Content uploaded by W. B. Karesh
Author content
All content in this area was uploaded by W. B. Karesh
Content may be subject to copyright.
VECTOR BORNE AND ZOONOTIC DISEASES
Volume 2, Number 2, 2002
© Mary Ann Liebert, Inc.
Research Paper
The Impact of Ecological Conditions on the Prevalence
of Malaria Among Orangutans
NATHAN D. WOLFE,
1,
* WILLIAM B. KARESH,
2
ANNELISA M. KILBOURN,
2
JANET COX-SINGH,
3
EDWIN J. BOSI,
4
HASAN A. RAHMAN,
5
ADRIA TASSY PROSSER,
6
BALBIR SINGH,
3
MAHEDI ANDAU,
4
and ANDREW SPIELMAN
1
ABSTRACT
Contemporary human land use patterns have led to changes in orangutan ecology, such as the loss of habitat. One
management response to orangutan habitat loss is to relocate orangutans into regions of intact, protected habitat.
Young orangutans are also kept as pets and have at times been a valuable commodity in the illegal pet trade. In
response to this situation, government authorities have taken law enforcement action by removing these animals
from private hands and attempted to rehabilitate and release these orangutans. In relocating free-ranging orang-
utans, the animals are typically held isolated or with family members for
,
48 h and released, but during the course
of rehabilitation, orangutans often spend some time in captive and semicaptive group settings. Captive/semicap-
tive groups have a higher density of orangutans than wild populations, and differ in other ways that may influ-
ence susceptibility to infectious disease. In order to determine the impact of these ecological settings on malaria,
the prevalence of malaria was compared between 31 captive and semicaptive orangutans in a rehabilitation pro-
gram at the Sepilok Orangutan Rehabilitation Centre and 43 wild orangutans being moved in a translocation pro-
ject. The prevalence of malaria parasites, as determined by blood smear and Plasmodium genus-specific nested-
polymerase chain reaction, was greater in the captive/semicaptive population (29 of 31) than in the wild population
(5 of 43) even when accounting for age bias. This discrepancy is discussed in the context of population changes
associated with the management of orangutans in captive/semicaptive setting, in particular a 50-fold increase in
orangutan population density. The results provide an example of how an ecological change can influence pathogen
prevalence. Key Words: Orangutans—Malaria—Ecology—Plasmodium—Pongo pygmaeus.
Vector Borne Zoonotic
Dis. 2, 97–103.
97
INTRODUCTION
O
RANGUTANS ARE AMONG our closest living
relatives. They are the descendants of
African apes that migrated to Europe and Asia
roughly 12–18 million years ago (Andrews and
Cronin 1982), and are the only great apes out-
side of Africa. Two genetically distinct sub-
species are recognized,
Pongo pygmaeus abelii
and
Pongo pygmaeus pygmaeus
(Zhi et al. 1996).
1
Department of Immunology and Infectious Disease, Harvard School of Public Health, Boston, Massachusetts.
2
Field Veterinary Program, Wildlife Conservation Society, Bronx, New York.
3
Faculty of Medicine & Health Sciences, Universiti Malaysia Sarawak, Kota Samarahan, Sarawak, Malaysia.
4
Department of Wildlife, Wisma Muis, Kota Kinabalu, Sabah, Malaysia.
5
Department of Health, Kota Kinabalu, Sabah, Malaysia.
6
Department of Anthropology, Harvard University, Cambridge, Massachusetts.
*Present address: Cameroon Program, Johns Hopkins University, Bloomberg School of Public Health, Baltimore,
Maryland.
The two subspecies of orangutans are currently
found in Indonesia and Malaysia, with
P. pyg-
maeus abelii
found in northwestern Sumatra
and
P. pygmaeus pygmaeus
found in Sabah and
Sarawak (Malaysian Borneo) and Kalimantan
(Indonesian Borneo). Estimates of numbers in
the wild have ranged from 10,000 (Harrison
1970) to 10,000–30,000 (International Union for
Conservation of Nature and Natural Resources
1976); a reasonable current estimate may be
roughly 10,000 (Andau 1994). They exist natu-
rally at low population densities, about two in-
dividuals per square kilometer (Mitani et al.
