Title: Phlebotomine sand flies (Diptera: Psychodidae) and diffuse cutaneous leishmaniasis in the Dominican Republic /
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00102795/00001
 Material Information
Title: Phlebotomine sand flies (Diptera: Psychodidae) and diffuse cutaneous leishmaniasis in the Dominican Republic /
Physical Description: Book
Language: English
Creator: Johnson, Richard Nicholas, 1956-
Copyright Date: 1984
 Record Information
Bibliographic ID: UF00102795
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: ltuf - ACN7945
oclc - 11908433

Full Text








To my parents,

Robert and Mary Johnson


This research could not have been done without the

advice, encouragement, and assistance of many people to whom

I am greatly indebted. My supervisory committee was helpful

in guiding my course of study: Dr. Jerry F. Butler, chair-

man; Drs. Donald W. Hall, Donald J. Forrester, and Ellis C.

Greiner; and especially Dr. David G. Young. Dr. Bryce C.

Walton, World Health Organization, first suggested the

project and was instrumental in securing financial support

under WHO grant #810314. Other support was generously given

by the Steffan Brown Foundation and the University of


I am also indebted to the personnel of the Dominican

Dermatology Institute, including Dr. Huberto Bogaert-Diaz

(Director), Dr. Denis de Martinez, Lic. Margarita de

Quinones, Lic. Marvis Lebron, Tomas Castro, and particularly

Francisco Castillo, for his friendship and assistance. Gulf

and Western Americas Corporation graciously provided facili-

ties at the Pedro Sanchez field station. The staff and

their families at this ranch provided the friendship that

turned it into a home for more than a year. Dr. Rodrigo

Zeledon and Juan Murillo, Instituto Costarricense de

Investigation y Ensenanza en Nutricion y Salud, were of


great help during their brief visit to the Dominican

Republic in 1983.

I wish to thank Dr. Eskild Petersen, University of

Arizona, for his advice and assistance. I am indebted to

Dr. Sam Telford for the careful instruction on the identifi-

cation of lizard parasites. Dr. Robert Woodruff, Florida

Division of Plant Industry, provided valuable advice which

facilitated working in the Dominican Republic.

Personnel at Walter Reed Army Institute of Research

were of much help in realizing the goals of the project;

these included MAJ Peter Perkins, Dr. Edgar Rowton, Spec. 4

Pedro Quintero, LTC Donald Roberts (Entomology): CPT Patrick

McGreevy, LTC Jonathan Berman, Dr. Eileen Franke (Experi-

mental Therapeutics).

Dr. Charles Woods of the Florida State Museum provided

information on the Republic and its mammalian fauna. Rick

Sullivan, who concurrently resided in the country, provided

extra traps, friendship, and, at times, mutual commissera-

tion. Dr. Steven Zam was instructive in laboratory cultur-

ing of Leishmania. I thank Ms. Diana Simon and Mrs. Debra

Boyd who handled many of the concerns that arose during my

absences from Gainesville. Ms. Edna Mitchell helped in

laboratory rodent and sand fly maintenance. Drs. G.B.

Fairchild and R.C. Wilkerson gave freely of their knowledge

of Latin America. Drs. Richard Endris, Peter Perkins,

Andrew Beck, Mr. Eric Milstrey, MAJ Phillip Lawyer, and CPT

Terry Klein offered their advice, assistance, and friendship

throughout varying times of acquaintance. Finally, I

gratefully acknowledge the encouragement and assistance

provided by my family and other friends who have been

supportive for the past 28 years.




LIST OF TABLES . . . . .


ABSTRACT . . . . . .



Literature Review . . . . .
Geography of the Dominican Republic
Current Status of Diffuse Cutaneous
Leishmaniasis in the Dominican
Republic . . . . . .
Study Site Selection . . . .
Study Site Descriptions . . .
Objectives . . . . . .

. . iii

. . viii

. . x


* . 12

* . 13
. . 12

. . 13
* . 20
* . 22
* . 27

CHRISTOPHEI . . . . . . .

Introduction . . . .
Methods and Materials . .

Field Studies .
Laboratory Rearing .
Sand Fly Dissections

Results . . . . .

Field Studies . .
Laboratory Studies .
Sand Fly Dissections .

Discussion . . . .

. . . 30
S . . 40
S . . 46

S . . 47

. . 65

Chapter Page

FLIES . . . . . . . . . 73

Introduction . . . . . . 73
Methods and Materials . . . . .. 74

Comparison of the Growth of Three
Strains of Leishmania . . . 74
Growth of Leishmania-Isabel Strain
in Laboratory Rodents . . .. 75
Growth of Leishmania-Isabel Strain
in Sand Fly . . . . . 77
Transmission of Leishmania-Isabel
Strain by Lutzomyia christophei 78

Results . . . . . . . ... 79

Comparison of the Growth of Three
Strains of Leishmania . . . 79
Growth of Leishmania-Isabel Strain
in Laboratory Rodents. . . 81
Growth of Leishman-Isabel Strain
in the Sand Fly . . . .. 84
Transmission of Leishmania-Isabel
Strain by Lutzomyia christophei 89

Discussion . . . . . . .. 90


Introduction . . . . . . 96
Methods and Materials . . . . .. 97
Results . . . . . . . 101
Discussion . . . . . . .. . 105

5 SUMMARY . . . . . . . . . 111

HISTORIES . . . . . . . .. 114

REPUBLIC, MAY 1981 AUAGUST 1983 ... .116

MAY 1981 AUGUST 1983 . . . . .. 118

REFERENCES . . . . . . . .. . . 119




Table Page

1-1 Sand fly species (Lutzomyia) which are
known or suspected vectors of leishmaniasis
in the New World, by country . . . . 5

1-2 Other mammalian hosts of Leishmania strains
which cause human leishmaniasis in the New
World, by country . . . . . . 10

2-1 Location and dates for flight trap samples
in the Dominican Republic . . . ... 35

2-2 Location, trap type, and dates for CDC traps
in the Dominican Republic . . . ... 36

2-3 Sites and dates for Disney traps used for
phlebotomine sand fly sampling in the
Dominican Republic . . . . . ... 39

2-4 Collection sites and dates for soil samples
examined for the presence of sand fly
larvae . . . . . . . . ... .41

2-5 Mean duration ( S.D.) in days of immature
stages of Lu. cayennensis hispaniolae at
three temperature regimes, according to sex
of sand fly . . . . . . ... 57

2-6 Mean duration ( S.D.) in days of immature
stages of Lu. christophei at two temperature
regimes, according to sex of sand fly . 63

2-7 Sites of collection for female Lu.
cayennensis dissected . . . . ... 66

3-1 Method of inoculation and number of
promastigotes used to infect laboratory
rodents with Leishmania-Isabel strain . 76


Table Page

3-2 Number of animals examined determined to be
infected with Leishmania-Isabel strain via
different isolation methods . . ... 82

3-3 The course of development of Leishmania-
Isabel strain in the sand fly, Lutzomyia
anthophora, based on daily dissections,
Days 1 to 7 post feeding . . . ... 85

4-1 Mammal specimens collected at six
leishmaniasis case sites in the Dominican
Republic . . . . . . . . . 99

4-2 Examination techniques used for mammals
collected during survey for reservoir hosts
of leishmaniasis in the Dominican Republic,
October 1981 to August 1983 . . ... .104


Figure Page

1-1 The Caribbean region, showing the location
of the Dominican Republic, on the island of
Hispaniola . . . . . . . ... 15

1-2 The geographic regions of the Dominican
Republic . . . . . . . ... .16

1-3 Case sites for patients with diffuse
cutaneous leishmaniasis (DCL) in the
Dominican Republic . . . . . ... 18

1-4 Typical terrain in the Cordillera Oriental,
Dominican Republic . . . . . ... 19

1-5 A young Dominican DCL patient with healed
lesions on wrist and upper arm ...... 19

2-1 Field collecting equipment and rearing
containers for phlebotomine sand flies . 31

2-2 Sand fly feeding cage a modified aquarium
with plaster of Paris bottom and back . 32

2-3 The author in front of a flight trap in a
cacao grove at Altos de Peguero, El Seibo
Prov. . . . . . . . . ... 34

2-4 A CDC trap and modified traps . . .. 34

2-5 A Disney trap, used for collecting rodent-
feeding sand flies ............ 38

2-6 Schematic diagram of laboratory techniques
for rearing of phlebotomine sand flies . 43

2-7 A modified microtiter plate for rearing
individual sand fly larvae . . .... 44

2-8 Collection sites for Lu. cayennensis
hispaniolae and Lu. christophei in the
Dominican Republic May 1981-August 1983 49

Figure Page

2-9 Weekly sample populations of Lu.
cayennensis at two study sites in the
Dominican Republic . . . . . .. 51

2-10 Eclosion time, in days after oviposition,
for Lu. cayennensis reared under constant
conditions . . . . . . . ... 58

2-11 Eclosion time, in days after oviposition,
for Lu. cayennensis reared under ambient
conditions, August-September . . ... 58

2-12 Eclosion time, in days after oviposition,
for Lu. cayennensis reared under ambient
conditions, January-February . . ... 59

2-13 Female Lu. christophei feeding on BALB/c
mouse . . . . . . . . . . 61

2-14 Eclosion time, in days after oviposition,
for F Lu. christophei reared under constant
condi ions . . . . . . . . 64

2-15 Eclosion time, in days after oviposition,
for F and F Lu. christophei reared under
constant conditions . . . . . .. 64

3-1 Daily estimated mean population of three
strains of Leishmania grown in Schneider's
medium, at 250C . . . . . ... 80

3-2 Leishmania-Isabel strain amastigotes in
spleen impression smear stained with Giemsa
(1000X) . . . . . . . ... 83

3-3 Promastigotes of Leishmania-Isabel strain
from the anterior midgut of the sand fly,
Lu. anthophora (1000X) . . . . ... 87

3-4 The course of Leishmania-Isabel strain
infection in the sand fly, Lutzomyia
anthopora . . . . . . . . 88

4-1 Sherman and Tomahawk live traps for small
mammals . . . . . . . ... .98

4-2 The three species of rodents trapped during
the survey . . . . . . . ... 102

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



Richard N. Johnson

August 1984

Chairman: Dr. Jerry F. Butler
Major Department: Entomology and Nematology

A survey for phlebotomine sand flies (Diptera:Psycho-

didae) in the Dominican Republic revealed that Lutzomyia

cayennensis hispaniolae (Fairchild and Trapido) was widely

distributed and fairly common. Lutzomyia christophei

(Fairchild and Trapido) was more limited in geographic

distribution. Specimens of the latter species were obtained

by light traps, flight traps, and aspirator collection from

human bait and resting sites. Laboratory colonies of both

species were established and life-cycle data were obtained.

Lutzomyia cayennensis females readily fed on Anolis lizards.

Female Lu. christophei readily fed on rodents and were

capable of experimentally transmitting a Dominican strain

(Isabel-WR336) of Leishmania to BALB/c mice seven to

ten days after biting infected mice. Development of the


parasite occurred in the anterior midgut in both Lu.

christophei and Lu. anthophora (Addis), a species that was

also experimentally infected. The course of development in

the sand fly was observed by dissecting 15 infected Lu.

anthophora on days 1-7 post-feeding. Development in this

species was parallel to that observed in the 17 Lu.

christophei. Promastigotes from flies four and five days

post-feeding were infective to hamsters, as determined by

xenodiagnosis with sand flies and spleen culture.

