Infectivity of seropositive dogs, showing different clinical forms of leishmaniasis, to Lutzomyia longipalpis phlebotomine sand flies

Veterinary Parasitology 147 (2007) 67–76 www.elsevier.com/locate/vetpar

Infectivity of seropositive dogs, showing different clinical forms of leishmaniasis, to Lutzomyia longipalpis phlebotomine sand flies E´rika Monteiro Michalsky a,b, Marı´lia Fonseca Rocha c, Ana Cristina Vianna Mariano da Rocha Lima a, Joa˜o Carlos Franc¸a-Silva d, Marize Quinhone Pires e, Fernanda Santos Oliveira e, Raquel Silva Pacheco e, ´ lvaro Jose´ Romanha a, Sara Lopes dos Santos a, Ricardo Andrade Barata a, A Consuelo Latorre Fortes-Dias f, Edelberto Santos Dias a,* a

Centro de Pesquisas Rene´ Rachou/FIOCRUZ, Av. Augusto de Lima, 1715, Barro Preto, Belo Horizonte, CEP 30190-002, MG, Brazil b Universidade Federal do Triangulo Mineiro, Uberaba, MG, Brazil c Secretaria Municipal de Sau´de, Montes Claros, MG, Brazil d Universidade Federal de Minas Gerais, Belo Horizonte, MG, Brazil e Instituto Oswaldo Cruz/FIOCRUZ, Rio de Janeiro, RJ, Brazil f Fundac¸a˜o Ezequiel Dias, Belo Horizonte, MG, Brazil Received 9 November 2006; received in revised form 28 February 2007; accepted 1 March 2007

Abstract Visceral leishmaniasis (VL) is a growing zoonosis with an increasing number of new cases and a rapid geographical spreading of the disease. In the present study, a canine survey was carried out in the city of Montes Claros (320,000 inhabitants), an endemic area of American visceral leishmaniasis in the state of Minas Gerais, Brazil. A total number of 4795 dogs were examined by serology, which showed a rate of seropositivity of 5%. Isoenzymatic analysis confirmed Leishmania infantum chagasi as the local aetiological agent of CVL. Canine tissues were assayed for the presence of Leishmania parasite DNA using different techniques. The infectivity of asymptomatic, oligosymptomatic and symptomatic seropositive dogs was tested by xenodiagnosis using laboratory reared Lutzomyia longipalpis. Rates of infection of 5.4%, 5.1% and 28.4% were found for the phlebotomine sand flies that fed in asymptomatic, oligosymptomatic and symptomatic dogs, respectively. Our results indicate that, under experimental conditions, symptomatic dogs are about four times more infective to VL vectors than oligosymptomatic or asymptomatic animals. The lower infectivity rates of dogs displaying any of the last two clinical forms of leishmaniasis, however, must be taken into account in the epidemiology of CVL. # 2007 Elsevier B.V. All rights reserved. Keywords: Canine visceral leishmaniasis; PCR; Leishmania chagasi; Lutzomyia longipalpis; Xenodiagnosis; Indirect immunofluorescence; Montes Claros

1. Introduction * Corresponding author. Tel.: +55 31 3349 7758; fax: +55 31 3349 7795. E-mail address: [email protected] (E.S. Dias). 0304-4017/$ – see front matter # 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.vetpar.2007.03.004

American visceral leishmaniasis (AVL) is a tropical disease of medical and veterinary importance that is transmitted through the bites of females of sand flies

68

E´.M. Michalsky et al. / Veterinary Parasitology 147 (2007) 67–76

Lutzomyia longipalpis (Lutz and Neiva, 1912) infected by the protozoan Leishmania infantum chagasi (Shaw, 2006). Visceral leishmaniasis is widespread throughout four continents and has become an increasing major public health problem (WHO, 1990). Typically regarded as a sylvatic disease of rural areas, it has undergone changes and become urbanized due to socio-environmental modifications, which have reduced the availability of wild animals for feeding sand flies, and to human migration from endemic areas to urban peripheries. Therefore, domestic dogs and humans have become alternative feeding sources for the sand flies. In Brazil, in the mid 1970s, VL was described as a sporadic disease from wild and rural areas, therefore affecting only humans or dogs that had had close contact with forested areas (Silva et al., 1997). In recent years, however, an increased prevalence of human and canine VL has been reported, concerning not only the number of cases but also the geographical dispersion of the disease. The disease is more prevalent among dogs than humans and it has been shown that the human cases are usually preceded by canine cases. In fact, CVL has been considered as a risk factor to human VL (Di Lorenzo et al., 2000). In the last decades, AVL has been regarded as a reemergent disease in our country. Between 1980 and 1991, the Brazilian Ministry of Health recorded a yearly mean of 1500 new notified cases of the disease, which increased to 2800 per year from 1992 to 1997. More than 90% of these cases were reported in the Northeast region, mainly in the state of Bahia (1254 cases of a total of 2572 cases for the whole country in 1997) (Marzochi and Marzochi, 1994; Vieira and Coelho, 1998). Cities like Montes Claros, Belo Horizonte, Porteirinha and Va´rzea Grande either have already faced or are facing human and canine VL epidemics (Silva et al., 2001; Barata et al., 2004; Margonari et al., 2004; Michalsky et al., 2005). Diagnosis of VL can be carried out by means of parasitological, serological or molecular methods, associated to clinical and epidemiological evidences. Another important technique available is the xenodiagnosis, which is based on the detection and isolation of a pathogen using its natural vector. Xenodiagnosis has been considered as an important diagnostic method for CVL and can be used to assess infection rates in dogs under different clinical forms of the disease (Molina et al., 1994). Although it is not routinely employed, due to intrinsic problems, xenodiagnosis can be a useful tool in epidemiological studies providing information on the clinical status as well as on the treatment of infected dogs (Alvar et al., 1994; Gradoni et al., 1987).

