JOURNAL
OF INVERTEBRATE
PATHOLOGY
58, 189-202 (1991)
Entomopathogens of Phlebotomine Sand Flies: Laboratory Experiments and Natural Infections ALON Medical
Entomology
Unit,
Malaria
WARBURG
Section, Department of Health, Bethesda,
of Health Maryland
and Human 20892
Services,
National
Institute
Received August 8, 1990; accepted October 26, 1990 The susceptibility of different geographical strains of Phlebotomus papatasi to a cytoplasmic polyhedrosis virus (CPV) was determined experimentally by feeding polyhedra to larvae. Of the Indian P. papatasi, 15.6% became infected, whereas Egyptian P. papatasi were mostly refractory. Infection rates were not augmented in colony flies from the Jordan Valley, 23.8% of which were naturally infected with CPV. The infectivity of Serratia marcescens and Beauvaria bassiana to P. papatasi were determined experimentally. A suspension of B. bassiana spores or S. marcescens bacteria, ingested by P. papatasi in sucrose solution, did not significantly augment mortality rates or reduce the number of eggs oviposited. However, B. bassiana spores smeared on a titter paper constituting 1 or 5% of the surface area available to flies induced 100% mortality of P. papatasi on days 5 and 4. respectively. Mortality in Lutzomyia longipalpis reached 100% on day 4. There were markedly lower mortality rates in the control groups and more eggs were produced by these females (P. papatasi: control = 48.5; experimental = 0.9-1.6 eggs/female: L. longipalpis: control = 17.1; experimental = 0 eggs/female). From wild-caught Colombian Lutzomyia spp., a nonfluorescent pseudomonas, an Entomophthorales fungus, and a Trypanosomatid protozoon (probably Leptomonas) were isolated in culture media. Gregarines (Ascogregarina saraviae) and nematodes (Tylenchida and Spirurida) were also recorded. In laboratory-reared flies, an ectoparasitic fungus was associated with high mortality rates of first instar Lutzomyia spp. larvae. Opportunistic ectoparasitic aggregates of bacteria, yeast, and fungi on the tarsi of colonized L. longipalpis and P. papatasi hindered their mobility and were associated with reduced colony vigor. Aspergillus jZavus, B. bassiana, and S. marcescens were isolated from laboratory-bred P. papatasi adults. o 191 Academic
Press. Inc.
KEY WORDS: Ascogregarina; Beauvaria bassiana; cytoplasmic polyhedrosis viruses; Diptera; entomopathogens; Entomophthorales; leishmaniasis; Leptomonas; nematodes; Lutzomyia gomezi; Lutzomyia towxsendi;
longipalpis: Lutzomyia
Lutzomyia lichyi; trapidoi; Phlebotomus
rurida; Trypanosomatidae;
Lutzomyia papatasi;
pia;
Lutzomyia
shannoni; Lutzomyia marcescens; Spi-
Psychodidae; Serratia
Tylenchida: yeast.
INTRODUCTION Phlebotomine sand flies, (Diptera: Psychodidae) of the genera Lutzomyiu (Americas) and Phlebotomus (Old World), transmit human leishmaniasis, bartonelosis, and sand-fly fever. The causative agents of these diseases are protozoan parasites, bacteria-like organisms, and phleboviruses, respectively (Adler, 1964; Killick-Kendrick, 1979; Tesh, 1988). Sand flies also transmit various other pathogens of vertebrates including Leishmania, Plasmodium, Hemogregarina, and Trypanosoma of reptiles and amphibians (Killick-Kendrick, 1979; McConnel and Correa, 1964).
Collecting sand fly larvae is an arduous task; only small numbers are found and their specie cannot be determined with confidence (Killick-Kendrick, 1987 and references therein). Consequently, the vast majority of entomogenous organisms (Young and Lewis, 1977, 1980) were found in adult sand flies dissected for leishmaniasis studies. The bias toward adults may explain why Phlebotomines appear to be parasitized to a lesser extent than dipterans with aquatic larvae (Adler, 1964; Ashford, 1974). However, limited pathogen dissemination in soil and the resultant low transmission rates between larvae may also contribute to this phenomenon. 189 0022-2011/91 $1.50 Copyright 0 1991 by Academic Press, Inc. AU rights of reproduction in any form reserved.
190
ALON
WARBURG
At least 80% (&5 weeks per generation, estimate based on laboratory breeding) of the life cycle of sand flies is spent in cool, dark, and humid locations. Eggs are oviposited in forest leaf litter, tree buttresses, caves, rodent burrows, cracks in walls, etc. (Killick-Kendrick, 1979, 1987). Larvae develop in these habitats and feed on moist, decaying organic matter and pupate in situ. Even adults select humid and dark diurnal resting sites. The prevailing conditions in such places are also conducive to the development and long term persistence of many entomopathogens. To ascertain the feasibility of biological control, phlebotomine larval biology requires further study, efficacious pathogens must be identified, and adequate delivery systems developed. This report attempts a preliminary assessment of the potential usefulness of some naturally occurring sand fly pathogens. The first part describes the results of infection studies with viruses, fungi, and bacteria isolated from colony-bred flies. Particular attention was paid to possible methods for field application. In the second part, observations were made on naturally occurring entomopathogens. The results and observations are discussed. emphasizing their possible implication for biological control of sand flies. MATERIALS
AND METHODS
Cytoplasmic polyhedrosis viruses (CPVs). The macerated guts of Phfebotomus papatasi naturally infected with CPV
were suspended in water and soaked in larval food. Groups of first instar larvae were kept on this food until it was completely consumed. Larvae were maintained routinely (Modi and Tesh, 1983) thereafter. Infections in adult flies’ midguts were monitored microscopically under phase-contrast illumination. Ingested Serratia marcescens and Beauvaria bassiana. Groups of 18-25 P. papatasi (females 24 hr after blood feeding) were confined in oviposition containers with plaster of paris (height 5 cm, diameter 6 cm)
and maintained at 24” + 1°C in 95 ? 5% relative humidity. A fresh suspension of infective units (B. bassiana spores or S. marcescens) in 30% (w/v) aqueous sucrose solution was offered to the flies daily for 8 days. Control groups were maintained on autoclaved 30% sucrose. B. bassiana spores on jilter papers. A 4or l-cm2 filter paper was smeared on both sides (respectively, about 5 and 1.3% of the surface area available to flies) with B. bassiana spores and suspended inside oviposition containers (as above). The flies were maintained on autoclaved sucrose solution. Sand flies. Laboratory colonies were maintained according to Modi and Tesh (1983). For pathogen isolation, flies were collected as previously described (Warburg et al., 1991). Prior to dissection, carbon dioxide-anesthetized flies were briefly immersed in 70% ethanol, washed in sterile water, immersed in 5% hypochlorite for 5 min, and washed in sterile water. They were dissected in sterile phosphate-buffered saline (PBS) and examined microscopically under phase-contrast illumination. Zsolation and culture. Infected insects were placed on appropriate media: nutrient agar (NA) and brain heart infusion for bacteria and Saboraud dextrose agar (SDA) and potato dextrose agar (PDA) for fungi. Media were fortified with 5% fetal bovine serum (FBS) or 5% yeast extract as necessary. Flagellates were seeded onto blood agar slants overlaied with PBS + 5% (v/v) FBS, 1000 IU ml-’ of penicillin, 1000 ug ml-’ of streptomycin, and 2000 kg ml-i of 5-fluorocytosine (5-FC). In subsequent passages 5-FC was not used and the concentrations of penicillin and streptomycin were reduced to 100 IU ml-’ and 100 pg ml-‘, respectively. Long term storage of fungi and bacteria was on PDA or NA slants under sterile mineral oil (Heckley, 1978). Flagellates were stored in liquid nitrogen. Identification of organisms. Standard keys were used to identify the pathogens (e.g., in Burgess, 1981; Poinar and Thomas,
SAND
FLY
1984). Morphological criteria (Vickerman, 1979; Wallace, 1979), monoclonal antibodies (McMahon-Pratt and David, 1981), and biochemical techniques (Saravia et al., 1985) were used for kinetoplastids. Experts were consulted for confirmation. Miscellaneous. Geographical locations of places in Colombia were given by Warburg et al. (1991). Various methodological details are given in relevant sections. RESULTS Infection
Experiments
Cytoplasmic polyhedrosis viruses. Polyhedra from spontaneously infected, colonybred P. papatasi (Jordan Valley) were used as the infectious inoculum for experiments with laboratory-bred P. papatasi from different geographical areas (Table 1). Artificial CPV infections induced in the Indian strain (Fig. I) were lower than, but comparable to, the spontaneous rates in the Jordan Valley strain, A significantly lower infection rate in Egyptian P. papatasi indicated the relative refractoriness of these flies (Table 1). Infection rates were not augmented in the Jordan Valley strain when temperatures were changed, larval diets were varied, and/or polyhedra were added to them (data not shown). S. marcescens. A strain of S. marcestens, isolated from laboratory-reared P. papatasi (Jordan Valley), was used for infection trials. A known concentration of NA-cultured S. marcescens was suspended in 30% sucrose from which flies fed. The TABLE CPV
Jordan Valley Egypt
India
mortality rate, the mean number of eggs per female, and the eclosion rate were monitored (Table 2). Accelerated mortality rates were evident beginning on day 6 of exposure, but at this stage the females had already oviposited, so there was no notable effect on their fecundity (Table 2). Eclosion rates were not affected. Red coloration, due to S. marcescens proliferation, spread to the entire body some 24 hr after death. B. bassiana spores in sugar solution. A calibrated suspension of spores was prepared in 30% sucrose solutions which were offered to flies. A slightly elevated mortality of flies on day 6 of exposure and thereafter was observed but the fecundity was not affected. Only 19-28% had mycelial growth after death (Table 2), indicating that infection rates were low. B. bassiana onfilterpapers. Spores were introduced into rearing containers on a piece of filter paper of known dimensions (see Materials and Methods). The longevity, average number of eggs per female, and eclosion rates were monitored (Table 3). Early mortality of gravid females prevented oviposition almost completely in both P. papatasi and Lutzomyia longipalpis (Table 3). A cottony white mycelium was observed covering the dead bodies (Figs. 2, 3) of most flies in these groups. Egg eclosion rates were similar in control groups and experimental ones. In larval-rearing containers into which mycellium-covered dead flies were introduced, first and second instar larvae were somewhat hindered physically by B. bassi1
INFECTIONS IN LABORATORY-REARED ADULT P. papatasi Total number of flies
Origin of strain
191
PATHOGENS
CPV-infected flies
Males
Females
Males
Females
% infection
130 42
80 28
28 2 4
22 0 16
23.8 2.9 15.6
78
50
Note. Infections in the Jordan Valley strain are spontaneous. The other two strains were infected experimentally by adding polyhedra suspensions to the larval food at an average ratio of one infected fly gut per five first instar larvae. In neither of the latter two strains were natural polyhedrosis infections detected.
