| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Phlebotomine sand flies (Diptera: Phlebotominae) are vectors of leishmaniasis, a disease caused by several species of the genus Leishmania (Kinetoplastida: Trypanosomatidae). Different species of the parasite are transmitted by specific sand fly vectors. Sanliurfa, located in southeastern Turkey on the border with Syria, is an endemic focus of anthroponotic cutaneous leishmaniasis (ACL) caused by genetically homogeneous Leishmania tropica.1,2 Phlebotomus sergenti, a vector of ACL, and P. papatasi, a vector of zoonotic CL, are the two most abundant sand fly species at this site.3–6
We report here more detailed studies of these sand flies in Sanliurfa to further understand aspects of their ecology by recording their spatial distribution in houses, the temperature and humidity in the sand fly habitats, and sources of blood for females. Small mammals that may serve as blood source and potential reservoirs of CL were trapped in this focus.
Sand flies were collected in five districts in the old part of the city during September 1999. Centers for Disease Control and Prevention (Atlanta, GA) miniature light traps were placed in different locations (yards, cellars, rooms, sheds, and roof tops) of the houses of patients with active or previous ACL. Typically, nine traps were set every night (three traps in each house) for 11 consecutive days. Traps were set before sunset and collected the next morning. Humidity and/or temperature were recorded using Bionaire Climate Check (Bionaire, Montreal, Canada) and a maximal-minimal thermometer (Exatherm, 
elezn
Brod, Czech Republic) placed near the traps. Species of live and dead fly specimens were identified as described elsewhere.6
Dissected guts from blood-fed females were smeared onto filter paper (No. 3; Whatman, Newton, MA), air-dried and stored at -20°C before use. The degree of blood digestion in the fly guts was graded into three levels: 1) fresh blood (bright red content in the midgut with intact erythrocytes visible under the microscope), 2) partially digested blood (dark red), and 3) extensively digested blood (brown).
The blood meals in fly guts were screened for their sources of origin by an enzyme-linked immunosorbent assay (ELISA) using commercially available antisera specific to different animal immunoglobulins. Filter papers containing blood meals from individual flies were eluted overnight at 4°C in 800 µl of Tris-buffered saline (TBS, 20 mM Tris, 150 mM NaCl, 0.05% Tween 20, 0.1 mM N
-tosyl-lysine-chloromethylketone, pH 7.6). Samples were homogenized mechanically and centrifuged at 9,000 x g for 10 minutes at 4°C to collect the supernatant. Microtitration plates (Gama, 
eské Bud
jovice, Czech Republic) were coated for one hour with primary antibodies against immunoglobulins of seven host species representing the readily available blood sources. Primary antibodies (swine anti-human, swine anti-bovine, swine anti-mouse, swine anti-rat, swine anti-goat, rabbit anti-chicken, and rabbit anti-horse immunoglobulins; Sevac, Prague, Czech Republic) were diluted in coating buffer (1 µg/ml of IgG in 20 mM TBS, pH 9). Plates were subsequently blocked by adding 3% rabbit serum for 15 minutes, then incubated with blood meal samples (100 µl/well) in TBS for two hours. The wells were then washed and incubated for one hour with peroxidase-conjugated anti-IgG antibodies (swine anti-human, swine anti-bovine, swine anti-mouse, swine anti-rat [Sevac], rabbit anti-chicken [A-9043; Sigma, St. Louis, MO], rabbit anti-horse [A-6917; Sigma], and monoclonal rabbit anti-goat/sheep [A-9452; Sigma]) at dilutions of 1:500-1: 2,000 in TBS. Wells without blood meals served as negative controls. The substrate buffer contained o-phenylenediamine in phosphate-citrate (0.11 M NaHPO4, 0.5 M citric acid, pH 5.5) and 0.03% hydrogen peroxide. The reaction was stopped with 10% sulfuric acid. Plates were read on Multiscan reader (Labsystems, Helsinki, Finland) at 492 nm using Multiscan Transmit software. In all experiments, the lower limit of a positive reaction was taken as 2-3 mean values of the negative samples.