There has been an increasing interest in developing characterizing techniques in order to allow identification of species and even strains of Leishmania parasites, as they may show different degrees of virulence and the infected patient may respond differently to chemotherapeutic agents (Marsden, 1979; Lainson, 1983; Peters et al., 1983). Therefore, the characterization of the infecting parasite has become essential for clinical and pathological investigations of VL. The present study was undertaken in the city of Montes Claros, state of Minas Gerais, Brazil, in order to gain better understanding of the role of the dog in the epidemiology of AVL. We have also included assessments of the infectivity of Lutzomyia longipalpis sand flies in seropositive dogs under different clinical forms, by using the xenodiagnosis technique, and identified the circulating Leishmania species in the municipality through isoenzymatic methods. In addition, canine tissues were tested for the presence of Leishmania DNA using different techniques. 2. Methodology This study was conducted in accordance with the ethical principles of animal experimentation adopted by the Brazilian College of Animal Experimentation (COBEA) and it was approved by the Ethical Committee on the Use of Animals (CEUA) of Fundac¸a˜o Oswaldo Cruz (FIOCRUZ), Rio de Janeiro, Brazil. 2.1. Geographical area under study Montes Claros (168430 4100 S, 438510 5400 W) is a medium size city with about 320,000 inhabitants (IBGE, 2002). It is located in the Northern region of the state of Minas Gerais, at 420 km from the state’s capital, Belo Horizonte (Fig. 1). The area is part of the basin of the Sa˜o Francisco River, within the so-called ‘‘dry-land polygon’’. The city occupies an area of 3.577 km2, corresponding to 0.59% of the state of Minas Gerais, and reaches a maximum altitude of 638 m. The climate is characterized as semi-humid tropical (hot and dry) with average temperatures around 25 8C and an extended dry season (approximately 5 months a year). Annual precipitation is around 520 mm3, with highest rainfall between October and March. The relative humidity ranges between 52% and 80%. 2.2. Survey of canine population A canine survey was carried out in 2002. Ten urban boroughs of the city were chosen based on previous data

E´.M. Michalsky et al. / Veterinary Parasitology 147 (2007) 67–76

69

Fig. 1. Geographical localization of the municipality of Montes Claros (state of Minas Gerais, Brazil) and the boroughs where the canine survey was carried out.

on increased incidence and prevalence of CVL (Rocha, 2002) and on the occurrence of human cases of AVL. Almost every resident dog in these boroughs was included (almost 5000 animals). For CVL diagnosis, canine blood samples were collected by health agents of the Zoonosis Control Center (ZCC) in Montes Claros, according to the recommendations of the Brazilian Ministry of Health. The test samples were collected through venous puncture of the auricular marginal vein using disposable microlancets and transferred to 3 cm  10 cm pieces of filter paper (Klabin number 25). After labeling with the data of the animal (sample code, gender, breed and age), of its owner (name and address) and the identification of the health agent, the samples were stored separated by cellophane paper to prevent cross contamination. The presence of anti-Leishmania immunoglobulin was assayed by indirect immunofluorescence reaction (Camargo and Rebonato, 1969). Test samples with fluorescence at 1:40 dilution were considered as positive and re-tested for confirmation with new blood samples (Mancianti et al., 1988; Costa et al., 1991). Each set of reactions was accompanied by positive and negative reaction controls. The seropositive animals were grouped according to the clinical signs of CVL into asymptomatic (with no clinical signs of infection), oligosymptomatic (with up to three clinical signs of infection, in general lymphoid adenopathy, slight decrease of weight and/or opaque hair) or symptomatic (with more than three severe signs of infection, i.e., cutaneous alterations such as depilation, furfuraceous eczema, and ulcers; onycogryphosis,

