PCR analyses. DNA extraction was performed using Gentra  Puregene  Tissue Kits (Qiagen), according to the manufacturer's instructions.
Enzyme-linked immunosorbent assay (ELISA) Serum samples were tested for Leishmania antibodies by ELISA (Voller et al., 1979). Soluble Leishmania antigen was produced from promastigote forms of Leishmania (Leishmania) infantum strain MHOM/BR/1967/BH46. Previously known positive (optical density [OD]: 0.340) and negative (OD: 0.060) sera were used as controls. The cutoff point was calculated as the median of positive and negative sera, plus four times standard value (Solano-Gallego et al., 2007). Values above OD: 0.955 were considered positive. All samples were tested in duplicate.

PCR assays
Leishmania kinetoplast DNA All ocular conjunctiva, and mouth mucosa cells were analyzed by PCR to detect Leishmania minicircle kinetoplast DNA (kDNA), generating a 120bp fragment as described by Degrave et al. (1994) and Passos et al. (1996). Briefly, PCR assays were performed in a total volume of 25 µL containing 5 µL DNA template, 0.2 mM DNTPs, 1.5 mM MgCl 2 , 1.5 µL DMSO 5%, 0.4 µM each primer, buffer (10 mM Tris-HCl, 50 mM KCl; pH 8.0) and 1.25 U of Taq DNA polymerase (Invitrogen™, Carlsbad, CA, USA). PCR conditions were as follows: initial denaturation at 94ºC for 5 min, followed by 35 cycles at 94ºC for 30 sec, 61ºC for 1 min, and 72ºC for 3 min, with a final extension at 72ºC for 10 min. Negative (no DNA) and positive (0.01 and 1.0 pg of Leishmania DNA) PCR controls were included in each set of reactions. Positive control for the PCR reactions included DNA extracted from promastigote forms of the L. infantum (MHOM/BR/74/PP75) strain.

Gel electrophoresis of PCR products
Amplification products were subjected to electrophoresis in 2% agarose gels, with a 100 bp DNA Step Ladder as a molecular weight size standard and stained with ethidium bromide (10 mg/mL). To identify Leishmania species ITS1 PCR products were digested with HaeIII (10 U/μL) using the conditions recommended by the supplier (Promega). Restriction digest profiles were analyzed by electrophoresis in 3% agarose gels, stained with ethidium bromide (10 mg/mL), and compared with Leishmania reference strains.

Sanger sequencing
Amplicons from positive samples by ITS1-PCR were purified using QIAquick PCR Purification kit (QIAGEN, Germantown, MD, USA), following the manufacturer's instructions. Purified fragments were sequenced using a Big Dye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA, USA) in a final reaction volume of 8 μL containing 10 ng of purified PCR product and 5.0 pmol of the forward and reverse primers. Sequences were generated using an ABI 3730xl DNA Analyzer, and the software Finch TV (Geospiza, Inc., USA) was used to analyze electropherograms and align sequences with others obtained from GenBank.
PCR RFLP ITS1 detected Leishmania DNA in five of eight kDNA positive samples. It was not possible to perform PCR analysis on DNA of three cats. At the species level, four samples were identified as positive for L. infantum and one for L. braziliensis, the latter detected in an oral swab. The five positive samples were subjected to sequencing Phylogenetic analysis We retrieved from GenBank database fifty-nine published sequence of internal transcribed spacer 1 and 5.8S ribosomal gene of Leishmania infantum and eighteen complete sequence of Leishmania braziliensis, sampled of cats, dogs and human in worldwide. The L. infantum sequences were collected from the Middle east (Turkey, Iran, Israel, Tunisia), Southeast Europe (Italy, Spain, Romania and Greece) and South America (Argentina and Brazil) while the L. braziliensis sequences were collected from the South America (Brazil and Venezuela) and North America (United States of America).

