Cutaneous and visceral leishmaniasis co-infection in dogs from Rio de Janeiro , Brazil : evaluation by specifi c PCR and RFLP-PCR assays

Introduction: During a diagnostic evaluation of canine visceral leishmaniasis (VL), two of seventeen dogs were found to be co-infected by Leishmania (Viannia) braziliensis and Leishmania (Leishmania) chagasi. Methods: Specifi c polymerase chain reaction (PCR) and restriction fragment length polymorphism-PCR (RFLP-PCR) assays were performed. Results: PCR assays for Leishmania subgenus identifi cation followed by RFLP-PCR analysis in biopsies from cutaneous lesions and the spleen confi rmed the presence of Leishmania (Viannia) braziliensis and Leishmania (Leishmania) chagasi in those fragments. Conclusions: This report reinforces the importance of using serological and molecular techniques in the epidemiological surveillance of canine populations in endemic areas in which both diseases are known to co-exist. In such cases, a reassessment of the control measures is required.

Visceral leishmaniasis (VL) and tegumentary leishmaniasis (TL) are zoonoses of great importance for public health.In the State of Rio de Janeiro, Leishmania (Viannia) braziliensis is the most prevalent species implicated in the epidemiological cycle of TL.Its transmission occurs in periurban areas in which primitive rain forest vegetation is being depredated due to disorderly human occupation 1 , and infections in man, dogs and horses have been reported 2,3 .Canine VL, which is caused by Leishmania (Leishmania) chagasi, is endemic in the Municipality of Rio de Janeiro.Dogs represent one of the main reservoirs in urban areas in which the disease has been observed due to several factors that infl uence the natural epidemiological scenario 4 .The overlapping transmission of TL caused by L. (V.) braziliensis and VL caused by L. (L.) chagasi has been reported in certain areas of the Municipality of Rio de Janeiro.Mixed infection with both parasites has already been reported in a patient 5 and in a dog 6 .Fortunately, according to Marzochi et al. 7 , no new human cases of co-infection have been registered since, but there is concern about the persistence of canine seroprevalence.Control measures are based on interrupting the transmission cycle, which involves the diagnosis and treatment of human cases and vector control through insecticides and serological screening, with the subsequent culling of dogs found to be seropositive 4 .
The present article discusses the detection of mixed TL and VL infections in two of seventeen dogs from endemic areas of Rio de Janeiro, Brazil, which tested seropositive by indirect immunofl uorescence (IIF) analysis of serum samples.The occurrence of canine cases with both diseases in the same geographic area impairs the diagnosis and implementation of control measures.
The animals included in this study were referred to the Zoonosis Service of the Instituto de Pesquisa Clínica Evandro Chagas-Fundação Oswaldo Cruz (IPEC-FIOCRUZ) with an indication for euthanasia according to the recommendations of the Brazilian Program for the Control of Leishmaniasis 4 after serological tests by IIF on serum samples, which were performed by the Epidemiology Service of the Municipality of Rio de Janeiro.This study was approved by the Ethics Committee on Animal Experimentation of the Fundação Oswaldo Cruz (CEUA/FIOCRUZ; program n o L-023/06).
All studied dogs were from urban and periurban areas of the Municipality of Rio de Janeiro and presented IIF titers ranging from 1:80 to 1:1,280.In certain animals, clinical symptoms of VL were evident, whereas others were asymptomatic.Cutaneous lesions were frequent (Table 1).
Fragments of the cutaneous lesions, intact skin from the scapular region, cervical lymph nodes and spleen from  all animals were collected after thiopental-overdose euthanasia and submitted for polymerase chain reaction (PCR) analyses using primers for the variable regions of kinetoplast DNA (kDNA) minicircles.Specifi c primers for the L. braziliensis complex (5'-GGGGTTGGTGTAATATAGTGG-3' and 5'-CTAATTGTGCACGGGGAGG-3') 8 and for the Leishmania donovani complex (5'-CCAGTTTCCCGCCCCG-3' and 5'-GGGGTTGGTGTAAAATAG-3') 9 were adopted, as previously described 10,11 .Then, the amplifi ed PCR products were visualized on agarose gels.Restriction fragment length polymorphism-polymerase chain reaction (RFLP-PCR) analyses were performed with a panel of four restriction enzymes (Msp I, Rsa I, Hinf I and Mbo I) to confi rm the specifi city of the amplifi ed kDNA minicircle products.
