Visceral Leishmaniasis (VL) is a form of zoonosis that is caused by protozoa of the genus Leishmania 1,2 . In Latin America, Leishmania (Infantum) chagasi is the predominant disease-causing species 8,24,25,31,47,48 . Infected dogs can exhibit a marked humoral immune response characterized by high levels of IgG anti-Leishmania antibodies, which, despite their upregulation, do not confer protection 20 . It is estimated that the time between infection and sero-conversion ranges from three months to seven years 6 .
Despite the wide range of tests already developed, the indirect fluorescent antibody test (IFAT) is the conventional method that is currently employed for epidemiological canine surveys. The euthanasia of dogs that test positive is a measure recommended by the Ministry of Health for the control of VL 16,34,37,43 . This measure was recommended as the result of canine populations constituting a major reservoir of the disease, representing the threat of canine-to-human disease transmission 3,35 . The determination of the precise CVL sero-conversion period in endemic areas therefore represents an important factor in the control of VL.
By IFAT, dogs with high antibody titers are associated with active parasitism, the presence of clinical symptoms 16 and an elevated probability of the presence of amastigotes as detected by parasitological examination 17,40,42 . It has therefore been suggested that the serological profile be used as an indicator for susceptibility or resistance to CVL 5,16 .
Dogs affected with VL generally present with renal changes 11 , manifested specifically by glomerulonephritis and interstitial nephritis 14 . Despite the high prevalence of renal pathologies, the development of renal failure is only evident when two thirds to three quarters of the renal tissue in both kidneys is compromised 27 .
Kidney disease in CVL can progress from mild proteinuria to nephrotic syndrome, developing progressively into renal failure, which is the leading cause of death among dogs 11 . The clinical evolution of CVL therefore promotes serological, parasitological and biochemical changes which directly correlate with the clinical status of the affected dog.
This study aims to define the time period in which dogs seroconvert during VL infection in the endemic area of Teresina in the state of Piauí, and to evaluate the functional and structural impairment of the kidneys during the early post-infection period.
MATERIALS AND METHODS
From a population of 88 dogs from the University Veterinary Hospital, 40 (70.1%) dogs which tested negative for VL were monitored in their homes every three months for one year. The dogs examined in this study ranged from eight months to 12.2 years of age and both males and females of undefined races were included. The animals were evaluated clinically for the diagnosis of CVL and were subjected to blood collection, bone marrow aspiration and popliteal lymph node biopsy under sedation and local anesthesia.
The detection of Leishmania using anti-Leishmania antibodies was performed by IFAT (lote 08ULC0127/Biomanguinhos/FIOCRUZ, Rio de Janeiro, RJ, Brazil) in sera diluted at 1:40, 1:80, 1:160 and 1:320. Parasitological examinations were performed on smears of popliteal lymph node aspirates and bone marrow samples. The uninfected dogs were monitored by clinical and parasitological examination and IFAT at the homes of their owners every three months for a maximal period of one year until the time of sero-conversion, whereupon the dogs were separated into the following three groups: 11 dogs with anti-Leishmania antibodies and more than three clinical symptoms for VL (seropositive), seven dogs with anti-leishmania antibodies and up to three clinical symptoms for VL (seropositive) and seven dogs exhibiting positive serology (seropositive) and no clinical symptoms for VL 41 . The control group was comprised of 11 uninfected dogs without clinical symptoms for VL. For clinical evaluation, we considered the following manifestations: skin lesions, lymphadenopathy, onychogryphosis, weight loss, alopecia, conjunctivitis, pale mucous membranes, muzzle keratinization, depigmentation of the muzzle, edema and apathy.
The characterization of Leishmania species was performed using PCR on kidney samples from dogs exhibiting positive serological profiles for VL, using sequences from the ITS (internal transcribed spacer) of the ribosomal RNA (rRNA) gene.
The experimental protocols used in this study were approved by the Ethics Committees on Animal Experimentation of the Federal University of Piauí under No. 015/2010.
After an interval of five months of sero-conversion, blood samples were further collected from nine dogs exhibiting positive serology to monitor renal function.
Serum samples were used to assess the levels of urea (Labtest No. 104-4), creatinine (kit Labtest No. 35), total protein (Labtest No. 99) and albumin (Labtest No. 19) using the semi-automated SB 190 method. The results were analyzed based on normal values for the species 22 .
We evaluated 40 dogs which tested negative for CVL and were distributed among various neighborhoods of the city of Teresina in the state of Piauí, including the districts of moderate transmission (Santo Antonio, Angelim and Satélite) 18 . The twenty-seven total neighborhoods in the city of Teresina visited to collect materials were distributed among five urban areas: Central, North, East, South and Southeast 45 .