In culture medium, Leishmania-Isabel strain grew at a

much slower rate than either of two strains of L. mexicana.

Hamsters and TRC mice, experimentally inoculated from

culture, showed no outward sign of infection until at least

2.5 months after inoculation with the Isabel strain.

In the Dominican Republic, 10 of the 21 known case

sites were visited. Coffee and cacao groves were charac-

teristic of these sites. Two female Lu. christophei were

captured while biting a patient. Four species of mammals

(170 specimens) were trapped from five of the case sites and

examined for leishmaniasis using various methods. None was

found to be infected, though 4 of 44 Rattus rattus from one

site were seropositive (1:16), as determined by indirect

fluorescent antibody test.

Lutzomyia christophei is most probably the vector of

diffuse cutaneous leishmaniasis in the Dominican Republic.

The identity of the reservoir remains unknown, but R. rattus

is the most likely suspect.




Literature Review

Phlebotomine sand flies* are biting members of the

dipteran family Psychodidae (Quate and Vockeroth, 1981).

They are suspected or confirmed vectors of various parasitic

agents including phleboviruses, such as sand fly fever

virus; bartonellosis (Adler and Theodor, 1957); saurian

malaria (Ayala and Lee, 1970); trypanosomes of amphibians,

lizards (Anderson and Ayala, 1968), and mammals (McConnell

and Correa, 1964); and leishmaniasis (Bray, 1974).

Leishmaniasis is a complex of diseases which is caused

by Leishmania spp. occurring in many parts of the world. In

1981, the World Health Organization estimated that there

were 400,000 new cases of leishmaniasis in the world,

annually. The disease occurs in the Americas, Europe,

Africa, and Asia. It has been considered the second most

important protozoan disease of man after malaria (Anonymous,

1981). The only known biological vectors are phlebotomine

sand flies (Bray, 1974). In the Americas, leishmaniasis

* In this presentation, sand fly will refer to members of
Diptera: Psychodidae: Phlebotominae.

occurs mainly among persons living in rural or forested

areas (Herrer et al., 1966). Leishmaniasis appears in three

basic clinical forms-visceral, mucocutaneous, and cutaneous.

In 1948, a subform of cutaneous leishmaniasis was

reported independently in Venezuela (Convit and Lapenta,

1948) and in Bolivia (Barrientos, 1948). At first, it was

considered a new form of the disease because of its charac-

teristic features (Convit et al., 1962; Convit and Kerdel-

Vegas, 1965). Bryceson (1969, p. 709) summarizes these

features as follows:

1. There is an initial lesion which spreads

locally, and from which the disease dissem-

inates to other parts of the skin, often

involving large areas.

2. The lesions are nodules and do not ulcerate.

3. There is a superabundance of parasites in the


4. The histology is characteristic in that

macrophages full of amastigotes predominate

in the lesion.

5. Internal organs are not invaded and there is

no history of visceral leishmaniasis.

6. The leishmanin (Montenegro) test is negative.

7. The disease progresses slowly and becomes


8. Treatment with antimony produces only slight

and temporary improvement.

By these criteria, other cases have been reported from

the USA in Texas (Simpson et al., 1968), Brazil, Ecuador,

Mexico, Ethiopia, and Tanzania (Bryceson, 1969). The

appearance of the lesions gave rise to the name diffuse (or

disseminated) cutaneous leishmaniasis (DCL). The causative

agent of the Venezuelan cases was first named Leishmania

pifanoi (Medina and Romero, 1962). Areas where cases

occurred were endemic for cutaneous leishmaniasis, but not

the visceral disease (Convit and Kerdel-Vegas, 1965).

Further work determined that rather than being a new para-

site, DCL was the result of a deficiency in the host's

cell-mediated immunity (Bryceson, 1970b; Convit et al.,

1971). The disease is caused by the same species of

Leishmania that causes ulcerative cutaneous leishmaniasis in

the same area, e.g., L. aethiopica in Ethiopia (Lemma et

al., 1970; Bray et al., 1973) and L. mexicana mexicana, L.

m. amazonensis, and L. m. pifanoi in Central and South

America (Lainson and Shaw, 1978).

In 1975, a new focus of DCL was reported in the Domin-

ican Republic (Bogaert-Diaz et al., 1975). Three humans,

siblings, were found infected. There have been 22 addi-

tional cases, none of which had ulcerating lesions, sup-

porting the concept that DCL is determined both by parasite

characteristics and a defect in host immunocompetence

(Walton, pers. comm.). Studies of some of the patients from

the Dominican Republic showed that this defect is a specific

inhibition of lymphocyte-proliferation responses by adherent

suppressor T-cells, which has a genetic basis (Petersen et

al., 1982). Many of the Dominican DCL patients have been

symptomatically cured using hot (450C) water treatment

(Neva, pers. comm.). The Leishmania causing DCL in the

Dominican Republic remains unnamed, but it appears to be

different from strains in the L. braziliensis complex and in

the L. mexicana complex based on enzyme electrophoretic

mobility assays (Kreutzer et al., 1983), excreted factor

serotype, growth in artificial media, and infectivity and

pathogenicity in laboratory rodents (Schnur et al., 1983).

Although leishmanaisis may be mechanically transmitted

under lab conditions by Stomoxys calcitrans, stable flies

(Lainson and Southgate, 1965), Rhipicephalus sanguineus,

brown dogs ticks (Sherlock, 1964), and Glossina morsitans,

tsetse flies (Lightner and Roberts, 1984), sand flies are

the only known biological vectors (Lainson and Shaw, 1978).

In the New World, 21 species of sand flies have been

reported as natural hosts of Leishmania spp. infecting man,

but only 6 species have been determined, at present, to be

natural vectors (Table 1-1).

Experimentally, New World Leishmania are capable of

infecting a number of sand fly species, most of which are

then capable of transmitting the infection to a susceptible

mammalian species (Killick-Kendrick, 1979). The amastigote

stage is parasitic in vertebrate macrophages (Bray, 1974).

Table 1-1. Sand fly species (Lutzomyia) which are known or
suspected vectors of leishmaniasis in the New
World, by country.

Suspected or Proven
Country Lutzomyia Vectors Leishmania Strain

Belize olmeca2 L.m.m.

Bolivia longipalpis3 L.d.c.

Brazil longipalpis L.d.c.

amazonensis L.b.b.

intermedia2 L.b.b.
migonei L.b.b.

paraensis3 L.b.b.
pessoai L.b.b.

wellcomei2 L.b.b.
whitmani L.b.b.

anduzei3 L.b. .

umbratilis2 L.b.q.

flaviscutellata3 L.m.

Colombia longipalpis3 L.d.c.

trapidoi3 L.b.

Table 1-1. Continued

Suspected or Proven
Country Lutzomyia Vectors Leishmania Strain

Costa Rica

El Salvador

French Guiana
































Table 1-1. Continued

Suspected or Proven
Country Lutzomyia Vectors Leishmania Strain

Paraguay longipalpis L.d.c.

Peru peruensis3 L.p.

verrucarum L.p.

Surinam umbratilis L.b._.

USA diabolica L.m.

Venezuela longipalpis3 L.d.c.

flaviscutellata3 L.m.a.

townsendi3 L.m.q.

Source: World Health Organization,

1984, pp.

Leishmania strains: L.m.m. = L. mexicana mexicana,
L.d.c. = L. donovani chagasi, L.b.b. = L. braziliensis
braziliensis, L.b._. = L.b. guyanensis, L.b.2. =
panamensis, L.p. = L. peruviana, L.m.a. = L.m.
amazonensis, L.m.q. = L.m. garnhami.
Proven vector.

Suspected vector.


Following ingestion by the sand fly, a telmophagic feeder,

the parasite exsheathes its flagellum to become the promas-

tigote, which multiplies initially in the hind or midgut,

depending on the parasite species. Members of the

L. braziliensis complex (Section Peripylaria) develop in the

posterior midgut and anterior hindgut before moving anter-

iorly to effect transmission. Leishmania donovani and

subspecies in the L. mexicana complex (Section Suprapylaria)

multiply initially in the anterior midgut (Killick-Kendrick,

1979). Transmission to a vertebrete occurs during the sand

fly bite, but the actual mechanism is unknown (Killick-

Kendrick, 1978).

Two extant species of sand flies are known from the

Dominican Republic (Fairchild and Trapido, 1950). Two fos-

sil species have been discovered recently in Dominican amber

(Johnson and Young, in preparation), reported to be from the

Oligocene Period, 40 to 60 million years old (Sanderson and

Farr, 1960). Prior to this study, practically nothing was

known abut the distribution or biology of the living spe-

cies. One of these, Lutzomyia christophei (Fairchild and

Trapido, 1950), belongs to the Lu. verrucarum species group

(Lewis, 1968), which also contains a number of man-biting

species (Young, 1979) and at least one vector of leishmania-

sis in Peru (Lainson and Shaw, 1979). Contrary to Lainson's

(1983) statement, nothing was known about the feeding habits

of Lu. christophei, prior to the author's study. The other

species, Lu. cayennensis hispaniolae (Fairchild and Trapido,

1950), is an endemic subspecies of a species that occurs

from Mexico south to Ecuador and French Guiana (Young,

1979). This species is known to feed on poikilothermic

vertebrates, though there is one report of females feeding

on bats in Venezuela (Deane et al., 1978). The specific

feeding habits of Lu. c. hispaniolae were not known prior to

this study.

Laboratory rearing of sand flies is a necessary part of

studying disease transmission, because it is an assured

method of obtaining uninfected flies. Colonization can also

lead to a better understanding of the life cycle of the

vector and parasites (Killick-Kendrick, 1978). Successful

rearing has been accomplished using various techniques and

several different formulations of larval food (Chaniotis,

1967; Endris et al., 1982; Gemetchu, 1976; Young et al.,


In the New World, leishmaniasis is a zoonotic disease,

with rodents serving as reservoir hosts for the L. mexicana

complex, canids for L. donovani, a variety of rodents,

procyonids, sloths, dogs, and primates for the L. brazilien-

sis complex (Table 1-2) (Lainson and Shaw, 1978). In the

Dominican Republic, ten species of rodents, two primates,

four to five insectivores, and four to six sloths are known

from fossils (Varona, 1974; Woods, pers. comm.). Some

survived until the time of Columbus (1492 AD) but only

two endemic terrestrial species exist today: an insecti-

vore, Solenodon paradoxus Brandt; and a capromyid rodent,

Table 1-2. Other Mammalian hosts of Leishmania strains
which cause human leishmaniasis in the New
World, by country.

Leishman a Nonhuman
Country Strain Mammalian Hosts





rodents: Heteromys, Nyctomys,

Ototylomys, Sigmodon(R)2

dog(R), foxes: Cerdocyon(R),


(rodents: Akodon, Oryzomys,


sloth: Choelopus; anteater(R),


rodents: Proechimys(R),


Costa Rica

French Guiana








(Dasyprocta, Heteromys, Neacomys,

Nectomys, Oryzomys, oppossums,

Marmosa, Caluromys, Metachirus?)


sloths: Bradypus(R), Choelopus

sloths: Choelopus


rodents: Heteromys, Nyctomys,

Ototylomys, Sigmodon



Table 1-2. Continued

Leishmania Nonhuman
Country Strain Mammalian Hosts

Panama L.b.p. sloths: Bradypus(R), Choelopus;

(primates: Aotus, Sanguinus;

procyonids: Bassaricyon, Nasua,


Peru L.R. (dog?)