192
ALON
WARBURG
FIG. 1. Cytoplasmic polyhedrosis viruses in a detached midgut cell of an experimentally infected, India-strain. P. paparasi female. Note virions (electron-dense dots) surrounding polyhedra. mv, microvilli; ph. polyhedra; vs. virogenic stroma. Transmission electron micrograph prepared as described by Warburg and Ostrovska (1987). Bar = 2 pm. FIG. 2. B. bassiana hyphae and fruiting bodies on an experimentally infected P. papnrasi female. The fly remained attached to a filter paper. Bar = 1 mm. FIG. 3. B. bassiana (white mycelium) and A. fravus (dark mycelium) on two P. papatasi reared in the same container. Bar = 1 mm.
ani a hyphae which grew on their food (rabbit feces and rabbit chow, mixed, fer:nted, dried, and autoclaved), but no inE tion of larvae was apparent. Many of the set :ond generation adults emerging from the :se containers exhibited B. bassiana
mycelial shown).
growth
after
death
(data
not
Pathogens of Wild-Caught Sand Flies Bacteria. A female Lutzomyia towns endi caught in Tulua (Colombia) was fount Ll to
193
SAND FLY PATHOGENS TABLE INFECTION
EXPERIMENTS
OF P. papatasi
No. of flies Pathogen Control
1
S. marcescens
% mortality on days’
B. bassiana
AND
M
F
Total
3
6
9
% with symptom@
Eggs per F
49
54
103
5.8
26.2
68.9
0
48.5
82.8
45 45
48 50
93 95
4.3 4.2
41.9 37.9
78.5 75.8
28 30.5
42.6 40.6
84.5 83.8
48 49
54 44
102 93
2.9 2.2
36.3 38.7
86.3 77.4
27.5 19.4
36.5 38.5
Group
No.
2 WITH
% eclosion
S. marcescens
1O’/ml lO’/ml
2
3
B. bassiana
4 5
10’ spores/ml 10’ spores/ml
82.5 78.3 Note. Sand flies (females 24 hr after a blood meal) were confined in oviposition containers (5 cm high, 6 cm diameter, T = 24 k 1°C. RH = 95 k 5%). B. bassiana spores or S. marcescens were suspended in sterile, 30% (w/v) sucrose solution. A fresh suspension was supplied daily for 8 days. Control groups were fed 30% sucrose. a Days after confinement and exposure to the pathogen. b B. bassiana mycellium or red coloration for S. marcescens evidence 5 days after death.
have atrophied ovaries characterized by a pink patch in each follicle. The ovaries were seeded on NA and slow growing, shiny white colonies were observed 3 days later. The bacterium was identified as a nonfluorescent pseudomonas. A definite connection between the red patches in the fly’s ovaries and the bacterium was not established. Flagellates. Two female Lutzomyia pia caught in Yotoco (Colombia) had about 20 flagellates each in their rectum. The dissected hindgut of one was seeded on blood agar slants overlaied with PBS + 5% FBS.
Growth was very slow at first but was much faster after a few passages. Morphologically, the cultured parasites could be divided into four types (Fig. 4). Type “A” represents the majority of individuals (about 80%): Promastigotes, 10-20 urn long, with or without basophiiic cytoplasmic granules and some with longitudinal body torsion. Type “B” was less common (10%): Slender promastigotes with particularly elongated posterior ends (20-40 urn). Type “C” (about 10%) were short widebodied promastigotes (5-7 urn) with pointed or rounded posterior ends. Type
TABLE INFECTION
OF P. papatasi
EXPERIMENTS
(GROUPS
No. of flies Group No.
M
F
l-3)
3 L. longipalpis
AND
% mortality on days”
Total
2
3
(GROUPS
4-6)
WITH
B. bassiana
4
5
% with symptomsb
Eas per F
% eclosion
12.4 100 49.5
26.7 100
0 97.9 99
48.5 1.6 0.9
82.8 80.4 86.4
16.4 100 100
43.6 -
17.1 0 0
92 -
P. papatasi
1 2 3
Control 8 cm’ 2 cm’
51 49 47
54 50 48
105 99 97
4.8 33.3 26.8
4 5 6
Control 8 cm’ 2 cm2
40 50 50
50 60 60
90 110 110
8.2 45.5 33.6
5.7 94.9 33 L. longipalpis
16.4 97.3 95.5
0 100 100
Note. Experimental procedure as for Table 2. Filter papers, smeared on both sides with B. bassiana were suspended inside the containers. Flies were maintained on sterile 30% sucrose. a Days after confinement and exposure to the pathogen. b B. bassiana mycellium or red coloration for S. marcescens evident 5 days after death.
spores,
194
ALON WARBURG
FIG. 4. Schematic drawing of stationary-phase NNN culture forms of a trypanosomatid isolated from L. piu. (A-D) Different promastigote morphotypes (for details see text). From Geimas-stained smear. Bar = 5 pm.
“D” were very short (3.5-5 pm) slender promastigotes represented by few individuals. The cultures were probed with 12 monoclonal antibodies which identify South American Leishmania spp. and differentiate between them. In addition, 13 of their enzymes were compared to marker Leishmania and Trypanosoma strains by thin starch-gel electrophoresis. There was no reaction with any of the antibodies and the isoenzyme profile was distinct from all the marker strains compared. Gregarines. Three female Lutzomyia lichyi collected in Villa Hermosa and Tulua (Colombia) were found infected with gregarines. Two to three gametocysts per infected female were attached to the hemocoel side of the genital accessory glands. Some of these were in the process of discharging oocysts into the glands. Oocystes were found in gametocysts, inside acces-
sory glands, and on eggs oviposited by an infected female. Oocysts were ellipsoidal 12.4 x 5.8 km (Fig. 5). The gregarine was described as a new species-Ascogregarina saraviae (Eugregarinorida: Lecudinidae) (Ostrovska et al., 1990). Fungi. A female L. pia, caught in Versalles (Colombia), exhibited mycelial growth in the thoracic muscles and the hemocoel. The fungus was seeded on PDA and SDA + 5% FBS; it grew only on the latter medium. Microscopical examination revealed thick hyphae (Fig. 6) characteristic of the Entomophthorales (gemera Entomophthora or Conidiobolus). Sporangiophores and ballistospores (Figs. 7, 8) were produced only after the cultures were irradiated with uv for 8 hr in a standard, sterile cabinet. The isolate was subsequently lost when it did not resume growth after 6 months under mineral oil.