The six dogs that were determined to be positive for VL by serological or parasitological examination and that presented more than three clinical symptoms for VL (seropositive) were enrolled for study with the permission of the owners, who signed a consent form. The dogs were transported to the sector of Animal Pathology of UFPI, where they were anesthetized with sodium thiopental 25 mg/kg intravenously 15,32 , euthanized and necropsied to collect kidney samples. The tissue samples were formalin-fixed, paraffin-embedded and then processed and stained with H&E, PAS and Massons trichrome stains 27 . Other renal tissue samples were prepared and subjected to immunohistochemical analysis using a mouse polyclonal, anti-Leishmania amazonensis antibody at a dilution of 1:400 and the EnVision + peroxidase system for the amplification reaction (Dako Corporation Carpinteria, CA, USA, catalog number K4002). The signal development was performed using 0.3 mg/mL 3,3-diaminobenzidine in PBS containing 0.06% hydrogen peroxide and counterstained with Harrys hematoxylin stain.
The renal abnormalities were classified according to the criteria of the World Health Organization to define the morphological patterns of glomerulonephritis 9,51 . The data were tabulated, graphed and analyzed using Sigma Stat statistical software (Bioestat, 5.0) for non-parametric tests (Mann-Whitney U test). p-values < 0.05 were considered statistically significant.
From a population of 88 dogs from the University Veterinary Hospital, 57 (64.7%) dogs that initially tested negative for CVL comprised the sero-conversion study group in our investigation. Of this group, 40 dogs (70.1%) were monitored in their homes every three months for one year.
Of the 40 dogs that initially tested negative for VL, 25 (62.5%) exhibited positive serologic tests during the study period. Of these dogs, 15 (60%) tested positive within three months (2nd serum collection), five (20%) tested positive within six months (3rd serum collection) and five (20%) tested positive within nine months (4th serum collection).
The PCR performed on samples of renal tissue from three dogs confirmed that the infection was caused by Leishmania (infantum) chagasi.
At the 12-month follow-up of the VL-positive dogs, 11 (44%) developed more than three clinical symptoms, seven (28%) presented up to three clinical symptoms and seven (28%) did not develop clinical symptoms. The percentage of dogs that tested positive for VL without manifesting clinical symptoms was therefore less (28%) than that of animals that presented clinical symptoms (72%). Clinical symptoms, such as lymphadenopathy and skin lesions, were the most frequent manifestations observed (Fig. 1).
Of the 18 dogs presenting clinical manifestations of VL (11 with more than three clinical symptoms and seven with up to three symptoms), 10 tested positive within an interval of three months (seven with more than three clinical symptoms and three with up to three clinical symptoms) and in the group of dogs without clinical symptoms, five also tested positive within an interval of three months. The largest number of dogs that tested positive for VL during the study period were those located within the urban neighborhoods of the North (24%) and South (24%), followed by the urban neighborhoods of the Central (20%), East (16%) and Southeast (16%) regions.
To evaluate the main indices of the health status of these animals that might contribute to disease progression, we performed serological and biochemical tests to evaluate renal function, followed by histopathological examination of the kidney tissue samples.
The majority of the dogs presented serological titers of anti-Leishmania antibodies between 1:40 and 1:80, regardless of their presentation of clinical symptoms. Although we found higher values of urea, creatinine and total protein in infected dogs than in uninfected dogs (Table 1), the differences that we observed were not significant. However, comparison of the data revealed that total protein (Mann-Whitney rank sum test, p = 0.0302) and globulin (t-test, p = 0.0214) values were significantly higher and that albumin values were (t-test, p = 0.0386) significantly lower in the infected dogs presenting with more than three clinical symptoms compared to the dogs in the control group (Fig. 2). The results of the biochemical measurements of the infected and uninfected dogs as well as the reference values are shown in Table 1.
|Clinical status of the dogs||Reference values*||Ages (year)||Seroconversion||Serology||District|
|Dogs presenting with more than three clinical symptoms||1||23.5||0.7||7.7||2.90||4.80||3.7||3 months||1:40||Central|
|Dogs presenting up to three clinical symptoms||12||57.0||1.1||9.1||3.39||5.71||6.7||9 months||1:40||East|
|Dogs with no clinical symptoms||19||38.4||1.0||10.8||2.87||7.95||7.11||3 months||1:40||North|
KANEKO et al., 2007.
Renal function tests revealed that among the 25 infected dogs, two (8%) presented with urea (dogs No. 20 and 23) and creatinine (dogs No. 2 and 10) levels that were above the normal range (Table 1). In general however, infected dogs presented with elevated blood urea nitrogen (BUN) and creatinine levels that were higher than in uninfected dogs (Fig. 3), although the differences were not significant.
Six newly infected dogs were euthanized for histological examination to evaluate the renal lesions. All of the dogs presented with two or more clinical symptoms of VL, such as weight loss, keratinization on the nose, conjunctivitis, skin lesions, seborrhea, enlarged lymph nodes, onychogryphosis and alopecia.