1991), and are mainly frugivorous. Males are
generally solitary, while females often travel
with an infant and/or juvenile offspring, al-
though group size can temporarily fluctuate in
relation to fruit availability.
Changes in wildlife ecology, such as habitat
loss and demographic modification, can play
an important role in the health and the ultimate
survival of endangered wildlife. Nevertheless,
few studies have documented the impact of hu-
man-induced wildlife population or habitat
changes on specific pathogens. Examination of
wildlife populations both before and after an
ecological modification would be a highly rec-
ommended though often politically difficult
and costly endeavor. Cross-sectional studies
comparing pathogen prevalence between pop-
ulations frequently provide a more timely and
feasible alternative. While there have been no
malaria studies of this sort in apes, a study that
compared the prevalence of malaria in wild
and rehabilitated South African jackass pen-
guins found a significantly greater prevalence
of malaria among temporarily captive pen-
guins (Graczyk et al. 1995). If it is found that
rehabilitated or previously captive populations
have a greater prevalence of infectious diseases
than their wild counterparts, this could play an
important role in estimating the health risks of
wildlife rehabilitation or reintroduction into
regions with existing wild populations (Karesh
and Cook 1995).
Orangutans are known to be infected by at
least two malaria parasites. The first,
Plasmod-
ium pitheci
, was described by Halberstaedter
and von Prowazek (1907) from an orangutan at
the Berlin Zoo. In the late 1960s and early 1970s
two expeditions were conducted to document
the malaria parasites of orangutans in Borneo
(Coatney et al. 1971, Peters 1973). These pro-
jects reported the presence of
P. pitheci
and led
to the discovery of
Plasmodium silvaticum
(Gar-
nham et al. 1972).
Here we used cross-sectional sampling and
a malaria model to examine the impact of re-
habilitation on a Bornean orangutan (
P. pyg-
maeus pygmaeus
) population in the Sepilok
Orangutan Rehabilitation Centre (SORC) by
comparing it with free-ranging orangutans in
the same region.
MATERIALS AND METHODS
Subjects
During 1996 and 1997, wild orangutans were
translocated (relocated) as part of a conserva-
tion project conducted by the Sabah Wildlife
Department and the Wildlife Conservation So-
ciety. Wild orangutans were translocated from
forest fragments slated for clearing in Eastern
Sabah to the Tabin Wildlife Reserve (Andau et
al. 1994) (Fig. 1). All forest fragments were
within 100 km of the Tabin Wildlife Reserve,
which consists of a 1,600-km
2
tract of protected
primary and secondary lowland tropical rain
forest. Habitat types were similar at source and
destination.
For comparison, blood samples were col-
lected from captive and semicaptive orang-
utans at the SORC, located on the northern
edge of the Sepilok Forest Reserve, a 400-km
2
reserve of protected primary and secondary
forest outside of Sandakan, Sabah in the north-
eastern part of the state (Fig. 1). Captive and
semicaptive individuals at the SORC are con-
fiscated or obtained from other captive situa-
tions. Upon arriving at the SORC, orangutans
are held in cages for quarantine for 6 months.
After quarantine, they are held in a range of
conditions from captivity to completely free-
ranging, depending on their ability to care for
themselves. Orangutans at the SORC are pro-
vided with daily feedings of bananas and milk;
infants receive additional nutritional supple-
ments, and older animals forage in the sur-
rounding forest and may or may not return
daily for the food provisions and interact with
staff and other orangutans.
WOLFE ET AL.
98
Both wild and captive/semicaptive orang-
utans were placed in four age classes, based on
weight and dental morphology. From experi-
ence with captive/semicaptive individuals of
known age at the SORC, these categories cor-
respond to approximate age groups: infant,
,
1
year; juvenile, 1–5 years; subadult, 6–10 years;
and adult,
.
10 years.
Blood sample collection
Free-ranging orangutans were immobilized
and transported directly to the release site at
Tabin Wildlife Reserve unless medical condi-
tions warranted short-term holding prior to re-
lease. Immediately following initial capture,
2–15 mL of blood was collected from each
orangutan (for details see Wolfe et al. 2001).