Venezuela L.m. (rodents: Heteromys, Proechimys,


Source: World Health Organization, 1984, pp. 56-61.

Leishmania strains: L.m.m. = L. mexicana mexicana,
L.d.c. = L. donovani chagasi, L.b.b = L. braziliensis
braziliensis, L.b.g. = L.b. guyanensis, L.b.p. = L.b.
panamensis, L.p. = L. peruviana.

R = Proven reservoir.

(Mammal?)--animal found infected in nature, extent of
infection not determined.

Plagiodontia aedium Cuvier, both of which are extremely

uncommon. Both animals are secretive and are found in

relatively undisturbed habitat (Woods, 1981). Introduced

mammals have replaced the endemic mammals in most areas.

Three species of Old World rodents now occur throughout the

island in cities, tropical rain forests, and deserts. These

are Rattus rattus alexandrinus (Geoffroy), the roof or black

rat; R. norvegicus (Berkenhout), the Norway or brown rat;

and Mus musculus brevirostris Waterhouse, the house mouse.

Geography of the Dominican Republic

The Dominican Republic occupies the eastern two-thirds

of the island of Hispaniola, with Haiti occupying the

western side. The island is a member of the Greater

Antilles, lying between latitudes 17030' to 2000' with the

Atlantic Ocean to the north and the Caribbean Sea to the

south (Fig. 1-1). The Republic has an area of 43,230km2, of

which roughly two-thirds is mountainous. Geographically the

country is broken up into four regions, based, in part, on

the mountain ranges (sierras and cordilleras). The Eastern

region includes the Cordillera Oriental with the Llano

Oriental (Eastern Plains) to the south. Directly west lies

the Cibao with the Cordillera Septentrional to the north,

just inland from the coast, the Cibao Valley, and the

Cordillera Central to the west and south. To the northwest

lies the Linea Noroeste, bordering Haiti. El Sur is the

region to the south west of the Cordillera Central; it also

borders Haiti, and contains two smaller mountain chains

Sierra de Neiba to the north of Sierra de Baoruco

(Fig. 1-2). Most of the 6.2 million inhabitants (1982

census) live on the valley floors that separate the

cordilleras, or on the rolling Eastern Plains. About 60% of

the Dominicans live in rural and agricultural areas. The

majority are small land owners. Much of the land has been

cleared for agriculture. The main agricultural products

include sugar cane, rice, coffee, cacao, cotton, tobacco,

corn, and beef. The country has a tropical maritime

climate; in the lower elevations, temperature average 220 to

280C with a range during July 1981 to August 1982 at Pedro

Sanchez in El Seibo Province, of 160 to 330C. Annual

rainfall averages 1397 to 1524mm with a range of 508 to

2413mm, depending on location. The highest rainfall usually

occurs in the eastern portion where the rainy season lasts

from May to November (Anonymous, 1977; Sholdt and Manning,


Current Status of Diffuse Cutaneous Leishmaniasis
in the Dominican Republic

Since the discovery of the first three cases of DCL in

the Dominican Republic in 1974 (Bogaert-Diaz et al., 1975),

an additional 22 cases have been diagnosed (Bogaert-Diaz,

unpublished data). Precise life histories are known only

for a few of the patients who have not moved to different

localities during the presumed evolution of the disease.

Personal data, when recorded, often did not denote the exact

Figure 1-1. The Caribbean region, showing the location of the Dominican Republic, on the
island of Hispaniola.

Atlantic Ocean


0* D

Figure 1-2. The geographic regions of the Dominican Republic
(Cordilleras and Sierras are mountain ranges).

location of residence. Many patients had moved from the

country to towns before being diagnosed. Others have lived

in several different areas for various periods of time;

thus, current place of residence may not have been where the

disease was contracted. Presumed site visitations and

interviews led to the confirmation of the probable locality

of infection for 15 patients. Brief descriptions of these

sites are given below and are shown as confirmed sites in

Figure 1-3; the remaining sites are shown as presumed. The

unknown incubation period and the relative benignancy of the

infection make it difficult to determine the probably date

of infection. Eight of the patients had had nodules or

plaques for four or more years before the disease was

diagnosed. At least 13 of the patients were under

four years of age when signs of the infection first

appeared, so epidemiological data must be based upon recol-

lections of parents or other relatives. The most out-

standing feature of the epidemiology of the disease is its

positive correlation to the Cordillera Oriental, as

19 patients (76%) resided in or near this region of the

country (Fig. 1-4). Most of the patients have been symptom-

atically cured of the disease with hot (46C) water treat-

ment (Neva, pers. comm.) (Fig. 1-5).

During the period March to September 1982, a serolog-

ical survey for leishmaniasis was undertaken by personnel of

the Instituto Dermatologico, in which the author assisted.

In the survey, blood samples were taken from residents




0- Presumed site 700 690

-- Confirmed site

0 51
, i a

L mkm
Figure 1-3. Case sites for patients with diffuse cutaneous leishmaniasis (DCL) in
the Dominican Republic (see Appendix 1-1 for name of numbered localities).


Figure 1-4.

Typical terrain in the Cordillera Oriental,
Dominican Republic.

Figure 1-5. A young Dominican DCL patient with healed
lesions on wrist and upper arm.

living in the area of 7 of the 21 case sites (Fig. 1-3;

cases 1-3, 7, 10, 17, 19 and 20, 21, and 22) and two control

sites (sites in regions where no cases of leishmaniasis had

been diagnosed). Site selection was based on accessibility

of the site to vehicular transport and on the probability of

obtaining an adequate number of volunteers from the local

residents. Indirect fluorescent antibody test (IFAT) was

performed on the sera, revealing that 26.0-48.0% of the

samples from case sites were seropositive for leishmanial

antibody, at 1:8 or higher titer. At the two control

sites 0% and 14% were determined to be seropositive (de

Quinones, unpublished data). The pertinent patient history

data are given in Appendix 1. Cases 1-3 are siblings, cases

14 and 15 are son and father, and cases 19 and 20 are

neighbors, but unrelated.

Study Site Selection

Four principal study sites were visited regularly

during the study. One, Pedro Sanchez, was not a leish-

maniasis case site, but was used as a control site, i.e. a

site with sand fly habitat but no known cases of leishman-

iasis. Pedro Sanchez was selected as a typical nonagricul-

tural wooded area. Due to extensive land clearing for

agricultural purposes, pasture, sugar cane planting, and

coffee/cacao groves, very little natural forest habitat

exists in the Dominican Republic. The Pedro Sanchez site

was representative of wooded river bank habitat in the

leishmaniasis endemic region of the Cordillera Oriental.

The field station (and author's residence) for the study was

located less than 0.5km distant, thus the site was con-

venient at all times of year. The other three principal

sites were leishmaniasis case sites, in all instances the

DCL patients were living at or near the sites. All

four patients (patients 19 and 20 were neighbors) were young

and had not lived elsewhere at the time the disease was

diagnosed. Interviews with the patients' parents led the

author to believe that the disease was contracted at the

sites. Very few changes had occurred at the sites in the

intervening years between the appearance of the first signs

of the disease and the beginning of the author's research,

this was not the case with several other presumed case


The system of roads in the Dominican Republic is

generally poor, particularly in mountainous regions and the

Eastern region, thus accessibility of the site was an

important criterion in study site selection. Certainty of

the site of disease contraction was another important

criterion. Many of the older DCL patients have lived in

various localities throughout their lives and do not

remember clearly where they were living when signs of the

disease first appeared, 20 years or more previous. Another

important factor was whether or not personnel of the

Institute Dermatologico were familiar with the location of

the case site. In a few cases, the patients had visited

rural public clinics where they had been diagnosed, their

place of residence at time of infection was not known except

in rather general terms. Eleven of the 21 DCL case sites

were never visited by the author, primarily due to the

factors mentioned above.

The three principal study case sites were approximately

35km west (Loma Pena Alta), 18km north (Morro de Miches),

and 13km east (Trepada de Jabilla) of the field station and

control study site at Pedro Sanchez. All four sites were

visited in excess of 100 man-hours during the study. The

remaining sites listed below were DCL case sites and were

visited one or more times, depending on proximity and

accessibility. Due to the concentration of cases in the

Cordillera Oriental, this area received the greatest amount

of attention by the author.

Study Site Description

Pedro Sanchez, El Seibo Province

Altitude: 76m

This site consists of a gallery forest along the banks

of the Rio Seibo on the southern outskirts of the village of

Pedro Sanchez. The Rio Seibo is 0.5 to 1.5m deep, depending

on the exact location and time of year, and 4 to 6m wide in

this region. The wooded border on the northwest bank was

20 to 50m wide and extended for several kilometers. It con-

tained several trees with diameter up to im, a moderately

thick shrub understory. A section of forest, with a length

of 40m, was used for this study, through August 1982; when

the site was revisited in May 1983, it had been extensively

altered for agricultural use.

The remaining study sites, listed alphabetically, all

are DCL case sites where a patient is living, or was living

at the presumed time of infection.

Altos de Peguero (5km E by 5km N of El Seibo), El Seibo

Province (Fig. 1-3, #17)

Altitude: 72m

The case site is on the northern side of an isolated

ridge on the southern edge of the Cordillera Oriental.

Several coffee and cacao groves are present in the area, the

largest of which is a 2ha cacao grove about 50m from the

former residence of the DCL patient. Coffee and cacao

groves also contain scattered larger hardwood trees to

provide light shade. There are no major streams or rivers

in the immediate vicinity, the closest being 1.5km distant.

Carrasco (10km S of Rio San Juan), Maria Trinidad Sanchez

Province (Fig. 1-3, #23)

Altitude: 15m

The area surrounding the patient's former residence is

flat pasture for at lease 2km, with the exception of a 1.5ha

cacao grove, 400m distant from the residence. A small

stream runs next to the grove and forms a small pool (7m

diameter) where the children of the area are said to swim.

La Culatica (10km S of Nisibon), Altagracia Province

(Fig. 1-3) #1-3)

Altitude: 150m

The patients' (three siblings) residence was located on

a small hill above a small (0.5ha) coffee grove. Other

coffee and cacao groves are in proximity to the house site,

some of which are owned by the patients' father. A small

wooded stream runs below the site, through parts of the

coffee grove. This locality is situated in the northeast

portion of the Cordillera Oriental.

Iguana Arriba (23km NE of Bani), Peravia Province (Fig. 1-3,


Altitude: 152m

The exact locality of the DCL patient's residence was

not known, but the area consists of extensive coffee plant-

ings throughout low mountain terrain. A few small streams

are present in the low areas between ridges. Larger trees

are found along the streams and in the coffee groves,

providing shade.

Loma Pena Alta (13km NW of Hato Mayor), El Seibo Province

(Fig. 1-3, #21)

Atltitude: 442m

The DCL patient lived at the top of the ridge, adjacent

to a 0.5ha coffee grove. The eastern side of the ridge was

shrub-overgrown pasture. Some coffee groves were present on

the western side, which was mostly lightly forested with

numerous rocky outcroppings.

