The role of Phlebotomus alexandri Sinton, 1928 in the transmission of kala-azar*

Bulletin of the World Health Organization, 64 (1): 107-112 (1986) i World Health Organization 1986 The role of Phlebotomus alexandri Sinton, 1928 in the transmission of kala-azar* GUAN LI-REN,1 XU YONG-XIANG, Li BAO-SHAN,3 & DONG JIANG4 Since 1968, kala-azar has been occurring sporadically in Meiyaogou, Turfan County, Xinjiang-Uygur Autonomous Region, wherefour species ofsandflies are known to exist. The present study ofsandflies collected in this areafrom May to August 1983 shows that Phlebotomus alexandri is the only anthropophilic and predominant species, accounting for 81.1% (7716/8843) of the sandfly population. After having beenfed on cotton rats or hamsters infected with Leishmania donovani, 93.9% (230/245) of P. alexandri were in- fected, the promastigotes not only developing in the stomach but also invading the pharynx, buccal cavity andproboscis. Of the 643 female P. alexandri collected in the wild and in houses, 13 (2.0%) were found to be naturally infected, with a distribution of promastigotes similar to that in the experimental infection. All the Chinese hamsters and golden hamsters that had been inoculated with 7 isolates ofpromastigotesfrom naturally infected sandflies developed visceral leishmaniasis. In addition, golden hamsters could be infected experimentally by the bite of P. alexandri which had an infection ofpromastigotes of L. donovani. This is the first demonstration of a species of Paraphlebotomus playing a role in the transmission of kala-azar in the Old World; studies for confirmation by biochemical identification of parasite isolates from naturally infected P. alexandri are now in progress. Since 1968, cases of kala-azar have been occurring sporadically in Meiyaogou (in Turfan County) in the Xinjiang (Sinkiang)-Uygur Autonomous Region of China. While the insect vectors Phlebotomus alexandri Sinton, 1928 and Sergentomyia minutus sinkiangensis Ting & Ho, 1962 were found in this area in 1973 (1) and a new species, S. turfanensis Xiong, Guan & Jin, 1981, was reported in 1975 (2), the three known vectors of kala-azar in China (P. chinensis Newstead,'1916, P. chinensis longiductus Nitzulescu 1931, and P. major wui Yang & Xiong, 1965) were not found. It was inferred that P. alexandri might be the vector of kala-azar in the study area (1) and the * From the Institute of Parasitic Diseases, Chinese Academy of Preventive Medicine, WHO Collaborating Centre for Malaria, Schistosomiasis and Filariasis, 207 Rui Jin Er Lu, Shanghai, China. The original version of this paper (in Chinese) was published in the Chinese Journal of Parasitology and Parasitic Diseases. This English version is published here with kind permission of the Editor of that journal. 1 Assistant Research Fellow, Department of Epidemiology, Institute of Parasitic Diseases. Requests for reprints should be sent to this author. 2 Senior Technician, Department of Epidemiology, Institute of Parasitic Diseases. 3Chief, Parasite Section, Anti-Epidemic Station, Urumqi, Xinjiang-Uygur Autonomous Region, China. 4 Physician, Parasite Section, Anti-Epidemic Station, Urumqi, Xinjiang-Uygur Autonomous Region, China. results of investigations carried out in 1983 on this possibility are described below. MATERIALS AND METHODS Leishmanin test and endemicity of kala-azar in the area The intradermal test (with leishmanin as antigen, prepared in our Institute) was carried out on the Uygur population and on immigrants from non: endemic regions between 1966 and 1983; case histories of kala-azar were reviewed to establish the present status and any endemic tendency of kala- azar. Distribution of various sandfly species Sandflies were collected extensively with tube aspirators or sticky paper for identification. Sandfly behaviour in relation to animals and man Female sandflies were collected before blood- sucking and kept in a cage. An anaesthetized hamster 44 -107- GUAN LI-REN ET AL. (homothermic) or lizard (poikilothermic) was put into the cage and exposed to the bite of the sandflies. After 2 to 6 hours, the sandflies were identified and classified as anthropophilic or cold-blooded feeders. These data were supplemented by taxonomic iden- tification of the sandflies collected from human dwellings and from the bodies of the occupants. Artificial infection of P. alexandri P. alexandri sandflies were collected at dusk from the field by means of human baits and introduced immediately into a cage. Cotton rats (Sigmodon hispidus) or golden hamsters (Mesocricetus auratus), infected with the Xinjiang strain of Leishmania donovani, were exposed to these sandflies. The animals