The orangutans at the SORC had been present
for days to years and were either manually re-
strained or anesthetized at the time of this
study for examination and sample collection
using the same methods as for the wild group.
Thick and thin blood films were prepared from
each individual together with triplicate blood
spots, each between 20 and 50 mL, which were
spotted directly onto filter paper (Whatman 3
MM chromatography paper). Blood films were
stained using Giemsa stain. Blood spots on fil-
ter papers were allowed to air-dry and were
placed individually in paper envelopes and
stored at room temperature until processed.
Extraction and genus-specific nested-polymerase
chain reaction (PCR) assay for parasite detection
Blood spots were clipped from filter papers
using an ethanol-flamed paper punch and
DNA template extracted for nested-PCR using
the InstaGene™ extraction described by Cox-
Singh et al. (1997). Extracted DNA was subject
to a
Plasmodium
genus-specific nested-PCR as-
say of the 18S small subunit rRNA gene (Singh
et al. 1999).
RESULTS
A total of 74 orangutans were examined for
the presence of
Plasmodium
as part of this study.
Of these, 43 were wild individuals captured
and immediately sampled in 13 distinct forest
fragments in eastern Sabah. The remaining 31
samples were collected from captive and semi-
captive orangutans, solely from the SORC.
Of the 74 samples examined by microscopy,
ORANGUTANS AND MALARIA PREVALENCE
99
FIG. 1. Map of study sites in Sabah, Malaysia.
30 slides were found to be positive for malaria:
27 from the captive/semicaptive group and
three from the wild orangutan group (Table 1).
Positive slides were examined for characteris-
tics used to distinguish
P. pitheci
and
P. sil-
vaticum
, including notable enlargement of in-
fected erythrocytes in
P. silvaticum
infection
(Garnham et al. 1972). In the present sampling,
only
P. pitheci
infection was observed; there
was no evidence of
P. silvaticum
.
Of the 74 samples examined by microscopy a
subset of 38 was further subjected to a sensitive
genus-specific nested-PCR assay (Table 1). Of the
38 samples examined by both microscopy and
nested-PCR, 16 were positive by microscopy,
and 20 were positive by nested-PCR. Nested-
PCR confirmed all of the microscopy-positive re-
sults in this subset and detected an additional
four infected animals. Taking all positives (mi-
croscopy and the additional four PCR-positive
microscopy-negative results), 93.5% (29 of 31) of
the captive/semicaptive orangutans and 11.6%
(5 of 43) of the wild orangutans were infected
with malaria, showing that the captive/semi-
captive population was more frequently infected
(
G 5
55.19;
p ,
0.01).
Individuals were sampled from both sexes
and all age groups (Table 2). Wild or cap-
tive/semicaptive group status was indepen-
dent of sex (
G 5
0.56;
p .
0.05) but not inde-
pendent of age group (
G 5
13.83;
p ,
0.01). The
captive/semicaptive population had a greater
proportion of juveniles and lesser proportion
of adults than the wild population. Since the
more frequent infections of the captive/semi-
captive group could potentially be attributable
to the skewed age distribution, a comparison
was undertaken among only juveniles. Among
juveniles only, the captive/semicaptive popu-
lation was also more frequently infected (
G 5
14.70;
p ,
0.01) than the wild population.
DISCUSSION
The prevalence of malaria differed signifi-
cantly between captive/semicaptive orang-
utans and wild orangutans. The increased
rate of malaria in the captive/semicaptive
group remained significantly higher when
adjusted for sex and age disparity between
the two groups.
WOLFE ET AL.
100
TABLE 1. PREVALENCE OF MALARIA AMONG ORANGUTANS FOR SAMPLES
ANALYZED BY MICROSCOPY OR MICROSCOPY AND NESTED-PCR
Total positive by
Number examined microscopy and/or
Host by microscopy No PCR PCR(2) PCR(1) No PCR PCR(2) PCR(1) PCR
Semicaptive 31 2 0 2 12 0 15 29
Wild 43 20 18 2 2 0 1 5
Total 74 22 18 4 14 0 16 34
Nested-PCR was conducted on a subset of those samples examined by microscopy. No PCR indicates results for
samples that were not examined by PCR.
