ArticlePDF Available

Serologic evidence for an epizootic dengue virus infecting Toque macaques (Macaca sinica) at Polonnaruwa, Sri Lanka

Authors:
Article

Serologic evidence for an epizootic dengue virus infecting Toque macaques (Macaca sinica) at Polonnaruwa, Sri Lanka

Abstract and Figures

Dengue is one of the most rapidly emerging diseases in the tropics. Humans are the principal reservoir of dengue viruses. It is unclear if nonhuman primates also serve as a reservoir of human dengue viruses under certain conditions. In this study, a cross-sectional serologic survey was carried out to characterize the pattern of transmission of a recently identified dengue virus among toque macaques in Sri Lanka. The results indicated that an epizootic dengue virus was active among the macaques. A single epizootic had taken place between October 1986 and February 1987 during which 94% of the macaques within the 3 km2 study site were exposed to the virus. The epizootic was highly focal in nature because macaques living 5 km from the study population were not exposed to the virus. The transmission of dengue viruses among macaques in the wild may have important public health implications.
Content may be subject to copyright.
300
Am. J. Trop. Med. Hyg., 60(2), 1999, pp. 300–306
Copyright
q
1999 by The American Society of Tropical Medicine and Hygiene
SEROLOGIC EVIDENCE FOR AN EPIZOOTIC DENGUE VIRUS INFECTING TOQUE
MACAQUES (MACACA SINICA) AT POLONNARUWA, SRI LANKA
ARAVINDA M. DE SILVA, WOLFGANG P. J. DITTUS, PRIYANI H. AMERASINGHE,
AND
FELIX P. AMERASINGHE
Section of Rheumatology, Department of Internal Medicine, Yale University School of Medicine, New Haven, Connecticut;
Department of Zoological Research, National Zoological Park, Smithsonian Institution, Washington, District of Columbia;
Institute of Fundamental Studies, Kandy, Sri Lanka; Department of Zoology, Faculty of Science, University of Peradeniya,
Peradeniya, Sri Lanka
Abstract. Dengue is one of the most rapidly emerging diseases in the tropics. Humans are the principal reservoir
of dengue viruses. It is unclear if nonhuman primates also serve as a reservoir of human dengue viruses under certain
conditions. In this study, a cross-sectional serologic survey was carried out to characterize the pattern of transmission
of a recently identified dengue virus among toque macaques in Sri Lanka. The results indicated that an epizootic
dengue virus was active among the macaques. A single epizootic had taken place between October 1986 and February
1987 during which 94% of the macaques within the 3 km
2
study site were exposed to the virus. The epizootic was
highly focal in nature because macaques living 5 km from the study population were not exposed to the virus. The
transmission of dengue viruses among macaques in the wild may have important public health implications.
Dengue is the name given to four closely related but dis-
tinct arboviruses (serotypes DEN-1, DEN-2, DEN-3, and
DEN-4) belonging to the genus Flavivirus in the family Fla-
viviridae. Over the past four decades dengue viruses have
emerged as a major threat to the health of people living in
many tropical areas of the world.
1,2
Human dengue viruses
are mostly active in urban areas where the virus is main-
tained through a cycle in which humans are the principal
reservoir host and Aedes aegypti is the principal mosquito
vector.
Human dengue infections have also been documented in
rural areas in Asia, although the prevalence of infection is
lower than in the urban setting.
3–5
In rural areas with a low
prevalence of human infection, how does the virus maintain
itself? The virus may persist in the vector through direct
vertical and horizontal transmission between mosquitoes.
Dengue-infected female mosquitoes have genital tract infec-
tions that sometimes lead to the infection of eggs during
oviposition.
6
Furthermore, venereal transmission has also
been noted from infected males to females.
7
Another mech-
anism used by human arbovirus for maintaining themselves
is to have additional nonhuman reservoir hosts. Yellow fever
virus, a Flavivirus that shares many similarities to dengue,
is maintained by an urban cycle involving Ae. aegypti and
humans as well as in a sylvatic cycle with Aedes mosquitoes
and nonhuman primates.
8
It remains to be resolved if human
dengue viruses also use nonhuman primates as reservoir
hosts under certain conditions.
Dengue viruses have been isolated from monkeys and pri-
matophilic vectors. Sylvatic circulation is common in parts
of West Africa where both DEN-1 and DEN-2 viruses have
been isolated from monkeys and sylvatic Aedes species.
8
However, the sylvatic cycle in West Africa does not to ap-
pear to cross-over to humans because dengue infection was
rare or absent among people living close to the sylvatic cy-
cle.
8
In fact, genetic analysis has shown that viruses isolated
from human epidemics in Africa are distinct from the syl-
vatic viruses.
9
Rudnick and Lim carried out studies on the ecology of
dengue in Malaysia.
10
These studies led to the isolation of
DEN-1, DEN-2, and DEN-4 viruses from sentinel monkeys
(Presbytis obscura and Macaca fascicularis). Dengue virus
was also isolated from Ae. albopictus, a rural vector found
at ground level, as well as from other primatophilic Aedes
species found at the canopy level. Based on these findings,
Rudnick has proposed that in Southeast Asia dengue viruses
are maintained in a primal cycle involving monkeys and vec-
tors of the Ae. niveus group that feed at the canopy level.
10
Rudnick also proposed that the rural dengue vector (Ae. al-
bopictus) may introduce the sylvatic virus into the human
population.
10
Much of the work in Malaysia was done in the
1960s and the sylvatic and human dengue viruses were not
sequenced to establish the relationship between human and
sylvatic dengue viruses in Asia.
Dengue is an important cause of human morbidity and
mortality in Sri Lanka.
11,12
While dengue viruses are primar-
ily active in the urban areas in the southwestern part of the
island, serologic surveys indicate low-level human exposure
in other parts of the country.
11
Four species of nonhuman
primates are present in Sri Lanka, and they often live in
proximity to people in rural villages.
13
The sociodemography
of a population of toque macaques (Macaca sinica)atPo-
lonnaruwa has been intensely studied by Dittus and others
since 1968.
13
Evidence for a sylvatic dengue cycle among
these macaques was recently reported by Peiris and others,
who found that 94% of the monkeys tested had antibodies
that neutralized DEN-2 virus.
14
The objective of the current
study was to further characterize the pattern of infection of
this virus among macaques at Polonnaruwa. We report here
that the virus causes epizootics among the macaques, rather
than being enzootic as was previously thought.
MATERIALS AND METHODS
Study site and macaque population. The population of
toque macaques used in the current study inhabit the natural
dry evergreen forest within the Nature and Archaeological
Reserve at Polonnaruwa (Figure 1).
