High Annual and Seasonal Variations in Malaria Transmission by Anophelines and Vector Species Composition in Dielmo, a Holoendemic Area in Senegal

Abstract

We conducted a three-year entomologic study in Dielmo, a village of 250 inhabitants in a holoendemic area for malaria in Senegal. Anophelines were captured on human bait and by pyrethrum spray collections. The mosquitoes belonging to the Anopheles gambiae complex were identified using the polymerase chain reaction. Malaria vectors captured were An. funestus, An. arabiensis, and An. gambiae. Anopheles funestus was the most abundant mosquito captured the first year, An. arabiensis in the following years. The annual entomologic inoculation rates calculated by enzyme-linked immunosorbent assay were 238, 89, and 150 for the first, second, and third years, respectively. Each year there was a peak of transmission at the end of the rainy season, but transmission occurred year round. The heterogeneity of transmission was found at four different levels: 1) the relative vector proportion according to the place and method of capture, 2) the human biting rate and relative proportion of vectors by month and year, 3) the infection rate of each vector by year, and 4) the number of infected bites for all vectors, and for each species, for the year. Our data show that even in areas of intense and perennial transmission, there exist large longitudinal variations and strong heterogeneity in entomologic parameters of malaria transmission. It is important to take these into account for the study of the variations in clinical and biological parameters of human malaria, and to evaluate this relationship, a very thorough investigation of transmission is necessary.

Full text

Am.
J.
Trop.
Med.
Hyg.,
56(3),
1997,
pp.
247-253
Copyright
8
1997
by The American Society
of
Tropical Medicine and Hygiene
HIGH ANNUAL AND SEASONAL VARIATIONS
IN
MALARIA TRANSMISSION
BY
ANOPHELINES AND VECTOR SPECIES COMPOSITION
IN
DIELMO, A
HOLOENDEMIC AREA
IN
SENEGAL
e,-*,
,%
DIDIER FONTENILLE, LAURENCE LOCHOUARN, NAFISSATOU DIAGNE,
CHEIKH
SOKHNA,
JEAN-JACQUES LEMASSON, MATHURIN DIATTA, LASSANA KONATE,
FARBA FAYE, CHRISTOPHE
ROGER,
AND
JEAN-FRANCOIS TRAPE
Laboratoire de Zoologie Medicale, Institut Francais de Recherche ScientiJique pour le Developpeinent en Cooperation
(ORSTOM), Dakar, Senegal; Institut Pasteur, Dakar, Senegal; Departement de Biologie Animale, Universite Cheikh Anta Diop,
Dakar, Senegal
Abstract.
We
conducted a three-year entomologic study in Dielmo, a village of 250 inhabitants in a holoendemic
area for malaria in Senegal. Anophelines were captured
on
human bait and by pyrethrum spray collections. The
mosquitoes belonging to the
Anopheles gambiae
complex were identified using the polymerase chain reaction. Malaria
vectors captured were
An. funestus,
An.
arabiensis,
and
An.
gambiae. Anopheles funestus
was the most abundant
mosquito captured the first year,
An. arabiensis
in the following years. The annual entomologic inoculation rates
calculated by enzyme-linked immunosorbent assay were 238, 89, and
150
for the first, second, and third years,
respectively. Each year there was a peak of transmission at the end of the rainy season, but transmission occurred
year round. The heterogeneity of transmission was found at four different levels: 1) the relative vector proportion
according to the place and method of capture, 2) the human biting rate and relative proportion of vectors by month
and year,
3)
the infection rate of each vector by year, and 4) the number of infected bites for
all
vectors, and for
each species, for the year. Our data show that even in areas of intense and perennial transmission, there exist large
longitudinal variations and strong heterogeneity in entomologic parameters of malaria transmission. It is important to
take these into account for the study of the variations
in
clinical and biological parameters of human malaria, and to
evaluate this relationship, a very thorough investigation of transmission is necessary.
The interpretation
of
malaria parameters such as parasit-
emia, morbidity, mortality, and associated immune responses
depends
on
having precise information and close follow-up
of variations in malaria transmi~sion.'~~ It is very well known
that the transmission of malaria
in
Africa is not homoge-
neo~s.~ The vector species and density, the
Plasmodium
spe-
cies, the number of infective bites per human per year (also
called the annual entomologic inoculation rate
[EX]),
and
the monthly
Em,
are changeable. Many studies have com-
pared transmission between villages in the same area:-8 but
few have shown that significant differences may also occur
within the same location over a several year follow-up study.
Such variations have to be taken into account in longitudinal
studies
on
the development
of
malaria immunity, but
so
far
these have been studied very little.