Microscopy-negative
Microscopy-positive
TABLE 2. NUMBER OF MALARIA-POSITIVES ACCORDING TO SEX AND AGE GROUP
Age of host
Host, sex Infant Juvenile Subadult Adult Total
Semicaptive (
n 5
31)
Female (0) 17 (17) 1 (1) 1 (1) 19 (19)
Male (0) 6 (7) 2 (2) 2 (3) 10 (12)
Wild (
n 5
43)
Female 1 (1) 1 (5) 0 (1) 2 (15) 4 (22)
Male 0 (2) 1 (3) 0 (2) 0 (14) 1 (21)
Total (
n 5
74) 1 (3) 25 (32) 3 (6) 5 (33) 34 (74)
Number in parentheses indicate number of individual orangutans examined.
Malaria detection was based on thick and
thin film microscopy, and a subset of samples
was included in a sensitive PCR assay. The ad-
ditional four positive results obtained by
nested-PCR were confirmed positive by re-
peated nested-PCR. The nested-PCR used here
can detect between 1 and 2 human malaria par-
asites/mL of blood compared with 20–30 para-
sites/mL by microscopy (Cox-Singh et al. 1997).
The results show that the nested-PCR assay
performed well in detecting orangutan malaria
and suggest its potential applicability to mon-
itoring other nonhuman primate malarias.
Orangutan rehabilitation projects are often
necessary to facilitate law enforcement opera-
tions. The process usually requires a period of
captivity and semicaptivity. Factors associated
with captive and semicaptive groups include:
increased host population density, greater
proximity to humans and human settlements,
decreased arboreality, decreased day ranges,
modified social structure and modified stress
levels, unnatural or modified ecological habi-
tat, and abnormal diets. A number of these
changes may be associated with increased
malaria transmission. It is possible that de-
creased arboreality and/or greater proximity
to human settlements may contribute to in-
creased transmission through the exposure to
new, more competent, and a higher abundance
of vectors. The construction and location of
SORC facilities, including drainage ditches,
rain gutters, and other sources of commonly
standing water, may contribute to higher vec-
tor abundance than found in some intact forest
areas where the wild orangutans sampled in
this study lived. This factor could be significant
in the findings presented here. Nevertheless,
wild orangutans move over large areas and
would undoubtedly spend time in swamp
forests, mangroves, or seasonally flooded ar-
eas. We did not speciate mosquito vectors pre-
sent in the various areas, or compare vector
abundance, but the observance of positive
malaria results from both sample populations
of orangutans indicates that both groups have
been exposed to an appropriate vector or vec-
tors.
In addition, a change in diet and/or stress
associated with capture may lead to decreased
immune function. Since chimpanzees and other
animals self-medicate in the wild (Clayton and
Wolfe 1993), the possibility that the low preva-
lence of malaria parasites in wild orangutans
may be influenced by consumption of anti-
malarial plants by captive/semicaptive ani-
mals should not be excluded, but this phe-
nomenon has not been documented or
suggested to occur in orangutans.
Perhaps the most striking difference between
these populations is in the orangutan popula-
tion density. Captive and semicaptive orang-
utans at the SORC have a population density
of
.
100/km
2
compared with the estimated two
individuals per square kilometer of wild pop-
ulations. Host population density does not nor-
mally weigh heavily in estimates of vectorial
capacity (VC), an epidemiological term de-
scribing the impact of various characteristics of
the vector such as host preference, longevity,
extrinsic incubation time, and abundance on
vector-borne disease transmission. There is
only a linear increase in VC associated with in-
creases in host density (Spielman and James
1990), but the magnitude of a 50-fold change in
host population density, as seen in the cap-
tive/semicaptive orangutan population, would
significantly influence VC. The over eightfold
increase in malaria prevalence of the cap-
tive/semicaptive population is consistent with
this expectation.
Previous studies based on microscopy found
a prevalence of
.
50% of captive/semicaptive
individuals at Sepilok infected with malaria
parasites [52.6% positive by Coatney et al.