13
The behavior, ecology,
demography, and genetics of these wild monkeys have been
intensively studied since 1968 by Dittus and others.
13,15–18
In
1995, the population comprised nearly 1,000 monkeys dis-
tributed among some 28 social groups. All the macaques in
the population have been individually identified and the
dates of birth and life histories of nearly all the animals are
301
DENGUE EPIZOOTIC AMONG MACAQUES IN SRI LANKA
F
IGURE
1. Location and map of study site (adapted from Hoelzer
and others
20
). The position of the study site and camp site at Polon-
naruwa are indicated.
known. The ages of animals whose births were not observed,
such as immigrants into the study population, were estimated
based on known relationships between morphologic devel-
opment and age.
19
Macaque serum samples. Blood was collected for ge-
netic and epidemiologic analyses by trapping the macaques
in cages and tranquilizing them with ketamine hydrochloride
as previously described.
20
Blood was collected by venipunc-
ture and separated into serum the same day and stored in
liquid nitrogen. Sera collected in 1995, 1987, and 1986 were
used in the current study. Control sera samples from labo-
ratory rhesus macaques (M. mullata) infected with dengue
or yellow fever viruses were provided by the Yale Arbovirus
Research Unit (YARU) (New Haven, CT).
Serologic tests. An ELISA was used to screen macaque
sera for dengue antibody. Each plate was coated for 2 hr at
37
8
C with the 4G2 anti-flavivirus monoclonal antibody (pro-
vided by YARU).
21
The plate was washed and incubated
with a blocking solution (5% horse serum in phosphate-buf-
fered saline [PBS]) for 15 min at 37
8
C. Next, the plates were
incubated overnight at 4
8
C with the virus antigen. Flavivirus
antigens (DEN-1, DEN-2, DEN-3, DEN-4, yellow fever,
Japanese encephalitis, and Zika viruses) were prepared and
provided by YARU. Since the antigens were prepared in
mouse brain, uninfected mouse brain antigen was used as a
negative control. The following day the unbound antigen
was washed away with PBS containing 0.5% Tween 20
(PBS/Tween) and the macaque sera were added at a 1:100
dilution and incubated for 1 hr at 37
8
C. The plates were
washed with PBS/Tween and incubated for 30 min at 37
8
C
with horseradish peroxidase–conjugated goat anti-monkey
immunoglobulin (Sigma Biosciences, St. Louis, MO). Fi-
nally, the plates were developed by adding commercially
prepared peroxidase substrate (Zymed Laboratories, San
Francisco, CA). Each serum sample was tested in duplicate
against each of the test antigens. Sera that gave an optical
density (OD) that was at least four times greater than the
normal mouse brain negative control were designated as pos-
itive sera.
RESULTS
Enzyme-linked immunosorbent assay for detecting
dengue antibody in monkeys. Flavivirus antigens are no-
torious for generating antibodies that cross-react between
Flaviviruses. Thus, it was important to determine if the ELI-
SA used to screen macaque sera was capable of distinguish-
ing between Flaviviruses. For this purpose sera from labo-
ratory monkeys infected with dengue or yellow fever were
used. As expected, the sera cross-reacted with Flavivirus an-
tigens other than the virus used in the original infection (Fig-
ure 2). However, for each of the sera the highest signal was
observed in the presence of antigen from the infecting virus
(Figure 2). Thus, although cross-reactivity was observed, it
was possible to clearly distinguish the infecting Flavivirus
from other Flaviviruses. These results validate the use of this
ELISA as a means of screening macaques for Flavivirus
antibodies, as well as characterizing at a serologic level the
relationship of macaque Flaviviruses to known Flaviviruses.
Analysis of 1995 sera. Two hundred forty-four serum
samples collected between July and October 1995 were
screened for the presence of antibodies against dengue-2 vi-
rus. Twenty-one percent (52 of 244) of the animals tested
positive for dengue virus antibody. The macaque sera were
also tested using antigens from yellow fever, Japanese en-
cephalitis, and Zika viruses. Most of the dengue-positive
sera also cross-reacted with these antigens, although the sig-
nal was consistently less than the signal for the dengue-2
virus antigen (Figure 3). The mean ODs for the dengue virus
antigen were 1.7, 3, and 3.3 times greater than the mean
ODs for the yellow fever, Japanese encephalitis, and Zika
virus antigens, respectively (Figure 3). None of the dengue
virus–negative sera bound to the other Flavivirus antigens.
These results indicate that the Flavivirus infecting the ma-
caques was antigenically most closely related to dengue vi-
rus and was distinct from the antigenic complexes that in-
clude Japanese encephalitis and Zika viruses.
The ELISAs were also performed to determine the dengue
virus serotype circulating among the macaques. Plates were
coated with antigen prepared with the different dengue virus
serotypes. Five representative positive sera were selected
and serial dilutions of each sera was tested with the different
dengue viral antigens. The positive sera cross-reacted with
the four antigens and, in spite of the serial dilutions, it was
not possible to single out any particular antigen as belonging
to the serotype responsible for the original infection.
Analysis by sex and social group (1995 sera). The ratio
of males to females in the study sample (1:1.13) was similar
to the ratio of males to females among the seropositive an-
imals (1:1.38), indicating that both sexes were equally sus-
ceptible to the virus. The 244 animals tested in this study
belonged to 16 of the 28 different social groups living within
the study area. The data were analyzed to determine the
prevalence of infection in the different social groups. Ani-
mals from all the groups tested had at least one infected
individual and the prevalence of infection in the different
groups ranged from 6% to 50% (Table 1), depending on the
number of group members nine years of age or older. Thus,
302
DE SILVA AND OTHERS
F
IGURE
2. The ELISA for detecting Flavivirus antibodies in macaque sera. The ELISA plates were coated with normal mouse brain (NMB)
and dengue-2 (D-2), Japanese encephalitis (JE), yellow fever (YF), and Zika virus antigens. Each plate was incubated with sera from a monkey
infected with yellow fever or DEN-2 viruses. Each serum and antigen combination was tested in duplicate. OD
5
optical density.