keratoconjunctivitis, loss of weight, rigidity of posterior limbs) (Mancianti et al., 1988). Forty dogs comprising 20 symptomatic, 10 oligosymptomatic and 10 asymptomatic ones were selected for diagnosis confirmation by the rapid test antiLeishmania donovani, TRALd (Corixa Co., Seattle, WA, USA). This test, in a dipstick format, employs a cloned antigen of Leishmania (rK39), which has been shown to be capable of detecting antileishmanial antibodies in acute VL sera with 99% sensitivity and 100% specificity for all members of the L. donovani complex (L. chagasi, L. donovani, L. infantum) (Badaro et al., 1996). The results were recorded and delivered to the ZCC of Montes Claros. Six dogs showing each clinical form of CVL, with diagnosis confirmed by TRALd, were kept in the ZCC kennel in order to be used in the xenodiagnosis experiments. The remaining animals were collected by health agents and euthanized, according to pre-established regulations. 2.3. Xenodiagnosis Xenodiagnosis was conducted with three groups of six dogs each, corresponding to the three clinical types of CVL. The dogs were sedated with 1% acepran. Twenty females F1 from laboratory reared L. longipalpis were placed in round plastic boxes (10 cm diameter  5 cm height) with an open side covered by a fine-mesh nylon screen. This side was placed over the skin of the internal ear of one dog and the whole set was covered with a piece of black fabric to achieve ideal

70

E´.M. Michalsky et al. / Veterinary Parasitology 147 (2007) 67–76

conditions for L. longipalpis blood feeding. After 30 min, the phlebotomine sand flies were transferred to holding cages that were kept at temperatures of 25– 28 8C and 90% of relative air humidity. On the fifth day after the blood meal, genomic DNA was extracted from the live females for molecular analysis. The dogs were anesthesized with thionembutal (30 mg/ml, via i.v.) and aspirates were taken from their bone marrow. After euthanasia by injecting potassium chloride (0.5 ml/kg of body weight, via i.v.) tissue samples were taken from the spleen and from the upper part of the ear skin, respectively. Aspirates and tissue samples were used for the culturing of parasites and for the preparation of smears and imprints, via glass slide and filter paper apposition, respectively. 2.4. Isolation of Leishmania from dogs after xenodiagnosis Fragments of skin and spleen and bone marrow aspirates were cultured in blood agar, in the presence of a biphasic medium Nicolle–Novy–McNeal (NNN)–LIT containing 10% of bovine calf serum (Camargo, 1964). The cultures were maintained at 25 8C and, once a week, they were examined for viability and motility of the flagellates. Subculturing was repeated for 5 weeks and then amplified by diluting 1:5 with fresh medium. Two weeks later, a new culture amplification was performed to a volume of 80 ml. The parasites in culture were counted in a Neubauer chamber and they were concentrated by centrifugation at 3800  g for 10 min at 4 8C. After two washings with 5 ml each of 0.1 M EDTA in 0.85% NaCl the promastigotes in the pellet were used for the identification of the infecting Leishmania strain by isoenzymes. 2.5. Detection of Leishmania DNA The surviving females of L. longipalpis used in the xenodiagnosis were individually transferred to conical microtubes. Each specimen was ground dry (under ice), added of 50 ml of lysis buffer (100 mM Tris–HCl, 100 mM NaCl, 25 mM EDTA, 0.5% SDS, pH 8.0) plus 1.0 ml of aqueous proteinase K solution (10 mg/ml) and incubated overnight at 37 8C. After addition of 70 ml of ultra-pure water, total DNA was extracted with 120 ml of phenol (saturated in buffer), followed by 120 ml of isoamyl:chloroform (24:1). DNA precipitation was carried out by adding 200 ml of cold absolute ethanol in the presence of 20 ml of 3 M sodium acetate pH 5.2. After incubation at 20 8C overnight, the DNA precipitate was washed with 100 ml of 70% cold ethanol,

resuspended in 30 ml of 1 TE buffer and stored at 20 8C. The concentration and degree of purity of the extracted DNA were estimated by spectrophotometer readings (GeneQuant, GE Health) at 260 and 280 nm. Infection of the phlebotomine sand flies was tested by the polymerase chain reaction (PCR) as described before (Michalsky et al., 2002). The technique was carried out using the following primers: 50 GGG GAG GGG CGT TCT GCG AA 30 , 50 CCG CCC CTA TTT TAC ACC AAC CCC 30 , 50 GGC CCA CTA TAT TAC ACC AAC CCC 30 , which amplify the region conserved in the minicircle of kDNA of Leishmania (Degrave et al., 1994). For the canine tissues, 10 ml of deionized water were added to the glass slides smears and the surface was scraped. The resuspended material was transferred to a microtube and the scraped surface was washed by addition of 10 ml of water. The same procedure was adopted for a second area of the slide. The four eluates (about 40 ml) were pooled together and heated at 70 8C for 10 min. After centrifugation at 13,000 rpm for 5 min at room temperature, the supernatant was stored at 4 8C until use for molecular analysis. The scraping and elution procedures were repeated for the second glass slide of the same dog and used for confirmation purposes. For the tissue imprints on filter paper, 5 mm2 circles were cut and transferred to a microtube. The adsorbed material was eluted with 50–100 ml of deionized water, followed by centrifugation at 13,000 rpm for 5 min at room temperature. The supernatant was stored at 4 8C until use. The presence of Leishmania DNA in the eluates from the canine tissue was tested by PCR, with a protocol similar to that employed for the phlebotomine sand flies. A negative (no DNA) and a positive control (purified kDNA from a reference strain of L. (V.) braziliensis, MHOM/BR/75/M2903) were included in every set of reactions. In addition, the presence of polymerase inhibitors in the tissue eluates was tested by repeating the amplification of all the negative samples after adding 100 fg of DNA of the reference strain. The presence of an amplicon with the same size and intensity as the positive control was taken as indicative of the absence of inhibitors. After amplification, the samples were submitted to electrophoresis on 6% polyacrylamide gels and the PCR products were visualized by silver impregnation. 2.6. Isoenzymatic characterization of the isolated strain of Leishmania The isoenzymatic study was done through electrophoresis in agarose gels (Momen and Salles, 1985).

