Detection of canine visceral leishmaniasis by conjunctival swab PCR

Introduction: Conjunctival swab PCR was evaluated as a tool to diagnose visceral leishmaniasis in dogs. Methods: Conjunctival swab PCR was compared to indirect immunofl uorescence antibody test and blood PCR. Results: Indirect immunofl uorescence was signifi cantly correlated with conjunctival swab PCR (p < 0.05), but not with blood PCR (p > 0.05). In addition, conjunctival swab PCR was signifi cantly associated with presence of clinical symptoms (p < 0.05), whereas blood PCR was associated with absence of clinical symptoms (p < 0.05). Conclusions: Results indicate that conjunctival swab PCR is useful in epidemiological surveys of canine visceral leishmaniasis.

Visceral leishmaniasis, a vector-borne zoonosis of global importance, is caused by the protozoan Leishmania infantum (syn.Leishmania chagasi) in Brazil (1) .The disease is transmitted to dogs, humans, and other hosts primarily through the bite of infected sand fl ies (1) .It is considered a serious and chronic illness in dogs, with clinical symptoms dramatically different from one animal to another in most cases (1) .For instance, dogs may present subclinical infection that is sometimes self-limiting, or may present severe symptoms that can lead to death.
Leishmania infection may be diagnosed by several methods.For PCR-based diagnosis, DNA can be extracted from various clinical specimens, including blood, skin biopsies, lymph nodes, bone marrow, and spleen.However, collection of such specimens is invasive, and requires skilled labor or appropriate facilities.In contrast, conjunctival swabs are fast and easy to collect (2) (3) .Thus, we assessed the suitability of conjunctival swabs as a biological sample for PCR-based diagnosis of canine visceral leishmaniasis.Its performance was compared to blood PCR and indirect immunofl uorescence antibody test.
Blood samples and conjunctival swabs were collected between July and August 2011 from 213 dogs in Ilha Solteira (20° 25′ 58′′ S and 51° 20′ 33′′ W), a city in the northwest region of the Brazilian State of São Paulo.During sample collection, each dog was evaluated for clinical symptoms consistent with visceral leishmaniasis.Animals were considered to be symptomatic if at least one of the following was observed: skin disorders, apathy, lymphadenomegaly, dry fur, alopecia, onychogryphosis, erosions, ulcers, prostration, and/or cachexia (1)  Indirect immunofl uorescence antibody test was performed according to published methods (4) based on canine anti-IgG conjugated to fluorescein isothiocyanate (Sigma-Aldrich, Bellefonte, PA, USA, Catalog No. F7884) and diluted 1:600.Sera were considered positive when samples were fl uorescently stained, with a 1:40 dilution as cutoff point (4) .
For PCR-based diagnosis, DNA was purified from conjunctival swabs and blood by phenol-chloroform (5) and salting-out (6) , respectively.DNA was stored at -20°C until analysis.A conserved 120 bp fragment in Leishmania spp.kinetoplast DNA minicircle was amplified according to a published protocol (7) , using two pairs of primers, including primers 13A (5′-dGTG GGG GAG GGG CGT TCT-3′) and 13B (5′-dATT TTA CAC CAA CCC CCA GTT-3′).Reactions consisted of 1 U Platinum® Taq DNA polymerase (Invitrogen, Camarillo, CA, USA), 15.25µL ultrapure water, 1× PCR buffer, 1.5mM MgCl 2 , 0.31mM each of dATP, dCTP, dGTP,  and dTTP, 0.26µM of each primer, and 2.5µL of extracted DNA in a fi nal volume of 25µL.Positive control reactions contained DNA extracted from an in vitro culture of L. infantum.Negative control reactions contained DNA extracted from blood and conjunctival swabs of dogs previously confi rmed to be uninfected.Amplifi ed products were resolved on 2% agarose, stained with ethidium bromide, and photographed with a Cybershot 7.2 megapixel digital camera (DSC-W70, Sony).All samples were tested in triplicate.
To eliminate false negatives due to issues with DNA loading, sample degradation, or PCR failure, a real-time PCR reaction was performed to amplify β-actin (8) .The gene was amplifi ed with forward primer 5′-dCTG GCA CCA CAC CTT CTA CAA-3′, reverse primer 5′-dGCC TCG GTC AGC AGC A-3′, and fl uorogenic probe 5′-CCAC GCG CAG CTC G-3′ (8) .To construct an absolute standard curve for this reaction, canine DNA was serially diluted nine times so that each point of the curve corresponded to one log.Template DNA concentration was estimated by absorbance at 260 and 280nm.
Notably, the sensitivity of conjunctival swab PCR was lower than that of other diagnostic approaches based on conjunctival swabs.For instance, previous studies demonstrated sensitivity as high as 92% (5) , and 91.7% (10) .In addition, 90% sensitivity was achieved by hybridization, and 83.3% by nested PCR (2) .L. infantum DNA was detected in conjunctival swabs from 95% of symptomatic dogs and 77.5% of asymptomatic dogs that also tested positive on other serological and parasitological tests (3) .These results, however, were obtained from highly sensitive techniques.In particular, hybridization can enhance the sensitivity of PCR and verify results (10) .However, this technique requires more elaborate infrastructure to handle radioisotopes.Similarly, nested PCR has been used to enhance sensitivity, but is more time consuming and is more susceptible to contamination (11) .
In line with previous results (12) , the adjusted kappa index between indirect immunofl uorescence and conjunctival swab PCR was 0.53, suggesting moderate (9) , but statistically signifi cant agreement (p < 0.05, Table 2).However, the number of seronegative dogs that tested positive on conjunctival swab PCR (n = 11) was similar to the number of seropositive dogs that tested negative on conjunctival swab PCR (n = 12).This result may refl ect parasite load, which may go above or below the limit of detection of each assay, depending on the phase of infection.Thus, the use of both tests would enhance diagnosis of visceral leishmaniasis in dogs.
For indirect immunofluorescence and blood PCR, the adjusted kappa index was 0.13, indicating slight agreement (9) .Indeed, these two tests were not significantly correlated (p > 0.05).Accordingly, blood and conjunctival swab PCR were also not signifi cantly correlated (p > 0.05).These results are summarized in Table 2, and are in line with published results (12) .
The authors declare there is no confl ict of interest.Clinical symptoms were apparent in 43 dogs.Of the 29 dogs that tested positive by indirect immunofl uorescence, 51.7% (15/29) were symptomatic.Similarly, symptoms were apparent in 57.1% (16/28) of dogs that tested positive by conjunctival swab PCR.In contrast, clinical symptoms were observed in only 28.6% (8/28) of dogs that tested positive on blood PCR.Clinical signs were also observed in 70.6% (12/17) of animals that tested positive on both conjunctival swab PCR and indirect immunofl uorescence.Three of fi ve animals that tested positive on all three tests were symptomatic.