T
ABLE
1
Prevalence of dengue antibody among macaques from different so-
cial groups in Polonnaruwa, Sri Lanka
Group ID No. tested No. positive (%)
22B
22D
22N
A
B
CH1
CF
D1
13
18
15
17
38
13
13
38
3
6
4
1
6
4
3
6
(23)
(33)
(27)
(6)
(16)
(31)
(23)
(16)
D2
F1
F2
F3
H2
J
M1
M2
9
14
6
6
9
4
25
6
2
4
3
1
2
1
5
1
(22)
(29)
(50)
(17)
(22)
(25)
(20)
(17)
Total 244 52 (21)
F
IGURE
3. Analysis of sera showing that flavivirus antibody–pos-
itive macaque sera bound best to the dengue virus antigen. The 52
flavivirus antibody-positive macaque sera were tested against normal
mouse brain (NMB), Dengue-2, Japanese encephalitis (JE), yellow
fever (YF), and Zika virus antigens. The mean value obtained from
all 52 positive sera against each of the antigens is displayed. Bars
show the mean
6
SD. OD
5
optical density.
the virus was active in an area encompassing the entire study
site, if not a larger area.
Analysis by age (1995 sera). The age of the animals test-
ed for dengue virus antibody ranged from less than one-
month-old infants to 33-year-old adults. The prevalence of
infection in the different age groups of macaques was ana-
lyzed. Figure 4 shows the number of animals tested in each
age cohort, as well as the number of animals exposed to the
virus. Very few (2.6%) animals in the eight year of age and
younger cohorts were exposed to the virus whereas 88% of
the animals in the nine year of age and older cohorts were
exposed to the virus. The pattern of exposure in the different
age cohorts is highly suggestive of an epizootic virus that
was active 8–9 years (between 1986 and 1987) prior to the
time when the blood was collected in 1995. Thus, all the
animals that were born after the epizootic in 1986–1987,
which includes 80% of the animals in the eight-year-old co-
303
DENGUE EPIZOOTIC AMONG MACAQUES IN SRI LANKA
F
IGURE
4. Prevalence of dengue antibody among macaques in different age groups (1995 sera).
T
ABLE
2
Dengue antibodies among macaques in the eight-year-old cohort
(1995 sera) in Polonnaruwa, Sri Lanka
Animal ID number Date of birth ELISA result
2739
2747
2941
2801
2884
2923
2952
2789
Aug 16, 1986
Dec 28, 1986
Jan 5, 1987
Jan 6, 1987
Feb 4, 1987
Feb 11, 1987
Feb 21, 1987
Feb 26, 1987
1
1
2
1
2
2
2
2
2700
3056
2845
2930
2781
2757
2925
2833
2759
3094
2920
3095
Mar 10, 1987
Apr 7, 1987
Apr 8, 1987
Apr 8, 1987
Apr 14, 1987
Apr 27, 1987
May 1, 1987
May 3, 1987
May 9, 1987
May 10, 1987
May 19, 1987
May 31, 1987
2
2
2
2
2
2
2
2
2
2
2
2
hort and all of the animals in the seven-year-old and younger
cohorts, were not exposed to the virus. The three positive
animals in the 0-year-old cohort are almost certainly due to
maternal antibody. In fact, two of the mothers of these in-
fants were tested and found to be positive animals more than
nine years of age. The mother of the third positive infant
was not tested in this study. Prior to the point at which trans-
mission stopped in 1986–1987, there might have been a sin-
gle epizootic that exposed almost all the individuals in the
population or the virus may have been enzootic among the
macaques.
Detailed analysis of eight-year-old animals (1995 sera).
To determine more accurately when this virus stopped cir-
culating among the macaques, all the eight-year-old animals
bled in 1995 (this sample included 10 animals from the orig-
inal 246 animal samples, as well as 10 animals not included
in the original sample, but also trapped in 1995) were tested
for dengue-2 virus antibody (Table 2). Three of the 20 ani-
mals in the eight-year-old cohort were positive for dengue
virus antibody. The dates of birth of the eight-year-old ani-
mals and their serologic status is presented in Table 2. It is
clear from these data that only the oldest animals in this
cohort (i.e., those born before January 6, 1987) were ex-
posed to this virus, while none of the animals born after
February 4, 1987 were exposed. These results suggest that
transmission abruptly ceased among the macaques starting
sometime between January and February 1987.
Analysis of 1986 and 1987 sera. The serologic analysis
of macaque samples collected in 1995 indicate that the an-
imals were exposed to a dengue virus whose transmission
abruptly ceased before February 1987. Prior to February
1987, the virus may have been responsible for a single large
epizootic that occurred towards the end of 1986 or beginning
of 1987 or the virus may have been enzootic among the
macaques. To further characterize the pattern of transmission
during the critical period between 1986 and 1987, blood
samples collected in September 1986 (pre-epizootic?) and
blood samples collected between June and August 1987
(post-epizootic) were tested for dengue virus antibody. These
results, which are presented in Table 3, confirm that 81% of
the macaques were infected due to a single epizootic that
304
DE SILVA AND OTHERS
T
ABLE
3
Prevalence of dengue antibody among macaques bled in 1986 and
1987 in Polonnaruwa, Sri Lanka
Date bled
1986 (September) 1987 (July–September)
No. of animals tested
No. positive (ELISA)
16
2* (12%)
44
41 (93%)
* The two positive animals were 15 and 20 years old at the time of bleeding in 1986.
T
ABLE
4
Rainfall at Polonnaruwa, Sri Lanka before, during, and after the
northeast monsoon of 1986–1987
Month Rainfall (mm) No. of days of rain
1986: August
September
October
November
December
21
6
276
125
350
3
1
10
8
18
1987: January
February
March
101
18
61
9
1
3
took place at the end of 1986. Of the 16 animals sampled
in September 1986, only two older animals were seroposi-
tive. In contrast, when the same population was sampled in
1987 the seroprevalence was 94% (41 of 44), indicating that
a major epizootic took place between 1986 and 1987. The
two positive animals in the 1986 sample indicate that some-
time in the past there may have been another epizootic, a
possibility that is supported by the observation that the two
positive animals (15 and 20 years of age) were among the
oldest of the animals in the 1986 sample.
Timing of infection in relation to weather. Most rainfall
at Polonnaruwa normally occurs between October and Jan-
uary due to inter-monsoonal convectional rains in October
and the northeast monsoon from November to January (Ta-
ble 4). February is normally dry with clear skies. The onset
of the virus infection among the macaques at Polonnaruwa
after September 1986 and its cessation before February 1987
coincided with the rainy season. The fact that transmission
took place during the rainy season is consistent with the
virus being transmitted by a mosquito vector.