E´.M. Michalsky et al. / Veterinary Parasitology 147 (2007) 67–76

71

Four enzymatic systems were used: ACON aconitate hydratase (E.C. 4.2.1.3), NH nucleotidase (E.C. 3.2.2.1), PEPD peptidase (L-proline) (E.C. 3.4.13.9) and GPI glycose-phospho-isomerase (E.C. 5.3.1.9). L. (V.) braziliensis M2903, L. (L.) amazonensis PH 8,0 and L. (L.) chagasi PP75 were loaded as positive controls. After staining and drying at room temperature, the gels were analysed according to Cupollilo (1992). 2.7. Statistical analysis Infection rate data were analyzed using the Kruskal– Wallis and Mann–Whitney tests ( p < 0.05) and the x2 test was employed as indicated under the corresponding table in the results section ( p < 0.05). 3. Results A total of 4795 dogs were examined in the municipality of Montes Claros (Fig. 1), from which 236 were shown to be seropositive for CVL (Table 1). The prevalence of the disease, in the 10 boroughs where the canine survey was carried out, varied from 1.84% (Village do Lago II) to 9.88% (Joa˜o Botelho), with an average value of 4.92%. Although 18 out of 40 seropositive dogs were selected for the xenodiagnosis, all 40 of them were shown to be positive by TRALd (CVL was confirmed in 100% of cases). After the xenodiagnosis, Leishmania parasites, at varying frequencies, were present in the skin, bone and spleen of those dogs, independently of the clinical forms of CVL (Table 2). Although not statistically significant, the overall positivity for Leishmania in both post-mortem Table 1 Survey of dogs resident in the municipality of Montes Claros (state of Minas Gerais, Brazil) in 2002 Boroughs

Number of dogs Examined

Seropositive for CVL

Prevalence of CVL (%)

Alterosa Chiquinho Guimaraes Joao Botelho Morrinhos Morro do Frade Santa Rita Vila Guilhermina Vila Mauriceia Vila Oliveira Village do Lago II

594 397 172 1169 625 248 470 307 324 489

20 9 17 74 39 12 17 11 28 9

3.64 2.26 9.88 6.33 6.24 4.83 3.61 3.58 8.64 1.84

Mean Total

– 4795

– 236

4.92 –

Fig. 2. PAGE analysis of the PCR products after amplification of total DNA of L. longipalpis with generic primers for Leishmania. Lanes: (1) molecular size marker fX174 HaeIII digested, (2) positive sample, (3) negative sample, (4) positive control (L. braziliensis/M2903), (5) negative control (no DNA). The arrow points to the generic band for Leishmania.

tissues and tissue cultures was always higher for the spleen. Leishmania DNA was detected in 100% of the skin samples of oligosymptomatic and symptomatic dogs and to a lesser extent (66.7%) in asymptomatic ones. Slide apposition seemed more effective than filter paper as a source of Leishmania DNA for amplification. An asymptomatic dog numbered 4 attracted our attention. Although neither Leishmania parasites nor DNA could be detected in tissue samples by any of the four methods employed, this animal was infective to L. longipalpis, as revealed by the presence of Leishmania DNA in the vector (Fig. 2). One hundred and sixty three (45.3%) out of the 360 females of L. longipalpis used in the xenodiagnosis remained alive, distributed as follows: 37/120 in the asymptomatic group, 59/120 in the oligosymptomatic group and 67/120 in the symptomatic group. Mean infection rates of 5.4%, 5.1% and 28.4% were obtained, respectively, for the phlebotomine sand flies fed on asymptomatic, oligosymptomatic and symptomatic dogs. The difference was statistically significant for the symptomatic group only. Although the infection rate of L. longipalpis was about four times higher for symptomatic dogs (28.4%), animals under any of the other two clinical forms of CVL were shown to be capable of infecting the vector as well, with infection rates of about 5%. The infectivity rates were determined as 33.3% (2/6), 16.7% (1/6) and 83.3% (5/6) for asymptomatic, oligosymptomatic and symptomatic dogs, respectively.

72

Table 2 Detection of Leishmania sp. in the phlebotomine sand flies and in tissues sampled from dogs after xenodiagnosis Clinical form of CVL

Dog no.