(1971); 84.6% positive by Peters et al. (1976)],
results comparable with the high prevalence
(93.5%) seen in the present study. The mainte-
nance of a high prevalence of parasite rate in
the captive/semicaptive population may be
reconciled with VC; however, the maintenance
of a unique parasite in a host species with such
a low density as found in wild orangutan pop-
ulations is difficult to understand. One possi-
bility is the presence of a hypnozoite stage,
which allows for “true” relapse in this parasite
and allowing infections to persist in a single
species host population where vector host con-
tacts are infrequent. Another possibility is that
malaria parasites commonly thought of as
orangutan parasites may in fact be parasites
with a broader host range that includes orang-
ORANGUTANS AND MALARIA PREVALENCE
101
utans, a finding supported by recent studies
showing that orangutans can be naturally in-
fected with
Plasmodium inui
, a malaria parasite
traditionally associated with macaques (Wolfe
et al., manuscript submitted for publication).
The present study shows that human induced
changes in a species’ ecological environment
(habitat and habitat use, population densities or
demography, movement or migration patterns,
etc.) can increase the prevalence of a vector-
borne pathogen. Although the impact of malaria
on orangutan health is still unclear, the results
indicate that for at least one vector-borne dis-
ease, holding orangutans in captivity for reha-
bilitation poses increased risks. It may be that
captive/semicaptive groups have substantially
increased risk for infection by a number of mi-
croorganisms. A combination of factors, includ-
ing increased host density, proximity to both
natural and human reservoirs, and stress and
diet changes, are all likely to contribute. The pre-
sent study provides additional evidence that in-
fectious disease threats should be factored into
wildlife management efforts, particularly when
maintaining groups of animals occurs in regions
with existing wild populations or abnormally
high densities of animals are being maintained
for future release into distant established wild
populations. The addition of regular health sur-
veillance for malaria and other microorganisms
has the potential to identify such threats prior
to the initiation of activities that could serve to
introduce pathogens to new areas or popula-
tions.
ACKNOWLEDGMENTS
This paper is dedicated to the memory of Dr.
Annelisa Kilbourn. While conducting this re-
search we received support from colleagues in
the United States and Malaysia. Foremost, we
thank the Sabah Department of Wildlife and
Sabah Health Department for their valuable as-
sistance in all stages of this research. Staff of
the SORC and the Sandakan Vector Control Di-
vision provided valuable field support and
helpful comments. William Collins, Sam
Telford III, and Mak Joon Wah provided valu-
able comments and assistance and kindly re-
viewed positive blood films. Helpful com-
ments and advice were provided by Ananias
Escalante, Altaf Lal, Richard Levins, Willy
Piessens, and Mary Wilson, and two anony-
mous reviewers for this journal. Fieldwork was
supported by grants to N.D.W. from the Taplin
Fellowship, National Science Foundation
Graduate Research Fellowship, Fulbright
Grant, Uwe Brinkmann Traveling Fellowship,
and Fredrick Sheldon Traveling Fellowship.
Fieldwork for A.M.K. was supported in part by
Mr. and Mrs. Renke Thye and the Morris Ani-
mal Foundation. This research was also sup-
ported in part by grant K01 TW00003-01 from
the Fogarty International Center, National In-
stitutes of Health to N.D.W.
ABBREVIATIONS
PCR, polymerase chain reaction; SORC, Sepi-
lok Orangutan Rehabilitation Centre; VC, vec-
torial capacity.
REFERENCES
Andau, PM. Orangutan survival programme Indonesia.
In: Ogden, J, Perkins, L, Sheeran, L, eds.
Proceedings of
the International Conference on “Orangutans: The Neglected
Ape.
” Fullerton, CA: Zoological Society of San Diego,
1994.
Andau, DD, Hiong, LK, Sale, JB. Translocation of pock-
eted orangutans in Sabah. Oryx 1994; 28:263–268.
Andrews, P, Cronin, JE. The relationships of Sivapithecus
and Ramapithecus and the evolution of the orang-utan.
Nature 1982; 297:541–546.
Clayton, DH, Wolfe, ND. The adaptive significance of
self-medication. Trends Evol Ecol 1993; 6:60–63.