A
longitudinal study be-
gan in 1990 to evaluate malaria infections and the mecha-
nisms of protective immunity in a population living
in
Diel-
mo, a village in a holoendemic area of Senegal. During a
four-month follow-up of the entire population conducted
during the 1990 rainy season, the cumulative prevalence of
P.
falciparum,
P.
malariae,
and
P.
ovale
were 98.6%,
50.5%,
and 40.3%, respe~tively.~ The preliminary studies
showed that the transmission was continuous throughout the
year
and
that the vectors were
Anopheles funestus
and
mos-
quitoes of the
An.
gambiae
complex.'0*'' Three species of
this complex were noted in the study area:
An. gambiae, An.
arabiensis,
and
An. melas.
Malaria transmission by these different vectors was stud-
ied from 1992 to 1995 using the polymerase chain reaction
(PCR),
which identifies the species of the
An.
gambiae
com-
plex.'*
In
this
paper, we have investigated the relative fre-
quencies of the different vector species
(An.
funestus,
An.
gambiae,
An.
arabiensis,
andAn.
melas)
collected within
the
study area according to the time, season, place, kd method
of capture, and
the
role of each vector in the transmission
of the different
Plasmodium
species during a three-year
lon-
gitudinal survey.
This
is the first study
in
Africa that pro-
vides data describing variation
of
the
EGR
according to
An.
gambiae
complex species.
MATERIALS
AND
METHODS
Study site.
The village of Dielmo (13"45'N, 16'25'W) is
situated in
an
area of Sudan-type savanna in the Sine-Saloum
region of Senegal, 280
km
southeast of Dakar and approx-
imately 15
km
north
of the Gambian border Figure
1).
The
rainy season lasts from June to mid-October. Over the last
20
years, the average annual rainfall has been approximately
700
mm.
It was
583
mm,
721
m,
and 657
mm
in 1992,
1993,. and 1994, respectively. Dielmo is situated
on
the
marshy bank of a
small
permanent stream that permits the
persistence of anopheline larval development sites year
round.
A
population of 250 inhabitants live
in
the village.
The study site was described in detail in a previous arti~le.~
Mosquito
collections.
Adult mosquitoes were collected
monthly from April 1992
to
March 1995 using the following
methods. The first was hourly outdoor and indoor human
bait catches from
7:OO
PM
to
7:OO
AM.
Captures were con-
ducted by the same male, adult volunteers for 12-18 person-
nights each month, with half of the volunteers staying out-
doors and half staying indoors, always at the same locations
in the village. The second was pyrethrum spray collections
in selected bedrooms and in storehouses. The human biting
rate
(HJ3R),
which is the number of mosquito bites per per-
son
per night, was calculated as the number of mosquitoes
captured
on
human bait during the month divided by the
number of person-nights.
Anopheline identification.
The identification of the ano-

c
'
248'
FONTENILLE AND
OTHERS
Saheb-sudanian domain
Isohyet
FIGURE
1.
Map
of
Senegal showing the
study
area.
phelines was made in the field following the GiIlies and
DeMeillon morphologic identification keys.I3 The mosqui-
toes of the
An. gambiae
complex were stored for further
identification by PCR
in
Dakar.
Field processing
of
anophelines.
The salivary glands of
mosquitoes collected from April 1992 to March 1994 on
volunteers were dissected and examined for malaria sporo-
zoites. When more than 40 anophelines were caught in an
hour, only 40 were randomly selected for dissection. Due
to
the lack of time, mosquitoes collected from April 1994 to
March 1995 were not dissected.
All
the anophelines caught
(dissected or not) were stored in 1.5-ml tubes with desiccant
for further laboratory analysis. These tubes were stored at
-20°C in Dakar.
Laboratory processing of anophelines.
The heads and
thoraces of
all
anopheline specimens were tested for circum-
sporozoite
(CS)
protein of
P.
falciparum,
P.
malariae,
and
P.
ovale
using an enzyme-linked immunosorbent assay
(ELI-
SA) described by Burkot and others14 and modified by Wirtz
and others,I5 and the
CS
protein rates were calculated.
Plas-
modium vivar
is not present in the area. The EJR was cal-
culated by multiplying the HBR calculated monthly by the
monthly CS protein rate (also referred to
in
this
article as
the infection rate). The mosquitoes belonging to the
An.
gambiae
complex 'were identified using the PCR technique
described by Scott and others, with minor modifications.16
Briefly, DNA was extracted from mosquito legs in the Dakar
laboratory, and the reactions were performed in
50
pl
of
PCR mixture using
An. gambiae
complex ribosomal DNA
intergenic spacer species-diagnostic primers. Some speci-
mens were processed by putting one leg directly
in
the re-
action mixture without extraction of the DNA. The length
of the amplified sequences was 315 nucleotides for
An. ar-
abiensis,
390 for
An. gambiae,
and 464 for
An. melas.
