Figure 1

Figure 1. Sandflies infected with Leishmania infantum from US foxhounds, showing blocked stomodeal valve. Development of L. infantum (MCAN/US/2001/ FOXYMO1) in laboratory-reared Lutzomyia longipalpis sandflies led to stomodeal valve blockage 10–13 days after...

A total of 300 laboratory-reared female Lu. longipalpis sandflies were allowed to feed on 2 hounds naturally infected with L. infantum, strain MCAN/US/2001/FOXYMO1 or a closely related strain. During 2007–2011, the hounds had been tested for infection with Leishmania spp. by ELISA, PCR, and Dual Path Platform Test (Chembio Diagnostic Systems, Inc. Medford, NY, USA (Table 1). L. infantum development in these sandflies was assessed by dissecting flies starting at 72 hours after feeding and every other day thereafter. Migration and attachment of parasites to the stomodeal valve of the sandfly and formation of a gel-like plug were evident at 10 days after feeding (Figure 1), indicating successful parasite development.

Next, to determine sandfly capacity to transmit the US strain of L. infantum to a susceptible vertebrate host, we allowed L. infantum–naive and L. infantum–infected sandflies to feed on 7 L. infantum–naive hamsters for 13 days. For confirmation of L. infantum infection, we dissected the alimentary tract of sandflies that fed on the hamsters (Table 2). A total of 30 sandflies were used for feeding on hamsters; 11 flies fed and were subjected to Leishmania detection by PCR, which confirmed Leishmania positivity for 5 sandflies. Hamster blood samples were collected 2 weeks after infection and monthly for 5 months. L. infantum DNA was detected in hamster blood by quantitative PCR (qPCR) and was present in hamster nos. 1 (at 2 mo), 2 (at 3 mo), 5 (at 4 mo), and 6 (at 4 mo) with cycle thresholds of 43.88, 28.27, 34.38, and 45 respectively. Cycle thresholds <45 were considered positive for L. infantum (5).

On hamster no. 5, a cutaneous lesion consistent with Leishmania infection persisted for 1 month. Tissue from this lesion was harvested to assay for Leishmania infection. Increased numbers of macrophages and granulocytes were present in the dermal layer. Bacteria found in the tissue probably represented secondary infection, a common sequela of canine VL. Cellular infiltrate was observed, indicative of inflammation and infection (Technical Appendix). No L. infantum parasites were observed on slides stained with hematoxylin and eosin, periodic acid–Schiff, or Giemsa, and lesion tissue was negative for L. infantum by qPCR (data not shown).

Figure 2

Figure 2. Visceralization of Leishmania infantum from US foxhounds, transmitted by sandflies into hamsters. Leishmania spp.–specific quantitative PCR was performed, and parasite load was calculated from a standard curve. Horizontal bars indicate mean...

qPCR was performed to quantify parasite load within common Leishmania-infected organs from all hamsters. The mean quantities of Leishmania DNA amplified from spleen, bone marrow, and lymph node from hamsters on which Leishmania-infected sandflies had fed were 12-, 22-, and 11-fold greater than that from hamsters on which Leishmania-naive sandflies had fed (Figure 2). According to extrapolation from a PCR standard curve similar to one previously used (5), the highest parasite load was in bone marrow, which contained an average of 1,238 (±282) parasites/mg tissue.

In the United States, parasites from VL-symptomatic, naturally infected hunting hounds remain highly infectious to Leishmania-competent Lu. longipalpis sandflies. Parasites that had fed on L. infantum–infected hounds were able to develop fully within sandflies and to be subsequently transmitted to and disseminated within hamsters.

The capacity of Lu. shannoni and Lu. longipalpis sandflies to acquire L. infantum from naturally infected dogs has been compared in leishmaniasis-endemic Brazil (13). Although lower infection rates were observed in Lu. shannoni (9%) than in Lu. longipalpis (36%) sandflies, the intensity of infection (200–500 promastigotes/fly) was higher in Lu. shannoni sandflies. This finding demonstrates a potentially lower threshold for acquiring infection from Lu. shannoni sandflies because the infectious dose per sandfly was greater. Lu. shannoni sandflies are commonly found within the United States and also in areas where L. infantum–infected hounds were reported (4). These data demonstrate the risk for vectorborne transmission of zoonotic VL from these dogs in the United States.

Despite our use of an L. infantum strain that is primarily, if not solely, transmitted via vertical transmission between dogs in the United States, we were able to measure substantial parasite loads in sandflies that fed on these dogs and in the bone marrow, spleen, and peripheral lymph nodes of hamsters on which infected sandflies had fed (Figures 1, 2). Parasite DNA was not amplified in the liver, possibly because of lower parasite loads in the liver during later infection, as demonstrated in experimental VL infections of mice (14). Therefore, the US strain of L. infantum that is circulating in North American hunting hounds has not lost virulence factors that facilitate adherence to sandfly gut and facilitate transmission, and subsequent dissemination, in a secondary host.

This study focused on the possibility that domestic hounds serve as reservoir hosts for L. infantum within the United States; however, other potential L. infantum reservoirs include coyotes, foxes, and opossums. When leishmaniasis was found to be reemerging among hounds in the United States in 2000, a total of 291 wild canids were trapped and tested (15). No serologic evidence of infection was found, but these studies were limited to the southeastern United States; further study is needed to rule out the possibility that enzootic cycles of transmission do not exist within wild canids. The range covered by Lu. shannoni sandflies overlaps that of reservoir species including coyotes, foxes, and hunting hounds. Occurrence of Leishmania vectors in areas of naturally infected hounds indicates a coalescence of components for establishment of a sylvatic and/or domestic cycle of L. infantum. Diagnostic testing and preventive measures should be considered for dog breeds known to harbor L. infantum.

In the United States, L. infantum is circulating among dogs. Despite the fact that vertical transmission maintains VL within the hound population (5), L. infantum was able to fully develop in sandflies and be further transmitted to a susceptible vertebrate host. Symptomatic hounds were highly infectious to sandflies. L. infantum strain MCAN/US/2001/FOXYMO1, similar to the common European zymodeme MON-1, circulating dog-to-dog in North America maintained all necessary requirements for complete development within sandflies. Overlap of sandfly infections (e.g., Lu. shannoni, and L. infantumi) in hounds may put companion dogs and humans at risk and could pose an emerging risk for L. infantum–triggered clinical disease in at-risk populations in North America.