Feline Leishmaniosis in Northwestern Italy: Current Status and Zoonotic Implications

Leishmaniasis remains one of the major neglected tropical diseases. The epidemiological profile of the disease comprises a wide range of hosts, including dogs and cats. Despite several studies about feline Leishmaniosis, the role of cats in disease epidemiology and its clinical impact is still debated. The present study raises awareness about the impact of leishmaniasis in cats from an endemic region in of Northwestern Italy (Liguria). A total number of 250 serum and 282 blood samples were collected from cats, then assessed for Leishmania infantum (L. infantum) serologically using western blot (WB) and molecularly using polymerase chain reaction (PCR). We also tested the association of Leishmania infection with some infectious agents like haemotropic Mycoplasma, Feline immunodeficiency virus (FIV) and Feline leukemia virus (FeLV) together with the hematobiochemical status of the examined animals. Interestingly, all tested animals were asymptomatic and out of 250 examined serum samples, 33 (13.20%) samples (confidence interval (CI) 95% 9.56–17.96%) were positive at WB for L. infantum, whereas of the 282 blood samples, 80 (28.36%) returned a positive PCR (CI 95% 23.43–33.89%). Furthermore, there was a statistical association between PCR positivity for L. infantum and some hematological parameters besides FIV infection as well as a direct significant correlation between Mycoplasma infection and WB positivity. Taken together, the present findings report high prevalence of L. infantum among cats, which reinforces the significance of such positive asymptomatic animals and confirms the very low humoral response in this species. In addition, the laboratory values provide evidence that infection by the parasite is linked to alteration of some hematological parameters and is correlated to some infectious agents. These data are of interest and suggest future research for accurate diagnosis of such zoonosis.


Introduction
Leishmaniasis is a group of neglected tropical diseases caused by an opportunistic intracellular protozoan of the genus Leishmania and transmitted to humans by the bite of female sandflies [1][2][3]. Among 15 well-recognized species of subgenus Leishmania, 13 species are zoonotic [3,4]. Nowadays, about 12 million people infected in 98 countries, and 350 million people are at risk [3,5,6]; the distribution of disease has traditionally been linked to tropical and subtropical regions besides being endemic in many areas worldwide such as the Mediterranean basin, East Africa and South America [1,3]. Three main forms of the disease are known: cutaneous leishmaniasis (CL), mucocutaneous leishmaniasis (MCL), and disseminated visceral infection (VL), with fatal prognosis in

Ethical Considerations
Our study complied with current Italian legislation on research and received ethical approval from the Department of Veterinary Sciences (Local ethical approval), University of Turin, which complies with all relevant CORDIS (European Commission).

Study Population
A total number of 346 domestic cats (n = 250 serum and n = 282 blood samples) from Liguria, a coastal region located in Northwestern Italy, were routinely sampled during veterinary clinic visits and underwent screening. The full details of the study cohort are shown in the Supplementary Materials (Table S1). To investigate Leishmania infection, 2 mL of blood was collected from the animal (through jugular vein puncture) in a clean sterile tube, both with Ethylene diamine tetraacetic acid (EDTA) and without anticoagulant. Samples were then transported to the laboratory of the division of parasitology, Department of Veterinary Sciences, University of Turin, Italy where the serum samples were centrifuged and stored at −80 • C until testing and the blood samples were processed to perform further tests and the experimental steps.

Western Blotting and SDS-Polyacrylamide Gel Electrophoresis
Serum samples were screened by WB using L. infantum promastigotes as antigen, according to the method previously described for different species, including cats [35,36]. Briefly, Two hundred micrograms (1 mg/mL) of promastigotes lysate was run on a 12% polyacrylamide gel (SDS-PAGE). Molecular weight markers were used on a separate track. Fractioned proteins were electroblotted (350 mA, 1 h) onto nitrocellulose sheets that were saturated (3 h) with 3% bovine serum albumin in Tris-Buffered Saline (TBS). Sera were diluted 1: 10 in TBS and incubated overnight at 4C in a Multiscreen Apparatus with reference positive and negative sera (Bio-Rad, Hercules, CA, USA). After washing three times with 0.05% Tween20-TBS, the secondary antibody (1:8000; horseradish peroxidase labelled goat anti-cat IgG (H + L); Novex by Life Tecnologies, Waltham, MA, USA) was incubated for 1 h. Then, the nitrocellulose membrane was washed three times, and anti-L. infantum antibodies were revealed using the ECL system (GE Healthcare, Chalfont St Giles, UK). Serum samples were screened by WB using L. infantum promastigotes as antigen, according to the method previously described for different species, including cats [35,36]. Briefly, Two hundred micrograms (1 mg/mL) of promastigotes lysate was run on a 12% polyacrylamide gel (SDS-PAGE). Molecular weight markers were used on a separate track. Fractioned proteins were electroblotted (350 mA, 1 h) onto nitrocellulose sheets that were saturated (3 h) with 3% bovine serum albumin in Tris-Buffered Saline (TBS). Sera were diluted 1: 10 in TBS and incubated overnight at 4C in a Multiscreen Apparatus with reference positive and negative sera (Bio-Rad, Hercules, CA, USA). After washing three times with 0.05% Tween20-TBS, the secondary antibody (1: 8000; horseradish peroxidase labelled goat anti-cat IgG (H + L); Novex by Life Tecnologies, Waltham, MA, USA) was incubated for 1 h. Then, the nitrocellulose membrane was washed three times, and anti-L. infantum antibodies were revealed using the ECL system (GE Healthcare, Chalfont St Giles, UK).
The resulting bands were compared using as a marker Prestained Protein Molecular Weight Marker (Fermentas International Inc, Ontario, ON, Canada) and Biotinylated SDS PAGE Standards broad range (BioRad Laboratories, California, CA, USA). Samples were considered positive by WB when at least two bands of 169, 115, 66, or 33 kDa could be detected [36].