SAND
FLY
Nematodes. Most of the abdominal cavity, part of the thorax, and some of the coxa of a Lutzomyia shannoni male captured in Versalles, Colombia, were congested with numerous juvenile Tylenchid nematodes (Figs. llA,B). A L. townsendi female from Tulua (Colombia) dissected for leishmanisis studies was found to contain a spirurid nematode (Figs. 12A,B) in its abdomen. It was not clear whether the nematode was inside the gut or in the hemocoel. Pathogens
of Laboratory-Bred
Sand Flies
Lutzomyia trapidoi and Lutzomyia gomezi larvae-the mixed first generation
progeny of wild-caught flies from Tumaco, Colombia-appeared sluggish and disoriented. Mortality of first and second instar larvae was high. In microscope preparations of larvae mounted in Hoyer’s medium, ectoparasitic fungi were seen clinging to the cuticle. They concentrated on the cutitular appendages of the head and the cauda1 setae (Figs. 9, 10). No penetrations of the cuticle were detected in whole mounts, histological sections, or electron micrographs of infected larvae. Attempts to culture the fungus failed and consequently it could not be positively identified. It does not belong to the Laboulbeniales-the major group of ectoparasitic entomogenous fungi (H. C. Evans, pers. commun.). Five saprophytic fungi were also isolated from surface-sterilized larvae of the same colony: Acremonium sp., Chaetomium globosum, Gliomastix mucorum, and Penicilliurn sp. Green-yellow fruiting bodies of Aspergillus jlavus and cottony white B. bassiana mycelium were observed growing on dead P. pupatasi (Jordan Valley) (Fig. 3). A. flavus did not cause any obvious morbidity but B. bassiana appeared to be associated with decreased colony vitality. The spread of infection could be controlled by the careful removal of dead flies and maintenance of general cleanliness. The B. bassiana isolate was subsequently used for infection experiments. S. marcescens was identified in
195
PATHOGENS
dead flies from the same colony. The bacterium was isolated and used for infection trials. Frequently, premature mortality in laboratory-reared flies was associated with accumulations of microbes on their tarsi. These sticky aggregates caused the flies to lose legs and eventually become stuck and die. The organisms multiplying on tarsi varied: In a P. papatasi colony mostly yeast were found (Fig. 13), whereas bacilli mixed with fungal hyphae and yeast were observed on the tarsi of L. longipalpis (Fig. 14). DISCUSSION CPV infections were previously observed to interfere with the development of concomitant Leishmania infections. Therefore, the virus was considered potentially useful for the biological control of leishmaniasis, despite its apparent lack of pathogenicity to sand flies (Warburg and Ostrovska, 1987). For the infection experiments, relatively small doses of polyhedra were available. This constraint may have contributed to the low infection rates (Table 1). The results clearly show that the Indian strain of P. papatasi is more susceptible than the Egyptian strain, but the overall infection rates remained low and the quantities of viral material were not sufficient to enable further studies. Interestingly, when CPVs were first described in a laboratory colony maintained at the Hebrew University, Israel, 95.9% (n = 244) of the flies were spontaneously infected (Warburg and Ostrovska, 1987). At the same time high rates (percentage unknown) were also recorded in colonized flies originating from the same stock but separated some 4 years earlier and maintained, using a different methodology, at Yale University. Rates subsequently subsided, more or less simultaneously in both colonies, without any apparent cause. S. marcescens commonly affects sand fly colonies of different species. Sugar solutions from which flies feed become pinkish
ALON
WARBURG
SAND
FLY
and dead flies turn red due to the proliferation of bacteria in their tissues (unpubl. observations). The phenomenon is not normally associated with a notable reduction in colony vigor or fecundity. In experiments, S. marcescens did not significantly reduce longevity or fecundity of flies, even when ingested in large quantities (Table 2). It is possible that sugar meals are not suitable vehicles for S. marcescens. However, a similar method was shown to be highly effective for the delivery of Bacillus thuringiensis israelensis endotoxin to both male and female sand flies (Yuval and Warburg, 1989). Like B. thuringiensis israelensis, S. marcescens initiates infections after being ingested. Hence, the most probable explanation for these observations is that the S. marcescens isolate< was of low toxicity to sand flies. Promastigote kinetoplastids, the most frequently reported parasites in sand flies, were routinely assigned to the genus Leishmania until biochemical and immunological techniques for parasite identification were introduced. Thus, at least some infections by other trypanosomatids were probably overlooked. The generic status of the parasite isolated from L. pia is uncertain. Biochemical, immunological, and morphological data show that it is neither a Leishmania sp. nor a Trypanosoma sp. The large size of some morphotypes and their longitudinal body torsion (Fig. 4) are characteristic of monoxenous-insect flagellates (genus Leptomonas) and of insect-transmitted plant parasites of the genus Phytomonas (Vickerman, 1979; Wallace, 1979). Since phle-
PATHOGENS
botomines
197
are not plant feeders, a Leptomthe most probable candidate. However, sand flies do ingest plant-derived sugars (Schlein and Warburg, 1985), making the presence of plant pathogens in their alimentary tract feasible. Furthermore, the absence of cysts in cultures of the L. pia-derived trypanosomatid could indicate its heteroxenous nature. Flagellar cysts are formed by Leptomonas and other monoxenous insect trypanosomatids but not by Phytomonas spp. (McGhee and Hanson, 1964; Vickerman, 1979; Wallace, 1979). Gregarines have been reported in over 20 species of sand flies (Young and Lewis, 1977, 1980). The genus Ascogregarina (=Ascocystis) comprises 15 described species including 3 from sand flies (previously Monocystis spp.) and 5 from mosquitoes (previously Lankesteria spp.; Levine, 1988). The best studied sand fly parasite is Ascogregarina chagasi in L. longipalpis, originally described by Adler and Mayrink (1961) and subsequently studied by others who performed cross infection experiments (Wu and Tesh, 1989) and characterized the role of the fly’s immune response in parasite vertical transmission (Warburg and Ostrovska, 1989). Frequently, high rates of infection are encountered in laboratory colonies but the effect on longevity and fecundity is apparently limited (Wu and Tesh, 1989). Ascogregarina spp. of sand flies are unique in that gametocysts adhere to the genital accessory glands and oocysts are injected into them, thereby facilitating vertical transmission (Adler and Mayrink, 1961; onas sp. must be considered
FIG. 5. Sporulated oocysts of A. saruviue from the accessory gland of I,. lichyi. Phase-contrast micrograph. Bar = 10 pm. FIGS. CS. SDA culture of Entomophthorales fungus isolated from L. pia. Phase-contrast micrographs. FIG. 6. Sterile hyphae. Bar = 20 pm. FIG. 7. Sporangiophore. Bar = 2 pm. FIG. 8. Ballistospore. Bar = 2 pm. FIGS. 9 AND 10. Ectoparasitic fungus on first instar Lutzomyiu spp. Phase-contrast micrographs of specimens mounted in Hoyer’s medium. FIG. 9. Whole larva with fungi on head and caudal setae (arrows). Bar = 0.5 mm. FIG. 10. Higher magnification of fungus on a caudal seta. Bar = 30 pm.
ALON
WARBURG
FIG. 11. Tylenchid nematodes in a wild-caught L. shannoni male. (A) External genitalia showing a worm squeezed out through the anus. (B) Abdomen containing numerous worms. Phase-contrast micrograph of specimen mounted in Hoyer’s medium. Bar = 0.1 mm. FIG. 12. Spimrid nematode dissected out of a L. rownsendi female. (A) Anterior portion, (B) posterior portion. Phase-contrast micrograph of a fresh preparation. Bar = 0.25 mm.
SAND
FLY
PATHOGENS
FIGS. 13 AND 14. Microbial aggregates on tarsi of laboratory-reared sand flies. Scanning electrc In micrographs (fixed in 3% glutaraldehyde for 2 hr, 1% osmium-tetroxide for 1 hr. and dehydrated Iin ethanol and freon). FIG. 13. Yeast and saprophytic fungi on P. papatasi. Bar = 20 km. FIG. 14. Bacteria and saprophytic fungi on L. longipalpis. Bar = 10 pm.
trovska et al., 1990; Warburg and Ostroca, 1989). infections contribute 1Zntomophthorales to televated mortality rates in mosquito popula .tions overwintering in caves and base1974, and references me :nts (Roberts, OS
vsj
therein). Cumulated reports indicate : that these fungi may constitute important I:bathogens affecting sand fly populations as well. McConnel and Correa (1964) encoun ttered high rates (14%) of fungal infections in eight cave-dwelling Panamanian sand fly splecies ,
200
ALON
WARBURG
The fungus was not identified, but a mycelium featured in their published micrograph is almost certainly of an Entomophthorales sp. I have observed numerous Entomophthorales infections in mounted specimens (courtesy of J. Murillo) of L. longipalpis caught in pit latrines in the town of Nazareth, Guanacaste, Costa Rica (unpubl. data). Maltese P. papatasi were reported infected with Entomophthora (Marett, 1915). Rioux et al. (1966) found resting spores in the hemocoel of a female Phlebotomos ariasi caught in the south of France. Many other reports of unidentified fungal infections in sand flies (Young and Lewis, 1977, 1980) may well have been due to Entomophthorales spp. Unfortunately, Entomophthorales fungi are particularly fastidious and difficult to culture, a trait which reduces their potential usefulness for biological control. Further studies are needed to identify possible means for mass propagation of Entomophthora spp. and to isolate strains more amenable to artificial culture. The L. pia isolate described here grew well on FBS-fortified SDA, indicating that it may have been a Conidiobolus sp. rather than an Entomphthora sp. (H. Evans, pers. commun.). B. bassiana spores suspended in sucrose solutions were not very infectious to P. papatasi; neither longevity nor fecundity were significantly reduced and only a low percentage of dead flies exhibited mycelial growth (Table 2). This observation is consistent with the fact that most entomopathogenic fungi infect the hemocoel after gaining access through the cuticle and not via the gut (Evans, 1989). In addition, a high sugar concentration (30% w/v) used for feeding sand flies may prevent spore germination, thereby contributing to the low infectivity of per OS-ingested B. bassiana.
In P. papatasi exposed to B. bassiana spores smeared on dry filter papers, the 100% mortality after 4-6 days of exposure contrasted sharply with a mere IO-25% observed in control groups. Preoviposition
death of most females was reflected by an extremely low egg count (Table 3). The same fungus was even more pathogenic to L. longipalpis. Mycelial growth was seen on virtually all of the dead flies and conidia were produced 4-6 days after death (Figs. 2,3). Many of the dead flies remained affixed to the mesh cover, plastic walls, and filter paper (Fig. 2). This phenomenon is commonly observed when insects that die of fungal infections assume elevated and exposed positions. They remain secured to the substrate by their stiff limbs and by hyphal strands (Evans, 1989). The saprophytic fungal isolates which were apparently associated with Lutzomyia spp. mortality had probably originated in the gut contents of the morbid larvae. Such fungi are normally present in larval-rearing media and may cause opportunistic secondary infections in flies affected by other pathogens or by adverse conditions. The bacterial and yeast aggregates found on the tarsi of colony-bred flies (Figs. 13,14) are probably propagated by contact with sugar solutions on which flies alight in order to feed. The maintenance of clean cages, frequent changing of sugar solutions, and correct humidity usually eliminate such problems. Tylenchid nematodes, described here in a L. shannoni male (Figs. llA,B), usually develop to third or fourth stage juveniles and leave the host via the anus or reproductive pore. Adult tylenchids live in the environment and mated females reinvade an insect host by direct penetration of the cuticle (Poinar, 1983). Heavy tylenchid infections have previously been recorded in Lutzomyia sanguinaria, nis, and Lutzomyia
Lutzomyia panamensis
vespertilio-
(McConell and Correa, 1964). Their life cycle renders tylenchids unsuitable for biological control. A possibly better candidate, a sand fly parasitic tetradonematid nematode, was studied by Killick-Kendrick et al. (1989). This pathogen delayed larval development, interfered with blood feeding by females, and induced male sterility.