For
technical reasons, if more than 30 anophelines from the
An.
gambiae
complex were caught each month using each meth-
od (outdoor human bait, indoor human bait, bedroom-resting
mosquitoes, and storehouse resting mosquitoes), only 30-50
were randomly selected for each of the methods for pro-
cessing.
If
less than
30
specimens were caught,
all
were
tested. Thus, the likely number of individuals per species
captured by method each month was calculated by extrap-
olation. Also,
all
mosquitoes found positive after salivary
gland dissection or CS protein ELISA were processed using.
the PCR.
RESULTS
Captures
of
mosquitoes.
From April 1992 to March
1995, 17,370 anophelines belonging to malaria vector spe-
cies were collected in resting sites and during 470 person-
night of captures on human volunteers. A total of 5,744
An.
funestus
and 11,626 anophelines from the
An. gambiae
com-
plex were collected, of which 2,337 were processed by the
PCR.
Anopheles arabiensis, An. gambiae,
and
An. melas
were present in the village. Nonvector species, such as
An.
coustani, An. freetownensis, An. pharoensis, An. &ìpes, An.
ziemanni,
and species from
Aedes,
Culex
and
Mansonia
gen-
era were also captured. The percentage of females by species
captured by each method is presented in Table 1 using ex-
trapolation for specimens belonging to the
An.
gambiae
com-
plex. Approximately the same number of anophelines vec-
tors were captured
on
humans indoors and outdoors.
Anopheles arabiensis
accounted for the majority (56.1%)
of the mosquitoes collected and was the species collected
most on human bait.
Anopheles arabiensis
was more exo-
phagic than
An. gambiae
and
An. funestus.
It was also more
exophilic:
59%
of the vector females caught feeding inside
on
human bait were
An. arabiensis
versus only 26.9%
of
the
females resting in bedrooms and caught by indoor spraying.
Anopheles funestus,
which represented 33.1% of the total
captures, was the most frequent species among the endo-
philic vectors captured, particularly
in
bedrooms.
Anopheles
gambiae
represented only
10.8%
of the total vectors cap-
tured. Only 12
An. melas
were caught, of which only one
was caught on human bait.

VARIATIONS IN MALARIA TRANSMISSION
IN
SENEGAL
249
TABLE
1
Number and percentage
of
malaria vectors caught
from
April
1992
to March
1995
by different methods in Dielmo, Senegal*
Feeding? Resting
Outdoors Indoors Bedrooms Storehouses Total
No.
of
An. funestus
captured
No.
of
An. gambiae
S.I.
captured
Total number of vectors captured
No.
of
An. gambiae
s.1. PCR-tested
%
An. ftcnestits
%
An. gambiae
%
An. arabiensis
%
An.
melas
Total
%
1,783
5,286
7,069
898
25.2
9.4
65.4
O
,100
2,099
5,039
7,138
773
29.4
11.6
59.0
o.o*
100
1,630
930
2,560
46
1
63.7
9.2
26.9
0.2
100
232
371
603
205
38.5
24.4
36.3
0.8
1
O0
5,744
1
1,626
17,370
2,337
33.1
10.8
56.1
o.
1
1
O0
*An.
=
Anopheles:
PCR
=
polymerase chain
reaction.
f
Caught
on
human bait.
t
Only
one
female
was captured
on
human bait indoors
(0.01%).
Variation
of
the
HBR.
The number and the relative pro-
portion of each of the vector species strongly varied during
the three year follow-up study (Figure 2). Among captures
on human bait,
An. funestus
was much more abundant than
the other vectors during the first year, while
An. arabiensis
dominated the two following years.
Anopheles gambiae
was
the least captured vector each year.
Figure 3 shows the variations of the
HBR,
rainfall, and
temperature for the three-year period.
Anopheles arabiensis
was captured during all 36 months, but showed a peak of
abundance each year from June
to
October. It was very
abundant in 1993 and 1994, with its
HBR
reaching 103 bites
per person per night in July 1993 and 164
in
August 1994.
Unlike
An. arabiensis,
more
An.
gambiae
females were cap-
tured the first year. The maximum
HBR
for this species was
observed in September 1992 with 25 bites per person per
night. Despite the presence of apparently favorable larval
development sites, this species was rare or absent during
each dry season (December
to
June).
Anopheles $mestus
was present throughout the survey, but
showed great variation. It had a classic peak of abundance
each year during the dry season (December to March).13
However, the first year of
the
survey it was
also
very abun-
dant in July and August, reaching 50 bites per person per
night in July 1992. This species became less common in the
second year but the number of specimens captured increased
again at the end of the 1994 dry season.