Analysis of camp animals. The tests performed with sera
from macaques living within the study site at Polonnaruwa
indicated that almost all the animals were exposed to a den-
gue virus towards the end of 1986. Was this a focal event
restricted to animals at the study site, or did the 1986 epi-
zootic represent an island-wide incident? To examine the
geographic extent of the epizootic, toque macaque sera col-
lected in 1986 and in 1995 from animals at a site 5 km from
the study site and separated from the study site by cultivation
(Figure 1, camp site) were tested for dengue virus antibody.
These animals have not been extensively studied and bio-
graphic data including the dates of birth were not available.
The ages of these animals was estimated using morphologic
criteria as previously described.
19
Two seropositive older an-
imals and two seronegative younger animals from the study
site were also tested in this experiment as controls. None of
the animals trapped at the camp site in 1986 were positive.
More importantly, none of the animals trapped in 1995,
many of whom were estimated to be more than nine years
of age, were positive for dengue virus antibody. This result
was in sharp contrast to the data obtained for animals at the
study site where nearly all the older animals were positive.
These data indicate that the 1986–1987 epizootic was focal
in nature and restricted to animals living in the immediate
proximity of the Polonnaruwa study site.
DISCUSSION
This study has provided serologic evidence for an epizo-
otic dengue virus among toque macaques in Sri Lanka. One
must be cautious when using serologic data to identify Fla-
viviruses because antibodies in response to Flavivirus infec-
tions are notorious for cross-reacting with other Flaviviruses.
However, the available data suggest that the macaque virus
is a dengue virus. First, while four Flavivirus antigens (den-
gue, Japanese encephalitis, Zika, and yellow fever viruses)
were tested in the current study, the highest signal was ob-
served consistently with the dengue antigen. Second, and
more importantly, in the previous study by Pieris and others,
they performed neutralization tests and observed that all the
positive sera neutralized dengue-2 virus (64 of 64), while
very few sera neutralized JE (3 of 64) or West Nile (1 of
64) viruses.
14
Based on these results, it is reasonable to des-
ignate the macaque virus as a dengue virus. The unambig-
uous identification of the agent is only possible once the
virus has been isolated from macaques. Unfortunately, at-
tempts to isolate the virus from macaque sera have so far
proven unsuccessful. As in humans, dengue is an acute in-
fection in monkeys and the virus does not persist in pri-
mates.
22
The failure to isolate virus from the sera is not sur-
prising since none of the samples were collected during the
actual epizootic between October 1986 and February 1987.
In the previous study by Pieris and others, a smaller sam-
ple of the same population of macaques was tested in 1987.
14
These investigators concluded that the macaques were grad-
ually exposed to the virus over time and that by six years
of age more than 90% of the animals had seroconverted.
These results were consistent with a highly active, enzootic
virus. The present study with blood samples collected from
the same population eight years later (1995), as well as blood
samples collected in 1986 and 1987, has revealed that this
is an epizootic virus characterized by outbreaks infecting
almost the entire population of macaques within at least a 3
km
2
area followed by prolonged (at least eight years for the
Polonnaruwa population) periods with no transmission. The
high seroprevalence observed by Pieris and others was due
to the fact that the population was sampled in 1987 imme-
diately after a major epizootic.
Although the virus was highly infectious to macaques, the
animals did not appear to have serious illness. These animals
have been intensively studied and no increased mortality or
obvious signs of morbidity were observed by field assistants
during the period of the epizootic (Dittus WPJ, unpublished
data). These findings are consistent with the agent being a
dengue virus since previous field and laboratory studies have
documented that infected monkeys are mainly asymptomatic
305
DENGUE EPIZOOTIC AMONG MACAQUES IN SRI LANKA
although they do develop viremia and can serve as a reser-
voir host.
8,22,23
The current study has raised the following important ques-
tions. What is the origin of this virus and how is the agent
maintained in Sri Lanka? Over a large part of the Amazon,
yellow fever virus moves in waves through monkey popu-
lations leading to epizootics in particular regions with a pe-
riodicity of 8–10 years.
8
In Senegal, epizootics of dengue
among monkeys were observed in 1974 and again in 1981.
24
The toque macaque is found in many forested areas of the
island, excluding the climatic extremes of arid and cold. Un-
like the rhesus macaque (M. mulatta) of India, however, it
does not cohabit with people in towns. It is conceivable that
this virus may sweep through the island infecting a troop of
macaques and then moving on to the next troop as immunity
builds up in the original group. A troop that has been in-
fected may again become susceptible to an epizootic as the
number of naive individuals increase in the population. The
population at the study site at Polonnaruwa may be poised
for another epizootic since the majority (79%) of the animals
tested in 1995 did not show evidence of exposure to this
virus.
When considering the reservoirs of the macaque virus, it
is important to consider other vertebrates, especially pri-
mates living on the island. In addition to toque macaques,
two other species of monkeys (Semnopithecus entellus and
Trachypithecus vetellus) and one species of prosimian, the
slender loris (Loris tardigradus), are found at the study site.
In areas of sympatry, the langurs are more common than the
macaques.
25
The virus may be enzootic among one of these
other primates and the toque macaques may simply serve as
an amplifying host. However, if the virus is enzootic in an-
other species at the study site, it is difficult to understand
why there was no evidence of infection among any of the
susceptible macaques over the eight-year period between
1987 and 1995. When considering the origin of the macaque
virus, it is important not to forget humans, the most abundant
primate on the island and a known reservoir of dengue vi-
ruses. It is possible that the epizootic in 1986 was due to
cross-over of a human dengue virus to the macaque popu-
lation. Future efforts will be directed towards isolating the
virus from macaques for unambiguous identification and to
determine the relationship of the macaque virus to dengue
viruses infecting humans in Sri Lanka.
Acknowledgments: During this study we received support from col-
leagues in Sri Lanka and the United States. Foremost, we thank Dr.
Rebeca Rico-Hesse for advice and support throughout the project.
We thank Shirley Tirrell-Peck of the Yale Arbovirus Research Unit
for guidance at various stages of the project. In Sri Lanka, we thank
Taya Dias, Wanaja Dittus, Sunil Goonatilake, Gayan Gunewardane,
Vajira Hettige, Don Melnick, Beatrice Sweeney-Perez, Anna Pethi-
yagoda, S. P. Ranasinghe, and Jennifer Wilcox for technical assis-
tance. Aravinda de Silva thanks Amy Weil for encouragement.
Financial support: Field research was supported by grants to Wolf-
gang P. J. Dittus from the National Science Foundation (BNS-
9104649, 9510894), the Harry and Frank Guggenheim Foundation,
Friends of the National Zoo, and Earthwatch. The laboratory studies
were supported by U.S. Army grant DAMD17-94-J-4004.
