Canine visceral leishmaniasis (CVL) is a worldwide zoonosis prevalent in approximately 50 countries in the Mediterranean basin, Middle East and South America [¹]. In Brazil, the parasite Leishmania infantum is the cause of CVL [²,³] and the sand fly Lutzomyia longipalpis is the principal blood-sucking vector. High infection rates are evident in areas of the country where environmental degradation and disorderly migration of people are related to the urbanization process [⁴,⁵]. Dogs are the principal reservoir for the parasite, playing a central role in transmission to humans, and therefore present a serious public health concern [⁶]. CVL infection is considerably more prevalent than clinical illness in endemic areas; this has been reported both in Brazil and Europe (Mediterranean basin countries) [^7-13]. Furthermore, clinical signs of the disease are highly variable and can include lymphadenopathy [¹⁴,¹⁵], skin lesions [^16-18], progressive weight loss [¹⁹], hepatosplenomegaly [²⁰,²¹], ocular and musculoskeletal abnormalities, renal disease and epistaxis [^22-24].

Gastrointestinal tract (GIT) disorders occur in human visceral leishmaniasis (HVL) [²⁵] and in dogs both in natural [^26-31] and experimental infection [³²,³³]. Hervás et al. [³⁴] described pathoanatomical alterations in a hemorrhagic stomach during necropsy of a sylvester reservoir in the jackal (Canis aureus); diagnosis was confirmed by the presence of macrophages parasitized with amastigotes forms of Leishmania. The majority of studies have described clinical signs and the subsequent inflammatory response in canidae predominantly infected with Leishmania infantum from the Mediterranean basin. In Brazil, Silva et al. [³⁵] described inflammatory lesions in one clinically unaffected dog that were comparable to those described in the literature. A chronic inflammatory exudate composed of macrophages, plasma cells and lymphocytes, with rare neutrophils and eosinophils, was found throughout the mucosa of the small and large intestine. Immunohistochemical analysis revealed that macrophages of the lamina propria were parasitized with immunolabelled intracellular amastigote forms of Leishmania. The authors reported a chronic inflammatory reaction along segments of the GIT involving all histological layers. Subsequently, Toplu and Aydogan [³⁶] carried out a study in 22 dogs naturally infected with Leishmania in Turkey and observed a mild to severe mononuclear cell infiltration, predominantly in the lamina propria, in the small and large intestines of all dogs as described by Silva et al. [³⁵]. However, parasites were evident in only 30 percent of cases.

The aim of this study was to provide a systematic pathological and parasitological overview of the GIT of dogs naturally infected with L. infantum from an endemic metropolitan area of Belo Horizonte, Minas Gerais, Brazil.

Gross examination revealed no severe changes in any part of the GIT mucosa of the dogs included in the study. However, five dogs (20%) (two asymptomatic and three symptomatic dogs) contained helminthes and showed minor focal hyperemia, but no hemorrhages or ulcers were evident.

In infected animals, an increased number of cells, focal or diffuse, was observed in lamina propria, muscularis mucosae and submucosa of GIT, while no changes were found in the control dogs. A chronic cellular exudate was observed in all cases of Groups I and II and was composed predominately by macrophages, plasma cells and lymphocytes with rare polymorphonuclear neutrophils or eosinophils (Figure 1A-E). Macrophages containing many Leishmania amastigotes were easily found in lamina propria of the mucosa and the submucosa in the majority of the cases. Many of the macrophages showed a pale and abundant cytoplasm and less dense nucleus typical of epithelioid cells (Figure 1F and 1G). Multinucleated giant cells and epithelioid cells were associated with areas in which mononuclear cells were more concentrated, but no typical granulomatous reaction was evident in the lamina propria (Figure 1H).

Under microscopic analysis, we also observed that in spite of the parasitism we did not find mucosal erosions. Independently of the defined clinical status of the infected dogs, microscopic analysis revealed the presence of amastigote forms of Leishmania in all GIT layers, particularly in the cells (macrophages) of lamina propria (Figure 2A-D) without severe GIT mucosal alteration. In fact, immunolabeled amastigotes were evident in all intestinal segments, but the parasitism was more evidenced (frequency) in the caecum and colon than in other GIT segments. In addition a higher parasite load of amastigote forms of Leishmania were evident in these segments (Table 1). However, no statistical difference was observed between Groups I and II (Table 2). Thus, in clinically unaffected as well as the affected dogs, parasites were most numerous in the colon, followed by the caecum, while the other GIT segments contained relatively few parasites (Tables 3 and 4, respectively).

Histology demonstrated an increased number of macrophages frequently parasitized with Leishmania amastigotes, plasma cells and lymphocytes throughout the GIT layers of all segments of the Groups I and II. However, parasites were predominantly located in the mucosa (lamina propria) of the GIT layers. These results are in accordance with Anderson et al. [²⁸], Longstaffe and Guy [⁴⁰] and Toplu et al. [³⁶].