201
SAND FLY PATHOGENS
The worm found
in L. townsendi
be-
longed to the Spirurida, an order of heteroxenous parasites of vertebrates infecting insects as intermediate hosts (Poinar, 1983). Encapsulated third-stage spirurid nematodes (rodent-infecting Masrophorus muris) have been recorded in P. ariasi (Killick-Kendrick et al., 1976). The effect of spirurid infections on sand flies is difficult to assess, but their heteroxenous life cycle precludes them from biological control. Demonstrating efficacy in laboratory trials is the essential prerequisite for identifying potential biocontrol agents. Clearly, the more challenging part is application in field situations. Control measures against phlebotomines should focus on large concentrations resting and breeding in accessible habitats. Some important vector species occupy such habitats. L. longipalpis, the vector of visceral leishmaniasis in South America, often breeds in caves and P. papatasi, the vector of cutaneous leishmaniasis in the Mediterranean basin and Russia, breeds in rodent burrows. As previously mentioned, these dark, humid, and cool (20-25%C) resting sites are also suitable for the long term persistence and germination of fungal spores. Within rooms or caves, the flies’ distribution is further limited to the darkest niches such as cracks, corners between ceiling and walls, and surfaces that do not receive direct light. Therefore, small pieces of paper smeared with fungal spores (e.g., a mixture of B. bassiuna and an effective pathogen of larvae) and hung in appropriate places may serve to deliver concentrated doses of spores that will remain viable for long periods. In less accessible places, spraying of spores may be more feasible. Adult sand flies, dying of the fungus, should facilitate the vertical transmission to larvae feeding below and horizontal transmission to other flies sharing the same resting sites. ACKNOWLEDGMENTS Part of this work-conducted at Centro Intemacional de Investigaciones Medicas, Cali, Colombiareceived support from Grant ID 840336 of the UNDPI
WORLD BANK/WHO special program for research and training in tropical diseases. Additional support was given by a Wellcome Trust grant (LSHTM, London). I am grateful to H. C. Evans of CIBC, UK, for identifying fungal and bacterial isolates; to scientists at CIDEIM for technical assistance; and to A. Chabaud, I. Landau, and D. Van Waerebeke of the Natural History Museum, Paris, for identifying nematodes.
REFERENCES ADLER, S. 1964. Leishmania. Adv. Parasitol., 2, 3% 96. ADLER, S., AND MAYRINK, W. 1961. A gregarine, Monocystis chagasi n. sp., of Phlebotomus iongipalpis: Remarks on the accessory gland of P. longipalpis.
Rev.
Med.
Trop.
Sao Paula,
3, 230-238.
ASHFORD, R. W. 1974. Sandflies (Diptera: Psychodidae) from Ethiopia: Taxonomic and biological notes. J. Med. Entomol., 11, 605-616. BURGESS, H. D. (Ed.) 1981. “Microbial Control of Pests and Plant Diseases 1970-1980.” Academic Press, New York. EVANS, H. C. 1989. Mycopathogens of insects of epigeal and aerial habitats. In “Insect-Fungus Interactions” (N. M. Collins, M. P. Hammond, and J. F. Webbes, Eds.), pp. 205-238. Academic Press, New York. HECKLEY, R. J. 1978. Preservation of microorganisms. AdI’. Appl. Microbial.. 24, l-53. KILLICK-KENDRICK, R. 1979. The biology of Leishmania in phlebotomine sandflies. In “Biology of the Kinetoplastida” (W. H. R. Lumsden and D. A. Evans, Eds.). Vol. 2, pp. 395-460. Academic Press, New York. KILLICK-KENDRICK, R. 1987. Breeding places of Phlebotomus ariasi in the Cevennes focus of leishmaniasis in the south of France. Parassitologia. 29, 181191. KILLICK-KENDRICK. R., KILLICK-KENDRICK, M.. QLJALA. N. A., NAWI. R. W.. ASHFORD, R. W., AND TANG. Y. 1989. Preliminary observations on a tetradonematid nematode of phlebotomine sandflies of Afghanistan. Ann. Parasitol. Hum. Camp., 64, 332-339. KILLICK-KENDRICK. R.. LEANEY. A. J., MOLYNEUX, D. H., AND RIOUX, J.-A. 1976. Parasites of Phlebotomus
ariasi.
Trans.
R. Sot.
Trop.
Med.
Hyg.
70,
22. Phylum LEVINE, N. D. 1988. “The Protozoan Apicomplexa.” Vol. 1. CRC Press, Boca Raton, Florida. MARETT. P. J. 1915. The bionomics of the Maltese phlebotomi. &it. Med. J.. 2, 172-173. MCCONNEL. E., AND CORREA, M. 1964. Trypanosomes and other microorganisms from Panamanian Phlebotomus sand flies. J. Parasitol., 50, 523-528. MCGHEE, R. B., AND HANSON, W. L. 1964. Comparison of the life cycle of Leptomonas oncopelti and Phytomonas elmassiani. J. Protozool.. 11, 555-562.
202
ALON
WARBURG
MCMAHON-PRATT, D., AND DAVID, J. R. 1981. Monoclonal antibodies that distinguish between New World species of Leishmania. Nature, 291, 581-583. MODI, G. B., AND TESH, R. B. 1983. A simple technique for mass rearing Lutzomyia longipalpis and Phlebotomus papatasi (Diptera: Psychodidae) in the laboratory. J. Med. Entomol., 20, 568-569. OSTROVSKA, K., WARBURG, A., AND MONTOYALERMA, J. 1990. Ascogregarina saraviae, N. sp. in Lutzomyia lichyi. J. Protozool., 37, 69-70. POINAR, G. 0. 1983. Nematode parasites of invertebrates. In “The Natural History of Nematodes” (G. 0. Pionar, Ed.), pp. 161-202. Prentice-Hall, Englewood Cliffs, New Jersey. POINAR, G. L., AND THOMAS, G. M. 1984. “Laboratory Guide to Insect Pathogens and Parasites.” Plenum, New York. ROBERTS, D. W. 1974. Coelomomyces, Entomophthora, Beauvaria and Metarhizium as parasites of mosquitoes. Misc. Pub. Entomol. Sot. Am., 7, MO155. RIOUX, J. A., MANIER, J. F., AND TOUR, S. 1966. Infestation a Entomophthora sp. chez Phlebotomus ariasi Tonnoir, 1921. Ann. Parasitol., 41, 251-253. SARAVIA, N. G., HOLGUIN, A. F., MCMAHONPRATT, D., AND D’ALESSANDRO, A. 1985. Mucocutaneous leishmaniasis in Colombia: Leishmania braziliesis subspecies diversity. Am. J. Trop. Med. Hyg., 34, 714-720. SCHLEIN, Y., AND WARBURG, A. 1985. Phytophagy and the cycle of feeding of Phlebotomus papatasi (Diptera: Psychodidae) under experimental conditions. J. Med. Entomol., 23, 11-15. TESH, R. B. 1988. The genus Phlebovirus and its vectors. Annu. Rev. Entomol., 33, 169-181. VICKERMAN, K. 1979. The diversity of the kinetoplas-
tid flagellates. In “Biology of the Kinetoplastida” (W. H. R. Lumsden and D. A. Evans, Eds.), Vol. 1, pp. l-34. Academic Press, New York. WALLACE, F. G. 1979. Biology of kinetoplastida of arthropods. In “Biology of the Kinetoplastida” (W. H. R. Lumsden and D. A. Evans, Eds.), Vol. 2, pp. 213-240. Academic Press, New York. WARBURG, A., MONTOYA, J., JARAMILLO, C., CRUZ, A. L., AND OSTROVSKA, K. 1991. Leishmaniasis vector potential of Lutzomyia spp. in Colombian coffee plantations. Med. Vet. Entomol., 5, 9-16. WARBURG, A., AND OSTROVSKA, K. 1987. Cytoplasmic polyhedrosis viruses in Phlebotomus papatasi inhibit development of Leishmania major. J. Parasitol., 73, 578-583. WARBURG, A., AND OSTROVSKA, K. 1989. An immune response-dependent mechanism for the vertical transmission of an entomopathogen. Experientia, 45, 710-172. Wu, W.-K., AND TESH, R. B. 1989. The infection of Old and New World phlebotomine sand flies (Diptera: Psychodidea) with Ascogregarina chagasi. J. Med. Entomol., 26, 237-242. YOUNG, D. G., AND LEWIS, D. J. 1977. Pathogens of psychodidae (phlebotomine sand flies). In “Bibliography on Pathogens of Medically Important Arthropods” (D. W. Roberts and J. M. Castillo, Eds.), Bull. WHO, SS(Suppl.), S9-S24. YOUNG, D. G., AND LEWIS, D. J. 1980. Pathogens of psychodidae (phlebotomine sand flies). In “Bibliography on Pathogens on Medically Important Arthropods” (D. W. Roberts and J. M. Castillo, Eds.), Bull. WHO, Xl(Supp1.) S9-Sl 1. YUVAL, B., AND WARBURG, A. 1989. Susceptibility of adult phlebotomine sandflies to Bacillus thuringiensis var. israeliensis. Ann. Trop. Med. Parasitol., 83, 165-196.