Preparation of Blood Samples and Extraction of DNA
Total genomic DNA was extracted from 200 µL of whole blood using GenomeElute commercial kit under conditions suggested by the manufacturer (Sigma-Aldrich). The DNA was then stored in sterile DNAse-and RNAse-free microtubes and kept at −20 • C. . Quantity and quality of DNA were assessed using photometric UV-based measurement of nucleic acids (Thermo Scientific™ NanoDrop 2000, ThermoFisher, Waltham, MA, USA) and agarose gel electrophoresis. Internal positive (L. infantum DNA from cultured promastigotes) and negative (double distilled water) controls were included in each reaction.. Fifteen randomly selected positive PCR products were sequenced (Macrogen Inc., Amsterdam, The Netherlands) and the resulting sequences were compared to those available in GenBank.

Detection of FeLV Antigen and FIV Antibody
To test the correlation between Retroviral and Leishmania infection, 87 samples were tested for the presence of FeLV antigen and 89 samples for FIV antibody. Detection of FeLV antigen (p27) and FIV antibody was performed using immunochromatographic BVTTM until 2006. From 2007 onwards, ELISA Test Snap IDEXXTM commercial assay kit (SNAP ® FIV Antibody/FeLV Antigen Combo Test; IDEXX Laboratories, Westbrook, ME, USA) was used.

Data Analysis
To identify possible associations between L. infantum prevalence and seroprevalence, we used generalized linear mixed models in which the result of the PCR and WB was the dichotomous response variable. The potential explanatory variables (covariates) considered were: individual factors (breed, age, sex), and hemato-biochemical parameters (WBC, MON, NEU, EOS, LYM, RBC, Hb, HCT, PLT, ALB, TP, CREA, BUN, ALT, AST, ALP, CHOL, TG, GLU, Ca, K and PHOS). Concurrent infection with FeLV, FIV and Mycoplasma were also investigated as potential risk factors. The Variance inflation Factor (VIF) was used to test and avoid multicollinearity among predictors [37]. Variables identified as significant factors (p ≤ 0.05) in the first univariate analysis were selected for further testing by multivariate linear regression. Best model selection was performed using AIC (Akaike information Criterion), while the goodness-of-fit of the final model was assessed by computing the area under the curve (AUC) of the receiver operating characteristic plots. All statistical analyses were performed using R Statistical Software (Foundation for Statistical Computing, Vienna, Austria) [38].

Results
As previously shown, all cats underwent clinical examination for any lesion and they were asymptomatic. Of the 250 sera tested with WB, 33 samples tested positive (CI 95% 9.56-17.96%) for L. infantum with prevalence of 13.20%, while of 282 blood samples tested with PCR, 80 samples were positive for L. infantum with a prevalence of 28.37% (CI 95% 23.43-33.89%) ( Table 1). It was possible to test a total of 186 cats by both PCR and WB. Of these, 10 tested positive by both PCR and WB (Table 1). Sequencing confirmed the specificity of the protocol used, as all the sequenced amplicons were identified as L. infantum (identity ≥ 98% to GenBank accession number: AB678348). On the other hand, 17 out of the 167 samples tested with PCR for various species of Mycoplasma (CI 95% 6.45-15.70%) were positive with a prevalence of 10.18%, including eight which were also positive for Leishmania using PCR and another three which were positive using WB and PCR. In addition, out of the 89 tested samples for FIV, 31 Table 2.