Leishmania parasites

Leishmania DNA(PCR)

Biopsy

Culture

a

Glass slide

Filter paper

L. longipalpis

Bone marrow

Spleen

Bone marrow

Spleen

Skin

Bone marrow

Spleen

Skin

Bone marrow

Spleen

1 2 3 4 5 6

     

     

+     

  +   

  +   

+  +  + +

+  +  + +

 + +   

  +   

  +   

  +   

  + +  

Oligosymptomatic

7 8 9 10 11 12

 + + +  +

 + +   +

 + + + + 

 +  

 +  

 + + +

 +  

 +  

+

+ + + + + +

+ +  +



+ + + + + +









 +    

Symptomatic

13 14 15 16 17 18

+  + + + 

  + + + 

+  + + + +

   +  

+  + + + 

+ + + + + +

 + +   

  + + + 

+ + + + + +

   + + +

  + + + 

+  + + + +

33.3 a

33.3 a

55.5 a

16.7 b

38.9 b

88.9 c

66.7 d

44.4 d

55.5 e

27.7 e

27.7 e

44.4

Asymptomatic

Overall positivity (%) b a b

The skin cultures were unsuccessful due to contamination. Equal letters mean no difference between the tissues tested (skin, bone marrow and spleen) for each methodology by x2 test ( p < 0.05).

E´.M. Michalsky et al. / Veterinary Parasitology 147 (2007) 67–76

Skin

E´.M. Michalsky et al. / Veterinary Parasitology 147 (2007) 67–76

73

Fig. 3. Isoenzymatic profiles of Leishmania promastigotes isolated from spleen cultures of seropositive dogs used in xenodiagnosis. (A) Peptidase Lproline (PEPD), (B) glycose-phospho-isomerase (GPI), (C) aconitate hydratase (ACON), (D) nucleotidase (NH). Lanes 1–4: test samples. Lanes 5– 7: reference strains for L. (L.) infantum chagasi, L. (V.) braziliensis and L. (L.) amazonensis, respectively.

Amongst the positive tissue cultures for Leishmania, isolation of parasites was successful in four spleen cultures. These samples showed the same isoenzymatic pattern as the reference strain of L. (L.) infantum chagasi (Fig. 3). 4. Discussion The phenomenon known as urbanization of visceral leishmaniasis (VL) has been reported since the 1950s (Deane and Deane, 1955). Environmental changes followed by migration from rural to urban areas, increasing urbanization processes and periodical and extended droughts have enabled the spread of the disease and its emergence in new foci, involving humans, dogs and vectors. Such phenomena led to a reduction of usual habitat for the disease, facilitating the occurrence of epidemics (WHO, 1992). Reports of epidemics in large urban centers in Brazil show how the migratory process to the urban areas has changed the epidemiological profile of AVL (Vieira and Coelho, 1998).

Several interrelated factors have been considered as determinants for the endemic levels of VL: farming, land exploration, interruption of epidemiological surveys, urbanization processes, increasing areas with poor housing and sanitary conditions and, consequently, the presence of infected dogs. All of them enable the adaptation of Leishmania to new ecological niches (Monteiro et al., 1994). Sherlock (1996) has observed that poverty, malnutrition, increased numbers of infected dogs and high density of phlebotomine sand flies in residential and peridomiciliar areas are associated with bad sanitary conditions and low social-economical levels in several Brazilian regions. Naturally, other determinant factors such as high density of vectors and high infection rates, besides host vulnerability and susceptibility to the disease, ought to be present for the disease to occur. An important aspect of vector-linked diseases is the presence of a host population, which is effectively responsible for the maintenance and dispersion of the disease (Woolhouse et al., 1997). Canine serological and entomological surveys carried out in a number of

74

E´.M. Michalsky et al. / Veterinary Parasitology 147 (2007) 67–76

endemic areas of VL have revealed that high prevalence rates of CVL and vector abundance increase the risk of VL transmission to man (Vieira and Coelho, 1998). In the Brazilian municipality of Montes Claros, a characteristic and favorable environment for VL occurrence is found. The dwellings are mostly poor with deficient garbage collection and minimal basic sanitary conditions. In some areas, many residents have low social-economical level and there is a high number of domestic animals. The accumulation of organic matter enables favorable conditions for the breeding of sand flies, and hence for VL transmission. Isoenzymatic analysis provides the main tool for the identification of Leishmania species (Cupollilo et al., 1994) and, as such, has been widely used for studies on genetic variability, identification and classification of Leishmania parasites in the New World. Our data confirmed that L. i. chagasi is the aetiological agent of CVL in Montes Claros. Indirect immunofluorescence was adopted by the Brazilian Foundation of Health (FNS) as the reference technique for canine serological surveys, displaying 98% of sensitivity and 70% of specificity (Costa et al., 1991). A higher sensitivity (100%) is reached with the alternative test TRALd with the advantages of low complexity, easy interpretation and adaptability to field studies (Badaro et al., 1996; Franc¸a-Silva et al., 2003; Rocha, 2002). However, its high cost is a main disadvantage and, in our case, it was used only for confirmatory purposes in a small number of dogs. PCR-based assays have been suggested to be very adequate for the detection of VL infection in its acute phase and for monitoring the parasitological cure of treated patients, due to their high sensitivity and specificity (Cortes et al., 2004). Our data are in agreement to that, especially for canine skin smears, which reached almost 90% of positivity (Table 2). Reithinger et al. (2004) demonstrated that the parasite burden in bone marrow, spleen and lymph node tends to be smaller due to the presence of inhibitors of the PCR that may affect its sensitivity. In the present work, extra precautions were taken to prevent false negative results in the eluates used in DNA amplification and the interference of inhibitors was not observed for any of the samples tested. The xenodiagnosis is a direct technique widely used to determine the epidemiological importance of dogs infected with Leishmania parasite. Although laborious and time-consuming, it allows the determination of feeding and infection rates of phlebotomine sand flies and provides, simultaneously, essential data on the transmission potential of the disease (Molina et al.,