OF INVERTEBRATE
PATHOLOGY
58, 189-202 (1991)
Entomopathogens of Phlebotomine Sand Flies: Laboratory Experiments and Natural Infections ALON Medical
Entomology
Unit,
Malaria
WARBURG
Section, Department of Health, Bethesda,
of Health Maryland
and Human 20892
Services,
National
Institute
Received August 8, 1990; accepted October 26, 1990 The susceptibility of different geographical strains of Phlebotomus papatasi to a cytoplasmic polyhedrosis virus (CPV) was determined experimentally by feeding polyhedra to larvae. Of the Indian P. papatasi, 15.6% became infected, whereas Egyptian P. papatasi were mostly refractory. Infection rates were not augmented in colony flies from the Jordan Valley, 23.8% of which were naturally infected with CPV. The infectivity of Serratia marcescens and Beauvaria bassiana to P. papatasi were determined experimentally. A suspension of B. bassiana spores or S. marcescens bacteria, ingested by P. papatasi in sucrose solution, did not significantly augment mortality rates or reduce the number of eggs oviposited. However, B. bassiana spores smeared on a titter paper constituting 1 or 5% of the surface area available to flies induced 100% mortality of P. papatasi on days 5 and 4. respectively. Mortality in Lutzomyia longipalpis reached 100% on day 4. There were markedly lower mortality rates in the control groups and more eggs were produced by these females (P. papatasi: control = 48.5; experimental = 0.9-1.6 eggs/female: L. longipalpis: control = 17.1; experimental = 0 eggs/female). From wild-caught Colombian Lutzomyia spp., a nonfluorescent pseudomonas, an Entomophthorales fungus, and a Trypanosomatid protozoon (probably Leptomonas) were isolated in culture media. Gregarines (Ascogregarina saraviae) and nematodes (Tylenchida and Spirurida) were also recorded. In laboratory-reared flies, an ectoparasitic fungus was associated with high mortality rates of first instar Lutzomyia spp. larvae. Opportunistic ectoparasitic aggregates of bacteria, yeast, and fungi on the tarsi of colonized L. longipalpis and P. papatasi hindered their mobility and were associated with reduced colony vigor. Aspergillus jZavus, B. bassiana, and S. marcescens were isolated from laboratory-bred P. papatasi adults. o 191 Academic
Press. Inc.
KEY WORDS: Ascogregarina; Beauvaria bassiana; cytoplasmic polyhedrosis viruses; Diptera; entomopathogens; Entomophthorales; leishmaniasis; Leptomonas; nematodes; Lutzomyia gomezi; Lutzomyia towxsendi;
longipalpis: Lutzomyia
Lutzomyia lichyi; trapidoi; Phlebotomus
rurida; Trypanosomatidae;
Lutzomyia papatasi;
pia;
Lutzomyia
shannoni; Lutzomyia marcescens; Spi-
Psychodidae; Serratia
Tylenchida: yeast.
INTRODUCTION Phlebotomine sand flies, (Diptera: Psychodidae) of the genera Lutzomyiu (Americas) and Phlebotomus (Old World), transmit human leishmaniasis, bartonelosis, and sand-fly fever. The causative agents of these diseases are protozoan parasites, bacteria-like organisms, and phleboviruses, respectively (Adler, 1964; Killick-Kendrick, 1979; Tesh, 1988). Sand flies also transmit various other pathogens of vertebrates including Leishmania, Plasmodium, Hemogregarina, and Trypanosoma of reptiles and amphibians (Killick-Kendrick, 1979; McConnel and Correa, 1964).
Collecting sand fly larvae is an arduous task; only small numbers are found and their specie cannot be determined with confidence (Killick-Kendrick, 1987 and references therein). Consequently, the vast majority of entomogenous organisms (Young and Lewis, 1977, 1980) were found in adult sand flies dissected for leishmaniasis studies. The bias toward adults may explain why Phlebotomines appear to be parasitized to a lesser extent than dipterans with aquatic larvae (Adler, 1964; Ashford, 1974). However, limited pathogen dissemination in soil and the resultant low transmission rates between larvae may also contribute to this phenomenon. 189 0022-2011/91 $1.50 Copyright 0 1991 by Academic Press, Inc. AU rights of reproduction in any form reserved.
190
ALON
WARBURG
At least 80% (&5 weeks per generation, estimate based on laboratory breeding) of the life cycle of sand flies is spent in cool, dark, and humid locations. Eggs are oviposited in forest leaf litter, tree buttresses, caves, rodent burrows, cracks in walls, etc. (Killick-Kendrick, 1979, 1987). Larvae develop in these habitats and feed on moist, decaying organic matter and pupate in situ. Even adults select humid and dark diurnal resting sites. The prevailing conditions in such places are also conducive to the development and long term persistence of many entomopathogens. To ascertain the feasibility of biological control, phlebotomine larval biology requires further study, efficacious pathogens must be identified, and adequate delivery systems developed. This report attempts a preliminary assessment of the potential usefulness of some naturally occurring sand fly pathogens. The first part describes the results of infection studies with viruses, fungi, and bacteria isolated from colony-bred flies. Particular attention was paid to possible methods for field application. In the second part, observations were made on naturally occurring entomopathogens. The results and observations are discussed. emphasizing their possible implication for biological control of sand flies. MATERIALS
AND METHODS
Cytoplasmic polyhedrosis viruses (CPVs). The macerated guts of Phfebotomus papatasi naturally infected with CPV
were suspended in water and soaked in larval food. Groups of first instar larvae were kept on this food until it was completely consumed. Larvae were maintained routinely (Modi and Tesh, 1983) thereafter. Infections in adult flies’ midguts were monitored microscopically under phase-contrast illumination. Ingested Serratia marcescens and Beauvaria bassiana. Groups of 18-25 P. papatasi (females 24 hr after blood feeding) were confined in oviposition containers with plaster of paris (height 5 cm, diameter 6 cm)
and maintained at 24” + 1°C in 95 ? 5% relative humidity. A fresh suspension of infective units (B. bassiana spores or S. marcescens) in 30% (w/v) aqueous sucrose solution was offered to the flies daily for 8 days. Control groups were maintained on autoclaved 30% sucrose. B. bassiana spores on jilter papers. A 4or l-cm2 filter paper was smeared on both sides (respectively, about 5 and 1.3% of the surface area available to flies) with B. bassiana spores and suspended inside oviposition containers (as above). The flies were maintained on autoclaved sucrose solution. Sand flies. Laboratory colonies were maintained according to Modi and Tesh (1983). For pathogen isolation, flies were collected as previously described (Warburg et al., 1991). Prior to dissection, carbon dioxide-anesthetized flies were briefly immersed in 70% ethanol, washed in sterile water, immersed in 5% hypochlorite for 5 min, and washed in sterile water. They were dissected in sterile phosphate-buffered saline (PBS) and examined microscopically under phase-contrast illumination. Zsolation and culture. Infected insects were placed on appropriate media: nutrient agar (NA) and brain heart infusion for bacteria and Saboraud dextrose agar (SDA) and potato dextrose agar (PDA) for fungi. Media were fortified with 5% fetal bovine serum (FBS) or 5% yeast extract as necessary. Flagellates were seeded onto blood agar slants overlaied with PBS + 5% (v/v) FBS, 1000 IU ml-’ of penicillin, 1000 ug ml-’ of streptomycin, and 2000 kg ml-i of 5-fluorocytosine (5-FC). In subsequent passages 5-FC was not used and the concentrations of penicillin and streptomycin were reduced to 100 IU ml-’ and 100 pg ml-‘, respectively. Long term storage of fungi and bacteria was on PDA or NA slants under sterile mineral oil (Heckley, 1978). Flagellates were stored in liquid nitrogen. Identification of organisms. Standard keys were used to identify the pathogens (e.g., in Burgess, 1981; Poinar and Thomas,
SAND
FLY
1984). Morphological criteria (Vickerman, 1979; Wallace, 1979), monoclonal antibodies (McMahon-Pratt and David, 1981), and biochemical techniques (Saravia et al., 1985) were used for kinetoplastids. Experts were consulted for confirmation. Miscellaneous. Geographical locations of places in Colombia were given by Warburg et al. (1991). Various methodological details are given in relevant sections. RESULTS Infection
Experiments
Cytoplasmic polyhedrosis viruses. Polyhedra from spontaneously infected, colonybred P. papatasi (Jordan Valley) were used as the infectious inoculum for experiments with laboratory-bred P. papatasi from different geographical areas (Table 1). Artificial CPV infections induced in the Indian strain (Fig. I) were lower than, but comparable to, the spontaneous rates in the Jordan Valley strain, A significantly lower infection rate in Egyptian P. papatasi indicated the relative refractoriness of these flies (Table 1). Infection rates were not augmented in the Jordan Valley strain when temperatures were changed, larval diets were varied, and/or polyhedra were added to them (data not shown). S. marcescens. A strain of S. marcestens, isolated from laboratory-reared P. papatasi (Jordan Valley), was used for infection trials. A known concentration of NA-cultured S. marcescens was suspended in 30% sucrose from which flies fed. The TABLE CPV
Jordan Valley Egypt
India
mortality rate, the mean number of eggs per female, and the eclosion rate were monitored (Table 2). Accelerated mortality rates were evident beginning on day 6 of exposure, but at this stage the females had already oviposited, so there was no notable effect on their fecundity (Table 2). Eclosion rates were not affected. Red coloration, due to S. marcescens proliferation, spread to the entire body some 24 hr after death. B. bassiana spores in sugar solution. A calibrated suspension of spores was prepared in 30% sucrose solutions which were offered to flies. A slightly elevated mortality of flies on day 6 of exposure and thereafter was observed but the fecundity was not affected. Only 19-28% had mycelial growth after death (Table 2), indicating that infection rates were low. B. bassiana onfilterpapers. Spores were introduced into rearing containers on a piece of filter paper of known dimensions (see Materials and Methods). The longevity, average number of eggs per female, and eclosion rates were monitored (Table 3). Early mortality of gravid females prevented oviposition almost completely in both P. papatasi and Lutzomyia longipalpis (Table 3). A cottony white mycelium was observed covering the dead bodies (Figs. 2, 3) of most flies in these groups. Egg eclosion rates were similar in control groups and experimental ones. In larval-rearing containers into which mycellium-covered dead flies were introduced, first and second instar larvae were somewhat hindered physically by B. bassi1
INFECTIONS IN LABORATORY-REARED ADULT P. papatasi Total number of flies
Origin of strain
191
PATHOGENS
CPV-infected flies
Males
Females
Males
Females
% infection
130 42
80 28
28 2 4
22 0 16
23.8 2.9 15.6
78
50
Note. Infections in the Jordan Valley strain are spontaneous. The other two strains were infected experimentally by adding polyhedra suspensions to the larval food at an average ratio of one infected fly gut per five first instar larvae. In neither of the latter two strains were natural polyhedrosis infections detected.