Discussion
The present study reports a high prevalence of infection with L. infantum, either serologically or molecularly, in examined cats from an endemic area. Likewise, our data provide more evidence about the correlation between infection of cats with L. infantum as widely known opportunistic pathogen and concomitant infection with some other pathogens like FIV, FeLV and three feline hemoplasma species that include Mycoplasma haemofelis, Mycoplasma haemominutum, and Mycoplasma turicensis. To the best our knowledge, this study is the first epidemiological investigation performed on leishmaniasis in cats in this endemic area in Northwestern Italy (Liguria). Given the fact that leishmaniasis is not as common in cats as in dogs, some previous reports have explored the occurrence of feline leishmaniasis worldwide, especially in some countries where the zoonotic form of the disease is present [28][29][30][31][39][40][41][42]. Our present data report that all examined cats were asymptomatic, which is consistent with the hypothesis stating that leishmaniasis is mostly subclinical in cats [20,24]. However, some scant clinical cases were reported elsewhere with typical cutaneous signs, including ulcer crusted dermatitis and nodular lesions on the nose, lips, ears, and eyelids [43]. Other groups may develop chronic ulceration, located particularly on the head and limbs [43]. In rare reported cases of VL, the infected cases showed visceral involvement of the liver, spleen, lymph nodes and kidneys together with some cutaneous manifestations [4,20,44,45].
In Europe, several clinical cases of feline leishmaniasis have been described since 1911 to date in France, Greece, Switzerland, Spain, Italy, and Portugal, with very controversial results among serological and molecular methods [19,20,28,32,41,[46][47][48][49]. In accordance with its prevalence in southern Europe, where the disease in dogs is endemic, several seroepidemiological studies have shown a prevalence of L.infantum in cats ranging from 0.6% to 68%, with differences among countries or even within the same country, in particular Italy [18,40,41,[46][47][48][50][51][52][53][54][55]. Diagnosis of L. infantum in combination with concurrent diagnosis of several opportunistic pathogens provides an important overview for better understanding the epidemiology of the disease. In this regard, western blotting is a core technique used to detect and quantify proteins that react with a specific antibody, among other available serological tests. Besides its high specificity and sensitivity, WB has shown greater sensitivity than immunofluorescence antibody test (IFAT) and Enzyme-linked immunosorbent assay (ELISA) in detection of infection with Leishmania, making it recommended mainly in doubtful cases [56][57][58]. In the present study, the prevalence of infection using WB yielded 13.20%. Nearly similar results have also been reported in some previous studies in the same country [52,59,60]. On the other hand, our present results are higher than previous studies in the same country; a previous study in Liguria and Tuscany, Italy recorded lower seroprevalence (0.9%) using IFAT [50]. Likewise, lower results were recorded in a nationwide survey of L. infantum in Italy using a combination of serological and molecular methods where the prevalence by serology was 3.3%, with a higher rate of cumulative prevalence (10.5%) in southern Italy than in the North (1.6%) [32]. In another study in southern Italy, a high cumulative serological prevalence was recorded using IFAT and molecular prevalence (25.8%) of L. infantum in cats [59]. Collectively, comparison of prevalence data even from the same contry is difficult, due to the differences among applied serological techniques which include either ELISA [18] IFAT [30,46,47,50,55,61], or direct agglutination test (DAT) [53,62]). This difference supports the use of WB in diagnosing both the clinical disease and subclinical infection [58,[63][64][65]. Geographic location, habit changes and expansion range of sandfly vectors represent other factors that might influence the degree of endemicity of the disease [12,13,[66][67][68], also taking into account that several serological studies have revealed that Leishmania-infected cats often develop a low level of humoral response or remain seronegative [47,50]. Furthermore, it should be borne in mind that different biological samples and PCR targets may affect the prevalence rate, which could be another explanation for the large variability in prevalence data reported so far [69,70].
Interestingly, polymerase chain reaction (PCR) is a widely accepted molecular tool for identification and quantification of Leishmania spp. in various tissues and body fluids in reservoir species or hosts [30,33,71]. Hence, several studies have recommended PCR as a highly efficient noninvasive tool for diagnosis and follow-up of the disease [72][73][74]. Regarding its prevalence in southern Europe, the positivity to Leishmania spp. by PCR in a previous studies ranged from 3% to 61%, explaining the possible influence of geographical location on the epidemiological pattern of the disease [32,47,75]. In the present study, we have reported positivity of 28.37% using PCR, which is similar to several previous studies in which the same method was applied [46]. In Southern Italy, a previous study detected a prevalence of infection (by PCR) of 25.8% [59]. On the other hand, lower results were reported in a previous nationwide survey in Italy for L. infantum using qPCR, where the overall cumulative prevalence was 0.8% [32]. In another study in cats in Northern Sardinia, Italy, molecular detection showed a prevalence rate of 5.5% for L. infantum in the population of tested cats [60]. As for sereoprevalence, previous reports of molecular prevalence in cats are subject to great variability among geographical areas [1,4,11,32].
The role of some other pathogens, viral infection such as FeLV, FIV and Mycoplasma, and their association to Leishmania infection, is still unclear and sometimes the results are controversial [50,76] As mentioned above, the occurrence of leishmaniasis in cats worldwide is usually asymptomatic; however, some of the affected cases were co-infected with FIV and/or FeLV, which may induce an impaired cellular immune response [20,77]. Based on our results, there was a statistical correlation between the positivity of sample to L. infantum using PCR and FIV, anemic status of an animal, blood urea nitrogen level, and neutrophilia, whereas using WB, there was a significant association between positivity to L. infantum and Mycoplasma infection. These data are consistent with several previous reports at both the national and international level [32,45,50,55,76,[78][79][80]. Our data suggest a role played by some viruses like FIV as a retroviral infection in development of some opportunistic pathogens like Leishmania [7,32,51]. However, it should be borne in mind that the sole presence of FIV is not a sufficient marker to demonstrate immunodeficient status; this requires additional immunological tests. On the other hand, there was no a clear relationship between infected cats with FeLV and contraction of infection by Leishmania, which is consistent with several previous results [18,47,48,52,54,81]. The possibility of identifing a set of multiparametric infectious/haematochemical values (i.e., FIV, BUN and RBC, which were included in the best fitting model of infection in cats) could be used as predictive tool that could allow clinicians to consider L. infantum in their differential diagnosis. Taken into account, the resistance of a cat to Leishmania infection probably depends on genetic factors, and is not solely related to cell mediated immunity [20,82,83]. As shown, the use of direct (PCR) and indirect (WB) methods yielded discrepant results, which has allowed us to show the major advantages provided by the combined use of both methods for detection of the prevalence of infection in cats as compared to serology alone. Nevertheless, the reported prevalence rate highlights the role played by cats in the transmission of disease, and this suggests cats could be either primary or secondary reservoir host for L. infantum, consistent with previous reports that revealed contraction of infection by sandflies in cats naturally infected with Leishmania [84,85]. The explanation for this data lies not in the difference in sensitivity of the two methods, but rather in the fact that in some species resistant to infection, like cats, the activated immune response is particularly of cell mediated rather than humoral immunity [78,82]. Interestingly, our present data might contribute to the hypothesis suggesting that the cat predominantly has a Th1 response, as evidenced by the high number infections (positivity to PCR) compared to the number of subjects with circulating antibodies (positive to WB), and the limited occurrence of non-specific clinical signs [78,86]. In addition, our present data give more information about the association between Leishmania infection and both retroviruses and hemoplasmosis, which seem to play a role in the pathogenesis of the disease and even in susceptibility to infection [80].