1994). These authors found no correlation between the clinical status of the dog and its infectivity to the vector. Others groups; however, reported that phlebotomine sand flies that fed on symptomatic dogs showed higher infection rate (Courtenay et al., 2002; Gradoni et al., 1987), a finding that is in accordance with our data. A mathematical model developed for VL transmission suggested low or even absence of infectivity for asymptomatic dogs, therefore attributing no important role for these animals in the transmission of the disease (Lanotte et al., 1979; Dye et al., 1992; Hasibeder et al., 1992). However, they have been considered as moderate sources of Leishmania for vectors by others (Sherlock, 1996; Travi et al., 2001). Our results confirm that both asymptomatic and oligosymptomatic dogs are able to infect phlebotomine sand flies, although at a lower rate than the symptomatic animals. It is important to note that this capability may be underestimated. Although the healthy skin of asymptomatic dogs may be attractive to sand flies (around 6% according to Guarga et al., 2000), it has been observed that the injured ear skin of more advanced clinical forms of CVL is still more attractive, reaching infection rates of 35% (Vexenat et al., 1994). Another point to be considered is that the mortality rate of the phlebotomine sand flies after the experiment was higher for the asymptomatic group of dogs. It might be due to a less efficient blood feeding as asymptomatic dogs remain less time sedated than symptomatic dogs, therefore resulting in a shortened period of contact between the food source and the sand flies. Our results of Leishmania detection may appear inconsistent when the different methods employed are compared as well as the subjects in a given clinical group. Several points may be raised. Firstly, the known variability in the sensitivity of the methods employed. Secondly, it has been demonstrated that at least three important variables interfere in the onset and subsequent stages of canine VL: the infectiousness of dogs to sand flies, the strength of the canine anti-Leishmania IgG response and the dog infectiousness through time (Courtenay et al., 2002). All these variables are related to the individual. For the third variable, however, they were able to follow dogs’ infectivity and concluded that there is a latent period around 6–7 months between infection and infectiousness, taking the seroconversion as the starting point. As we used dogs from the field, the individual starting points were unknown and, consequently, any knowledge of the time-driven stage of infectiousness was absent. Nevertheless, although under our experimental conditions, symptomatic dogs were shown to be about

E´.M. Michalsky et al. / Veterinary Parasitology 147 (2007) 67–76

four times more infective to AVL vectors than oligosymptomatic or asymptomatic ones, the lower infectivity rates of dogs displaying any of the last two clinical forms of leishmaniasis may be taken into account in the epidemiology of AVL in endemic areas of the disease. Acknowledgements To the technicians of the Center of Zoonosis Control (CZC) of the Department of Health of Montes Claros for the support in the canine survey and xenodiagnosis experiments. To Dr. Michael Richardson for kindly reviewing the language. This work was financed by Centro de Pesquisas Rene´ Rachou (CPqRR/FIOCRUZ) and FAPEMIG. References Alvar, J., Molina, R., San Andres, M., Tesouro, M., Nieto, J., Vitutia, M., Gonzalez, F., San Andres, M.D., Boggio, J., Rodriguez, F., 1994. Canine leishmaniasis: clinical, parasitological and entomological follow-up after chemotherapy. Ann. Trop. Med. Parasitol. 88, 1–8. Badaro, R., Benson, D., Eulalio, M.C., Freire, M., Cunha, S., Netto, E.M., Pedral-Sampaio, D., Madureira, C., Burns, J.M., Houghton, R.L., David, J.R., Reed, S.G., 1996. rK39: a cloned antigen of Leishmania chagasi that predicts active visceral leishmaniasis. J. Infect. Dis. 173, 758–761. Barata, R.A., Franc¸a-Silva, J.C., Costa, R.T., Fortes-Dias, C.L., Silva, J.C., Paula, E.V., Prata, A., Michalsky, E.M., Dias, E.S., 2004. Phlebotomine sand flies in Porteirinha, an area of American Leishmaniasis transmission in the State of Minas Gerais. Braz. Mem. Inst. Oswaldo Cruz 99, 481–487. Camargo, E.P., 1964. Growth and differentation in Trypanossoma cruzi I. Origen of metaciclic Trypanossoma in liquid medium. Rev. Inst. Med. Trop. Sa˜o Paulo 6, 43–100. Camargo, M.E., Rebonato, C., 1969. Cross-reactivity in fluorescence tests for Trypanossoma and Leishmania antibodies. Am. J. Trop. Med. Hyg. 18, 500–505. Cortes, S., Rolao, N., Ramada, J., Campino, L., 2004. PCR as a rapid and sensitive tool in the diagnosis of human and canine leishmaniasis using Leishmania donovani s.l.—specific kinetoplastid primers. Trans. R. Soc. Trop. Med. Hyg. 98, 12–17. Costa, C.A., Genaro, O., Lana, M., Magalha˜es, P.A., Dias, M., Michalick, M.S.M., Melo, M.N., Costa, R.T., Magalha˜es-Rocha, M.N., Mayrink, W., 1991. Leishmaniose visceral Americana Canina: avaliac¸a˜o da metodologia sorolo´gica utilizada em inque´ritos epidemiolo´gicos. Rev. Soc. Bras. Med. Trop. 24, 21–25. Courtenay, O., Quinnel, R.J., Garcez, L.M., Shaw, E.E., Dye, C., 2002. Infectiousness in a cohort of brazilian dogs: why culling fails to control visceral leishmaniasis in areas of high transmission. J. Infect. Dis. 186, 1314–1320. Cupollilo, E., 1992. Perfil enzima´tico e ana´lise nume´rica de variantes do geˆnero Leishmania encontradas no Novo Mundo. Mem. Inst. Oswaldo Cruz 93, 663–668.