192
ALON
WARBURG
FIG. 1. Cytoplasmic polyhedrosis viruses in a detached midgut cell of an experimentally infected, India-strain. P. paparasi female. Note virions (electron-dense dots) surrounding polyhedra. mv, microvilli; ph. polyhedra; vs. virogenic stroma. Transmission electron micrograph prepared as described by Warburg and Ostrovska (1987). Bar = 2 pm. FIG. 2. B. bassiana hyphae and fruiting bodies on an experimentally infected P. papnrasi female. The fly remained attached to a filter paper. Bar = 1 mm. FIG. 3. B. bassiana (white mycelium) and A. fravus (dark mycelium) on two P. papatasi reared in the same container. Bar = 1 mm.
ani a hyphae which grew on their food (rabbit feces and rabbit chow, mixed, fer:nted, dried, and autoclaved), but no inE tion of larvae was apparent. Many of the set :ond generation adults emerging from the :se containers exhibited B. bassiana
mycelial shown).
growth
after
death
(data
not
Pathogens of Wild-Caught Sand Flies Bacteria. A female Lutzomyia towns endi caught in Tulua (Colombia) was fount Ll to
193
SAND FLY PATHOGENS TABLE INFECTION
EXPERIMENTS
OF P. papatasi
No. of flies Pathogen Control
1
S. marcescens
% mortality on days’
B. bassiana
AND
M
F
Total
3
6
9
% with symptom@
Eggs per F
49
54
103
5.8
26.2
68.9
0
48.5
82.8
45 45
48 50
93 95
4.3 4.2
41.9 37.9
78.5 75.8
28 30.5
42.6 40.6
84.5 83.8
48 49
54 44
102 93
2.9 2.2
36.3 38.7
86.3 77.4
27.5 19.4
36.5 38.5
Group
No.
2 WITH
% eclosion
S. marcescens
1O’/ml lO’/ml
2
3
B. bassiana
4 5
10’ spores/ml 10’ spores/ml
82.5 78.3 Note. Sand flies (females 24 hr after a blood meal) were confined in oviposition containers (5 cm high, 6 cm diameter, T = 24 k 1°C. RH = 95 k 5%). B. bassiana spores or S. marcescens were suspended in sterile, 30% (w/v) sucrose solution. A fresh suspension was supplied daily for 8 days. Control groups were fed 30% sucrose. a Days after confinement and exposure to the pathogen. b B. bassiana mycellium or red coloration for S. marcescens evidence 5 days after death.
have atrophied ovaries characterized by a pink patch in each follicle. The ovaries were seeded on NA and slow growing, shiny white colonies were observed 3 days later. The bacterium was identified as a nonfluorescent pseudomonas. A definite connection between the red patches in the fly’s ovaries and the bacterium was not established. Flagellates. Two female Lutzomyia pia caught in Yotoco (Colombia) had about 20 flagellates each in their rectum. The dissected hindgut of one was seeded on blood agar slants overlaied with PBS + 5% FBS.
Growth was very slow at first but was much faster after a few passages. Morphologically, the cultured parasites could be divided into four types (Fig. 4). Type “A” represents the majority of individuals (about 80%): Promastigotes, 10-20 urn long, with or without basophiiic cytoplasmic granules and some with longitudinal body torsion. Type “B” was less common (10%): Slender promastigotes with particularly elongated posterior ends (20-40 urn). Type “C” (about 10%) were short widebodied promastigotes (5-7 urn) with pointed or rounded posterior ends. Type
TABLE INFECTION
OF P. papatasi
EXPERIMENTS
(GROUPS
No. of flies Group No.
M
F
l-3)
3 L. longipalpis
AND
% mortality on days”
Total
2
3
(GROUPS
4-6)
WITH
B. bassiana
4
5
% with symptomsb
Eas per F
% eclosion
12.4 100 49.5
26.7 100
0 97.9 99
48.5 1.6 0.9
82.8 80.4 86.4
16.4 100 100
43.6 -
17.1 0 0
92 -
P. papatasi
1 2 3
Control 8 cm’ 2 cm’
51 49 47
54 50 48
105 99 97
4.8 33.3 26.8
4 5 6
Control 8 cm’ 2 cm2
40 50 50
50 60 60
90 110 110
8.2 45.5 33.6
5.7 94.9 33 L. longipalpis
16.4 97.3 95.5
0 100 100
Note. Experimental procedure as for Table 2. Filter papers, smeared on both sides with B. bassiana were suspended inside the containers. Flies were maintained on sterile 30% sucrose. a Days after confinement and exposure to the pathogen. b B. bassiana mycellium or red coloration for S. marcescens evident 5 days after death.
spores,
194
ALON WARBURG
FIG. 4. Schematic drawing of stationary-phase NNN culture forms of a trypanosomatid isolated from L. piu. (A-D) Different promastigote morphotypes (for details see text). From Geimas-stained smear. Bar = 5 pm.
“D” were very short (3.5-5 pm) slender promastigotes represented by few individuals. The cultures were probed with 12 monoclonal antibodies which identify South American Leishmania spp. and differentiate between them. In addition, 13 of their enzymes were compared to marker Leishmania and Trypanosoma strains by thin starch-gel electrophoresis. There was no reaction with any of the antibodies and the isoenzyme profile was distinct from all the marker strains compared. Gregarines. Three female Lutzomyia lichyi collected in Villa Hermosa and Tulua (Colombia) were found infected with gregarines. Two to three gametocysts per infected female were attached to the hemocoel side of the genital accessory glands. Some of these were in the process of discharging oocysts into the glands. Oocystes were found in gametocysts, inside acces-
sory glands, and on eggs oviposited by an infected female. Oocysts were ellipsoidal 12.4 x 5.8 km (Fig. 5). The gregarine was described as a new species-Ascogregarina saraviae (Eugregarinorida: Lecudinidae) (Ostrovska et al., 1990). Fungi. A female L. pia, caught in Versalles (Colombia), exhibited mycelial growth in the thoracic muscles and the hemocoel. The fungus was seeded on PDA and SDA + 5% FBS; it grew only on the latter medium. Microscopical examination revealed thick hyphae (Fig. 6) characteristic of the Entomophthorales (gemera Entomophthora or Conidiobolus). Sporangiophores and ballistospores (Figs. 7, 8) were produced only after the cultures were irradiated with uv for 8 hr in a standard, sterile cabinet. The isolate was subsequently lost when it did not resume growth after 6 months under mineral oil.