Monte Claro (16km NE of Cotui), Sanchez Ramirez Province

(Fig. 1-3, #24)

Altitude: 76m

A 0.5ha coffee grove is 10m distance from the site

where the DCL patient lived at the time of infection. The

area consists of rolling hill pasture, with the coffee grove

being the only wooded area in a radius of 1km from the

house. The Rio Chacuey has a lightly wooded border, and is

about 1km distant from the site.

Morro de Miches (10km S of Miches), El Seibo Province

(Fig. 1-3, #19, 20; Fig. 1-4)

Altitude: 305m

Two unrelated patients live at this site, approximately

100m apart and separated by the peak of the ridge.

One patient lived besided a 0.25ha coffee grove which

contained several large shade trees. The grove is separated

from nearby wooded areas by pasture and cultivated plots. A

small stream runs down the ridge, 100m from the house site.

The other patient's residence was located on the northern

edge of a rather extensive mixed planting of coffee and

cacao which had a length of about 0.5km and a width of


Najayo Arriba (20km NW of San Cristobal), San Cristobal

Province (Fig. 1-3, #5)

Altitude: 400m

The patient's residence is located on the side of a

ridge immediately above a small (0.5ha) coffee grove. A

small stream runs along the bottom of the ridge, about 75m

from the house. The site is located on the southern edge of

the Cordillera Central.

Rio Llano (10km W by 30km N of Higuey), Altagracia Province

(Fig. 1-3, #14 and 15)

Altitude: 250km

The patients (father and son) lived in a house 15m from

a small stream. A small (0.5ha) coffee grove is situated on

the opposite side of the stream. Rio Guancho is less than

0.5km distant. A gallery forest, 10-30m wide, runs along

the banks of the river. This site is in the eastern portion

of the Cordillera Oriental, about 10-15km (straight line

distance) south of La Culatica.

Trepada de Jabilla (2.8km N of Las Cuchillas), El Seibo

Province (Fig. 1-3, #7)

Altitude: 130m

The patient's residence was approximately 10m from the

edge of a rather extensive coffee grove ( 3ha). As with

other coffee groves, there are some larger shade trees

present. The terrain is rolling hills, primarily pasture,

south of the Cordillera Oriental. The Rio Soco borders on

side of the coffee grove and is 0.5 to 2.0m deep, depending

on location and season, and generally 10 to 12m wide.

Several other coffee groves exist in the area.


In 1975, Bogaert-Diaz et al. reported the discovery of

three human cases of diffuse cutaneous leishmaniasis (DCL)

in the Dominican Republic, the first autochthonous cases of

cutaneous leishmaniasis known in the country or in the

entire West Indies, except Trinidad, a continental island.

Over the succeeding four years, a concentrated search for

additional cases, carried out under the National Leprosy

Program, revealed 15 more cases of DCL. As of March 1984,

25 cases of DCL have been diagnosed. Surprisingly, none of

the patients had ulcerating lesions. Interest in this

unique situation led to grant support from the World Health

Organization to the Instituto Dermatologico Dominicano for

epidemiological studies of DCL in the Dominican Republic,

beginning in 1981. Some objectives of the author's

research, listed below, were included in the project. Field

studies were performed in the Dominican Republic during the

periods 19-24 May 1981, 27 July 1981 to 2 August 1982, and

18 May to 3 August 1983. Laboratory studies were performed

at a field station (Pedro Sanchez) and at the Instituto

Dermatologico in Santo Domingo, the University of Florida,

and Walter Reed Army Institute of Research.

A. Field Studies:

1. To survey for the presence of phleboto-

mine sand flies at selected locations in

the Dominican Republic.

2. To study the ecology of sand flies

including host preference, resting

sites, population dynamics, and seasonal

and geographic distribution for species

located in the survey.

3. To identify the wild and/or peridomestic

reservoir host(s) of leishmaniasis.

B. Laboratory Studies:

1. To establish laboratory colonies of

indigenous phlebotomine sand flies.

2. To study the life cycle of Leishmania in

the sand fly.

3. To determine the vector potential of

these flies.

4. To further elucidate some of the biolog-

ical differences between the Dominican

Leishmania and other known Leishmania





Phlebotomine sand flies, the only known biological

vectors of leishmaniasis, also transmit other parasitic

agents of man and other vertebrates (Adler and Theodor,

1957). A new focus of leishmaniasis was discovered in 1974

in the Dominican Republic (Bogaert-Diaz et al., 1975), the

eastern portion of the island of Hispaniola. Only two extant

species of sand flies are known from Hispaniola, Lutzomyia

cayennensis hispaniolae (Fairchild and Trapido, 1950) and

Lu. christophei (Fairchild and Trapido, 1950). Both species

were collected from tree trunks and buttresses, primarily,

but little was known about the habits and range of these

species. Lutzomyia cayennensis hispaniolae is an endemic

subspecies of Lu. cayennensis (Floch and Abonnenc) that

ranges from Ecuador and French Guiana north to Mexico. Most

members of the Lu. cayennensis species group are reported as

reptile feeders (Young, 1979). Lutzomyia christophei is a

member of the Lu. verrucarum species group that contains an

number of man-biting species (Lewis, 1968) including one

suspected vector of leishmaniasis (Lainson and Shaw, 1979).

This study was conducted to determine the geographic

range, the habits, and life cycle of the two species.

Methods and Materials

Field Studies

Chaniotis (1978) and Young (1979) reviewed techniques

used for sampling phlebotomine sand flies. The methods used

in the present study were aspirator collections at resting

sites, man-biting collections, sticky paper traps, flight

traps, CDC light traps (Sudia and Chamberlain, 1962), and

Disney traps (Disney, 1966). Specific equipment preparation

and field techniques for aspirator collections were given by

Endris et al. (1982) (Fig. 2-1). Aspirator collections were

made at various sites around the country, mostly from tree

trunks, but also from tree holes and rock crevices.

Cigarette smoke was occasionally used to disturb resting

sand flies from the latter two types of resting sites. Live

wild-caught sand flies were transported to the field station

at Pedro Sanchez where they were either held in a feeding

cage (Fig. 2-2) for host preference studies or kept

individually for oviposition and later dissected for


The other techniques were used at five leishmaniasis

case sites or the study site at Pedro Sanchez (see Chap-

ter 1). Attempts to collect sand flies from human bait at






1.^' '.-





Figure 2-1.

Field collecting equipment and rearing
containers for phlebotomine sand flies (from
Endris et al., 1982). (A) Collecting con-
tainers. (B) Oviposition/larval rearing
vial. (C) Aspirator.



Figure 2-2.

Sandy fly feeding cage--a modified aquarium with
plaster of paris bottom and back (from Endris et al.,

Trepada de Jabilla were made various times during the day

and at dusk during November 1981 to July 1982 and May 1983.

Similar attempts were also made at Loma Pena Alta at dusk on

various days in June and July 1983. A sticky-paper trap of

double-sided tape on a 25cm square wooden frame was used at

Pedro Sanchez from 21-23 May 1981. The trap was placed in

the crotch of a tree (Im high) known to harbor resting sand

flies. Flight traps (Fig. 2-3) and CDC light traps

(Fig. 2-4) were set at various sites (Tables 2-1, 2-2). Two

modified CDC light traps were also employed. In one, a UV

light source was substituted for the normal light source;

the second type had no light source, instead a cage

containing a hamster, Mesocricetus auretus, was suspended

above the trap (Fig. 2-4). The light traps were run on

nights when the moon was less than one-half full, or when

the sky was mostly overcast. CDC traps were turned on in

late afternoon and taken down the following morning. Flight

traps were set up in coffee or cacao groves, with one

exception (lightly wooded stream bank at Hato Mayor). The

flight traps were emptied every two to four days. The trap

collections were checked for sand flies with the aid of a

dissecting microscope (7-30X). Disney traps (Fig. 2-5) were

used at two sites with hamsters serving as bait (Table 2-3).

The bottom of cake pans (22.5 x 30.0cm) or cookie sheets

(27.0 x 38.0cm) were covered with a thin layer of mineral

oil or castor oil. The condition of the hamsters was

K' m

Figure 2-3.

The author in front of a flight trap in a cacao
grove at Altos de Peguero, El Seibo Prov.

Figure 2-4. A CDC trap and modified traps (from left CDC
with UV light source, normal CDC, and hamster-
baited trap).

Table 2-1.

Location and dates for flight trap samples in
the Dominican Republic.



Altos de Peguero cacao grove

Carrasco cacao grove

Hato Mayor wooded stream bank

Loma Pena Alta coffee grove



Claro coffee grove

de Miches coffee grove

Pedro Sanchez wooded river bank

Trepada de Jabilla coffee grove

9-25 Mar, 1-28 May 1982

3-10 Jul 1982

17 Aug 1981

24-27 Jul 1982
9 Jun-3 Aug 1983

19-26 June 1982

7-14 Aug 1981
5 Oct-3 Nov 1981

20-22 May 1981

6 Nov-14 Dec 1981
18 Jan-26 Jul 1981
23 May-18 Jun 1983

Table 2-2.

Location, trap type, and dates for CDC in the
Dominican Republic.

1 Date2
Location # And Trap Type Date

Altos de Peguero


La Culatica

Loma Pena Alta

Monte Claro

Morro de Miches
















2-3 Apr 1982
27-28 May 1982

9-10 Jul 1982

7-8 Jun 1983

26-27 Jul 1982

2-3 Jun 1983

23-24 Jun 1983

9-10 Jul 1983

17-18 Jul 1983

23-24 Jul 1983

30-31 Jul 1983

25-26 Jul 1982

21-22 May 1983

22-23 Sep 1981

18-19 Jul 1982

29-30 May 1983

Table 2-2. Continued

Location # And Trap Typel Date2

Pedro Sanchez 2 CDC 20-21 May 1981

1 CDC 7-8 Aug 1981

2 CDC 5-6 Jun 1982

Trepada de Jabilla 2 CDC 18-19 Jan 1982

1 CDC 16-17 Apr 1982
21-22 May 1982
30-31 May 1982
9-10 Jun 1982
30 Jun-
1 Jul 1982
14-15 Jul 1982

2 CDC 26-27 May 1983

1 CDC 30-31 May 1983

1 CDC 18-19, 19-20
Jul 1983

CDC-UV with blacklight
trap, no light (see Fig.

source, CDC-HB -

hamster baited

2From approximately 1830 hrs Day 1 1000 hrs Day 2

Figure 2-5. A Disney trap, used for collecting rodent-
feeding sand flies.

Table 2-3. The sites and dates for Disney traps used for
phlebotomine sand fly sampling in the Dominican

Site # traps Date

Loma Pena Alta 2 8 July 3 August 1983

Trepada de Jabilla 2 28 May 15 June 1982

4 23 May 2 June 1983

2 7 9 June 1983

checked twice a day by local personnel and food and water

were provided at all times. Hamsters were exposed for three

to four days and then replaced by others.