Authors’ addresses: Aravinda M. de Silva, Department of Microbi-
ology and Immunology, University of North Carolina, Campus Box
7290, Chapel Hill, NC 27599. Wolfgang P. J. Dittus, Institute of
Fundamental Studies, Hantana Road, Kandy, Sri Lanka. Priyani H.
Amerasinghe and Felix P. Amerasinghe, Department of Zoology,
Faculty of Science, University of Peradeniya, Peradeniya, Sri Lanka.
Reprint requests: Aravinda M. de Silva, Department of Microbiology
and Immunology, University of North Carolina, Campus Box 7290,
Chapel Hill, NC 27599.
REFERENCES
1. Monath TP, 1994. Dengue: the risk to developed and developing
countries. Proc Natl Acad Sci USA 91: 2395–2400.
2. Halstead SB, 1992. The XXth century dengue pandemic: need
for surveillance and research. World Health Stat Q 45: 292–
298.
3. Silarug N, Foy HM, Kupradinon S, Rojanasuphot S, Nisalak A,
Pongsuwant Y, 1990. Epidemic of fever of unknown origin
in rural Thailand caused by influenza A (H1N1) and dengue
fever. Southeast Asian J Trop Med Public Health 21: 61–67.
4. Rudnick A, 1965. Studies on the ecology of dengue in Malaysia:
a preliminary report. J Med Entomol 2: 203–208.
5. Mehendale SM, Risbud AR, Rao JA, Banerjee K, 1991. Out-
break of dengue fever in rural areas of Parbhani district of
Maharastra (India). Indian J Med Res: 93: 6–11.
6. Rosen L, Shroyer DA, Tesh RB, Freier JE, Lien JC, 1983.Tran-
sovarial transmission of dengue viruses by mosquitoes: Aedes
albopictus and Aedes aegypti. Am J Trop Med Hyg 32: 1108–
1119.
7. Rosen L, 1987. Sexual transmission of dengue viruses by Aedes
albopictus. Am J Trop Med Hyg 37: 398–402.
8. Rodhain F, 1991. The role of monkeys in the biology of dengue
and yellow fever. Comp Immunol Microbiol Infect Dis 14: 9–
19.
9. Rico-Hesse R, 1990. Molecular evolution and distribution of
dengue viruses type 1 and 2 in nature. Virology 174: 479
493.
10. Rudnick A, Lim TW, 1986. Dengue fever studies in Malaysia.
Inst Med Res Malaysia Bull 23: 127–147.
11. Vitarana, T. 1982. Viral diseases in Sri Lanka: a national over-
view. Mackenzie JS, ed. Viral Diseases in South East Asia
and the Western Pacific. London: Academic Press, 198–204.
12. Vitarana T, 1990. Dengue haemorrhagic fever. Ceylon Med J
35: 83–87.
13. Dittus WPJ, 1977. The socioecological basis for conservation
of the Toque monkey (Macaca sinica) of Sri Lanka. Prince
Rainier III of Monaco, Bourne GH, eds. Primate Conserva-
tion. New York: Academic Press, 237–265.
14. Peiris JS, Dittus WPJ, Ratnayake CB, 1993. Seroepidemiology
of dengue and other arboviruses in a natural population of
Toque macaques (Macaca sinica) at Polonnaruwa, Sri Lanka.
J Med Primatol 22: 240–245.
15. Dittus WPJ, 1979. The evolution of behaviors regulating density
and age-specific sex ratios in a primate population. Behaviour
69: 265–302.
16. Dittus WPJ, 1988. Group fission among wild toque macaques
as a consequence of female resource competition and envi-
ronmental stress. Anim Behaviour 36: 1626–1645.
17. Cheverud JM, Dittus WPJ, 1992. Primate population studies at
Polonnaruwa. II. Heritability of body measurements in a nat-
ural population of toque macaques (Macaca sinica). Am J
Primatol 27: 145–156.
18. Keane B, Dittus WPJ, Melnick DJ, 1997. Paternity assessment
in wild groups of toque macaques (Macaca sinica) at Polon-
naruwa, Sri Lanka using molecular markers. Mol Ecol 6: 267–
282.
19. Dittus WPJ, Thorington Jr RW, 1981. Techniques for aging and
sexing primates. Techniques for the Study of Primate Popu-
lation Ecology. Washington, DC: Subcommittee on Conser-
vation of Natural Primate Populations, National Academy of
Sciences Press, 81–134.
20. Hoelzer GA, Dittus WPJ, Ashley MV, Melnick DJ, 1994. The
local distribution of highly divergent mitochondrial DNA hap-
lotypes in Toque macaques Macaca sinica at Polonnaruwa,
Sri Lanka. Mol Ecol 3: 451–458.
21. Sithiprasasna R, Strickman D, Innis BL, Linthicum KJ, 1994.
ELISA for detecting dengue and Japanese encephalitis viral
antigen in mosquitoes. Ann Trop Med Parasitol 88: 397–404.
306
DE SILVA AND OTHERS
22. Halstead SB, Shotwell H, Casals J, 1973. Studies on the patho-
genesis of dengue infection in monkeys. 1. Clinical laboratory
responses to primary infection. J Infect Dis 128: 7–14.
23. Halstead SB, Shotwell H, Casals J, 1973. Studies on the patho-
genesis of dengue infection in monkeys. II. Clinical labora-
tory responses to heterologous infection. J Infect Dis 128: 15–
22.
24. Saluzzo JF, Cornet M, Adam C, Eyraud M, Digoutte JP, 1986.
Dengue 2 in eastern Senegal: serologic survey in simian and
human populations. 1974–1985. Bull Soc Pathol Exot Filial
79: 313–322.
25. Eisenberg JF, Muckenhirn NA, Rudran R, 1972. The relation
between ecology and social structure in primates. Science
176: 863–874.
... There was significant intra-group heterogeneity in the serology group (I 2 : 97.82%), and pooling of results from both serology and RT-PCR methods was not supported as given significant inter-group heterogeneity was noted (p = 0.000). There were 15 studies involving a forest setting [22,[24][25][26][27][28][29][30][31][32][33][34][35][36][37], 12 of which were based solely in a forest. Of 7 studies involving the urban/rural setting [26,31,[38][39][40][41][42], only 5 were exclusively based at such. ...