When comparing the GIT segments of all infected dogs we demonstrated the highest parasite load in the ceacum and colon. These results are consistent with Keenan et al. [⁴¹] González et al. [³³] Ferrer et al. [³⁰] and Adamama-Moraitou et al. [³¹]. Keenan et al. [⁴¹] observed comparable parasitological loading in all GIT segments of two distinct groups of German shepherd dogs experimentally infected with L. chagasi and L. donovani. González et al. [³³] observed a severe chronic inflammatory process in the mucosa and submucosa of the colon and rectum of infected beagles where they described the surface and the epithelium of the crypts of Lieberkϋhn with progressive degeneration characterized by cellular swelling. As a result, focal micro-erosions developed in the mucosal surface reducing the area of the large bowel available for absorption and causing diarrhea. Therefore, the authors concluded that chronic colitis in beagles was caused by L. infantum, and that the diarrhea was consistent with a disorder prevailing in the large bowel. However, Ferrer et al. [³⁰] reported chronic colitis and diarrhea in two dogs naturally infected with L. infantum, but questioned whether the concurrent lympho-plasmacytic infiltrate of the colonic mucosa was caused by Leishmania (detected by immunohistochemistry) or was an incidental finding, particularly relevant if the parasite load is low. The authors concluded that there was no evidence of a definitive pathogenic correlation between CVL and chronic colitis. Moreover, Adamama-Moraitou et al. [³¹] demonstrated that 32.3% of clinically affected dogs naturally infected with L. infantum presented with asymptomatic colitis. Endoscopic examination of colonic mucosa demonstrated focal hyperemia, edema and mild erosions in infected dogs, which were comparable to those observed in dogs with CVL or other infections including Histoplasma, Salmonella and Yersinia, or infected with parasites such as Giardia, Trichuris, Ancylostoma, Entamoeba and Balantidium. However, the authors excluded these infections after fecal examination. As reported by Chiapella [⁴²], it is important to consider idiopathic large bowel inflammatory diseases such as plasmacytic-lymphocytic colitis and histiocytic-ulcerative colitis before making a diagnosis.

No severe macroscopic and microscopic lesions in the mucosa of the GIT during necropsy were evident during the present study. In some dogs the mucosa was somewhat hyperemic, and helminth parasites were observed in less than 20% of cases (three clinically affected and two unaffected dogs). Despite the presence of helminth parasites in the guts of some dogs, it was concluded that this did not interfere with the histological analysis of GIT segments from dogs with CVL, as no severe macroscopic or microscopic related alterations were evident, such as hemorrhages, ulcers or pyogenic granulomas. In the present study, an increased number of macrophages, frequently parasitized with Leishmania amastigotes, and plasma cells and lymphocytes, were observed throughout the GIT layers. However, as depicted in Figure 2, despite the presence of Leishmania amastigotes in macrophages of the lamina propria, only discrete atrophy of the mucosa epithelium cells without any severe lesions (necrotic or erosive mucosal lesions) was observed. González et al. [³³] described the surface and the epithelium of the crypts of Lieberkϋhn as showing progressive degeneration characterized by cellular swelling, but with only focal micro-erosions, in the mucosa of the colon and rectum. Epithelioid and multinuclear cell formations were observed in some cases, although typical granulomas were not observed. However, Adamama-Moraitou et al. [³¹] reported that a granulomatous inflammatory pattern was the most common histological feature in the colonic mucosa and that this could be associated with a mildly eroded colonic mucosa.

In our study, clinically affected dogs harbored more parasites in the caecum and colon than in other GIT segments. This is consistent with other reports Anderson et al. [²⁸]; Ferrer et al. [³⁰]; González et al. [³³]; Silva et al. [³⁵]; Longstaffe and Guy [⁴⁰]; Keenan et al. [⁴¹] where the difference in parasitic loading between specific GIT segments was observed in clinically affected dogs. However, it is important to reiterate that in our studies also clinically unaffected dogs harbored parasites in all GIT segments.

We observed a high parasite burden throughout the GIT mucosa, but the pathological changes were relatively mild. Thus, this may led us to consider whether Leishmania gains an advantage from the intestinal immunoregulatory response (immunological tolerance). Such question will require further profound research and will help to elucidate the mechanisms underlying Leishmania infection.

Research was supported by grants Conselho Nacional de Desenvolvimento da Pesquisa Tecnológica e Científica (CNPq/473601/2009-5) and Fundação de Amparo a Pesquisa do Estado de Minas Gerais (FAPEMIG/CBB 00219/09-80 and APQ-01355-09), Brazil. Students fellowship financed by Conselho Nacional de Desenvolvimento da Pesquisa Tecnológica e Científica (CNPq) and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Brazil. We do not have any kind of commercial interesting or competition with other researches. It is only an academic manuscript.

AJWP, MMF and FLS did all the clinical exams, necropsies and histology. AJWP, MMF, TM did morphometrical and statistical analysis. MSMM was responsible for all serological exams. WLT, WLT advisors, revise all the histological analysis and the manuscript.

All authors read and approved the final manuscript.

The authors thank Dr. Hélio Chiarini-Garcia (Associate Professor), Departamento de Morfologia, Instituto de Ciências Biológicas da UFMG, Belo Horizonte, Brasil, for photographic support.