CONFLICT OF INTEREST
Notably, statistical analysis (Table 3) indicated that a positive result on conjunctival swab PCR is significantly associated with the presence of clinical signs (p < 0.05), while a positive result on blood PCR is signifi cantly associated with the absence of clinical signs (p 0.05).Indeed, conjunctival swab PCR has been demonstrated to have high sensitivity for Leishmania in clinically diseased (12) (13) , and asymptomatic dogs (2) .We note, however, that higher sensitivity was achieved in asymptomatic dogs by other groups (2) .

Committee in Animal Experimentation and Animal Welfare of the Faculty of Veterinary Medicine and Animal Science from the University of Sao Paulo, and was compliant with national guidelines (Law No. 11.794, 8/10/2008).
. The study (Protocol No. 2203/2011) was approved by the Ethics

TABLE 1 -Sensitivity, specifi city, positive predictive value, and negative predictive value of conjunctival swab and blood PCR for canine visceral leishmaniasis, using indirect immunofl uorescence antibody test as gold standard. A total of 213 dogs were tested by all three methods. Parameter IFAT × CS-PCR IFAT × blood PCR
IFAT: immunofl uorescence antibody test; PCR: polymerase chain reaction; CS-PCR: conjunctival swab-polymerase chain reaction.

TABLE 3 -Association of a positive test on indirect immunofl uorescence, conjunctival swab PCR, and blood PCR with presence or absence of clinical symptoms. Clinical symptoms
IFAT: immunofl uorescence antibody test; PCR: polymerase chain reaction.