Soil samples (dirt, humus, and leaf litter) were

collected from tree buttresses and tree holes at five sites

where sand flies were common (Table 2-4). Samples

(1-4 liters in volume) were placed in plastic bags for

transport back to the field station. For observation, a

sample was transferred to a white plastic tray, and the

material was examined with the aid of a dissecting micro-

scope (7-30X). Larger leaf matter was checked for the

presence of larvae or pupae and then discarded. Recovered

larvae were held in larval rearing vials (Fig. 2-1b) with a

small amount of larval food (Young et al., 1981), and held

until adult emergence for species identification. The rest

of the sample was returned to the plastic bag and periodi-

cally checked for emerged adults or immature stages. The

bags were held at ambient temperature (19-30C) for up to

two months.

Laboratory Rearing

Techniques used for laboratory rearing of Dominican

sand flies (Fig. 2-6) were those described by Endris et al.

(1982). Individuals from an egg batch were raised together

or individually, as described by Perkins (1982), in a

modified tissue culture plate (Fig. 2-7) Newly hatched

(within 6 hrs) first instar larvae were transferred from

Table 2-4.

Collection sites and dates for soil samples
examined for the presence of sand fly larvae.

Site Date # Samples Microhabitat

Altos de Peguero

leaf matter
at tree base-
cacao grove

Loma Pena Alta

Monte Claro

Pedro Sanchez

2 Jul 1982

20 Jul 1982

25 Jul 1982

15 Aug 1981

as above

leaves and
humus at tree

as above

leaves and
humus at tree

as above

as above

leaves and
humus at tree

27 Aug 1981

22 Sep 1981

30 Oct 1981

Trepada de Jabilla

25 May 1982

15 Jun 1982

2 Dec 1981

as above

as above

leaves and
humus at tree

leaves and
humus at tree
base and in
tree hole-
coffee grove

18 May 1982

8 Jun 1982 3

28 May 1982

as above

Figure 2-6.

Schematic diagram of laboratory techniques for rearing of phlebotomine
sand flies. (1) Plaster of Paris at bottom of rearing vial in moistened
with tap water. (2) Wild-caught or lab-reared gravid females are placed
individually in vials and drop of sugar solution is placed on each screen
lid. (3) A solid lid replaces the screen lid after eggs are deposited.
(4) Larval food is sprinkled on the plaster of Paris anytime before the
larvae hatch. (5) Additional larval food is added as larvae grow.
(6) Lidless vials containing pupae are placed in the feeding cage; fruit
slices provide a sugar source for emerging adults. (7) For a bloodmeal
source, a vertebrate is placed in the feeding cage (Endris et al., 1982).










Figure 2-7.

A modified microtiter plate for rearing
individual sand fly larvae (approximately
0.5cm layer of plaster of Paris in each well).

oviposition vials to wells in the tissue culture plate

(one larva per well). The larvae were checked at approxi-

mately the same time every day. Individual rearing was

performed under different temperature-humidity regimens for

both species of sand flies. Lutzomyia christophei larvae

were held in a Hotpack chamber (Hotpack, Inc., Philadelphia,

PA) at 23C or 28C. High humidity was maintained by

placing the plates with moistened plaster of Paris in

tightly sealed plastic boxes lined with moist paper towels.

Lutzomyia cayennensis were reared under three sets of

conditions. The first group was reared in a Hotpack chamber

at 28C, as above. The second and third groups were held at

ambient conditions in the Dominican Republic of 24 to 330C

and 60 to 95%RH (August-September), and 16 to 280C with 30

to 90%RH (January-February), respectively.

Eclosion was checked in the vials or plates once per

day. Adults were released into the feeding chamber. At the

field station, various types of locally available fruit were

provided as a sugar source. These included grapefruit,

orange, or sweet lemon sections (skin removed); cashew

fruit; and peeled mango skin. Lutzomyia christophei reared

at Walter Reed Army Institute of Research (WRAIR) were

provided with apple slices. For Lu. christophei females, an

anesthetized Rattus rattus, Syrian hamster (Mesocricetus

auretus), or BALB/c mouse (Mus musculus) was provided as a

blood source. An Anolis sp. lizard was regularly provided

as a blood source for female Lu. cayennensis; occasionally a

human hand or hamster was offered. The lizard was either

restrained in a small cylinder of hardware cloth or was

allowed to be free. If free, the mouth was taped shut to

prevent the ingestion of sand flies.

To determine the age at first feeding for either

species, all flies emerging in a six hour period were held

separate by species. For Lu. cayennensis, an Anolis lizard

was provided until all females were fed or dead. If neces-

sary, the lizard was exchanged every two to three days for a

similar sized, conspecific lizard. The lizards were not

restrained. For Lu. christophei females, an anesthetized

hamster was provided for one hour two times per day, approx-

imately 0903 hrs and 1530 hrs.

Sand Fly Dissections

A sample of both male and female sand flies from

various survey sites was routinely dissected for species

determination, using the methods given by Young (1979), or

by simply cutting off the head and the last few abdominal

segments and mounting them in a drop of Hoyer's mounting

medium (Young, pers. comm.). In addition, wild-caught and

lab-fed sand flies were dissected upon death to examine the

digestive tract for parasites. The dissections were done in

normal saline or in Medium 199 (GIBCO, Grand Island, NY) on

a microscope slide and observed with the aid of a compound

microscope (200X or 450x). Permanent slides were made by

removing the cover slip and allowing the liquid to dry, then

fixing with absolute methanol and staining with Giemsa for

20 minutes. After drying, a drop of Euparal mounting medium

was added and a cover slip placed over it.


Field Studies

Adult Lu. cayennensis were aspirator-collected from

tree trunks (10 cm diameter and larger) at many localities

in the Dominican Republic (Fig. 2-8, Appendix 2). At all

sites, there was sufficient tree and shrub vegetation to

provide a "forest-floor" type habitat, with little ground

cover vegetation. The sites were shaded, to varying

degrees. Many sites in El Seibo Province were visited more

than once. This species was also recovered from flight trap

samples at three sites (Appendix 2). None was recovered by

any other method. Adult Lu. christophei were recovered at

seven sites by flight trap, CDC trap, or aspirator

collection from rock and tree crevices, primarily of ceiba

trees (Ceiba pentandra) (Fig. 2-8, Appendix 2). The use of

cigarette smoke facilitated their capture from deeper

crevices. Six of these sites were leishmaniasis case sites;

the seventh was about 5km distance from a case site. No Lu.

christophei were collected by hamster-baited CDC traps,

Disney traps, or man-biting collections. At Loma Pena Alta,

both Disney traps were within lm of resting sites of Lu.

christophei females. The man-biting collections attempted

Figure 2-8. Collection sites for Lu. cayennensis hispaniolae and Lu. christophei
in the Dominican Repuglic from May 1981 August 1983 (See Appendix 2-1,

t- no Lutzomyia recovered

* Lu. cayennensis only

* Lu. cayennensis and
Lu. christnphei

at this site were performed within 2 to 3m of tree crevices

known to harbor resting sand flies. Two female sand flies

were recovered while biting DCL patient #25, in September

1983; these were later identified as Lu. christophei, by the


Populations of Lu. cayennensis at Pedro Sanchez varied

throughout the year and the population at Trepada de Jabilla

followed the same trends from January through July 1982

(Fig. 2-9). The population sample was based on the total

number of flies aspirator-collected from 10 marked trees.

The sample from Pedro Sanchez was always larger than the

same week's sample from Trepada. At Pedro Sanchez, the

samples varied from a high of 67 flies to a low of 2 flies;

the total for the 39 samples was 986 flies (536 males and

449 females). At Trepada, the high was 39 flies and the low

was 0 flies; the total for the 22 samples was 269 flies

(156 males and 113 females). Fewer flies were collected

during the period February to mid-May 1982, which corres-

ponds to the dry season and the first two weeks of the rainy

season. The population level began to rise approximately

two and one-half weeks after the beginning of the rainy

season (3 May 1982). The male:female ratio, on dates

when females were collected, ranged from 0.90 to 3.00

(x = 1.90) males/female (8 out of 22 samples had no

females). Blood-fed or gravid females comprised 0 to 100%

(x = 31.3%) of the females at Pedro Sanchez and 0 to 75%

--------- Trepada de Jabilla
Pedro Sanchez

I a
\ ",


So Oc

De Ja Fe Mr

Je Jy

Figure 2-9.

Weekly sample populations of Lu. cayennensis at two study sites in the
Dominican Republic (based on tree trunk resting collections).


2 50
t 40

8 30



(x = 48.9%) at Trepada. Due to the scarcity of Lu. christ-

ophei adults, seasonal abundance could not be monitored for

this species.

During the wettest months (May to December), Lu.

cayennensis adults could be found on tree trunks up to a

height of 2m. Normally when disturbed, the sand flies would

fly only a short distance (less than 10cm); however, on

rain-wetted tree trunks the sand flies readily flew off the

trees. During the dry season, the flies were generally

found at the tree base, often where the ground had separated

from the tree trunk in fissures, or in the moss at the base.

Smaller tree holes also harbored Lu. cayennensis at this

time of year. Female Lu. cayennensis were commonly observed

feeding on Anolis lizards during the day, principally on

Anolis distichus, but also on A. cybotes. Lizard identifi-

cation was based on Cochran (1941).

Adult Lu. christophei were usually recovered from deep

tree crevices and ground level tree holes in large shade

trees in coffee groves. Two types of trees provide crevices

which could serve as resting sites for Lu. christophei, the

ceiba and the strangler fig (Ficus spp.) The former may

have very large buttresses and the latter usually has

numerous crevices of various sizes as it entwines its host

tree. Cigarette smoke puffed into the crevices forced the

sand flies to the entrance, where they were more easily

captured. Most of the Lu. christophei collected were

obtained at Loma Pena Alta, where a total of 81 flies

(51 males and 22 females) were captured by aspirator,

7 flies (2 males and 5 females) were captured in a flight

trap, and 23 flies (10 males and 13 females) were captured

in CDC or CDC-UV traps. In one tree hole, 14 sand flies

were found in association with a rat nest constructed of

leaves. Of eight females, three were blood-fed or gravid

and were the only fed Lu. christophei females recovered in

aspirator collections. A common characteristic of the other

resting sites was the presence of land snails and millipede

feces. On rare occasions at Loma Pena Alta, male Lu.

christophei were found resting on tree buttresses near the

entrance to tree crevices, in association with Lu.

cayennensis. Lutzomyia christophei was the more active of

the two species.

Five fourth-instar larvae were recovered from a soil

sample at Monte Claro. The microhabitat was humus and leaf

litter which had accumulated in the buttress of a large

shade tree in a coffee grove. The larvae were reared to the

adult stage and were Lu. cayennensis. No other phlebotomine

larvae were seen or recovered from the other soil samples

(Table 2-3).

Laboratory Studies

Wild-caught and lab-reared Lu. cayennensis adults

behaved similarly. Females fed readily on Anolis lizards in

the feeding chamber, but showed no interest in feeding on

human, hamster, or rat (R. rattus). Over 50 wild-caught

females were exposed to a skink, Mabuya mabouya, in a

feeding chamber, but none fed or was seen probing. Twenty

lab-reared females were exposed to tree frogs, Hyla sp., and

toads, Bufo marinus, but the flies showed no interest in

feeding. The preferred feeding site on lizards was on the

mid-dorsal region to the tail base. Some females were

occasionally observed feeding on the top of the head and the

shoulder region. Females rarely probed more than one spot

before feeding to repletion. Feeding time ranged from

62.5min to 82.5min (n = 26, x = 73.Omin6.25min (S.D.)).