Vector infection rates.
The
CS
protein rate was calcu-
lated for each species each month. The
ELISA
method de-
tected
l
.9 times more Plasmodium-positive mosquitoes than
did dissection.
Plasmodium falcipanim
was found in 92% of
the positive mosquitoes,
P.
malariae
in
7.3%,
and
P.
ovale
in 1.7%. One
An.
arabiensis
was positive for
P.
falciparum
and
P.
malariae,
and one was positive for
P.
ovale
and
P.
malariae.
The infection rate for each species varied greatly,
not only according to the season, as it classically does, but
also according to the year.
Each year
An.
funestus
had the highest infection rate for
P.
falciparum
(Table 2). Its mean rate calculated over the
three years was 2.62%; it decreased by a factor of 2.5 be-
tween the first and the third year. These differences are high-
ly significant
(P
=
0.001,
by
xz
test). The mean rate for
An.
gambiae
was 1.46%. It was higher the first year and lower
the second year.
Anopheles arabiensis
had a mean rate of
0.6% and decreased between the first and the third year.
However, for
An. gambiae
and
An. arabiensis
the variations
in the infection rate observed, depending on the year, were
not significant
(P
=
0.18
and
P
=
0.73,
respectively).
Variation
of
the
EIR.
The
P.
falciparum
total
EIR,
cal-
culated as described in the Materials and Methods, for the
April 1992
-
March 1993
April 1993
-
March 1994
3399
mosquitoes
3923
mosquitoes
6885
mosquitoes
April 1994
-
March 1995
FIGURE
2.
Relative proportions
of
the
three
vectors collected by outdoor and indoor human bait catches during a three-year period in
Dielmo, Senegal.

FONTENILLE AND OTHERS
Rainfall
Minimum temp.
-
Maximum temp.
AMJ JASONDJFMAMJJASOND JFMAMJJASONDJPM
25
20
15
10
5
O
AMJ
J
AS
ONDJFMAMJJ AS ONDJ FMAMJ JASONDJ
FM
2:
50
45
40
35
30
25
20
15
10
5
O
AMJ
J
AS ONDJFMAMJ
J
ASOND JFMAMJ JAS OND
J
FM
1992 1993 1994 1995
Human biting
rate
O
R,
45
CIO
35
30
25
20
15
10
5
O
-
Monthly entomologic inoculation rate
FIGURE
3.
Rainfall, temperature (temp.), monthly human biting rate, and monthly entomologic inoculation rate calculated by enzyme-linked
immunosorbent assay for all three
Plasmodium
species for each vector from April
1992
to March 1995
in
Dielmo, Senegal.

VARIATIONS IN MALARIA TRANSMISSION
IN
SENEGAL
25
1
TABLE
2
Infection rate, calculated by enzyme-linked immunosorbent assay from the head-thoraces
of
mosquitoes captured
on
human bait,
for
each
Plasmodium
species and each vector species by year*
First year Second year Third year
tested
P$ P.m. P.O.
tested
PJ
P.m.
P.O.
tested
P$
P.m.
P.O.
No.
No. No.
An.
gambiae
618
2.10 0.16
O
509
0.78
O
O
314 1.27
0.32
O
An.
arabiensis
797
0.75
0.13
0.13 2,993 0.63
0.10
0.03
4,410 0.54 0.09
O
An.
fiinestus
1,977
3.49 0.1
0.05
363
2.20
O
O
1,277 1.40 0.08
O
*
PJ
=
infection rate for
P.
falciponrm;
P.m.
=
infection rate for
P.
malariae;
P.O.
=
infection rate
for
P.
ovale; An.
=
Anopheles.
three-year period was 437 infected bites per person: 223 the
first year, 79 the second year, and 135 the third year. Despite
the low infection rate for
P.
malariae
and
P.
ovale,
their
mean annual
EIRs
were 10.8 and 2.5 infected bites per per-
son, respectively (Table
3).
Table
4
shows the total
ER
for
P.
falciparum
for each of the 12 quarters, and the confidence
interval calculated following the Poisson di~tribution.'~ The
maximum
ElR
observed was approximately 95 (95% con-
fidence interval
=
69-133) infected bites during the rainy
season from July to September 1992, while the
EIR
was
O
from April to June 1994 during the dry season when no
ELISA-positive mosquitoes were found. For each season,
large variations in the EIR were observed during the díffer-
ent years.
Figure
3
shows the variation in the
EIR
by month for each
of the three vectors. The
EIR
was calculated by ELISA for
each vector species. During the first year, transmission was
higher, reaching 53 infected bites per person in July 1992.