All of the canine biopsy fragments, except for two from cutaneous lesions, produced the expected 800bp diagnostic bands after PCR with primers D1/D2.Those fragments of cutaneous lesions that tested negative were submitted to PCR assays with the primers B1/B2, and the expected 750bp diagnostic bands were observed (Figure 1A).
PCR was performed in combination with RFLP-PCR to confi rm the presence of L. (V.) braziliensis deoxyribonucleic acid (DNA) in the cutaneous lesion biopsies and L. (L.) chagasi DNA in the spleen and lymph node fragments from two dogs (dogs A43 and A63, Table 1).In Figures 1B, 1C, 1D and 1E, the PCR/RFLP results confi rming the mixed infection are observed.Because previous results from our group 12,13 have demonstrated that the restriction enzymes Msp I, Rsa I, Hinf I and Mbo I are the most appropriate for typing Leishmania species from the subgenera Viannia and Leishmania, these enzymes were adopted in the present study.The Msp I restriction enzyme linearizes the kDNA minicircles, displaying a major band of approximately 750bp in the case of Leishmania (V.) species.In contrast, a polymorphic restriction profi le is always observed in L (L.) chagasi 12 .
Herein, the Hinf I restriction patterns of the amplifi ed parasitic DNA from cutaneous lesions were similar to the patterns of the L. (V.) braziliensis (MHOM/BR/75/M2903) reference strain kDNA.The Rsa I restriction patterns obtained with the amplifi ed products from the spleen and lymph node fragments were also similar compared with the patterns of the L. (L.) chagasi (MHOM/BR/74/PP75) reference strain kDNA (Figure 1C).Both subgenera were confi rmed after digestion with Msp I (Figure 1E).RFLP-PCR analysis with the restriction enzymes Msp I, Rsa I and Mbo I corroborated the PCR results, justifying the use of such a technique in Leishmania species identifi cation.
In clinical-epidemiological surveillance, emphasizing the need to perform extensive sampling in dogs with cutaneous lesions is important, particularly in endemic areas in which both diseases overlap, such as certain rural areas in Rio de Janeiro.In these areas, where closely related etiological groups are present, the interpretation of serological data is a limiting aspect due to possible serological cross-reactions.Although Trypanosoma cruzi infection is unknown in the Municipality of Rio de Janeiro, a new species, Trypanosoma caninum, was recently described in dogs 14 .
In this scenario, a reassessment of control measures is required.The control of leishmaniasis is relatively complex, particularly in areas both the tegumentary and visceral forms of the disease co-exist.As a control measure, the Brazilian government usually culls seropositive dogs 4 .However, according to a recent review 15 , the strategy of killing dogs is hampered for several reasons, including the low accuracy of the methods used to assess the infectivity of dogs and the high replacement rate of these animals.In this scenario, a search for sensitive and specifi c molecular tools is needed to distinguish dogs infected with L. (V.) braziliensis, thus preventing unnecessary sacrifi ce.The results presented here show the usefulness of specifi c PCR assays and the RFLP technique for differentiating between L. (V.) braziliensis and L. (L.) chagasi and may contribute to providing support for control programs.Conversely, dogs with TL and VL co-infection would be subjected to euthanasia according to the guidelines of the Ministry of Health.Concerning control measures, including the detection and treatment of human cases, the disposal of dogs with VL, the efforts to control vectors with systematic indoor and outdoor spraying and the use of collars and mosquito nets impregnated with insecticides, the latter measure alone would likely be more effi cient than the fi rst two measures together.Finally, the development of human vaccines should also be cons idered as a high priority.
The authors declare that there is no confl ict of interest.

Pires
MQ et al -Cutaneous and visceral leishmaniasis co-infection in dogs

TABLE 1 -
The serological titers, clinical status and PCR results of the 17 dogs from Rio de Janeiro.PCR with specifi c primers for Leishmania donovani PCR: polymerase chain reaction; A: absent; *mixed infection.