Article
Full-text available
Dengue is a rapidly spreading mosquito-borne flavivirus infection that is prevalent in tropical and sub-tropical regions. Humans are known to be the main reservoir host maintaining the epidemic cycles of dengue but it is unclear if dengue virus is also maintained in a similar enzootic cycle. The systematic review was conducted in accordance to Cochrane's PRISMA recommendations. A search was done on PubMed, EMBASE, Scopus and Cochrane Library. Key data on animal dengue positivity was extracted and classified according to animal type and diagnostic modes. Of the 3818 articles identified, 56 articles were used in this review. A total of 16,333 animals were tested, 1817 of which were positive for dengue virus by RT-PCR or serology. Dengue positivity was detected in bats (10.1%), non-human primates (27.3%), birds (11%), bovid (4.1%), dogs (1.6%), horses (5.1%), pigs (34.1%), rodents (3.5%), marsupials (13%) and other small animals (7.3%). While majority of dengue positivity via serology suggests potential enzootic transmission, but regular dengue virus spillback cannot be excluded. With the exception of bats, acute infection among animals is limited. Further investigation on animals is critically required to better understand their role as potential reservoir in dengue transmission.
... The IgG ELISAs we used for screening produce reliable negative results and are frequently used in experimental arboviral vaccine models involving NHPs to prove freedom from exposure [47][48][49][50]. They have proved to be reproducible and sensitive [51] in arboviral studies involving various NHPs in Africa, Asia Total 513 122 and the Americas [38][39][40][41][42][52][53][54]. The ELISAs are well suited for large-scale screening for previous infections in field studies because wild infected NHPs mount a robust antibody response although only viraemic for 1-7 days and show no obvious clinical signs [4]. ...
Article
Full-text available
Background Dengue, chikungunya and Zika viruses (DENV, CHIKV and ZIKV) are transmitted in sylvatic transmission cycles between non-human primates and forest (sylvan) mosquitoes in Africa and Asia. It remains unclear if sylvatic cycles exist or could establish themselves elsewhere and contribute to the epidemiology of these diseases. The Caribbean island of St. Kitts has a large African green monkey (AGM) (Chlorocebus aethiops sabaeus) population and is therefore ideally suited to investigate sylvatic cycles. Methods We tested 858 AGM sera by ELISA and PRNT for virus-specific antibodies and collected and identified 9704 potential arbovirus vector mosquitoes. Mosquitoes were homogenized in 513 pools for testing by viral isolation in cell culture and by multiplex RT-qPCR after RNA extraction to detect the presence of DENV, CHIKV and ZIKVs. DNA was extracted from 122 visibly blood-fed individual mosquitoes and a polymorphic region of the hydroxymethylbilane synthase gene (HMBS) was amplified by PCR to determine if mosquitoes had fed on AGMs or humans. Results All of the AGMs were negative for DENV, CHIKV or ZIKV antibodies. However, one AGM did have evidence of an undifferentiated Flavivirus infection. Similarly, DENV, CHIKV and ZIKV were not detected in any of the mosquito pools by PCR or culture. AGMs were not the source of any of the mosquito blood meals. Conclusion Sylvatic cycles involving AGMs and DENV, CHIKV and ZIKV do not currently exist on St. Kitts.
... In Sri Lanka, almost all the NHPs studied in Polonnaruwa in 1987 had neutralising antibody to DENV-2 ( Table 2) [56]. Further analysis revealed a highly focal epizootic of DENV had occurred in the population ( Table 2) [57], which was not associated with a concurrent human outbreak. On the island of Borneo (Malaysia), nearly a third of wild and semi-captive orangutans (Pongo pygmaeus) sampled were seropositive to DENV-2 ( ...
Article
Full-text available
Abstract Arboviruses infecting people primarily exist in urban transmission cycles involving urban mosquitoes in densely populated tropical regions. For dengue, chikungunya, Zika and yellow fever viruses, sylvatic (forest) transmission cycles also exist in some regions and involve non-human primates and forest-dwelling mosquitoes. Here we review the investigation methods and available data on sylvatic cycles involving non-human primates and dengue, chikungunya, Zika and yellow fever viruses in Africa, dengue viruses in Asia and yellow fever virus in the Americas. We also present current putative data that Mayaro, o’nyong’nyong, Oropouche, Spondweni and Lumbo viruses exist in sylvatic cycles.
... predominantly Aedes ageypti and Aedes albopictus. 1 Humans are the principal reservoir for DENV infections and the only host to develop clinical disease following natural infections. 2 DENV causes a self-limiting acute febrile illness known as dengue fever (DF) to the more severe dengue hemorrhagic fever (DHF) and dengue shock syndrome (DSS). 3 The latest study reports 390 million dengue infections per year, of which 96 million infections become apparently severe. ...
Article
Full-text available
Background Dengue is a global arboviral threat to humans; causing 390 million infections per year. The availability of safe and effective tetravalent dengue vaccine is a global requirement to prevent epidemics, morbidity, and mortality associated with it. Methods Five experimental groups (6 mice per group) each of 5-week-old BALB/c mice were immunized with vaccine and placebo (empty plasmid) (100 µg, i.m.) on days 0, 14 and 28. Among these, four groups (one group per serotype) of each were subsequently challenged 3 weeks after the last boost with dengue virus (DENV) serotypes 1-4 (100 LD50, 20 µl intracerebrally) to determine vaccine efficacy. The fifth group of each was used as a control. The PBS immunized group was used as mock control. Serum samples were collected before and after subsequent immunizations. EDIII fusion protein expression was determined by Western blot. Total protein concentration was measured by Bradford assay. Neutralizing antibodies were assessed by TCID50-CPE inhibition assay. Statistical analysis was performed using Stata/IC 10.1 software for Windows. One-way repeated measures ANOVA and Mann-Whitney test were used for neutralizing antibody analysis and vaccine efficacy, respectively. Results The recombinant EDIII fusion protein was expressed adequately in transfected 293T cells. Total protein concentration was almost 3 times more than the control. Vaccine candidate induced neutralizing antibodies against all four DENV serotypes with a notable increase after subsequent boosters. Vaccine efficacy was 83.3% (DENV-1, -3, -4) and 50% (DENV-2). Conclusion Our results suggest that vaccine is immunogenic and protective; however, further studies are required to improve the immunogenicity particularly against DENV-2.
Chapter
Dengue viral infection (DVI) is a debilitating arthropod-borne disease caused by four closely related strains of dengue viruses (DENV). The spectrum of disease varies from mild self-limiting illness to dengue fever (DF) to more severe and fulminating forms such as dengue hemorrhagic fever (DHF) and dengue shock syndrome (DSS). In recent years with the geographical spread of dengue and more involvement of adults, there have been increasing reports of DVI with unusual manifestations, termed as “expanded dengue syndrome” (EDS). These isolated organopathies include hepatic, renal, cardiac, respiratory, and neurological involvements and could be explained as complication of severe profound shock or associated with underlying host conditions or co-infections. This chapter describes the basics of DVI along with introduction and concept of EDS.