The comparative role
of
each vector according to each
month is clearly identified. The first year,
An. funestus
was
the main vector, accounting for 76% of all transmission, with
180 infected bites per person. At the end of the rainy season,
An. gambiae
also played a prominent role, whereas
An. ar-
abiensis
was only a secondary vector. In contrast, during the
following two years,
An. arabiensis
was responsible for the
most transmission, sometimes with sudden variations from
one month to another, as in June and July 1994, when the
monthly
EIR
for
An. arabiensis
increased from
O
to 44 (Fig-
ure 3).
DISCUSSION
This study shows the high annual and seasonal variations
of malaria transmission in Dielmo. Mosquitoes were present
throughout the year because of the permanent stream and
the presence of different kinds of anopheline larval devel-
opment sites. One aim of the.study was to leam as accurately
as possible the rate of transmission of malaria by the differ-
ent vectors to determine the relationship between the varia-
tions in transmission and the variations in clinical incidence,
parasitemia, and immunologic parameters. Great care must
be taken in the analysis of these variations in clinical and
biological data due to the great variation in transmission
found at four different levels: 1) the relative vector frequen-
cy according to the place and method of capture,
2)
the
abundance and relative frequency of vectors according to the
month, as is classically observed, but also according to the
year,
3)
the infection rate of each vector by year, and
4)
the
number of infected bites for all vectors, and for each species,
for the year.
The identification of all mosquito specimens captured,
particularly those caught on humans, is necessary. Since
1990, a PCR that identifies five species of the
An. gambiae
complex from individual dried mosquitoes has been
used.I2.
l6
This technique makes it possible to determine the
species of every specimen collected, particularly those
caught on humans, and not just the half-gravid ones, as is
the case with cytogenetic methods.
Prior
to this study, we
demonstrated that the PCR technique gave the same identi-
fication as cytogenetics on West African mosquitoes of the
An. gambiae
complex.I8 Previous cytotaxonomic studies
have shown two main cytotypes in the
An. gambiae
speci-
mens
of
this region: savanna and
bis sa^.'^
Comparative studies on transmission using only cytoge-
netics
on
half-gravid females captured as indoor resting mos-
quitoes for the determination of specimens of the
An. gam-
biae
complex are often biased because this type of sample
leads to an overrepresentation of
An. gambiae,
which is well
known for its endophilic behavior, as our results clearly
show. Among specimens from the
An. gambiae
complex,
only 14.4% of the mosquitoes captured on humans were
An.
gambiae,
whereas this rate was
25.4%
and 39.6% for mos-
quitoes collected by spraying
in
bedrooms and storehouses,
respectively.
There are few studies on the transmission of malaria that
have analyzed differences in transmission according to the
TABLE
3
Annual entomologic inoculation rate calculated by enzyme-linked immunosorbent assay foi the three
Plasmodium
species by vector species*
First year Second year Third year
Bf.
P.m.
P.O.
Bf.
Rm.
BO. Bf.
P.m.
P.O.
An.
gambiae
32.5
2.5
O
10.1
O
O
8.4
1.7
O
An.
arabiensis
17.5
2.5 2.5
48.2 7.6 2.6
81.0 10.7
O
An.
funestus
172.8
5
2.5
20.2
O
O
45.8
2.5
O
Total
222.8
10
5
78.5 7.6 2.6 135.2 14.9
O
*
PJ
=
P.
falcipancm: P.m.
=
P.
malariae;
P.O.
=
P.
ovale; An.
=
Anopheles.

FONTENILLE AND OTHERS
252
TABLE
4
Entomologic inoculation rate
for
Plasmodium falciparum
calculated by enzyme-linked immunosorbent assay with
95%
confidence intervals
(values in parentheses), by quarter from April
1992
to March
1995
in Dielmo, Senegal
April
to
June
July
to
Sept
Oct
to
Dec
Jan
to
March
Total
April
1992
to March
1993
27.5 95.3 47.5 52.5 222.8
April
1993
to March
1994
5.2
23.1
40.
I
10.1 78.5
April
1994
to March
1995
O
68.8 25.7 40.7 135.2
(14-50)
(69-133) (29-75)
(33-80) (179-274)
(1-18)
(1
1-44) (23-66)
(3-26) (53-1
11)
(0-9) (49-93)
(
15-43)
(23-65) (106-173)
month by the different vector species.