Feeding only occurred under lighted conditions. Females

would feed at any time of day in the feeding chamber,

provided that the chamber was left in a well lighted room.

Generally, females first fed at age 3.0 to 4.5 days (n = 50)

posteclosion; however, on various occasions lab-reared

flies, less than 48hrs old, fed on lizards. Mating was

observed before, during, or after bloodfeeding, with the

pair remaining in copula for up to 21min. Females began

oviposition four to five days post-bloodfeeding. Complete

oviposition usually required less than one day. Wild-caught

females generally died within 24hrs after laying eggs.

Holding the flies in a Hotpack environmental chamber helped

to increase the survivorship of lab-reared females.

Approximately 17 to 33% of the females of every generation

survived to take a second blood meal; 50 to 75% of these

survivors subsequently laid a second batch of eggs. No

males or unfed females lived more than six days.

Female Lu. cayennensis were anautogenous, each female

laying up to 60 eggs. Most unmated bloodfed females died

without ovipositing (20 out of 23 flies), but three laid

partial egg batches of 5 to 9 eggs. The size of a full egg

batch for wild-caught females was 38 to 60 eggs (n = 50,

x = 47.76.6 (S.D.)), based on females collected with a full

blood meal that had no eggs retained at death. For lab-

reared females, a full egg batch contained 39 to 60 eggs

(n = 50, x = 46.66.3 (S.D.)). Of 103 F1 females held

singly in vials, only 19 laid full egg batches, 61 females

laid partial egg batches of 7 to 29 eggs, and 23 females

died without laying eggs. The percent egg hatch, for full

egg batches from wild-caught and lab-reared females ranged

from 20.4-100% (n = 100, x = 68.4%29.7% (S.D.)). Percent

egg hatch for partial egg batches ranged from 0 to 100% (n =

50, x = 61.2%33.2 (S.D.)). Percent egg hatch was not

statistically different between the two groups (t-test, p =

0.05). When three or fewer eggs were laid by a female, none


All hatching from an egg batch occurred within a 12hr

period. Under ambient conditions at the field station, the

incubation time for eggs was 8 to 12 days, depending on time

of year (ambient temperature). Longer incubation times were

associated with cooler temperatures, especially in January

and February (Table 2-5).

Larvae of Lu. cayennensis exhibited burrowing behavior.

Larvae tended to stay under their food, except when the food

was very damp. When individually reared, male sand fly

larvae developed faster than did females, under all three

temperature-humidity regimes (t-test, p = 0.05) (Table 2-5).

Males developed faster under ambient conditions in August-

September (24-33C, 60-95%RH) than males under constant

conditions (270C, 85%RH). Both groups developed faster than

those under ambient conditions in January-February in the

Dominican Republic (16-280C, 30-90%RH) (t-test, p = 0.05).

The time from oviposition to adult eclosion, for males and

females, is presented in Table 2-5 and Figures 2-10, 2-11,


A closed colony of Lu. cayennensis was maintained for

six generations, or for almost one year.

Wild-caught and lab-reared Lu. christophei behaved

similarly in the laboratory. Females readily fed on

anesthetized rodents, including a wild-caught R. rattus,

laboratory mice and hamsters. They readily probed on a

human hand. The females showed no feeding site preference,

feeding equally well on the ears, paws, or eyelids of the

offered rodent host. Females often probed more than one

spot and after initiating feeding, many females moved to a

second site to complete feeding. Females of this species

fed equally well under light or dark conditions. Feeding

time ranged from 2.75min to 4.95min (n = 25, x = 3.35min

0.50min (S.D.)). Some lab-reared females fed as early as

24hrs after emergence, though most did not feed until 48 to

72hrs after eclosion. Females took a very large bloodmeal

Table 2-5. Mean duration ( S.D.) in days of immature stages of Lu. cayennensis
hispaniolae at three temperature regimes, according to sex of sand fly.

Larval Instars
Condition Sex Egg 1 2 3 4 Pupa (egg-adult) n

Constant male 8 5.11.0 3.60.8 3.00.4 6.5+1.6 7.10.3 33.71.9 12
female 8 5.01.0 3.20.8 4.41.3 6.5+2.3 7.50.8 36.12.5 11

(48) (31) (28) (28) (25) (23)

Ambient male 8 4.70.5 2.30.5 2.70.5 6.0+0.7 6.70.6 30.3+1.3 23
female 8 5.00.4 2.40.5 3.10.3 6.4+0.7 7.70.7 32.51.0 24

(48) (48) (48) (48) (48) (47)

Ambient male 12 5.00.0 3.60.6 3.10.7 8.30.6 9.1+0.3 43.6+1.6 17
female 12 5.41.0 3.3+0.8 3.60.7 8.90.8 9.40.5 45.51.6 24

(48) (47) (46) (46) (42) (41)

Note: Number in ( ) indicates individuals surviving each stage.

X, 343+1 .7


i ~

x =37



0 31 32 33 34 35 36 37 38
Oays after ovioosit on

Figure 2-10.

Eclosion time in days after egg deposition for
Lu. cayennensis reared under constant condi-
tions (280C, 90%RH).

Days after oviposition

Figure 2-11.

Eclosion time in days after egg deposition for
Lu. cayennensis reared under ambient condi-
tions August-September (24-330C, 60-90%RH).

1 5


I .

39 40

m m






x = 45.5+1.6

9 x 43 6+1 .6
40 to 85%RH)6
J 3
1 I ],
40 41 42 43 44 45 46 47 48 49
Days after ovioosition

Figure 2-12. Eclosion time, in days, after egg deposition,
for Lu. cayennensis reared under ambient
conditions, January-February (16 to 280C,
40 to 85%RH).

(Fig. 2-13), but were very active afterwards. Mating, only

rarely observed, occurred before or after feeding, with

coupling lasting up to 28.5min. Oviposition began at least

five days after blood-feeding, but for a few females, it was

delayed up to ten days post-feeding; it was usually com-

pleted in less than one day. Approximately 17% of the

females that survived five or more days post-feeding (8 of

46 F1 and F2 females) did not exhibit ovarian development

after their first blood meal. By seven days however, they

had excreted the blood meal remnants in their feces and were

capable of refeeding and developing eggs. In the F3 genera-

tion, 8 of 52 females (15.8%) developed partial egg batches

of eight or fewer eggs before refeeding. All eight died

without ovipositing.

Only 2 of the 11 lab-fed wild-caught females laid full

egg batches of 39 and 49 eggs. The nine other flies laid 0

to 35 eggs (x = 7.311.3 (S.D.)). The size of a full egg

batch for lab-reared females was 35 to 87 eggs (n = 10, x =

50.712.9 (S.D.)). Of 64 Fl to F3 females which survived

five or more days post-feeding, only 14 (21.9%) laid full

egg batches, 24 females (37.5%) laid partial egg batches of

5 to 27 eggs, and 26 females (40.6%) died without ovipos-

iting. The number of eggs laid, plus the number retained at

death (a measure of reproductive potential in the above

64 females) ranged 36-88 eggs and/or ovarioles/female (x =

64.1%24.3% (S.D.)). The percent egg hatch for full egg

batches ranged from 12.6% (11/87 eggs) to 100% (36/36 eggs)


Figure 2-13. Female Lu. christophei feeding on BALB/c

(n = 12, x = 64.4%31.1% (S.D.)). Percent egg hatch for

partial egg batches ranged from 0% (0/18 eggs) to 100%

(28/28 eggs) (n = 20, x = 56.8%38.1% (S.D.)). Percent egg

hatch was not statistically different between the two groups

(t-test, p = 0.05). All hatching from an egg batch occurred

within a 24hr period. The time period for egg incubation

was 10 to 17 days, depending, in part on the temperature

maintained (Table 2-6).

Lutzomyia christophei larvae exhibited burrowing

behavior, prefering to remain under the food in the larval

rearing containers. In the individual and group rearing

experiments, development time from egg to adult ranged from

51 to 69 days for the F1 generation (27C) and from 57 to

73 days for the F2 and F3 generations (230C). Males devel-

oped at a faster rate than did females (t-test, p = 0.05).

The data from these rearing experiments are presented in

Table 2-6 and Figures 2-14 and 2-15.

The closed colony of Lu. christophei is being main-

tained at Walter Reed Army Institute of Research, Washing-

ton, D.C.

Sand Fly Dissections

Dissections of 319 wild-caught parous Lu. cayennensis

were made. The flies came principally from five sites, of

which 46.4% (148 flies) of the total were from leishmaniasis

case sites. Most females had a blood meal evident at

capture and did not die, or were not killed, until the blood


G: tr D co in
N (( 3-

4- m (N 0
4- .
4J r-4l
cr +1 +1 +1 +1
-H a r- In L.o r-
E-i U7 [- M (N in
1 CD 1.0 m a
0 U

OI *- o

H 14- 4
* I I 01
UI +1 +1 0 +I +1

C v f -1 x I Ir
Q : ca m m-

S4-- H I I
r0 n ( 1
rn Im I C

1 I I r-
S* I I *

C ( N I I *v C
EU M +1 +1 I I
*t (0 ui o oo *- *roj

-0 Nl I I O
S*m >

0 I I

I> I n

(a s 0 c N I I N m

C I I 04

a V C-
* 0) 4 4N (N o *
0 4 I I 0
4l+1 +1 < I I
+1 M in a% I s
a, i- I* I (U

S* *3 -
O 5 H- C 0
-H W +1 + -l

40 4 O Cd *N d ) C
102 2 ) o o 2 4

C> 1. 4-)'
'0 1:

I X r-- I z

44 4-4-Q

M 4-j c Z fa Z *

S-4 4-1 u d0 0( ( O)*

C: .C C co co NC N co 041
) 0 0 ON [ O 0O
E- ( U L' 2 z

H | x = 57.7+6.3
S57.6.3 x 63.5+6.5

51 53 5 57 659 63 5 7 69 71 73
Days after oviposition

Figure 2-14.

Eclosion time, in days after oviposition, for
F1 Lu. christophei reared under constant

Days after oviposition

Figure 2-15.

Eclosion time, in days after oviposition, for
F and F Lu. christophei reared under con-
stant coAditions.

had been digested and the residue excreted. The totals for

the sites, the percent blood-fed or gravid, and the time of

year collected are presented in Table 2-7.

Dissections were examined of 23 wild-caught Lu. christ-

ophei from Loma Pena Alta. Nine of these took bloodmeals in

the laboratory, two on a R. rattus captured at the site and

seven on a hamster. None of the 23 lab-fed females was

positive for parasites. The results of feeding lab-reared

Lu. christophei on leishmanial-infected BALB/c mice are

presented in Chapter 3.


Lutzomyia cayennensis hispaniolae has a widespread

geographic distribution in the Dominican Republic. It

occurs in wooded areas within 100m of the coast and in

coffee groves in the interior (Loma Pena Alta, elevation

427m). Throughout the sugarcane growing regions, Lu.

cayennensis was found along streams and rivers wherever

there was a sufficient number and density of trees to

support a "forest-floor" type habitat (i.e. leaf and other

organic matter present). This species was found at or near

eight of the DCL case sites visited during the survey. In

the wild, Lu. cayennensis appears to be very sedentary, as

very few were collected by the flight traps, often despite

close proximity of the trap to known resting sites. At

Monte Claro, where 13 flies were recovered from the flight

Table 2-7. Site of collection for female Lu. cayennensis dissected.