Article
Full-text available
Flaviviruses continue to cause globally relevant epidemics and have emerged or re-emerged in regions that were previously unaffected. Factors determining emergence of flaviviruses and continuing circulation in sylvatic cycles are incompletely understood. Here we identify potential sylvatic reservoirs of flaviviruses and characterize the macro-ecological traits common to known wildlife hosts to predict the risk of sylvatic flavivirus transmission among wildlife and identify regions that could be vulnerable to outbreaks. We evaluate variability in wildlife hosts for zoonotic flaviviruses and find that flaviviruses group together in distinct clusters with similar hosts. Models incorporating ecological and climatic variables as well as life history traits shared by flaviviruses predict new host species with similar host characteristics. The combination of vector distribution data with models for flavivirus hosts allows for prediction of global vulnerability to flaviviruses and provides potential targets for disease surveillance in animals and humans.
Article
Full-text available
Dengue virus, a RNA virus of family Flaviviradae is considered non-neurotropic. Increasing studies and case reports reveal neurological manifestations of dengue virus. In our case series, we have evaluated magnetic resonance imaging (MRI) findings of 3 patients with dengue fever diagnosed by positive dengue NS1 antigen with neurological symptoms, which revealed nonspecific imaging features of dengue encephalitis in two cases and dengue meningoencephalitis in one case. Autopsy findings are also correlated in 2 patients who succumbed to their disease. This case series underlines the consideration of dengue encephalitis in patients of dengue fever with neurological symptoms and relevant imaging findings.
Article
The aim of this study was to determine the prevalence of several viral antibodies in non-human primates from two zoological gardens from Venezuela. Only two out of 66 sera were positive for antibodies to dengue virus by hemagglutination inhibition. Six out of 62 sera exhibited antibodies against Hepatitis B virus (HBV) core antigen. Viral DNA was detected by nested PCR in one positive serum and in three negative without serological evidence of HBV infection. Sequence analysis showed the circulation of HBV genotype F, predominant in Venezuela. Antibodies against rotavirus antigens were found in 87% (20/23) of Old World primates and in 50% (13/26) of New World primates. Both the prevalence of antibodies and the mean O.D. value by ELISA were significantly lower in New World primate sera. These results suggest that non-human primates do not seem to represent an important reservoir for dengue virus infection, highly endemic in Venezuela. Anthropozoonotic infection of HBV seems to occur among primates not only from the Old but also from the New World, reinforcing the importance of vaccination of species at risk. This study also suggests a lower frequency of infection by rotavirus of non-human primates from the New World, compared to primates from the Old World.
Article
Full-text available
Outbreak of dengue fever in Chikalthana, Pimpalgaon and Waloor villages in Parbhani district of Maharashtra (India) were investigated. Clinically, the illness was typical of dengue fever except for the absence of maculopapular rash. A total of 42 acute, 14 late acute, 73 convalescent and 19 sera from contacts were collected. Of the 15 virus isolates, 12 were identified as dengue virus type 2 and 1 as dengue virus type 1. Serological tests confirmed the etiological role of dengue virus in the outbreak. House-to-house survey was carried out in Chikalthana and Pimpalgaon villages. Overall, 15.09 per cent of the surveyed population was affected during the outbreak and attack rate was higher at Pimpalgaon. A tendency of water storage was observed in the households and concomitant entomological studies proved Aedes aegypti breeding. Higher prevalence of dengue fever was noted among larger families and in families that had two or more patients, the commonest duration between the first and the last patient was often less than 5 days.
Article
In einer vorher veröffentlichten Arbeit (DITTUS, 1977a) wurden die Beziehungen zwischen Daten betreffend Verhalten, Ökologie und Demographie einer Population von etwa 450 Hutaffen, Macaca sinica, untersucht. Diese Daten weisen darauf hin, dass die Grössen von Trupps (und daher der Population) dem Nahrungsangebot angepasst sind, und durch Ausgleiche zwischen Geburtsraten und Mortalität entstehen. Mortalität wird durch Wettbewerb um Nahrung und weibliche Paarungspartner verursacht. Zugang zu diesen Gütern (oder Rang in der Dominanzhierarchie) und daher die sozial bedingte Mortalität unterscheiden sich je nach Alter, Geschlecht und Verwandtschaft. In dieser Weise bestimmt Sozialverhalten die Alters- und Geschlechtszusammenstellung und die Bevölkerungsdichte der Sozietät und der Population. Die Evolution dieser Sozialverhaltensweisen wurde in dieser Veröffentlichung be- sprochen. Unterschiede im Durchschnittsgrad der Verwandtschaft (r) zwischen Tieren unter- schiedlichen Alters und Geschlechtes innerhalb eines Trupps wurden entsprechend ihrer einzelnen Wanderungsraten geschätzt. Erwachsene Weibchen innerhalb eines Trupps sind miteinander und mit allen Säuglingen und Jungtieren verwandt. Die Evolution von Sozialverhalten, welches die Mortalität je nach Alter und Geschlecht unterschiedlich bestimmt, wird im Lichte von klassischer und verwandtschaftsbedingter Auslese be- sprochen. Unterwerfendes Verhalten wird durch dominante Tiere erzwungen, und wird unter klassischer Auslese allein als anpassungsfähig betrachtet. Der Verlust in Fitness durch eine untergeordnete Tat muss daher nicht durch einen gleichgrossen oder grösseren Gewinn von “inclusive fitness” kompensiert werden um sich in der Evolution zu ent- wickeln. Unter Verwandten gewinnt der Untergeordnete jedoch zusätzliche “inclusive fitness”. Unterwerfendes Verhalten zwischen Verwandten gilt als “erzwungene Selbst- losigkeit” (imposed altruism) oder “erzwungenes Verwandteninvestment”. Duldung von Nachkommen oder von untergeordneten Verwandten gelten als Elterninvestment und “willkürliches Verwandteninvestment”, in dieser Reihenfolge. Beziehungen zwischen dominanten und untergeordneten Tieren haben meistens einen Austausch von Futter oder anderen Gütern zur Folge, welche das Überleben und Fitness beeinflussen. Daher wird angenommen, dass der Grad in welchem Individuen unter- schiedlichen Alters und Geschlechtes sich gegenseitig drohen oder ausbeuten, dem Grad von Elterinvestment oder Verwandteninvestment zwischen ihnen direkt oder umgekehrt entspricht. Solches Investment ist dem Geschlechtsunterschied der Fort- pflanzungsstrategie angepasst, und dient dazu um maximale Fitness der Eltern und Verwandten zu gewinnen. Bevor Männchen die Geschlechtsreife und Fortpflanzungs- bereitschaft erreichen, zeigen sie grössere Ansprüche für ihr Wachstum und in ihrer Mortalität, besonders während der Pubertät nachdem sie den Muttertrupp ver- lassen haben, als weibliche Altersgenossen. Eltern und Verwandte vermeiden diesen möglichen Fitnessverlust durch ihre männlichen Nachkommen, indem sie mehr in Männchen vor dem Auswanderungsalter investieren, auf Kosten weiblicher Jungtiere.