In
Africa. the only
published work
is
that of Taylor and others,2O but the deter-
mination of the species of the
An. gambiae
complex was
done only
on
anophelines caught resting indoors by radio-
labeled DNA probes. In our study, only 12
An. melas
were
caught. The larval development sites of this halophilic and
zoophilic species are located near the mangrove forest
6
km
west of Dielmo, and probably very few specimens reached
the village. The infection rate was calculated by ELISA be-
cause this method allowed the processing of
aIl
mosquitoes
and the comparison of the results using the same supplies
and technique. To limit the overestimation of the sporozoite
index by
this
method, only the head-thoraces were tested.
While dissection in the field was done by different people
and sensitivity might have varied, the results of the ELISA
always remained the same throughout the survey and made
possible the comparison from month to month.
In
our study,
the ELISA detected 1.9 times more positive mosquitoes than
dissection.
This
higher value is similar to the results of other
studies.8V2'
Using
a
standardized protocol for collecting and analyzing
mosquitoes for three years, we observed very large varia-
tions in malaria transmission in Dielmo. As is classically
observed in Africa, the EIR for the three vectors varied
greatly according to the month, with a peak of transmission
during and at the end of the rainy season from July to Sep-
tember. More surprising were the great variations in the en-
tomologic components of transmission, such as the HBR and
the mosquito infection rate, as well as the number and rel-
ative proportion of the three vectors over the three years.
From April to June 1994,
no
mosquitoes carrying
CS
pro-
tein were found in 179 captured, whereas two years before
during the same season, 28 infected bites were recorded.
Likewise, transmission was higher from January to March
1993 in the dry season than during the following rainy sea-
son
from July to September. These data show that in Dielmo,
where larval development sites
are
permanent, there
is
no
justification for dividing the year into only two seasons (dry
and rainy) to assess transmission and to correlate it with
clinical and immunologic data because there
are
too many
variations between the years.
The role of each vector species was different for each of
the three years. The first year,
An.
funestus
had a much great-
er effect
on
transmission than
An. gainbiae
and
An.
arabien-
sis.
This was due
to
two
factors: a higher HBR and a higher
mosquito infection rate due to this species longer life ex-
pectancy and its higher anthropophilic rate (Fontenille D,
unpublished data). Its role in transmission in Dielmo was
particularly significant since it was the main vector during
the dry season and it ensured continuous transmission
throughout the year. For these reasons, this species probably
plays an important role in maintaining immunity in Dielmo.
In the second and the third years,
An. arabiensis,
which is
generally considered to be a less efficient vector, was the
major vector due to its very high HBR, despite its low in-
fection rate.
Anopheles arabiensis
accounted for lo%,
66%,
and 61%, respectively, of the transmission over the three-
year period. Transmission by
An. gainbiae
was the lowest
except in the
first
year, during which its
ER
was twice that
of
An. arabiensis.
Over the three-year period,
An.
gambiae
was responsible for only 12% of the transmission.
These data differ from those obtained in other regions of
West Africa, where it is generally considered that
An.
gam-
biae
and
An. funestus
are more efficient vectors than
An.
arabiensis.22
In
the Gambia,
30
km
south of Dielmo, studies
carried out during the rainy season between 1979 and 1981
and
in
1988 showed that
An.
gambiae
was largely domi-
nant.6.19
This
was also the case in Burkina Faso
in
the region
of Bobo Dioulass0.2~ To our knowledge,
no
study has found
such an important role for
An.
funestus
in West Africa, as
was found in Dielmo. Some years, for as yet unidentified
reasons,
An.
funestus
can be very abundant and transmission
can be higher during the dry season than during the rainy
season. This species is, however, present in the entire
Su-
danese and Guinean climatic region, and is an important vec-
t0r.2~
In
East Africa and Madagascar, it can play a major
During the three-year study, we were unable to find any
explanation, for example, temperature or rainfall, that could
predict the variation of the observed transmission and the
involvement of different vector species.
These detailed data
on
transmission are necessary to cor-
relate with data
on
parasitemia, incidence of clinical attacks,
associated immune responses, human genetic factors, and
variation of
Plasmodium
polymorphism. Recent studies
show that several different
Plasmodium
genotypes exist in
Dielmo.26.
27
These genotypes could have varying pathoge-
nicities and may not be transmitted with equal efficiency by
the different vector species. For these reasons, only a very
thorough study of transmission would provide useful data
for associated research.
role.8.20.25
Acknowledgments: We thank Malick Faye and Frederic Simard
for
technical assistance, Frank Collins (Centers
for
Disease control and
Prevention, Atlanta, GA)
for
providing
Plasmodium
monoclonal an-
tibodies and
for
advice on the PCR, Vincent Robert and Tovi Leh-
mann for comments on the manuscript, and Pauline Roussilhon
for
help with the English translation
of
the manuscript.