Results
Of 64 feline serum samples collected, 19 were positive for Leishmania antibodies based on ELISA results, representing 29.80% seropositivity.
PCR analysis revealed that oral swabs from four animals (6.25%) were positive for Leishmania, only one of which was also positive by ELISA. PCR analysis of the ocular conjunctiva revealed that two cats were positive in the right eye, two in the left eye, and one animal positive in both eyes, totaling five positive cats (7.81%), one of which was also positive by PCR from an oral swab sample. In total, eight cats were positive by PCR (12.5%). There was poor agreement between the serological and molecular test results (k = 0.16). Results are summarized in Table 1. Phylogenetic analysis (Figure 1) showed that Leishmania spp. found in this study form two monophyletic groups. The first group is composed of L. infantum isolates, and the subdivision of isolates has no relation to origin or with the species in which the parasite was found. However, in the first group, composed basically of L. infantum, a clade of L. braziliensis isolates from Brazilian dogs and a human from the USA was formed.
analysis, which confirmed four as L. infantum and one as L. braziliensis. L. infantum was found in both oral and ocular conjunctival swabs. Results and GenBank accession numbers for the nucleotide sequences are summarized in Table 2.  In addition, the second group was formed exclusively by L. braziliensis isolates from humans from Brazil, Venezuela and the USA.
Although, the L. infantum isolates from cats detected in this study form a single cluster, similar to the human isolate from Greece and the isolates found in dogs in Brazil, they are phylogenetically distant from the L. infantum strains found in domestic cats from other countries as Italy, Turkey and Iran (Supplementary Material Figure S1 A). However, the L. braziliensis isolate found in a Brazilian domestic cat is very similar to the strains circulating in Venezuela (Supplementary Material Figure S1 B).