SAND
FLY
Nematodes. Most of the abdominal cavity, part of the thorax, and some of the coxa of a Lutzomyia shannoni male captured in Versalles, Colombia, were congested with numerous juvenile Tylenchid nematodes (Figs. llA,B). A L. townsendi female from Tulua (Colombia) dissected for leishmanisis studies was found to contain a spirurid nematode (Figs. 12A,B) in its abdomen. It was not clear whether the nematode was inside the gut or in the hemocoel. Pathogens
of Laboratory-Bred
Sand Flies
Lutzomyia trapidoi and Lutzomyia gomezi larvae-the mixed first generation
progeny of wild-caught flies from Tumaco, Colombia-appeared sluggish and disoriented. Mortality of first and second instar larvae was high. In microscope preparations of larvae mounted in Hoyer’s medium, ectoparasitic fungi were seen clinging to the cuticle. They concentrated on the cutitular appendages of the head and the cauda1 setae (Figs. 9, 10). No penetrations of the cuticle were detected in whole mounts, histological sections, or electron micrographs of infected larvae. Attempts to culture the fungus failed and consequently it could not be positively identified. It does not belong to the Laboulbeniales-the major group of ectoparasitic entomogenous fungi (H. C. Evans, pers. commun.). Five saprophytic fungi were also isolated from surface-sterilized larvae of the same colony: Acremonium sp., Chaetomium globosum, Gliomastix mucorum, and Penicilliurn sp. Green-yellow fruiting bodies of Aspergillus jlavus and cottony white B. bassiana mycelium were observed growing on dead P. pupatasi (Jordan Valley) (Fig. 3). A. flavus did not cause any obvious morbidity but B. bassiana appeared to be associated with decreased colony vitality. The spread of infection could be controlled by the careful removal of dead flies and maintenance of general cleanliness. The B. bassiana isolate was subsequently used for infection experiments. S. marcescens was identified in
195
PATHOGENS
dead flies from the same colony. The bacterium was isolated and used for infection trials. Frequently, premature mortality in laboratory-reared flies was associated with accumulations of microbes on their tarsi. These sticky aggregates caused the flies to lose legs and eventually become stuck and die. The organisms multiplying on tarsi varied: In a P. papatasi colony mostly yeast were found (Fig. 13), whereas bacilli mixed with fungal hyphae and yeast were observed on the tarsi of L. longipalpis (Fig. 14). DISCUSSION CPV infections were previously observed to interfere with the development of concomitant Leishmania infections. Therefore, the virus was considered potentially useful for the biological control of leishmaniasis, despite its apparent lack of pathogenicity to sand flies (Warburg and Ostrovska, 1987). For the infection experiments, relatively small doses of polyhedra were available. This constraint may have contributed to the low infection rates (Table 1). The results clearly show that the Indian strain of P. papatasi is more susceptible than the Egyptian strain, but the overall infection rates remained low and the quantities of viral material were not sufficient to enable further studies. Interestingly, when CPVs were first described in a laboratory colony maintained at the Hebrew University, Israel, 95.9% (n = 244) of the flies were spontaneously infected (Warburg and Ostrovska, 1987). At the same time high rates (percentage unknown) were also recorded in colonized flies originating from the same stock but separated some 4 years earlier and maintained, using a different methodology, at Yale University. Rates subsequently subsided, more or less simultaneously in both colonies, without any apparent cause. S. marcescens commonly affects sand fly colonies of different species. Sugar solutions from which flies feed become pinkish
ALON
WARBURG
SAND
FLY
and dead flies turn red due to the proliferation of bacteria in their tissues (unpubl. observations). The phenomenon is not normally associated with a notable reduction in colony vigor or fecundity. In experiments, S. marcescens did not significantly reduce longevity or fecundity of flies, even when ingested in large quantities (Table 2). It is possible that sugar meals are not suitable vehicles for S. marcescens. However, a similar method was shown to be highly effective for the delivery of Bacillus thuringiensis israelensis endotoxin to both male and female sand flies (Yuval and Warburg, 1989). Like B. thuringiensis israelensis, S. marcescens initiates infections after being ingested. Hence, the most probable explanation for these observations is that the S. marcescens isolate< was of low toxicity to sand flies. Promastigote kinetoplastids, the most frequently reported parasites in sand flies, were routinely assigned to the genus Leishmania until biochemical and immunological techniques for parasite identification were introduced. Thus, at least some infections by other trypanosomatids were probably overlooked. The generic status of the parasite isolated from L. pia is uncertain. Biochemical, immunological, and morphological data show that it is neither a Leishmania sp. nor a Trypanosoma sp. The large size of some morphotypes and their longitudinal body torsion (Fig. 4) are characteristic of monoxenous-insect flagellates (genus Leptomonas) and of insect-transmitted plant parasites of the genus Phytomonas (Vickerman, 1979; Wallace, 1979). Since phle-
PATHOGENS
botomines
197
are not plant feeders, a Leptomthe most probable candidate. However, sand flies do ingest plant-derived sugars (Schlein and Warburg, 1985), making the presence of plant pathogens in their alimentary tract feasible. Furthermore, the absence of cysts in cultures of the L. pia-derived trypanosomatid could indicate its heteroxenous nature. Flagellar cysts are formed by Leptomonas and other monoxenous insect trypanosomatids but not by Phytomonas spp. (McGhee and Hanson, 1964; Vickerman, 1979; Wallace, 1979). Gregarines have been reported in over 20 species of sand flies (Young and Lewis, 1977, 1980). The genus Ascogregarina (=Ascocystis) comprises 15 described species including 3 from sand flies (previously Monocystis spp.) and 5 from mosquitoes (previously Lankesteria spp.; Levine, 1988). The best studied sand fly parasite is Ascogregarina chagasi in L. longipalpis, originally described by Adler and Mayrink (1961) and subsequently studied by others who performed cross infection experiments (Wu and Tesh, 1989) and characterized the role of the fly’s immune response in parasite vertical transmission (Warburg and Ostrovska, 1989). Frequently, high rates of infection are encountered in laboratory colonies but the effect on longevity and fecundity is apparently limited (Wu and Tesh, 1989). Ascogregarina spp. of sand flies are unique in that gametocysts adhere to the genital accessory glands and oocysts are injected into them, thereby facilitating vertical transmission (Adler and Mayrink, 1961; onas sp. must be considered
FIG. 5. Sporulated oocysts of A. saruviue from the accessory gland of I,. lichyi. Phase-contrast micrograph. Bar = 10 pm. FIGS. CS. SDA culture of Entomophthorales fungus isolated from L. pia. Phase-contrast micrographs. FIG. 6. Sterile hyphae. Bar = 20 pm. FIG. 7. Sporangiophore. Bar = 2 pm. FIG. 8. Ballistospore. Bar = 2 pm. FIGS. 9 AND 10. Ectoparasitic fungus on first instar Lutzomyiu spp. Phase-contrast micrographs of specimens mounted in Hoyer’s medium. FIG. 9. Whole larva with fungi on head and caudal setae (arrows). Bar = 0.5 mm. FIG. 10. Higher magnification of fungus on a caudal seta. Bar = 30 pm.
ALON
WARBURG
FIG. 11. Tylenchid nematodes in a wild-caught L. shannoni male. (A) External genitalia showing a worm squeezed out through the anus. (B) Abdomen containing numerous worms. Phase-contrast micrograph of specimen mounted in Hoyer’s medium. Bar = 0.1 mm. FIG. 12. Spimrid nematode dissected out of a L. rownsendi female. (A) Anterior portion, (B) posterior portion. Phase-contrast micrograph of a fresh preparation. Bar = 0.25 mm.
SAND
FLY
PATHOGENS
FIGS. 13 AND 14. Microbial aggregates on tarsi of laboratory-reared sand flies. Scanning electrc In micrographs (fixed in 3% glutaraldehyde for 2 hr, 1% osmium-tetroxide for 1 hr. and dehydrated Iin ethanol and freon). FIG. 13. Yeast and saprophytic fungi on P. papatasi. Bar = 20 km. FIG. 14. Bacteria and saprophytic fungi on L. longipalpis. Bar = 10 pm.
trovska et al., 1990; Warburg and Ostroca, 1989). infections contribute 1Zntomophthorales to televated mortality rates in mosquito popula .tions overwintering in caves and base1974, and references me :nts (Roberts, OS
vsj
therein). Cumulated reports indicate : that these fungi may constitute important I:bathogens affecting sand fly populations as well. McConnel and Correa (1964) encoun ttered high rates (14%) of fungal infections in eight cave-dwelling Panamanian sand fly splecies ,
200
ALON
WARBURG
The fungus was not identified, but a mycelium featured in their published micrograph is almost certainly of an Entomophthorales sp. I have observed numerous Entomophthorales infections in mounted specimens (courtesy of J. Murillo) of L. longipalpis caught in pit latrines in the town of Nazareth, Guanacaste, Costa Rica (unpubl. data). Maltese P. papatasi were reported infected with Entomophthora (Marett, 1915). Rioux et al. (1966) found resting spores in the hemocoel of a female Phlebotomos ariasi caught in the south of France. Many other reports of unidentified fungal infections in sand flies (Young and Lewis, 1977, 1980) may well have been due to Entomophthorales spp. Unfortunately, Entomophthorales fungi are particularly fastidious and difficult to culture, a trait which reduces their potential usefulness for biological control. Further studies are needed to identify possible means for mass propagation of Entomophthora spp. and to isolate strains more amenable to artificial culture. The L. pia isolate described here grew well on FBS-fortified SDA, indicating that it may have been a Conidiobolus sp. rather than an Entomphthora sp. (H. Evans, pers. commun.). B. bassiana spores suspended in sucrose solutions were not very infectious to P. papatasi; neither longevity nor fecundity were significantly reduced and only a low percentage of dead flies exhibited mycelial growth (Table 2). This observation is consistent with the fact that most entomopathogenic fungi infect the hemocoel after gaining access through the cuticle and not via the gut (Evans, 1989). In addition, a high sugar concentration (30% w/v) used for feeding sand flies may prevent spore germination, thereby contributing to the low infectivity of per OS-ingested B. bassiana.
In P. papatasi exposed to B. bassiana spores smeared on dry filter papers, the 100% mortality after 4-6 days of exposure contrasted sharply with a mere IO-25% observed in control groups. Preoviposition
death of most females was reflected by an extremely low egg count (Table 3). The same fungus was even more pathogenic to L. longipalpis. Mycelial growth was seen on virtually all of the dead flies and conidia were produced 4-6 days after death (Figs. 2,3). Many of the dead flies remained affixed to the mesh cover, plastic walls, and filter paper (Fig. 2). This phenomenon is commonly observed when insects that die of fungal infections assume elevated and exposed positions. They remain secured to the substrate by their stiff limbs and by hyphal strands (Evans, 1989). The saprophytic fungal isolates which were apparently associated with Lutzomyia spp. mortality had probably originated in the gut contents of the morbid larvae. Such fungi are normally present in larval-rearing media and may cause opportunistic secondary infections in flies affected by other pathogens or by adverse conditions. The bacterial and yeast aggregates found on the tarsi of colony-bred flies (Figs. 13,14) are probably propagated by contact with sugar solutions on which flies alight in order to feed. The maintenance of clean cages, frequent changing of sugar solutions, and correct humidity usually eliminate such problems. Tylenchid nematodes, described here in a L. shannoni male (Figs. llA,B), usually develop to third or fourth stage juveniles and leave the host via the anus or reproductive pore. Adult tylenchids live in the environment and mated females reinvade an insect host by direct penetration of the cuticle (Poinar, 1983). Heavy tylenchid infections have previously been recorded in Lutzomyia sanguinaria, nis, and Lutzomyia
Lutzomyia panamensis
vespertilio-
(McConell and Correa, 1964). Their life cycle renders tylenchids unsuitable for biological control. A possibly better candidate, a sand fly parasitic tetradonematid nematode, was studied by Killick-Kendrick et al. (1989). This pathogen delayed larval development, interfered with blood feeding by females, and induced male sterility.