75

Cupollilo, E., Grimaldi Jr., G., Momen, H., 1994. A general classification of New World Leishmania using numerical zymotaxonomy. Am. J. Trop. Med. Hyg. 50, 296–311. Deane, L.M., Deane, M.P., 1955. Leishmaniose visceral urbana (no ca˜o e no homem) em Sobral. Ceara´. O. Hosp. 47, 75–87. Degrave, W., Fernandes, O., Campbell, D., Bozza, M., Lopes, U., 1994. Use of molecular probes and PCR for detection typing of Leishmania—a mini review. Mem. Inst. Oswaldo Cruz 89, 463–469. Di Lorenzo, C., Proietti, F.A., Assunc¸a˜o, R.M., 2000. A urbanizac¸a˜o da leishmaniose visceral no Brasil—uma breve revisa˜o. Rev. Soc. Bras. Med. Trop. 33, 316–317. Dye, C., Killick-Kendrick, R., Vitutia, M.M., Walton, R., KillickKendrick, M., Harith, A.E., Guy, M.W., Canavate, M.C., Hasibeder, G., 1992. Epidemiology of canine leishmaniasis: prevalence, incidence and basis reproduction number calculated from a crosssectional serological survey on the Island of Gozo. Parasitology 105, 35–41. Franc¸a-Silva, J.C., Costa, R.T., Siqueira, A.M., Machado-Coelho, G.L.L., Costa, C.A., Mayrink, W., Vieira, E.P., Costa, J.S., Genaro, O., Nascimento, E., 2003. Epidemiology of canine visceral leishmaniosis in the endemic area of Montes Claros Municipality, Minas Gerais State. Braz. Vet. Parasitol. 111, 161–173. Gradoni, L., Maroli, M., Gramiccia, M., Mancianti, F., 1987. Leishmania infantum infection rates in Phebotomus perniciosus fed on naturally infected dogs under antimonial treatment. Med. Vet. Entomol. 1, 339–342. Guarga, J., Moreno, J., Lucientes, J., Gracia, M., Peribanez, M., Alvar, J., Castillo, J., 2000. Canine leishmaniasis transmission: higher infectivity amongst naturally infected dogs to sand flies is associated with lower proportions of T helper cells. Res. Vet. Sci. 69, 249–253. Hasibeder, G., Dye, C., Carpenter, J., 1992. Mathematical modeling and theory for estimating the basic reproduction number of canine leishmaniasis. Parasitology 105, 43–53. Lainson, R., 1983. The American leishmaniasis some observations on their ecology and epidemiology. Trans. R. Soc. Trop. Med. Hyg. 77, 569–596. Lanotte, G., Rioux, J.A., Perieres, J., Vollhardt, Y., 1979. Ecologie des leishmaniases dans le sud de la France10 Les formes e´volutives de la leishmaniose vesce´rale canine. Elaboration d’ une typologie bio-clinique a` finalite´ e´pide´miologique. An. Parasitol. Hum. Comp. 54, 277–279. Mancianti, F., Gramicia, M., Gradoni, L., Pieri, S., 1988. Studies on canine leishmaniasis control. Evolution of different clinical forms of canine leishmaniasis following antimonial treatment. Trans. R. Soc. Trop. Med. Hyg. 82, 566–567. Margonari, C.S., Pessanha, J.E., Barata, R.A., Michaslky, E.M., Costa, D.C., Dias, E.S., 2004. Study on phlebotomine sandy fly (Diptera: Psychodidae) fauna in Belo Horizonte, state of Minas Gerais. Braz. Mem. Inst. Oswaldo Cruz 99, 795–803. Marsden, P.D., 1979. Leishmaniasis. New Engl. J. Med. 300, 350–352. Marzochi, M.C.A., Marzochi, K.B.F., 1994. Tegumentary and visceral leishmaniases in Brazil—Emerging anthopozoonosis and possibilities for their control. Cad. Sau´de Pu´bl. 10 (Supl. 2), 359–375. Michalsky, E.M., Fortes-Dias, C.L., Pimenta, P.F.P., Secundino, N.F.C., Dias, E.S., 2002. Assessment of PCR in the detection of Leishmania spp. in experimentally infected individual phlebotomine sand flies (Diptera: Psychodidae: Phlebotominae). Rev. Inst. Med. Trop. Sa˜o Paulo 44, 255–259. Michalsky, E.M., Franc¸a-Silva, J.C., Costa, R.T., Costa, D.C., Barata, R.A., Paula, E.V., Machado-Coelho, G.L.L., Rocha, M.F., Dias,