Microscopic examination revealed glomerulonephritis (GN) in five dogs, minimal change GN in two dogs (dogs Nos. 14 and 16), focal segmental glomerulosclerosis in one dog (dog No. 13) (Fig. 4A) and membranoproliferative GN (Fig. 4B) in two dogs (dog No. 4 and 8), according to previously described morphological characteristics 12 . The glomerular lesions were focal and of medium intensity. Perivascular mononuclear inflammatory infiltrates were observed in the corticomedullary and cortical regions that were proximal to blood vessels and tubules, ranging in intensity from minimal to average (Fig. 4C). Hyaline casts (Fig. 4D) were present in two dogs, one with minimal change GN and one with focal segmental glomerulosclerosis, which exhibited normal urea and creatinine levels. The PAS staining revealed the presence of medium intensity tubular atrophy with focal distribution in one dog. Immunostaining of the Leishmania antigen was observed in glomerular cells, in tubular epithelial cells (Fig. 4E) and in inflammatory cells of all the recently infected dogs.
Of the 40 dogs that tested negative, it was possible to monitor the 25 dogs that subsequently tested positive for VL during the study period. The loss of follow-up of 15 dogs was attributed to the death of some dogs, the owners absence due to travel which precluded the collection of material, the change of owners from other cities and even the refusal of owners to continue their participation in our study. Although such obstacles have been previously described by other studies 49 , these difficulties hampered the monitoring and evaluation of all of the animals.
We observed that the dogs in the endemic areas of Teresina became infected within a short period of time, with 60% seroconverting after three months of follow-up. Increased periods of sero-conversion were observed in similar studies. In the endemic areas of Rio de Janeiro, 60 dogs were examined and 50% became seropositive over one year 49 . A study of 23 dogs in an endemic area of Italy revealed that 19 dogs tested PCR- or IFAT-positive at eight months and that four dogs tested positive after 12 months 39 . The results of sero-conversion in this study represent a unique epidemiological parameter for the endemic areas of Teresina, contributing to the improved control of VL. This new information can serve as a guideline for the sampling frequency for analysis, for the faster identification of seropositive dogs and for the expedited removal of the animals from the areas transmission. Because of the findings highlighted above, similar studies have been conducted in other regions 1,7,30,36,37,38,49 . However, the results were unique to each particular endemic area because each area has its own specific characteristics related to the canine population, socio-economic conditions of the human population, health conditions, sanitation and vector populations with different densities.
During the one-year follow-up, we found that most of the 25 seropositive dogs presented with more than three clinical symptoms. A study of 12 dogs experimentally infected with L. infantum followed over a period of six months showed that all dogs became positive within four months after infection. The dogs presented with clinical symptoms such as lymphadenopathy and lesions of skin 29 , which are similar to the results of this study in which dogs were naturally infected.
Our results therefore suggest that even with short infection times, the presence of clinical manifestations contributes to the transmission of VL, as has been demonstrated in other studies 33,44,50 . The dogs in Teresina can contribute greatly to the intensification of the infectious cycle of the disease, which is demonstrated by the high incidence of CVL in this city.
According to our results, a larger number of infected dogs were domiciled in the North and South urban districts of Teresina city. Currently, more intense control measures (vector control by spraying with insecticides) have been used in six districts of the urban zone of Teresina, two of which are located in the North district (Santa Maria da Codipi, Pedra Mole), three of which are in the South district (Santo Antonio, Angelim, Esplanada) and one in the Southeast district (Satellite). These districts are reported to have an average of three to five cases/year of human visceral leishmaniasis 18 , confirming its high transmissibility. The factors related to the high incidence of disease in these urban areas 46 were associated with poor sanitation systems. Of the 169,771 households in Teresina, only 22,108 (13.06%) have a general network of sewers 18 . These areas are characterized by high levels of poverty, low-income families, households with poor living conditions, unpaved streets, close proximity to natural lakes surrounded by slums as observed in the Northern region and poor infrastructure, all of which contribute to the high transmission rates of VL 18 .
In general, the dogs that become seropositive for VL and that are more susceptible to the disease have high titers of anti-Leishmania antibodies 20 . In the present study, we found low titers of anti-Leishmania antibodies among the dogs examined. In this regard, there are two serological profiles for dogs with VL. The first is a profile in which the animals initially have low antibody titers which remain low, whereas in the other profile the dogs start with low antibody titers but they tend to increase over the period of infection 49 . Our study is consistent with these profiles because most of the dogs examined exhibited low antibody titers when they subsequently tested positive for VL, despite later developing clinical manifestations. The fact that most of the dogs in our study did not exhibit a positive parasitological examination can be explained by the low antibody titers observed, considering that a positive parasitological test is directly related to an increase in antibody titers 28,40,42 . This issue still requires further clarification because the dogs without clinical symptoms can exhibit high titers of antibodies without presenting parasitological positivity in the cultures of popliteal lymph node aspirate 49 .