201
SAND FLY PATHOGENS
The worm found
in L. townsendi
be-
longed to the Spirurida, an order of heteroxenous parasites of vertebrates infecting insects as intermediate hosts (Poinar, 1983). Encapsulated third-stage spirurid nematodes (rodent-infecting Masrophorus muris) have been recorded in P. ariasi (Killick-Kendrick et al., 1976). The effect of spirurid infections on sand flies is difficult to assess, but their heteroxenous life cycle precludes them from biological control. Demonstrating efficacy in laboratory trials is the essential prerequisite for identifying potential biocontrol agents. Clearly, the more challenging part is application in field situations. Control measures against phlebotomines should focus on large concentrations resting and breeding in accessible habitats. Some important vector species occupy such habitats. L. longipalpis, the vector of visceral leishmaniasis in South America, often breeds in caves and P. papatasi, the vector of cutaneous leishmaniasis in the Mediterranean basin and Russia, breeds in rodent burrows. As previously mentioned, these dark, humid, and cool (20-25%C) resting sites are also suitable for the long term persistence and germination of fungal spores. Within rooms or caves, the flies’ distribution is further limited to the darkest niches such as cracks, corners between ceiling and walls, and surfaces that do not receive direct light. Therefore, small pieces of paper smeared with fungal spores (e.g., a mixture of B. bassiuna and an effective pathogen of larvae) and hung in appropriate places may serve to deliver concentrated doses of spores that will remain viable for long periods. In less accessible places, spraying of spores may be more feasible. Adult sand flies, dying of the fungus, should facilitate the vertical transmission to larvae feeding below and horizontal transmission to other flies sharing the same resting sites. ACKNOWLEDGMENTS Part of this work-conducted at Centro Intemacional de Investigaciones Medicas, Cali, Colombiareceived support from Grant ID 840336 of the UNDPI
WORLD BANK/WHO special program for research and training in tropical diseases. Additional support was given by a Wellcome Trust grant (LSHTM, London). I am grateful to H. C. Evans of CIBC, UK, for identifying fungal and bacterial isolates; to scientists at CIDEIM for technical assistance; and to A. Chabaud, I. Landau, and D. Van Waerebeke of the Natural History Museum, Paris, for identifying nematodes.
REFERENCES ADLER, S. 1964. Leishmania. Adv. Parasitol., 2, 3% 96. ADLER, S., AND MAYRINK, W. 1961. A gregarine, Monocystis chagasi n. sp., of Phlebotomus iongipalpis: Remarks on the accessory gland of P. longipalpis.
Rev.
Med.
Trop.
Sao Paula,
3, 230-238.
ASHFORD, R. W. 1974. Sandflies (Diptera: Psychodidae) from Ethiopia: Taxonomic and biological notes. J. Med. Entomol., 11, 605-616. BURGESS, H. D. (Ed.) 1981. “Microbial Control of Pests and Plant Diseases 1970-1980.” Academic Press, New York. EVANS, H. C. 1989. Mycopathogens of insects of epigeal and aerial habitats. In “Insect-Fungus Interactions” (N. M. Collins, M. P. Hammond, and J. F. Webbes, Eds.), pp. 205-238. Academic Press, New York. HECKLEY, R. J. 1978. Preservation of microorganisms. AdI’. Appl. Microbial.. 24, l-53. KILLICK-KENDRICK, R. 1979. The biology of Leishmania in phlebotomine sandflies. In “Biology of the Kinetoplastida” (W. H. R. Lumsden and D. A. Evans, Eds.). Vol. 2, pp. 395-460. Academic Press, New York. KILLICK-KENDRICK, R. 1987. Breeding places of Phlebotomus ariasi in the Cevennes focus of leishmaniasis in the south of France. Parassitologia. 29, 181191. KILLICK-KENDRICK. R., KILLICK-KENDRICK, M.. QLJALA. N. A., NAWI. R. W.. ASHFORD, R. W., AND TANG. Y. 1989. Preliminary observations on a tetradonematid nematode of phlebotomine sandflies of Afghanistan. Ann. Parasitol. Hum. Camp., 64, 332-339. KILLICK-KENDRICK. R.. LEANEY. A. J., MOLYNEUX, D. H., AND RIOUX, J.-A. 1976. Parasites of Phlebotomus
ariasi.
Trans.
R. Sot.
Trop.
Med.
Hyg.
70,
22. Phylum LEVINE, N. D. 1988. “The Protozoan Apicomplexa.” Vol. 1. CRC Press, Boca Raton, Florida. MARETT. P. J. 1915. The bionomics of the Maltese phlebotomi. &it. Med. J.. 2, 172-173. MCCONNEL. E., AND CORREA, M. 1964. Trypanosomes and other microorganisms from Panamanian Phlebotomus sand flies. J. Parasitol., 50, 523-528. MCGHEE, R. B., AND HANSON, W. L. 1964. Comparison of the life cycle of Leptomonas oncopelti and Phytomonas elmassiani. J. Protozool.. 11, 555-562.
202
ALON
WARBURG
MCMAHON-PRATT, D., AND DAVID, J. R. 1981. Monoclonal antibodies that distinguish between New World species of Leishmania. Nature, 291, 581-583. MODI, G. B., AND TESH, R. B. 1983. A simple technique for mass rearing Lutzomyia longipalpis and Phlebotomus papatasi (Diptera: Psychodidae) in the laboratory. J. Med. Entomol., 20, 568-569. OSTROVSKA, K., WARBURG, A., AND MONTOYALERMA, J. 1990. Ascogregarina saraviae, N. sp. in Lutzomyia lichyi. J. Protozool., 37, 69-70. POINAR, G. 0. 1983. Nematode parasites of invertebrates. In “The Natural History of Nematodes” (G. 0. Pionar, Ed.), pp. 161-202. Prentice-Hall, Englewood Cliffs, New Jersey. POINAR, G. L., AND THOMAS, G. M. 1984. “Laboratory Guide to Insect Pathogens and Parasites.” Plenum, New York. ROBERTS, D. W. 1974. Coelomomyces, Entomophthora, Beauvaria and Metarhizium as parasites of mosquitoes. Misc. Pub. Entomol. Sot. Am., 7, MO155. RIOUX, J. A., MANIER, J. F., AND TOUR, S. 1966. Infestation a Entomophthora sp. chez Phlebotomus ariasi Tonnoir, 1921. Ann. Parasitol., 41, 251-253. SARAVIA, N. G., HOLGUIN, A. F., MCMAHONPRATT, D., AND D’ALESSANDRO, A. 1985. Mucocutaneous leishmaniasis in Colombia: Leishmania braziliesis subspecies diversity. Am. J. Trop. Med. Hyg., 34, 714-720. SCHLEIN, Y., AND WARBURG, A. 1985. Phytophagy and the cycle of feeding of Phlebotomus papatasi (Diptera: Psychodidae) under experimental conditions. J. Med. Entomol., 23, 11-15. TESH, R. B. 1988. The genus Phlebovirus and its vectors. Annu. Rev. Entomol., 33, 169-181. VICKERMAN, K. 1979. The diversity of the kinetoplas-
tid flagellates. In “Biology of the Kinetoplastida” (W. H. R. Lumsden and D. A. Evans, Eds.), Vol. 1, pp. l-34. Academic Press, New York. WALLACE, F. G. 1979. Biology of kinetoplastida of arthropods. In “Biology of the Kinetoplastida” (W. H. R. Lumsden and D. A. Evans, Eds.), Vol. 2, pp. 213-240. Academic Press, New York. WARBURG, A., MONTOYA, J., JARAMILLO, C., CRUZ, A. L., AND OSTROVSKA, K. 1991. Leishmaniasis vector potential of Lutzomyia spp. in Colombian coffee plantations. Med. Vet. Entomol., 5, 9-16. WARBURG, A., AND OSTROVSKA, K. 1987. Cytoplasmic polyhedrosis viruses in Phlebotomus papatasi inhibit development of Leishmania major. J. Parasitol., 73, 578-583. WARBURG, A., AND OSTROVSKA, K. 1989. An immune response-dependent mechanism for the vertical transmission of an entomopathogen. Experientia, 45, 710-172. Wu, W.-K., AND TESH, R. B. 1989. The infection of Old and New World phlebotomine sand flies (Diptera: Psychodidea) with Ascogregarina chagasi. J. Med. Entomol., 26, 237-242. YOUNG, D. G., AND LEWIS, D. J. 1977. Pathogens of psychodidae (phlebotomine sand flies). In “Bibliography on Pathogens of Medically Important Arthropods” (D. W. Roberts and J. M. Castillo, Eds.), Bull. WHO, SS(Suppl.), S9-S24. YOUNG, D. G., AND LEWIS, D. J. 1980. Pathogens of psychodidae (phlebotomine sand flies). In “Bibliography on Pathogens on Medically Important Arthropods” (D. W. Roberts and J. M. Castillo, Eds.), Bull. WHO, Xl(Supp1.) S9-Sl 1. YUVAL, B., AND WARBURG, A. 1989. Susceptibility of adult phlebotomine sandflies to Bacillus thuringiensis var. israeliensis. Ann. Trop. Med. Parasitol., 83, 165-196.