had been anaesthetized by intraperitoneal injection of urethane and their abdominal skins were well shaven before exposure. They were taken out of the cage after 2 hours and those sandflies that were engorged with fresh blood were transferred the next morning into glass tubes with plaster at the bottom, fed 5% glucose solution, and kept at room tempera- ture (26-28 °C). Batches of these sandflies were dissected daily from the 4th day onwards to observe the growth and development of promastigotes of L. donovani in their alimentary tracts. Natural infection of P. alexandri Sandflies were collected from the field or house- holds and those with their blood meals partially or completely digested were dissected and examined for signs of natural infection (promastigotes) in their alimentary tract. Promastigotes, when present, were isolated and inoculated into the peritoneal cavity and skin of normal hamsters (M. auratus and Cricetulus barabensis). About 3 months later, the hamsters were sacrificed and examined for amastigotes and their localization. Experimental transmission of L. donovani to normal hamsters through the bite of P. alexandri P. alexandri, after being fed on cotton rats infected with L. donovani, were kept for 11-12 days and released into a cage; an anaesthetized uninfected golden hamster was then introduced and kept in the cage overnight. The sandflies were examined for blood-taking and infectivity the next day, while golden hamsters that had been exposed to their bites were dissected 5 months later and examined micro- scopically for signs of visceral leishmaniasis. RESULTS Leishmanin test and endemicity of kala-azar in the area The intradermal test with leishmanin was per- formed on the local population in Meiyaogou and the proportion of positive reactions appeared to vary with the duration of residence in the area (Table 1). Among infants who had experienced only one sandfly season, 6 out of 17 showed a positive reaction. The positive rate of this test in the Uygur population was significantly higher than that in the immigrant Han population, owing to the outdoor sleeping habits among the Uygur during the sandfly season. Two autochthonous cases of kala-azar occurred in this area between October 1982 and February 1983: one was a 2+-year-old Uygur child and the other a Han child aged 6 years and 3 months. Both had never been out of this area and were cured by treatment with sodium stibogluconate. Distribution of various sandfly species A total of 8843 sandflies were collected from Meiyaogou and identified. Of the four species dis- Table 1. Results of the leishmanin test on the local population (Uygur and Han) in Meiyaogou, by the duration of residence in the area Uygur population Han population No. of years x2 test of residence No. tested No. positive No. tested No. positive (P value) 1-5 21 7 (33.3)a 59 7 (11.9) <0.05 6-10 48 31 (64.6) 145 20 (13.8) <0.001 11-15 55 36 (65.5) 90 29 (32.2) <0.001 16 10 9 (90.0) 8 3 (37.5) <0.05 Total 134 83 (61.9) 302 59 (19.5) <0.001 a Figures in parentheses are percentages. 108 TRANSMISSION OF KALA-AZAR BY PHLEBOTOMUS ALEXANDRI Table 2. Distribution of various sandfly species in Meiyaogou S. minutus Total number P. alexandri P. mongolensis S. turfanensis sinkiangensis of sandflies Locality captured No. % No. % No. % No. % Wild 6914 5386 77.9 11 0.2 1141 16.5 376 5.4 Houses 1929 1790 92.8 8 0.4 103 5.3 28 1.5 Total 8843 7176 81.1 19 0.2 1244 14.1 404 4.6 covered, two belonged to Phlebotomus (subgenus Paraphlebotomus) P. alexandriand P. mongolensis - and two to Sergentomyia, namely S. turfanensis and S. minutus sinkiangensis. P. alexandri was the predominant species, accounting for 81.1 % of the sandflies captured (Table 2). In Potaogou, a non-endemic area, 138 sandflies were caught on four occasions. They comprised 67.4% S. minutus sinkiangensis, 25.4% P. mongo- lensis, and only 7.2% P. alexandri. The composition of the sandfly population here was thus very different from that in Meiyaogou. Sandfly behaviour in relation to animals and man When kept in a cage with one hamster or two lizards, P. alexandri was found to prefer the hamster (blood-sucking rate, 79.0%) and to refuse the lizard's blood. On the other hand, S. turfanensis and S. minutus sinkiangensis fed only on the lizard's blood (Table 3). Both species of Sergentomyia were ecologically associated with the lizard (Alsophylax microtis) in nature, and nucleated lizard erythrocytes were often found in their stomach. From June to August, out of a total of 837 sandflies collected by human baits in the field at dusk, 835 (99.8/o) were P. alexandri and only 2 (0.2%) were S. turfanensis. Of the 1929 sandflies