Discussion
Studies in Brazil and other countries have investigated the occurrence of feline leishmaniasis; however, many questions remain to be answered (Maia & Campino, 2011;. The present study was conducted in Belo Horizonte and Confins, Minas Gerais, which are both regions where cases of leishmaniasis have been registered in humans and dogs. Although dogs are the main hosts and reservoir for Leishmania infantum (Maia & Campino, 2011), our study demonstrated the presence of DNA from this parasite in conjunctiva swabs from cats and, for the first time in Brazil, in oral mucosa swabs, calling attention to the presence of parasite DNA in these sites and suggesting that these tissues could be sampled for diagnosis. Moreover, sequence analysis identified L. braziliensis in one oral swab and L. infantum in an eye conjunctiva sample.
Leishmania infection results in different clinical manifestations depending on the immune response of the host and the protozoan species detected, being L. infantum associated with visceral and cutaneous Leishmaniasis (unusual) and L. braziliensis with cutaneous manifestation of the disease (tegumentary leishmaniasis) (Loeuillet et al., 2016;Pace, 2014). Cats with dermatological lesions might present visceral leishmaniasis (Vides et al., 2011) and both cited species have been identified in cats . Positive cats subjected to xenodiagnosis assays can infect the Leishmania vector (Maroli et al., 2007) and Leishmania braziliensis can be detected in skin samples from experimentally chronically infected cats (Simões-Mattos et al., 2005). Recently, in the same region of the present study, L. braziliensis DNA was detected in conjunctival swab of dogs (Souza et al., 2019), highlighting the importance of detecting Leishmania species not only in dogs but also in cats, to better understand the role of these hosts in the epidemiology of tegumentary leishmaniasis.
In our study, 40.63% of evaluated cats were seropositive for antibodies against Leishmania by ELISA. The frequency of feline seropositivity varies between regions, and caution is advised when comparing this result with those of other studies, since variables, such as serological technique, the level of endemicity, and cross-reaction with other Leishmania species or parasites, can interfere with the results. The seropositivity rate for Leishmania antibodies has been reported as 16.1% in Spain (Martín-Sánchez et al., 2007), 0% in Angola (Lopes et al., 2017), 6.7% in Jerusalem (Nasereddin et al., 2008), 24.27% in Iran (Mohebali et al., 2017), 3.87% to 21.6% in Greece (Diakou et al., 2009;Chatzis et al., 2014), and 25.3% in Italy (Spada et al., 2013). In Brazil, the reported frequency of seropositivity for feline Leishmania antibodies determined by ELISA varies from 3.9% in Pernambuco (Silva et al., 2014) to 11.50% and 15.23% in São Paulo (Costa et al., 2010;Sobrinho et al., 2012). The discrepancies are due to different levels of endemicity, type of feline populations included in the study and, mainly, different methodologies, especially validation of the serological test (Pennisi, 2015). A previous study conducted in Minas Gerais, showed that 54% of the cats were seropositive for Leishmania antibodies by immunofluorescence antibody test (Coura et al., 2018). The high frequency of feline Leishmania seropositivity observed in the present study can be explained by possible occurrence of co-infections with other pathogens (Sobrinho et al., 2012), or the assay performed may have exhibited cross-reaction with antibodies against other parasites (Andrade et al., 2009;Luciano et al., 2009;Soares et al., 2016), although cats do not mount a humoral response like dogs (Solano-Gallego et al., 2007).
Studies of feline leishmaniasis using PCR have used DNA from blood (Martín-Sánchez et al., 2007;Maia et al., 2010;Sherry et al., 2011;Spada et al., 2013;Akhtardanesh et al., 2017), ocular conjunctiva (Chatzis et al., 2014;Oliveira et al., 2015b), urine  and lymphoid organs (Coelho et al., 2011;Noe et al., 2015;Mohebali et al., 2017;Otranto et al., 2017); however, this is the first report of Leishmania DNA identification in oral mucosa swabs from cats in Brazil, supporting recently found results . This finding is important, since there is no consensus on the ideal biological sample for use in cytological or molecular diagnosis of feline leishmaniasis (Braga et al., 2014;Chatzis et al., 2014;Noe et al., 2015;Akhtardanesh et al., 2017). Low production of antibodies by cats (Solano-Gallego et al., 2007) and cross-reactions with other microorganisms that may occur highlights the need for different diagnostic tools for feline leishmaniasis (Noe et al., 2015). Furthermore, we performed sequencing analysis of five DNA samples, which confirmed the presence of L. infantum and L. braziliensis DNA in swab samples evaluated, indicating that infections with these two species might occur in the region of the study.
Phylogenetic analyzes for L. infantum showed low divergence between parasite isolates in cats from Minas Gerais, Brazil. However, there was no direct correlation between the isolates of this study and the other L.infantum isolates from domestic cats described in countries such as Italy, Iran or Turkey (Can et al., 2016;Spada et al., 2016;Mohebali et al., 2017). These results show the genetic complexity of Leishmania spp. populations which circulate in a different geographical area but in the same species (Cupolillo et al., 1998;Ortuño et al., 2019). On the other hand, isolated samples from Brazilian domestic cats were phylogenetically close to isolated samples from domestic dogs, which indicates that possibly L. infantum circulating in Brazil is present in both species. Mohammadiha et al.(2013) identified similar results for dogs and humans, demonstrating the parasite's ability to overcome the interspecies barrier.
Regarding L. braziliensis, the feline sample found in the country is very similar to the isolate described in Venezuela. Aguilar et al. (1989) have already observed that the L. braziliensis strain responsible for an outbreak of cutaneous leishmaniasis that affected humans, dogs and horses in Venezuela was the same strain that circulated in Brazil. The circulation of L. braziliensis between the countries occurs through the presence of the hematophagous vectors Lutzomyia panamensis in Venezuela and Lutzomyia intermedia in Brazil, which perform the blood support in horses and eventually dogs. These hosts transit between Brazil and Venezuela border and have peridomestic habits, allowing the maintenance of the parasite in the environment. Because of this, the strains of Leishmania spp. circulating between the hosts of the two countries have high genetic similarity.
Our study found poor agreement between the results generated using ocular conjunctiva and those from oral mucosa swabs. Other studies have reported inconsistent results when using different types of tissue samples for PCR-based diagnosis compared with parasitological and serological data (Akhtardanesh et al., 2017;Baldini-Peruca et al., 2017;Braga et al., 2014;Chatzis et al., 2014;Diakou et al., 2009;Maia et al., 2010;Noe et al., 2015;Oliveira et al., 2015b;Sherry et al., 2011), indicating that sampling from different tissues together with serological analysis is required when screening for feline leishmaniasis. The study was conducted with 64 cats and use ELISA for serological tests, and these factors should be considered as limitations for further conclusions of cat role in the epidemiology of leishmaniases.

Conclusion
Our findings revealed a high seroprevalence for leishmaniasis in cats and parasite DNA was detected in feline oral and conjunctival swabs. Moreover, sequence analysis detected L. infantum and L. braziliensis in these sites. These results highlight that oral mucosa swabs can also be used for diagnosis, alongside another diagnostic tool. Poor agreement between serological and molecular results indicates that positive serology cannot be used as indicative of Leishmania infection in cats. The results of phylogenetic analyzes show that L. infantum circulating in Brazil is capable of infecting different hosts, demonstrating the parasite's ability to overcome the interspecies barrier. In addition, the L. braziliensis transits between hosts with peridomestic habitats, present on the Brazil and Venezuela border, allowing the maintenance of the parasite in the environment. There is a need to continue the standardization of diagnostic procedures for feline leishmaniasis and for further investigation of the role of cats as reservoirs of Leishmania.