The anti-immunoglobulin antisera were prescreened to verify their antigenic specificity by reactions with blood samples from different animals. These samples were prepared exactly as described for the sand fly blood meals, including the steps of smearing on filter paper, storage, and elution. Cross-reactions were noted only between closely related animals, i.e., anti-chicken antibody cross-reacted with turkey (Meleagris gallopavo) blood, anti-goat antibody with sheep blood, and anti-rat (Rattus norvegicus) antibody with black rat (Rattus rattus) blood. The anti-rat antiserum did not cross-react with mouse blood. This specificity was further verified with the gut contents of Lutzomyia longipalpis fed on mice or black rats. The anti-rat antibodies showed a positive reaction with black rat blood, but not with mouse blood in the fly guts and vice versa. Statistical tests were performed using Statistica 6.0 (STATSoft, Tulsa, OK) and Statgraphics 4.2 (StatPoint, Englewood Cliffs, NJ) software.
Rodents and insectivores were collected at the same sites for fly collections in August 1998. Snap traps (n = 130 trap-nights) with standard baits (pieces of wick fried in fat with flour), and wooden and live traps (Tomahawk Live Trap Co., Tomahawk, WI) (n = 155 trap-nights) were set in kitchens, basements, and cellars of houses where sand fly collection was under way. Traps were also set in fig orchards and small fields in the vicinity of the town (70 snap traps). Biopsy samples were obtained from the ears, tail, lymph nodes, liver, and spleen of trapped animals for culture in SNB-9 blood agar at 25°C to check for the presence of Leishmania. The specimens were weighed and measured; species were identified according to Harrison and Bates monograph.7 These collections are now kept in the Department of Zoology at Charles University (Prague, Czech Republic).
More than 99% (2,129) of the sand flies trapped were P. sergenti and P. papatasi, with the former species twice as abundant (1,388) and representing 65% of the total. Phlebotomus sergenti predominated in the sand fly fauna sampled from 1998 to 2000.5,6 The male:female ratio was 1.34 (
2 = 29, P < 0.0001) for P. sergenti and 0.94 for P. papatasi (2 = 0.7, P = 0.4).
Sand flies of both species were most abundant in sheds, followed by cellars, and less so in rooms, yards, and roof tops (F = 5.7, P = 0.0004, by analysis of variance [ANOVA]; Table 1
). However, the ratio of P. sergenti to P. papatasi differed significantly in these locations (2 = 231, P < 0.0001). Phlebotomus papatasi was more abundant in sheds, representing 56% of trapped flies, while P. sergenti was more abundant in rooms (82%). This difference was even more pronounced for females (2 = 200, P < 0.0001). Phlebotomus sergenti females represented only 32% of the females trapped in sheds, but they represented 88% of 130 females trapped in rooms. Previously, P. sergenti was also found to be more abundant in houses than in stables;3 thus, this behavior may enhance the probability of its feeding on humans, thereby increasing the transmission of L. tropica in this focus.
|
Of the 588 dissected females, 120 were found to contain blood in their guts. Of these, the percentage of fed P. papatasi was significantly higher (43 of 157, 27%) than that of fed P. sergenti (76 of 430, 18%; 2 = 6.7, P = 0.01). One fed female was identified as P. brevis. Blood meal samples were tested by ELISA for their sources of origin from different animals with seven different antisera. Sixty-four samples (53%) were positive with at least one antiserum. Identifiable blood meals relative to the total samples examined were proportionally similar for both species (58% in P. sergenti and 46% in P. papatasi; 2 = 1.43, P = 0.23). However, the efficacy of the ELISA differed significantly for the two fly species in relation to the levels of blood digestion (Table 2). This efficacy did not affect the identification in P. papatasi (2 = 2.0, P = 0.37), but did so in P. sergenti (2 = 39.1, P < 0.00001). In fact, the origin of the blood meals could not be identified in the latter species beyond level three of digestion. Likewise, Lu. longipalpis blood meals could no longer be identified when they turned dark.8 Our sandwich ELISA is comparable in sensitivity to the competitive ELISA previously used for P. perniciosus, in which 59% of the blood meals are identifiable.9 However, only freshly engorged females may have been tested in this study because they were not dissected, as in other previous studies.10,11 This finding, together with the use of different detection methods (microplate precipitin test, Ouchterlony’s double diffusion test), makes comparison of these results difficult. In samples with level 1 blood digestion, we were able to identify 73% of the blood meals tested, similar to the 83% and 74% of positive rates reported in previous studies.10,11 Countercurrent immuno-electrophoresis, which was used in another study with P. ariasi,12 identified 89% of the blood meals, which was comparable to our results for freshly engorged P. sergenti. The results of our experiments showed that the ELISA used is simple and sensitive, yielding a high percentage of positive identifications even in females with partially digested blood, and after the guts have been checked microscopically for the presence of promastigotes.