Conclusions
Our present data reporting high prevalence of L. infantum in asymptomatic cats represent an important parameter to further take into account in epidemiological studies. Interestingly, our data may contributes to the hypothesis stating that cats may be a Leishmania reservoir, and suggest further studies to explore their role in maintaining the epidemiological foci of VL in the Mediterranean area. Furthermore, our data suggest performing routine and preferably combined (serological and molecular) testing for leishmaniasis in cats, especially in endemic areas, regardless of the presence or absence of clinical signs. Furthermore, the identfied set of multiparametric infectious/haematochemical values, mainly FIV, BUN and RBC, could be used as predictive tool for diagnosis of L. infantum in cats. Importantly, more strict alternative prophylactic strategies may be essential to reduce the risk of infection and promote identification of new models of leishmanial transmission.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/vetsci8100215/s1, Table S1: The full details of the study cohort for each of the enrolled cat's sex, age, breed and living habits.
Author Contributions: E.K.E., E.F., S.Z., A.A. and A.T. designed the idea of the conception and contributed their scientific advice and supervision besides revision of the manuscript. E.K.E., S.Z., M.P. and A.T. performed the methodology, formal analysis and data curation. E.K.E. and S.Z. drafted the manuscript. K.F.A. contributed their scientific advice, prepared the manuscript for publication and revision. All authors have read and agreed to the published version of the manuscript.
Funding: Non applicable.

Institutional Review Board Statement:
The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Institutional Review Board of the Department of Veterinary Sciences (Local ethical approval), University of Turin, Italy. The study was approved on 2 November 2012.
Informed Consent Statement: Blood samples from cats were collected by licenced veterinarians within routine clinical diagnostic activities and after informed consent of the animals' owner.

Data Availability Statement:
The data that support the findings of this study is contained within the article.