# Blood-Fed
Site or Gravid #Parous Total Time of year

Pedro Sanchez 93 34 127 May 1981,
Aug 1981-Jul 1982
Trepada de Jabilla 50 21 71 Nov 1981-Jul 1982

Altos de Peguero 25 9 34 Mar-Jul 1982

Loma Pena Alta 13 7 20 Jul 1982

Monte Claro 12 8 20 June 1982

Morro de Miches 1 2 3 Aug 1981

Miscellaneous 37 7 44 Aug 1981-Jul 1982

trap collection, the trap was set so that the side of one of

the end panels was in contact with a tree on which were

found over 70 resting sand flies. The absence of Lu.

cayennensis in light trap collections suggests that this

species was not attracted to light. Both wild-caught and

lab-reared flies exhibited a positive phototaxic response in

the feeding chamber, which was noted by changing the

position of a desk lamp used to illuminate the chamber.

Later, as sand flies were collected from small tree holes,

it became clear that this was a probable response to light

as an escape attempt, with the path towards the light

replacing the path out of a tree hole. In the lab and in

the wild, this species was only diurnally active, as were

its hosts, Anolis lizards. Because these lizards are very

common, there is no great need for extensive host-seeking


Although there may be some wind-borne dispersal of

adults, most of the distance an individual female travels

probably occurs during the 60+min that it is feeding on a

lizard. This length of time is much longer than that of

similar-sized mammalian feeding species, such as Lu.

anthophora (Endris, 1982) and Lu. christophei; however, the

lizards are not capable of reaching the sand flies to

dislodge them. Lutzomyia vexator (Coq.), also a lizard-

feeder that occurs in the USA and is about twice the size of

Lu. cayennensis, feeds to repletion in only 10min (Perkins,

unpublished data). Chaniotis (1967) reported 60+min feeding

times for some lizard-feeding sand flies in California.

The collection of larvae at Monte Claro represents the

first recovery of larvae of Lu. cayennensis in the field.

Hanson (1968) was unsuccessful in locating Lu. cayennensis

larvae in Panama. Monte Claro supported a very large

population of sand flies in a very small area. Many of the

trees sampled in the coffee grove had 50+ sand flies resting

on them; thus it does not seem surprising to find the

larvae in such a situation. Hanson (1968) noted that larvae

burrowed in culture, but to what depth they occur in nature

is unknown. In the current study, the soil samples were

jostled in transport back to the field station, so the depth

of the five recovered larvae was not known. The microsite

where they were collected had an unusually deep layer of

humus, 4 to 6cm deep. An injury to the tree trunk may have

been the cause of the ooze that started 0.5m above ground

level and continued to the ground, perhaps enriching the

soil at this spot. The soil samples taken from both Pedron

Sanchez and Trepada had very little humus.

The population levels of Lu. cayennensis are related to

wet and dry seasonal periods. Ambient temperature probably

influences the population level as well. In the eastern

region, the dry season usually lasts from late February to

the beginning of May. May is the wettest month, but from

June through December the ground remains fairly well satura-

ted. The population level of Lu. cayennensis, although

somewhat erratic on the weekly basis, remained fairly high

from August to December 1981 and from late May to July 1982

(Fig. 2-9).

The short lifespan of wild caught and lab-reared flies

at the field station was probably due to low humidity.

Other sand flies reared in the laboratory, and kept in the

environmental chamber at 85%RH, often lived 13 to 15 days

after eclosion. Early female death was the probable cause

when only partial egg batches were laid. Sugar feeding was

never observed in nature and only rarely in captivity. Lack

of carbohydrate may have been a contributing cause to early

mortality. As Lu. cayennensis is a very sedentary species,

the natural sugar source must be very close to the resting

sites, perhaps secretions from the trees.

The rearing times observed for Lu. cayennensis were

similar to those reported for Lu. anthopora, a similar sized

species (Endris, 1982), but much shorter than those for the

sympatric Lu. christophei. The individual rearing gave

somewhat misleading results, in that most flies emerged

within a seven day period. When larvae from a single egg

batch were reared together in a vial, emergence of adults

occurred during a period of up to 22 days. This delay could

be the effect of overcrowding or interference, as has been

reported among mosquito larvae (Ikeshoji and Mulla, 1970),

but not previously for sand flies.

Lutzomyia christophei appears to have a more limited

distribution than Lu. c. hispaniolae, as Lu. christophei

were recovered at seven sites, all of which were leishmania-

sis case sites, or in close proximity to such sites. As

resting Lu. christophei were secreted in tree crevices,

their apparent scarcity may have been partially an artifact

of the amount of time spent at the survey sites and the

collection methods used. It was not until cigarette smoke

was used to flush the flies from the crevices, that speci-

mens of this species were collected with any regularity.

With the exception of various locations in the Pedro Sanchez

area, including the study site and a small 15h woods, few

non-case sites received more than 10hrs of searching during

the study period. The habitat of this species, crevices and

ground-level tree holes in large shade trees primarilyy Ceiba

pentandra) in coffee groves and rock crevices in forested

areas, also may be a factor limiting its distribution. In

the Dominican Republic, very few virgin rain forests remain,

coffee and cacao groves, however, may simulate forest

conditions, as heavy leaf litter and much shade are charac-

teristic of these areas. Typically, land snails and milli-

pede feces were observed in the Lu. christophei-inhibited

crevices, it is quite likely that snail and millipede

byproducts enrich the soil of these crevices and possibly

provide sand fly larvae with an adequate diet. Lutzomyia

christophei may have evolved as a nest inhabiting species

with one or more of the tree hole-inhabiting mammals that

were present before the arrival of the Spaniards (1492 AD),

much the way Lu. anthophora is associated with woodrat

(Neotoma) nests in the USA (Young, 1972). Plagiodontia

aedium, the only endemic rodent in the D.R., is a tree hole

inhabitant (Woods, 1981). As the introduced rats replaced

the endemic rodents, the sand fly may have moved into the

tree hole nest of R. rattus or Mus musculus. Along with

this move may have been the development of a new reservoir

host for an endemic Leishmania; although there is discussion

as to whether the Dominican Leishmania is native to the

country or introduced (Walton, pers. comm., Zeledon, pers.


Although sand flies were never encountered in large

numbers in crevice resting sites, residents at several case

and non-case sites reported them to be a major annoyance at

times. On Cuba, man-biting sand flies have been reported

from caves (Avila et al., 1969); although the species was

not determined, these flies were probably Lu. orestes

(Young, pers. comm.), a very close relative of Lu.

christophei. In the Dominican Republic, very few of the

visited DCL case sites had caves or rock outcroppings.

"Erisos" were reported to be common in June and July, at

Altos de Peguero, Loma Pena Alta, and Trepada de Jabilla,

though much more so in past years than in the three summers

(1981-1983) that the author was present. "Erisos" were

originally described to the author as pale-colored flies,

smaller than a mosquito, which start biting around dusk and

continue into the late evening, inside the house as well as

outside. The bite was described as being as painful as a

mosquito's. "Erisos" also hold their wings aloft, and tend

to hop around on the person before biting. All descriptions

were similar and all described typical sand fly behavior.

In September 1983, two "erisos" were collected while

feeding, on one of the leishmaniasis patients. These were

confirmed to be female Lu. christophei by the author and

Dr. D. G. Young.

Thus Lu. christophei is regarded as the probable vector

of leishmaniasis in the Dominican Republic. It fulfills the

requirements of readily feeing on man and rodents, the

probably reservior hosts, and is capable of experimentally

transmitting the Dominican Leishmania (see Chapter 3). One

of the difficulties of in studying this sand fly is based on

its long life cycle of more than two months. Futher work

needs to be done on determining the relationship of the long

cycle to the natural habitat. Much also remains to be

determined on the bionomics of this species. The epidemio-

logical data, such as vector efficiency and infection rates

in the field, need to be studied further. The sample of

23 wild-caught female flies available for examination during

this study was insufficient. A long-range program, per-

formed by personnel who could visit the sites at various

times over a two or three year period, is needed.




New World cutaneous leishmaniasis is caused by members

of the Leishmania braziliensis and L. mexicana complexes

(Bray, 1974), but diffuse cutaneous leishmaniasis (DCL) has

been associated only with the L. mexicana complex in the

Americas (Schnur et al., 1983) and with L. aethiopica in

Africa (Bray and Bryceson, 1969). Current knowledge of DCL

is based primarily on studies of L. aethopica in Ethiopia

(Bryceson, 1969, 1970a, b, c) and L. mexicana pifanoi in

Venezuela (Convit et al., 1971). The focus of DCL in the

Dominican Republic is unique owing to the complete absence

of human cases with ulcerating lesions (Bogaert-Diaz,

unpublished data). This is one of the features that has

sparked interest in the identity of this Leishmania.

Schnur et al. (1983) and Kreutzer et al. (1983) believe that

the Dominican parasite differs from other known subspecies

in the L. braziliensis and L. mexicana complexes, but

appears to be closer to strains in the L. mexicana complex.

Lainson (1983) reported that the Dominican parasite

developed in the

anterior midgut (Suprapylaria) of experimentally infected

Lutzomyia longipalpis from Brazil. Schnur et al. (1983)

described some of the biological characters of the parasite

in culture and lab animals, but only in subjective terms.

Since the identity of the Dominican Leishmania remains

undetermined, the commonly used strain, isolated from a

14-year-old female patient from the Dominican Republic, is

designated Leishmania-Isabel strain (Petersen et al., 1982).

The purpose of this study was to quantitate the growth

of the Dominican parasite in comparison to other strains of

Leishmania, to determine the course of infection in suscep-

tible laboratory rodents, to describe the course of infec-

tion in susceptible sand flies, and to effect transmission

with the probable natural vector species.

Methods and Materials

Comparison of the Growth of Three Strains of Leishmania

Stock culture of L. mexicana amazonensis was obtained

from Dr. K. P. Chang, Rockefeller University, inoculated

into a Syrian hamster, Mesocricetus auretus, and later

reisolated and passage one time on Schneider's Drosophila

Medium (GIBCO Laboratories, Grand Island, New York) supple-

mented with 20% (v/v) heat-inactivated (560C, 30min) fetal

bovine serum (FBS) (Hendricks and Wright, 1979). The stock

culture of L. mexicana-Texas (WR-411) was provided by

Dr. Larry Hendricks, Walter Reed Army Institute of Research

(WRAIR) and was handled as above. The culture of Leish-

mania-Isabel strain (WR-336) was provided by Dr. Eileen

Franke, WRAIR, but was not passage through animals. For

each strain, 15 tubes of enriched Schneider's medium were

inoculated so that the Day 0 populations were approximately

2.50 x 105 log phase promastigotes/ml. The tubes were held

in a Hotpack environmental chamber (Hotpack, Inc., Philadel-

phia, PA) at 24.000.50C. The populations were checked

daily for 16 days using a hemacytometer with the aid of a

compound microscope (200X)

Growth of Leishmania-Isabel Strain in Laboratory Rodents

Prior to inoculation in rodents, the Isabel strain was

passage one time on NNN medium (Mansour et al., 1974).