Financial support: This work was supported by the French Ministry

VARIATIONS IN MALARIA TRANSMISSION
IN
SENEGAL
,
-
I.
253
of Cooperation (grant 93837). the Institut Francais de Recherche
Scientifique pour le Developpement en Cooperation (ORSTOM),
and the Institut Pasteur de Dakar.
Authors’ addresses: Didier Fontenille, Laurence Lochouam, Nafis-
satou Diagne, Cheikh Sokhna, Jean Jacques Lemasson, Mathurin
Diatta, Farba Faye, and Jean-Francois Trape, ORSTOM, BP 1386,
Dakar, Senegal. Lassana Konate, Departement de Biologie Animale,
Universite Cheikh Anta Diop, Dakar, Senegal. Christophe Rogier,
Institut Pasteur, BP 220, Dakar. Senegal.
Reprint requests: Didier Fontenille, ORSTOM, BP 1386, Dakar,
Senegal.
REFERENCES
1.
Hviid L, Theander TG, 1993. Seasonal changes in human im-
mune responses to malaria.
Parasito1 Today 9:
26-27.
2.
Beier JC, Oster CN, Onyango FK, Bales
JD,
Sherwood JA,
Perkins PV, Chumo DK, Koech DV, Whitmire
RE,
Roberts
CR, Diggs CL, Hoffman SL, 1994.
Plasmodium falciparum
incidence relative to entomologic inoculation rates at a site
proposed for testing malaria vaccines in western Kenya.
Am
J
Trop Med Hyg
50:
529-536.
3. McElroy PD, Beier JC, Oster CN, Beadie C, Sherwood JA,
0100
AJ,
Hoffman SL, 1994. Predicting outcome in malaria:
correlation between rate of exposure to infected mosquitoes
and level of
Plasmodium falciparum
parasitemia.
Am
J
Trop
Med Hyg
51:
523-532.
4.
Mouchet
J,
Carnevale
P,
Coosemans M, Fontenille D, Ravaon-
janahary C, Richard A, Robert V, 1993. Typologie du pal-
udisme en Afrique.
Cah Sante 3:
220-238.
5.
Mbogo CNM,
Snow
WB,
Khamala CPM, Kabiru EW Ouma
JH, Githure
JI,
Marsh K, Beier JC, 1995. Relationships be-
tween
Plasmodium falciparum
transmission by vector popu-
lations and the incidence of severe disease at nine sites
on
the
Kenyan coast.
Am
J
Trop Med Hyg 52:
201-206.
6. Lindsay SW, Alonso PL, Armstong Schellenberg
JRM,
Hem-
ingway
J,
Thomas PJ, Shenton FC, Greenwood BM, 1993. A
malaria control trial using insecticide-treated bed nets and tar-
geted chemoprophylaxis in a rural area of The Gambia, West
Africa. 3. Entomological characteristics of the study area.
Trans
R
Soc Trop Med Hyg 87 (suppl
2):
19-23.
7. Shiff CJ, Minjas
JN,
Hall T, Hunt
RH,
Lyimo
S,
Davis JR, 1995.
Malaria infection potential of anopheline mosquitoes sampled
by light trapping indoors in coastal Tanzanian villages.
Med
Vet Entomol9:
256-262.
8.
Beier JC, Perkins PV, Onyango FK, Gargam TP, Oster CN,
Whitmire
RE,
Koech DV, Roberts CR, 1990. Characterization
of malaria transmission by
Anopheles
(Diptera: Culicidae) in
Western Kenya in preparation
for
malaria vaccine trials.
J
Med Entomol27:
570-577.
9. Trape
JE
Rogier C, Konate L, Diagne
N,
Bouganali H, Canque
B, Legros
F,
Badji A, Ndiaye Ndiaye
P,
Brahimi K, Faye
O,
Druilhe
P,
Pereira Da Silva L, 1994. The Dielmo project: a
longitudinal study of natural malaria infection and the mech-
anisms of protective immunity in a community living in a
holoendemic area of Senegal.
Am
J
Trop Med Hyg
51:
123-
137.
10.
Konate L, Diagne N, Brahimi K, Faye
O,
Legros
E
Rogier C,
Petrarca V, Trape
JF,
1994. Biologie des vecteurs et trans-
mission de
Plasmodium fakiparum, P. malariae,
et
P. ovale
dans un village de savane d’Afrique de l‘Ouest (Dielmo, Sen-
egal).
Parasite
I:
325-333.
11. Diagne N, Fontenille D, Konate L, Faye
O,
Traore-Lamizana
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
M, Legros
F,
Molez JF, Trape
JF,
1995. Les anopheles
du
Senegal. Liste commentee et illustree.