Article
The heritability of quantitative traits, or the proportion of phenotypic variation due to additive genetic or heritable effects, plays an important role in determining the evolutionary response to natural selection. Most quantitative genetic studies are performed in the laboratory, due to difficulty in obtaining genealogical data in natural populations. Genealogies are known, however, from a unique 20-year study of toque macaques (Macaca sinica) at Polonnaruwa, Sri Lanka. Heritability in this natural population was, therefore, estimated. Twenty-seven body measurements representing the lengths and widths of the head, trunk, extremities, and tail were collected from 270 individuals. The sample included 172 offspring-mother pairs from 39 different matrilineal families. Heritabilities were estimated using traditional mother-offspring regression and maximum likelihood methods which utilize all genealogical relationships in the sample. On the common assumption that environmental (including social) factors affecting morphology were randomly distributed across families, all but two of the traits (25 of 27) were significantly heritable, with an average heritability of 0.51 for the mother-offspring analysis and 0.56 for the maximum likelihood analysis. Heritability estimates obtained from the two analyses were very similar. We conclude that the Polonnaruwa macaques exhibit a comparatively moderate to high level of heritability for body form.
Article
At Polonnaruwa, Sri Lanka, four out of 29, groups of toque macaques, Macaca sinica, divided in a period of 16 years. Temporary peripheral subgroups of varying sizes and compositions preceded fission by 9–40 months. Fission crystallized within a month through an increase and stabilization of subgroup membership and permanent division. All members in the newly seceded groups had been frequent participants in pre-fission subgroups, and belonged to subordinate matrilineages. Subgroups, and hence group divisions, were initiated by cores of mutually loyal females and occurred mostly along kinship lines. In the year of fissions, the rate of change in female dominance relations was significantly greater among groups that divided than among those that did not. It is hypothesized that low-ranking females secede to form new groups when the costs, especially of intragroup competition for food resources, outweigh the benefits of group membership. Such seceding females were easily available and familiar mates for group males that had recently lost rank. Final division, therefore, resulted from a coalition of subordinate females and males acting according to their respective interests. It was triggered in this population by rapid growth of some groups to large size and by environmental stress (the reduction and fragmentation of food resources caused by drought and a cyclone), which accentuated the costs of resource competition. Male aggression, such as infanticide, which negatively affects female fitness, might also have contributed to one group fission.
Article
By the last decade of the XXth century Aedes aegypti and the 4 dengue viruses had spread to nearly all countries of the tropical world. Some 2 billion persons live in dengue-endemic areas with tens of millions infected annually. Dengue pandemics were also documented in the XVIIIth and XIXth centuries; they were contained by organized anti-Aedes aegypti campaigns and urban improvements. The XXth century dengue pandemic has brought with it the simultaneous circulation of multiple serotypes and in its aftermath, endemic dengue haemorrhagic fever/dengue shock syndrome (DHF/DSS). Nearly 3 million children have been hospitalized with this syndrome in the past 3 decades, mainly in South-East Asia. Recent outbreaks of DHF/DSS in the Pacific Islands, China, India, Sri Lanka, Cuba and Venezuela are indicators of the high intensity and rapid spread of dengue transmission. The magnitude of the XXth century dengue pandemic requires urgent improvements in early warning surveillance by WHO Member States and the development of the capacity to study underlying mechanisms of the disease. A key research question is why does DHF/DSS not occur with all second dengue infections? Two answers have been suggested: (1) a human resistance gene. Data from the 1981 DHF/DSS epidemic in Cuba have demonstrated the existence in blacks of a resistance gene. The effect of such a gene in reducing disease susceptibility of American and African blacks requires more study. (2) The existence of dengue "biotypes". Some, but not all biotypes may cause DHF/DSS during a second dengue infection.(ABSTRACT TRUNCATED AT 250 WORDS)
Article
During the past several decades, dengue viruses have progressively extended their geographic distribution, and are currently some of the most important mosquito-borne viruses associated with human illness. Determining the genetic variability and transmission patterns of these RNA viruses is crucial in developing effective control strategies for the disease. Primer-extension sequencing of less than 3% of the dengue genome (across the E/NS1 gene junction) provided sufficient information for estimating genetic relationships among 40 dengue type 1 and 40 type 2 virus isolates from diverse geographic areas and hosts. A quantitative comparison of these 240-nucleotide-long sequences revealed previously unrecognized evolutionary relationships between disease outbreaks. Five distinct virus genotypic groups were detected for each of the two serotypes. The evolutionary rates of epidemic dengue viruses of types 1 and 2 were similar, although the transmission pathways of these viruses around the world are different. For dengue type 2, one genotypic group represents an isolated, forest virus cycle which seems to have evolved independently in West Africa. This is the first genetic evidence of the existence of a sylvatic cycle of dengue virus, which is clearly distinct from outbreak viruses.
In the late summer (rainy season) of 1987, a sharp outbreak of fever of unknown origin (FUO) in rural southern Thailand was investigated by a field epidemiology team. In a random survey of households, 40 percent of the children and 20 percent of adults were reported to have had febrile illnesses within the last month. There was at least one death, possibly from Reye's syndrome. Testing 34 pairs of acute and convalescent sera showed significant HI antibody titer rises to influenza A (Taiwan/(H1N1) (9 cases) and dengue virus (12 cases). Testing 79 single sera with the antibody capture ELISA test for dengue, revealed that 23 percent had high titers in the IgM serum fraction suggesting recent infection. There were also six antibody titer rises to coxsackie B viruses, three from well controls. Dengue has previously been observed as a cause of FUO in rural areas in the tropics, but finding a combined epidemic of dengue and influenza was unexpected. With cooperative villagers, adequate personnel and laboratory support, especially the antigen capture ELISA test for dengue infections, it is feasible to successfully investigate disease outbreaks with serologic methods in remote villages.