Subadult or young Syrian hamsters and IRC strain mice

(5-7 weeks old) were inoculated via intracardial (IC),

intraperitoneal (IP), or subcutaneous (sub Q) routes. the

dosages used are given in Table 3-1. The animals were

maintained by inoculation group. Two animals from each

species/inoculation group were examined at 30 and 60 days

post-inoculation. Four subcutaneously inoculated hamsters

were examined via xenodiagnosis with sand flies, Lutzomyia

anthophora, at 3, 7, and 11 months post-inoculation; at

15 months, they were killed and assayed, as outlined below,

using Schneider's medium for cultures.

Before killing the animal, 0.2-0.5ml of blood was taken

via heart puncture and added to 4ml RPMI medium (80% RPMI

Table 3-1 Method of inoculation and number of promastigotes
used to infect laboratory rodents with
Leishmania-Isabel strain.

Species Method of Inoculation Size of Inoculum

Hamster subcutaneous 9.00 x 104 promastigotes

Hamster intraperitoneal 2.25 x 105 promastigotes

Hamster intracardial 9.00 x 104 promastigotes

Mouse subcutaneous 1.13 x 105 promastigotes

Mouse intraperitoneal 5.65 x 105 promastigotes

Medium 1640 (GIBCO) + 20% FBS (v/v)). A subcutaneous

aspirate was taken from the left hind paw and placed in

another tube of RPMI. After death, tissue samples were

taken of hind paw skin, liver, and spleen. Impression

smears were made of each on microscope slides. The tissue

samples were then placed in a Petri dish of normal saline

which contained 4000U Penicillin G Sodium (U.S. Biochemical

Corporation, Cleveland, OH) and 1.5mg/ml Streptomycin

Sulfate (U.S. Biochemical Corp.), with one Petri dish/

animal. The Petri dishes were left in a refrigerator (40C)

for 24hrs. The method was adapted from Herrer and Christen-

sen (1975). The following day the Petri dishes were removed

and uncovered in a biological hood. Each tissue sample was

washed twice with normal saline. A small portion of tissue

(9mm2 skin, 27mm3 liver or spleen) was then placed in 2ml

sterile normal saline in a sterile mortar and ground by

pestle until macerated. The solution was allowed to settle

for 1 min then lml of supernant was drawn off and added to a

tube of RPMI Medium. The culture tubes were held at room

temperature for eight days then checked for the presence of

promastigotes, with the aid of a compound microscope (200X).

Impression smears were fixed in absolute methanol, stained

with Geimsa for 30 minutes, then observed with a microscope


The Growth of Leishmania-Isabel Strain in the Sand Fly

Four-month-old BALB/c mice were inoculted in the hind

foot pads with approximately 2.5 x 105 promastigotes of

Leishmania-Isabel strain. The mice were held for four to

six weeks to allow development of the histiocytoma, during

which time the foot pad grew to two to three times normal

size. Four to seven-day-old laboratory-reared Lutzomyia

anthophora, a species from Texas and Mexico, were allowed to

feed on the infected hind foot pads of the mice. The body

and tail of the mouse was covered so that only the hind feet

were exposed to the sand flies. After feeding, the sand

flies were held in groups of 20 to 50 flies, in 40dr rearing

vials, and held in a Hotpack environment chamber at

23.00.50C and 705%RH. A sample of flies were dissected on

each day (Day 1 to 7 post-feeding) so that a total of

15 infected flies were observed for each day. After

dissecting out the digestive tract in Medium 199 (GIBCO),

the mouthparts, head, and digestive tract were examined for

the presence of leishmanial promastigotes, with the aid of a

microscope (100X, 200X). The number of infected flies was

noted along with location, number and shape of the para-

sites. To determine if the parasites observed were infec-

tive, a sample (pooled by for Day 3, 4, and 5) was

inoculated into the hind foot pads of two six-week-old


Transmission of Leishmania-Isabel Strain by Lutzomyia

Female F3 generation Lu. christophei were allowed to

feed on the swollen hind foot pads of Leishmania-Isabel

(WR-336) infected BALB/c mice. The sand flies fed on the

mice five to seven weeks post-inoculation. The flies were

then held individually in 7dr vials in a Hotpack chamber,

23.000.50C and 805%RH, until death or until they were

ready to refeed, usually seven or more days after the first

bloodmeal. Females that died one to seven days post-feeding

were dissected and examined, to correlate the infection in

Lu. christophei with that in Lu. anthophora, performed as

above. For their second bloodmeal, the flies were released

into a feeding chamber with an anesthetized noninfected

BALB/c mouse, covered with a cloth sleeve except for the

hind feet and tail. The refed flies were then recaptured

and held individually in 7dr vials until death. Upon death,

the flies were dissected to determine the state of

infection. The mice used for refeeding the sand flies were

maintained in the laboratory for three weeks before attempt-

ing to xenodiagnose leishmanial infection with Lu. antho-

phora and/or diagnosing through culturing of spleen and

liver tissue sample and subcutaneous aspirate in Schneider's

medium, performed as above. Female Lu. anthophora were used

for xenodiagnosis due to the unavailability of female Lu.

christophei at the time.


Comparison of the Growth of Three Strains of Leishmania

The two strains of Leishmania mexicana, L. mexicana-

Texas and L. m. amazonensis grew at rates that were not

statistically different, but their growth rates were much

faster than that of the Leishmania-Isabel (t-test, p = 0.05)

(Fig. 3-1). The L. mexicana strains maintained log phase

growth until Day 6, stayed in a stationary phase until

Day 10, and then decreased rapidly. Cultures of the Isabel

strain did not achieve peak population growth until Day 12,

after which they declined slowly. Post-peak populations

were difficult to estimate due to the large number of dead

or inactive promastigotes present in the samples, only

motile promastigotes were counted.


*- Leishmania-Isabel
Sa- L. mexicana-Texas
4* L. m. amazonensis

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16


Figure 3-1. Daily estimated mean population of three
strains of Leishmania grown in Schneider's
medium, at 25.00C.

Growth of Leishmania-Isabel Strain in Laboratory Rodents

Leishmania-Isabel-inoculated hamsters and IRC strain

mice examined at 30 and 60 days post-inoculation showed no

externally visible signs of infection. Four hamsters were

inoculated via subcutaneous route and maintained for over

60 days, by 75 days post-inoculation, only one hamster

exhibited a very slight swelling of the hind paw. All four

were xenodiagnosed using lab-reared Lu. anthophora and were

infected at this time. By 7 months post-inoculation, the

cutaneous infection had apparently self-cured because none

of 30 sand flies feeding on the "infected" foot of each

hamster developed promastigote infections. Aspirate

cultures taken at seven months were also negative. When the

four hamsters were killed at 15 months, no parasites were

isolated by aspirate, liver, or spleen culture.

The most sensitive method for determination of infec-

tion, besides xenodiagnosis, was by culturing of splenic

tissue. No parasites were observed in the cultures of heart

blood or in skin impression smears. Active infections were

recovered from all animals killed at 30 days, by spleen

culture and from all, but one mouse, via liver culture. The

results were much more variable at 60 days (Table 3-2). At

60 days, no leishmaniae were observed or isolated from one

mouse (IP) and two hamsters (1IP, 1IC). Animals inoculated

via subcutaneous injection were the most readily confirmed

as infected by the different methods of culturing and

culturing and impression smears (Table 3-2, Fig. 3-2).

Table 3-2 Number of animals examined determined to be infected with Leishmania-
Isabel strain via different isolation methods.

Cultures Smears
Inoculation Day Heart Foot
Species Method Examined Blood Aspirate Skin Spleen Liver Skin Spleen Liver

Mouse sub Q 30 0 0 0 2 1 0 1 0

60 0 0 1 1 1 0 1 1

IP 30 0 0 0 2 0 0 1 1

60 0 0 0 2 0 0 2 0

Hamster sub Q 30 0 0 1 2 2 0 2 2

60 0 1 1 2 2 0 2 2

IP 30 0 0 0 2 2 0 2 2

60 0 0 0 1 0 0 0 0

IC 30 0 0 0 2 2 0 2 1

60 0 0 0 1 0 0 1 1

Note: n = 2, except Mouse-sub Q-60 days where n = 1.



Figure 3-2. Leishmania-Isabel strain amastigotes in a
spleen impression smear stained with Giemsa

Growth of Leishmania-Isabel Strain in the Sand Fly

Promastigotes were observed in some of the Lu.

anthophora females in each sample from Day 1 to Day 7

post-feeding on a Leishmania-Isabel infected BALB/c mouse.

The percent infected increased with time, but the percent

with bacterial contamination (Leishmania infection undeter-

minable) decreased with time, because of the high death rate

in these flies. In the sample from Day 1, only one promas-

tiogote was observed in 30 fly dissections; however, a few

amastigote-infected macrophages were observed in the blood

meals in five of the dissections which were later stained

and examined with the aid of microscope (1000X). For the

remaining days, infected flies represented 53.6% to 75.0% of

the sample dissected each day. The infections were of

several hundred to several thousand promastigotes/fly

(Table 3-3). Promastigotes were first observed only in the

anterior midgut, (Fig. 3-3) near the stomodael valve in

Day 2 and Day 3 flies. By Day 4, parasites were observed in

the posterior pharynx, as well. On Day 5, promastigotes

were observed in the mouthparts in 4 of the 15 infected

flies. Mouthpart infections were observed 12 of 15 and 13

of 15 infected Day 6 and Day 7 flies, respectively

(Fig. 3-4). The shape of the promastigotes was highly

variable in young infections (Day 2 and 3); however, only

elongate forms were observed anterior of the stomodael valve

Table 3-3. The course of development of Leishmania-Isabel strain in the sand fly,
Lutzomyia anthophora, based on daily dissections, Days 1 to 7 post-

Day post- Status of Ovarian # infected Location, number &
feeding blood meal status flies/sample shape of parasites

RBC's intact,
meal dark red

RBC's intact
meal black

Blood remnants

Meal totally
excreted in most

Meal excreted
in all flies





eggs fully






anterior MG+ w/ blood meal,
only 1 promastigote ob-
served in dissections,
stumpy form, amastigotes
observed in macrophages in
5 dissections

anterior MG w/ blood meal,
100's observed in all
infected flies, primarily
stumpy forms

anterior MG, forward to
stomodaeal valve, 100's
observed, stumpy and some
elongate forms

1000 in anterior MG, 50-
100 in pharynx, elongate

1000's in anterior MG, 50-
100 in pharynx, 10-20 in
cibarial region, 5-10 in
mouthparts, elongate forms

Table 3-3. Continued.

Day post- Status of Ovarian # infected Location, number &
feeding blood meal status flies/sample shape of parasites

6 as Day 5 commenced 15/21 as Day 5

7 as Day 5 finished 15/20 1000's in anterior MG,
oviposition 100's in pharynx, 20-50 in
mouthparts, elongate forms

* parasites probably still in amastigote stage, infection not determinable

+ MG = midgut

Figure 3-3. Promastigotes of Leishmania-Isabel strain from
the anterior midgut of the sand fly, Lu.
anthophora (1000X).

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - Version 2.9.9 - mvs