Bull Soc Pathol
Exot
Paskewitz SM, Collins
FH,
1990. Use of polymerase chain re-
action to identify mosquito species of the
Anopheles gambiae
complex.
Med Vet Entomol 4:
367-373.
Gillies MT, DeMeillon B, 1968. The Anophelinae of Africa
South of the Sahara. Second edition. Johannesburg: South Af-
rican Institute of Medical Research.
Burkot TR, Williams JL, Schneider
I,
1984. Identification of
Plasmodium
falciparum-infected mosquitoes by a double
an-
tibody enzyme linked immunosorbent assay.
Am
J
Trop Med
Wirtz
RA,
Zavala
E
Charoenvit
Y,
Campbell GH, Burkot TR,
Schneider
I,
Esser
KM,
Beaudoin
RL,
Andre RC, 1987. Com-
parative testing of monoclonal antibodies against
Plasmodium
falciparum
sporozoites for ELISA development.
Bull World
Health Organ 65:
139-145.
Scott JA, Brogdon WG, Collins FH, 1993. Identification of
sin-
gle specimens of the
Anopheles gambiae
complex by poly-
merase chain reaction.
Am
J
Trop Med Hyg 49:
520-529.
Liddell FDK, 1984. Simple exact analysis of the standardised
mortality ratio.
J
Epidemiol Community Health
38:
85-88.
Fontenille D, Faye
O,
Konate L, Sy N, Collins FH, 1993. Com-
paraison des techniques PCR et cytogenetique pour la deter-
mination des membres du complexe
Anopheles gambiae
au
Senegal.
Ann Parasitol Hum Comp 68:
239-240.
Bryan JH, Di Deco MA, Petrarca V, Coluzzi M, 1982. Inversion
polymorphism and incipient speciation in
Anopheles gambiae
s.sb
in The Gambia, West Africa.
Genetica 59:
167-176.
Taylor
KA,
Koros JK, Nduati J, Coperland RS, Collins
FH,
Brandling-Bennett DA, 1990.
Plasmodium falciparum
infec-
tion rates in
Anopheles gambiae, An. arabiensis,
and
An.
fu-
nestus
in Western Kenya.
Am
J
Trop Med Hyg 43:
124-129.
Robert V, Verhave
JP,
Ponnudurai
T,
Louwe L, Scholtens
P,
Car-
nevale
P,
1988. Study of the distribution of circumsporozoite
antigen in
Anopheles gambiae
infected with
Plasmodium fal-
ciparum
using the enzyme-linked immunosorbent assay.
Trans
R
Soc Trop Med Hyg 82:
389-391.
Coluzzi
M,
1984. Heterogeneities of the malaria vectorial sys-
tem in tropical Africa and their significance in malaria epi-
demiology and control.
Bull World Health Organ 62:
107-
113.
Robert
V,
Petrarca
V,
Carnevale P, Ovazza L, Coluzzi
M,
1989.
Analyse cytogenetique du complexe
Anopheles gambiae
dans
la region de Bobo-Dioulasso (Burkina Faso).
Ann Parasitol
Hum Comp 64:
290-31 1.
Robert V, Gazin
P,
Boudin C, Molez JE Ouedraogo
V,
Carnevale
P,
1985. La transmission du paludisme en zone de savane
arboree et en zone rizicole des environs de Bobo-Dioulasso
(Burkina Faso).
Ann
Soc
Belg Med Trop
65
(suppl 2):
201-
214.
Severini
C,
Fontenille D, Ramiakajato MR, 1990. Importance
d’dnopheles funestus
dans la transmission do paludisme au
hameau de Mahitsy, a Tananarive, Madagascar.
Bull
Soc
Pathol Exot
83:
114-1 16.
Ntoumi
F,
Contamin H, Rogier C, Bonnefoy
S,
Trape JF, Mer-
cereau-Puijalon
O,
1995. Age-dependent carriage of multiple
Plasmodium falciparum
merozoite surface antigen-2 alleles in
asymptomatic malaria infections.
Am
J
Trop Med Hyg
52:
Daubersies
P,
Sallenave-Sales
S,
Magne
S,
Trape
fi,
Rogier C,
Contamin
H,
Fandeur T, Mercereau-Puijalon
O,
Druilhe
P,
1996. Rapid turn-over of
Plasmodium falciparum
populations
in asymptomatic individuals living in a high transmission
area.
Am
J
Trop Med Hyg 54:
18-26.
87:
267-277.
Hyg
33:
783-788.
81-88.

VOLUME
56
NUMBER
3
The
American
Journal
of
OFFICIAL ORGAN OF
THE AMERICAN SOCIETY
OF
TROPICAL MEDICINE AND HYGIENE