caught from households, P. alexandri accounted for 92.8% (Table 2). Forty- two (80.8%) blood-meal samples out of 52 obtained from P. alexandri had human blood, as identified by the precipitation test: 29 showed a positive reaction with human antiserum, 13 with antisera from man and sheep, and 1 with antiserum from sheep, 9 samples being negative with the above-mentioned antisera. The results indicate that P. alexandri is an anthro- pophilic species of epidemiological significance in the transmission of kala-azar in the study area; the role of S. minutus sinkiangensis and S. turfanensis as vectors of human leishmaniasis can be ruled out owing to their preference for lizard's blood. Artificial infection of P. alexandri P. alexandri sandflies were dissected between the 4th and 11th days after exposure to cotton rats or hamsters infected with the Xinjiang strain of L. donovani and the result showed a high infection rate among the sandflies (around 90/0o) (Table 4). The promastigotes persisted after the digestion of the blood meal in the stomach of the sandflies. Thus, P. alexandri appears to be a favourable host for the development of L. donovani at 4-5 days after feeding, the promastigotes being often attached to the stomach wall of the sandfly by the tip of their flagella. On the 6th to 7th day, the blood meal had been completely digested in many sandflies and the stomach and proventriculus were filled with pro-. mastigotes, which then invaded the oesophagus and Table 3. Feeding habit of different species of sandf lies P. alexandri S. turfanensis S. minutus sinkiangensis No. No. No. No. No. No. Animal exposed fed exposed fed exposed fed Hamster 400 316 (790)a 32 0 21 0 Lizard 96 0 35 34 (91.7) 27 27 (100) a Figures in parentheses are percentages. 109 ~~~~~~~~~GUANLI-REN ET AL. Table 4. Artificial infection of P. alexandri with L. donovani Distribution of promastigotes in sandflies Days after No. No. Pro- Buccal Eso- Pro- Mid- Hind- feeding dissected infected boscis cavity Pharynx phagus ventriculus gut gut 4 2 2 0 0 0 0 1 2 0 5 1 5 13(86.7)a 0 0 0 6 13 13 3 6 22 21 (95.5) 0 0 1 (4.8) 12 21 21 1 7 27 25 (92.6) 0 0 3 (12.0) 19 25 25 2 8 47 44 (93.6) 0 0 12 (27.3) 36 44 44 2 9 43 39 (90.7) 2 1 13 (33.3) 33 39 39 3 10 50 49 (98.0) 4 0 19 (38.8) 44 49 49 0 1 1 39 37 (94.9) 6 2 19 (51.4) 31 37 37 2 Total 245 230 (93.9) Figures in parentheses are percentages. pharynx. The rate of pharygeal infection increased with time. Rosette formation of the promastigotes was also frequently seen. On the 9th day, the promastigotes had migrated to the proboscis (Table 4). In some heavily infected sandflies, a few rosette- like clusters of promastigotes were found at the junction of the Malpighian tubules and the hindgut, but the rectum was not involved. It was demonstrated that the promastigotes of L. donovani migrated forwards in the digestive tract of P. alexandri, but were not discharged from the anal cave. Natural infection of P. alexandri A total of 643 female P. alexandri sandflies were collected from different places and examined; 13 (2.0%/) were found to be infected with promastigotes. Among these, 12 had completely digested their blood meal, and 6 of them were heavily infected with promastigotes which flocked in the stomach and invaded the pharynx and even the buiccal cavity and probosc'is (Table 5). Seven isolates of promastigotes obtained from naturally infected P. alexandri were inoculated intraperitoneally and subcutaneously into normal hamsters which developed visceral leishmaniasis later. The promastigotes found in naturally infected P. alexandri can tentatively be considered to belong to L. donovani (Table 6), pending biochemical characterization of the isolates from this sandfly. The monthly natural infection rate of P. alexandri was observed from May to August. No infection was found in May (0/13), the first infection appeared on 25 June, and the infection rates in June, July and August were 0.50%(1/213), 2.50/. (4/160) and 3. 10/ (8/257), respectively. This shows that the risk of kala- azar transmission began at the end of June, with July and August as the major transmission seasons in this area. In addition, the examination of 58 S. turfa- nensis and 30 S. minutus sinkiangensis female sand- flies showed no promastigotes in the stomach. Heart blood and liver tissue from 17 lizards were culti- vated in NNN (Novy-Nicolle-MacNeal) culture medium and showed no growth of Leishmania. Experimental transmission of L. donovani to normal hamsters through the bite of P. alexandri Two batches of 18 and 23 sandflies were fed on a cotton rat that had been infected with L. donovani. Table 5. Distribution of promastigotes found in naturally infected P. alexandri Locality of No. No. Buccal collection dissected infected Proboscis cavity Pharynx Esophagus Proventriculus Midgut Dry gully 262 8 (3.1)a 1 1 3 4 7 8 Wild caves 154 4 (2.6) 0 0 2 3 4 4 Human dwellings 227 1 (0.4) 0 0 1 1 1 1 Figures in parentheses are percentages. 