|
). Poultry is the predominant source of blood found for P. sergenti, representing 65% of the identified blood meals. This source represents only 18% of the blood meals for P. papatasi (2 = 16.3, P = 0.00005).
|
2 = 3.0, P = 0.55). Of the 15 samples with multiple blood meals, 14 contained either mouse or rat blood. Rodents have been shown to display efficient defensive behavior against Culex nigripalpus mosquitoes,13 and we believe that this behavior is also displayed against sand flies, which interrupts their feeding and causes them to feed consecutively on another host. The presence of domestic animals and wild rodents at the same sites than explains 23% frequency of multiple meals. Other sand fly species were also shown to feed on multiple hosts, e.g., 16% of P. perniciosus and 5% of P. ariasi tested.9,12 Our results suggest that the local population of P. sergenti is highly ornithophilic. However, the host-feeding pattern described refers only to the relative frequency of blood source detection in the blood meal samples, which does not necessarily imply a higher preference for a particular host. Phlebotomus sergenti was abundant in cellars where poultry was kept; chicken were shown to display the lowest defensive behavior among several avian and mammalian species.13 Its availability and low irritability affects frequent feeding on poultry; however, the role of poultry as a risk factor in the local focus of ACL is not clear. Poultry kept in cellars represent a rich source of blood for P. sergenti females to sustain their high population numbers. The abundance of sand flies increases vector-host contact and thus the chance for transmission. In addition, we suspect that these chicken cellars may serve as an important breeding place for sand flies in Sanliurfa as they do for other sand fly species.14 Conversely, certain components of bird blood, namely, nucleated turkey erythrocytes, were shown to inhibit the development of L. major in P. papatasi.15 Therefore, a negative effect of chicken blood on L. tropica in P. sergenti cannot be excluded.
Four species of small mammals were trapped, with black rats and house mice being especially abundant (Table 4). Tristram’s jirds were also trapped near human settlements, representing potential reservoir host for zoonotic CL. Results of ELISAs for the blood meals from collected flies showed that they feed on mice and rats. Phlebotomus sergenti and P. papatasi originating in Sanliurfa and raised in laboratory colonies feed on various mammals, including mice and black rats. Moreover, both laboratory mice16 and feral black rats (Svobodová M, unpublished data) are susceptible to L. tropica from the same site. The potential role of these rodents in the transmission of L. tropica is thus suggested, although none of those caught in the field was positive for Leishmania, as assessed by cultivation.
|
Received May 6, 2002. Accepted for publication August 29, 2002.
Acknowledgments: We thank Feridun Akkafa (Harran University, Sanliurfa, Turkey), Seray Ozensoy and Yusuf Ozbel (Ege University, Izmir, Turkey), Jan Vot
Financial support: This work was supported by the Ministry of Education of the Czech Republic (grants GAUK 78/1998/BBio and J13/ 981131-B4) to Petr Volf and Milena Svobodová, and by an anonymous source to K.-P. Chang.
pka, Eva Dvoráková (Charles University, Prague, Czech Republic), and Kadri Bulut (Harrankapi Health Center, Sanliurfa, Turkey) for assistance in the field.
Reprint requests: Milena Svobodová, Department of Parasitology, Charles University, Faculty of Science, Vinicná 7, CZ-128 43 Prague 2, Czech Republic, Telephone: 420-2-2195-1814, Fax: 420-2-2491-9704, E-mail: milena{at}natur.cuni.cz
Authors’ addresses: Milena Svobodová, Jovana Sádlová, and Petr Volf, Department of Parasitology, Faculty of Science, Charles University, Vinicná 7, CZ-128 43 Prague 2, Czech Republic. K.-P. Chang Department of Microbiology/Immunology, University of Health Sciences/Chicago Medical School, North Chicago, IL 60064.
This article has been cited by other articles:
|
|
 |
|
  N. Haouas, B. Pesson, R. Boudabous, J.-P. Dedet, H. Babba, and C. Ravel Development of a Molecular Tool for the Identification of Leishmania Reservoir Hosts by Blood Meal Analysis in the Insect Vectors Am J Trop Med Hyg, December 1, 2007; 77(6): 1054 - 1059. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