Coatney, GR, Collins, WE, Warren, M, Contacos, PG.
The
Primate Malarias
. Bethesda, MD: U.S. Department of
Health, Education, and Welfare, 1971.
Cox-Singh, J, Mahayet, S, Abdullah, MS, Singh, B. In-
creased sensitivity of malaria detection by nested poly-
merase chain reaction using simple sampling and DNA
extraction. Int J Parasitol 1997; 27:1575–1577.
Garnham, PCC, Rajapaksa, N, Peters, W, Killick-Ken-
drick, R. Malaria parasites of the orang-utan,
Pongo pyg-
maeus
. Ann Trop Med Parasitol 1972; 66:287–294.
Graczyk, TK, Brossy, JJ, Plos, A, Stoskopf, MK. Avian
malaria seroprevalence in jackass penguins (
Spheniscus
demersus
) in South Africa. J Parasitol 1995; 81:703–707.
Halberstaedter L, von Prowazek, S. Untersuchungen über
die Malariaparasiten der Affen. Arb K GesundhAmt
1907; 26:37–43.
Harrison, T. The orang-utan situation. Biol Conserv 1970;
3:45–46.
WOLFE ET AL.
102
International Union for Conservation of Nature and Nat-
ural Resources.
Red Data Book
:
Update Sheets
. Morges,
Switzerland: International Union for Conservation of
Nature and Natural Resources, 1976.
Karesh, WB, Cook, RA. Applications of veterinary medi-
cine to in situ conservation efforts. Oryx 1995; 29:244–252.
Mitani, JC, Grether, GF, Rodman, PS, Priatna, D. Associ-
ations among wild orang-utans: sociality, passive ag-
gregations or chance? Anim Behav 1991; 42:33–46.
Peters, W. Blood films from patients with benign tertian
malaria in Eastern Sabah: is this a zoonosis? Trans R
Soc Trop Med Hyg 1973; 67:2.
Peters, W, Garnham, PCC, Rajapaksa, N, Cheong, WH, et
al. Malaria of the orang-utan in Borneo. Philos Trans R
Soc Lond [Biol] 1976; 275:439–482.
Singh, B, Bobogare, A, Cox-Singh, J, Snounou, G, et al. A
genus- and species-specific nested polymerase chain re-
action malaria detection assay for epidemiologic stud-
ies. Am J Trop Med Hyg 1999; 60:687–692.
Spielman, A, James, AA. Transmission of vector-borne
disease. In: Warren, KS, Mahmoud, AAF, eds.
Tropical
and Geographic Medicine
, 2nd edit. New York: McGraw-
Hill, 1990:146–158.
Wolfe, ND, Kilbourn, AM, Karesh, WB, Rahman, HA, et
al. Sylvatic transmission of arboviruses among Bornean
orangutans. Am J Trop Med Hyg 2001; 64:310–316.
Zhi, L, Karesh, WB, Janczewski, DN, Frazier-Taylor, H, et
al. Genomic differentiation among natural populations
of orang-utan (
Pongo pygmaeus
). Curr Biol 1996;
6:1326–1336.
Address reprint requests to:
Dr. Nathan D. Wolfe
Cameroon Program
Johns Hopkins University
Bloomberg School of Public Health
624 North Broadway #217
Baltimore, MD 21205
E-mail:
nwolfe@jhsph.edu
ORANGUTANS AND MALARIA PREVALENCE
103
This article has been cited by:
1. Michael E. Steiper, Nathan D. Wolfe, William B. Karesh, Annelisa M. Kilbourn, Edwin J. Bosi, Maryellen
Ruvolo. 2005. The Population Genetics of the α-2 Globin Locus of Orangutans (Pongo pygmaeus). Journal
of Molecular Evolution 60:3, 400. [CrossRef]
2. Yuji TAKENOSHITA. 2005. Infectious Diseases Problem in Great Ape Conservation. Primate Research
21:1, 47. [CrossRef]
3. Noko KUZE. 2004. The Present State of Research and Conservation of Orangutans in Sabah, Malaysia.
Primate Research 20:1, 77. [CrossRef]