110 TRANSMISSION OF KALA-AZAR BY PHLEBOTOMUS ALEXANDRI Table 6. Result of inoculation of hamsters with promastigotes from naturally infected P. alexandri Amastigotes in hamster Days from Sandfly Experimental inoculation Bone- Lymph- Subcutaneous No. animal to dissection Spleen Liver marrow node tissue 1 Cricetulus barabensis 97 + + + + 2 C. barabensis 88 + + + + 6 Mesocricetus auratus 86 + + + - - 7 C. barabensis 88 + + + + + 10 M. auratus 90 + 11 C. barabensis 93 + + 13 C. barabensis 93 + + + + + After 11 and 12 days, respectively, two normal golden hamsters were exposed to their bite. Only 3 sandflies took blood from No. 1 hamster and 7 from No. 2. As these sandflies were gravid, the blood volume en- gorged corresponded only to a third to a half of their abdomen's capacity. All the 10 sandflies containing fresh blood were dissected and promastigotes were found in 6 of them. Except for one sandfly which had been fed on hamster No. 1 and showed light infection with promastigotes in the midgut only, the other 5 sandflies were heavily infected with promastigotes in the stomach, proventricules and pharynx, and even in the proboscis (in one of them). Hamster No. 1 was dissected on day 147 after having been bitten by artificially infected P. alexandri and hamster No. 2 on day 145. Splenomegaly devel- oped in hamster No. 1 and hepatomegaly in No. 2. Smears of liver, spleen, bone-marrow, lymphnodes, and subcutaneous tissue were prepared. Amastigotes could be found on Giemsa-stained smears of the liver and spleen from both animals and in the lymphnodes of hamster No. 1. This result shows that the golden hamster could get visceral leishmaniasis through the bite of P. alexandri which had been infected with promastigotes of L. donovani. DISCUSSION In 1957 it was reported that members of the P. major group were responsible for kala-azar in the Old World (3), and a few years later species of the "Synphlebotomus" complex (closely related to the P. major group) were reported to be vectors of the disease in East Africa (4, 5). In general, species of the subgenus Paraphlebotomus were incriminated as vectors for the transmission of cutaneous leishmaniasis only. For instance, based on epidemio- logical data, the role of P. alexandri in the trans- mission of cutaneous leishmaniasis was recognized in certain regions of Central Asia (USSR) and Turkey (6, 7), and its natural infection with an unidentified species of Leishmania in Khuzestan province of Iran has also been reported (8). In the area of Heishanhu of Jiuquan County, Gansu province of China, where L. gerbilli was present, natural infection with flagel- lates was observed in P. alexandri which was considered to be a vector of gerbil leishmaniasis (9). In the same area, in P. alexandri sandflies that had experimentally been infected through the biting of hamsters parasitized with L. donovani, it was re- ported that promastigotes developed well in the stomach and could invade the pharynx and, oc- casionally, the proboscis; this species of sandfly was therefore considered to possess the character- istics essential for the transmission of kala-azar (10). In the present study, P. alexandri has for the first time been shown to be a vector of kala-azar in the Old World by field observations and laboratory experimentation. The region of Meiyaogou occupies a gully formed by running water from the Tianshan mountain cutting the stone desert at the foot of the mountain, the banks of the gully being rocky precipices. This is an isolated area far from the old endemic foci of kala- azar. Before 1966, only at the end of the gully were there a few residents of Uygur nationality. With the development of this area, both Uygur and Han populations gradually moved in. In 1968, the first case of autochthonous infantile kala-azar occurred in an 8-month-old Han child. Thereafter, 1-8 cases of kala-azar occurred every year after the sandfly season. Man became infected with kala-azar after entering these originally desert areas for production III 112 GUAN LI-REN ET AL. and other activities. In 1983, only two baby dogs were identified which had not experienced the sandfly season; they were therefore not likely to be the source of natural infection of P. alexandri. It is presumed that L. donovani existed in this area long before its occupation by people, probably in wild animals yet to be discovered. Further