Most of the dogs that seroconverted, especially the group presenting with more than three clinical symptoms, exhibited hyperproteinemia that was attributed to hyperglobulinemia and was significantly increased compared to the control dogs. These results revealed that even dogs that were recently infected exhibited changes in serum protein levels, indicating the importance of hyperproteinemia and particularly of hyperglobulinemia as potential diagnostic markers for CVL, consistent with previous studies 14,19,43 . In contrast, previous studies have also shown that dogs without clinical manifestations exhibited reduced antibody titers, lower levels of total protein and of globulin and increased albuminglobulin ratios 4 , which differs from our findings of hyperglobulinemia and hypoalbuminemia in infected dogs without clinical symptoms and is similar to dogs with clinical manifestations.
The increased levels of total protein, globulin, urea and creatinine and the lower levels of albumin observed in four infected dogs compared to control dogs indicate that dogs with VL tend to present with alterations in renal function at early stages of infection without the subsequent development of renal failure. Renal failure is a predominant cause of death in dogs with progressive VL 11 . Studies of chronically infected dogs of the endemic areas of São José de Ribamar in the state of Maranhão found high levels of urea, without changes in creatinine levels 14 . These higher levels of urea in a greater number of dogs might be attributed to pre-existing liver disease 26 , which is a possibility that we cannot exclude for the dogs examined in this study since VL can cause liver damage 44 . AMUSATEGUI et al. 4 showed that at advanced stages of CVL, the mean values of urea and creatinine were high, whereas the dogs in the initial phase of infection exhibited lower levels of antibodies. Furthermore, the levels of urea and creatinine did not differ from those found in control dogs, which is consistent with our findings. We suggest that the biochemical assessment of serum from dogs with VL should be taken into account together with other biochemical analyses to aid in the diagnosis and assessment of the progression of infection, as has also been proposed in other studies 14,23 .
The six infected dogs were euthanized for the collection of kidney tissue samples and the subsequent analysis of renal lesions. These dogs presented with clinical symptoms and exhibited positive serological or parasitological tests. The limited number of animals euthanized has been justified by difficulties also observed by SILVA et al. 49 , who reported that the elimination of dogs without clinical symptoms of VL is difficult, as the owners do not easily relinquish their dogs for euthanasia. Furthermore, we observed that many owners of infected dogs preferred treatment over euthanasia.
Using immunohistochemistry, the Leishmania antigen was observed in glomerular cells, tubular epithelial cells and interstitial inflammatory cells, confirming the relationship between Leishmania and renal lesions, which is consistent with our previous studies 11,12,13 . Two dogs with normal levels of urea and creatinine exhibited hyaline casts (Dogs. No. 4 and 16) in renal tubules, demonstrating that proteinuria can occur even in the absence of severe structural changes of the kidneys, as observed by light microscopy in the cases of minimal change GN and of medium intensity membranoproliferative GN 51,52 present in these dogs.
The description of the patterns of glomerular lesions diagnosed in this study was previously described by COSTA et al. 11 . The results revealed that the glomerular lesions are present in the early phase of natural infection by Leishmania (Infantum) chagasi, although we cannot completely exclude the possibility of other pre-existing infections, even when examining resident dogs. In general, the glomerular and tubular interstitial changes were of low intensity, with mild elevation of BUN and creatinine levels in infected dogs compared to uninfected dogs. However, these changes were insufficient to cause renal failure, which is consistent with the results from nephropathy studies, emphasizing that the functional alterations of the kidneys are observed only when 2/3 or 3/4 of tissue from both kidneys is impaired 10,21,26 .
The infected dogs monitored for one year that were not euthanized after six months were again subjected to biochemical blood analyses to assess renal function. Based on biochemical evidence, there was no worsening of renal function. The renal changes that occur during the course of CVL progression are therefore likely to evolve slowly, as these changes were not detected by any biochemical tests regardless of the post-infection time period, but rather were detected by histopathological examination in the recently infected dogs.
Based on the results of this study, we found that Teresina seroconversion to CVL occurs within a short time period (within three months of exposure) and that the infected dogs are therefore quickly included in the epidemiological pool of VL transmitters, contributing to the maintenance and spread of the disease. Most newly infected dogs presented with clinical symptoms of VL but did not exhibit high titers of anti-Leishmania antibodies. The infected dogs exhibited changes in serum proteins as well as high levels of urea and creatinine compared to uninfected dogs, which are likely to result from the renal lesions which must be taken into account for the early diagnosis of VL. Finally, we suggest that the sero-epidemiological surveys for CVL in Teresina must be performed within shorter time periods than it is performed currently (annually).