76

E´.M. Michalsky et al. / Veterinary Parasitology 147 (2007) 67–76

E.S., 2005. Leishmaniose visceral Americana: estudo de flebotomı´neos e infecc¸a˜o canina em Montes Claros, Minas Gerais. Bras. Rev. Soc. Bras. Med. Trop. 38, 147–152. Molina, R., Amela, C., Nieto, J., San-Andres, M., Gonzalez, F., Castillo, J.A., Lucientes, J., Alvar, J., 1994. Infectivity of dogs naturally infected with Leishmania infantum to colonized Phlebotomus perniciosus. Trans. R. Soc. Trop. Med. Hyg. 88, 491–493. Momen, H., Salles, C.A., 1985. Enzyme markers for Vibrio cholerae; identification of classical El Tor and enviromental strains. Trans. R. Soc. Trop. Med. Hyg. 79, 773–776. Monteiro, S.P., Lacerda, M.M., Arias, J.R., 1994. Controle da Leishmaniose Visceral Americana no Brasil. Rev. Bras. Med. Trop. 27, 67–72. Peters, W., Evans, D.A., Lanham, S.M., 1983. Importance of parasite identification in cases of leishmanioses. J. R. Soc. Trop. Med. 76, 540–542. Reithinger, R., Coleman, P.G., Alexander, B., Vieira, E.P., Assis, G., Davies, C.R., 2004. Are insecticide-impregnated dog collars a feasible alternative to dog culling as a strategy for controlling canine visceral leishmaniasis in Brazil? Int. J. Parasitol. 34, 55–62. Rocha, M.F., 2002. Validac¸a˜o de teste ra´pido para detecc¸a˜o de anticorpos anti-Leishmania donovani (TRALd) no diagno´stico da Leishmaniose Visceral canina em Montes Claros, Minas Gerais, Brasil. Dissertac¸a˜o de Mestrado na Universidade Federal de Sa˜o Paulo. Shaw, J., 2006. Further thoughts on the use of the name Leishmania (Leishmania) infantum chagasi for the aetiological agent of American visceral leishmaniasis. Mem. Inst. Oswaldo Cruz 101, 577–579.

Sherlock, I.A., 1996. Ecological interactions of visceral leishmaniasis in the State of Bahia. Mem. Inst. Oswaldo Cruz 91, 671–683. Silva, A.R., Viana, G.M., Varonil, C., Pires, B., Nascimento, M.D., Costa, J.M., 1997. Visceral leishmaniasis (kala-azar) on Ilha de Sa˜o Luı´s, Maranha˜o Brazil: its evolution and outlook. Rev. Soc. Bras. Med. Trop. 30, 359–368. Silva, E.S., Gontijo, C.M.F., Pacheco, R.S., Fiuza, V.O.P., Brazil, R.P., 2001. Visceral Leishmaniasis in the Metropolitan region of Belo Horizonte, State of Minas Gerais. Braz. Mem. Inst. Oswaldo Cruz 96, 285–291. Travi, B.L., Tabares, C.J., Cadena, H., Ferro, C., Osorio, Y., 2001. Canine visceral leishmaniasis in Colombia: relationship between clinical and parasitological status and infectivity for sand flies. Am. J. Trop. Med. Hyg. 64, 119–124. Vexenat, J.A., Castro, J.A., Cavalcante, R., Tavares, J.P., Silva, M.R., Batista, W.H., Campos, J.H., Howard, M.K., Frame, I., McNerney, R., 1994. Visceral leishmaniasis in Teresina, State of Piaui Brazil: preliminary observations on the detection and transmissibility of canine and sand fly infections. Mem. Inst. Oswaldo Cruz 89, 131–135. Vieira, J.B.F., Coelho, G.E., 1998. Leishmaniose visceral ou calazar: aspectos epidemiolo´gicos e de controle. Rev. Soc. Bras. Med. Trop. 31 (Supl. 2), 85–92. Woolhouse, M.E.J., Dye, C., Etard, J.F., 1997. Heterogeneities in a transmission of infectious agents: implications for the design of control programs. Proc. Natl. Acad. Sci. 94, 338–342. World Health Organization, 1990. Lucha contra las leishmaniasis. Informe. Te´cnico. Se´r. 793, 177. World Health Organization, 1992. Control of leishmaniasis: report of WHO Expert Committee. Series WHO 793, 139–158.