studies are needed to elucidate the existence of a natural reservoir from an epidemiological point of view. In 1974 another focus of kala-azar in some newly established farms on the borders of the Takelamagan desert in Xinjiang was reported (11). The sole local sandfly was the wild species P. major wui, and the natural infection rate of promastigotes of L. dono- vani during the sandfly season was 2.9%. Feeding on hamsters infected with L. donovani, 85.1% of P. major wui were infected and the promastigotes could invade the pharynx. It was considered that this species of sandfly was the vector of kala-azar in this dry desert area (11). However, in the very different stony desert area of Meiyaogou P. alexandri was clearly the most predominant species and P. major wui was absent. These results indicate that sandflies of different species may be active in kala-azar trans- mission in different regions, which needs to be con- firmed by further study. In conclusion, the present findings on P. alexandri meet most of the criteria for vector incrimination laid down by the WHO Expert Committee on Leish- maniasis (12); studies on the biochemical identifi- cation of parasite isolates from this sandfly are now being pursued for confirmation. RESUMt LE ROLE DE PHLEBOTOMUS ALEXANDRI SINTON, 1928 DANS LA TRANSMISSION DU KALA-AZAR Depuis 1968 des cas sporadiques de kala-azar s'observent au Meiyaogou, district de Turfan, Region Autonome Ouigoure du Xinjiang. Quatre especes de phl6botomes, a savoir Phlebotomus alexandri Sinton, 1928, P. mongolensis Sinton, 1928, Sergentomyia turfanensis Xiong, Guan et Jin, 1981 et S. minutus sinkiangensis Ting et Ho, 1962 ont e captur&es dans cette region du mois de mai au mois d'aoiut 1983. P. alexandri, 1'espece predominante, represente 81,1/o (7716/8843) de la population totale de phlebotomes. C'est aussi la seule espce anthropophile. Apres s'etre nourries sur des sigmodons ou des hamsters infestes par Leishmania donovani, 93,9% (230/245) des P. alexandri ont ete infestes a leur tour et les promastigotes se sont developpes non seulement dans 1'estomac mais ont aussi enhavi le pharynx, la cavite buccale et la trompe. Parmi les 643 P. alexandri femelles collect6es a 1'ext6rieur et dans les habitations, 13 (2,0%) ont e trouv&es infest&es de promas- tigotes, avec une distribution analogue a celle de l'infestation experimentale. Tous les hamsters chinois et les hamsters dores inocules avec 7 isolements de promastigotes provenant des phl6botomes naturellement infestes ont fait une leishmaniose viscerale. C'est la premiere fois que l'on demontre qu'une espece du sous-genre Paraphlebotomus peut jouer un role dans la transmission du kala-azar dans I'ancien monde. Des etudes sont en cours en vue d'identifier du point de vue biochimique des isolements de parasites provenant de P. alexandri naturellement infestes. REFERENCES 1. ANTI-EPIDEMIC STATION OF XINJIANG-UYGUR AUTO- NOMOUS REGION. Collection of public health and pre- ventive medicine papers, No. 10: 48-53 (1973) (in Chinese). 2. HSIUNG, K. H. ET AL. New records of sandflies and description of a new species, Sergentomyia turfanensis sp. nov. in north-western China. Acta entomologica Sinica, 24: 430-435 (1981) (abstract in English). 3. ADLER, S. & THEODOR, 0. Transmission of disease agents by phlebotomine sandflies. Annual review of entomology, 2: 203-226 (1957). 4. HEISCH, R. B. ET AL. In pursuit of the vector of kala- azar in Kenya. British medical journal, 1: 1456-1458 (1962). 5. MANSON-BAHR, P. E. C. ET AL. Recent research on kala-azar in East Africa. Journal of tropical medicine and hygiene, 67: 79-84 (1964). 6. PETRISCEVA, P. A. The natural focality of leishmaniasis in the USSR. Bulletin of the World Health Organization, 44: 567-576 (1971). 7. HouIN, R. ET AL. Phlebotomes du sud de la Turquie. Annales de parasitologie humaine et comparee, 46: 633-652 (1971). 8. JAVADIAN, F. ET AL. Natural leptomonad infection of sandflies with its first occurrence in Ph. alexandri in Khuzestan Province, Iran. Protozoological abstracts, 3: 229 (1979). 9. WANG, J. ET AL. The leishmania infection found in Rhombomys opimus, its transmission and relation to man. Acta parasitologica Sinica, 1: 17-23 (1964) (abstract in English). 10. XIONG, G. H. ET AL. Experimental infection of Ph. alexandri and Ph. mongolensis with Leishmania donovani. Acta zoologica Sinica, 15: 607-610 (1963) (abstract in English). - 11. HSIUNG, K. H. ET AL. Study on vectors of kala-azar transmission in Xinjiang. Journal of control and research of epidemic diseases, No. 4: 327-334 (1974). 12. WHO Technical Report Series, No. 701, 1984 (The leishmaniases: report of a WHO Expert Committee).