Am. J. Trop. Med. Hyg., 61(2), 1999, pp. 230–239
Copyright q 1999 by The American Society of Tropical Medicine and Hygiene
MITOCHONDRIAL DNA VARIATION AMONG ANOPHELES ALBIMANUS
POPULATIONS
ANA MARIA P. DE MERIDA, MARGARITA PALMIERI, MARIA MARTA YURRITA, ALVARO MOLINA,
EDUVIGES MOLINA, AND WILLIAM C. BLACK IV
Universidad del Valle de Guatemala, Guatemala City, Guatemala; Department of Microbiology, Colorado State University,
Fort Collins, Colorado
Abstract. Barriers to gene flow between Pacific and Atlantic coast populations of Anopheles albimanus were
reported in an earlier study of variation in the intergenic spacer of the nuclear ribosomal DNA. We examined the
distribution of mitochondrial DNA haplotypes among A. albimanus populations to test for gene flow barriers with an
independent genetic marker. A region of the NADH dehydrogenase subunit 5 gene was amplified by the polymerase
chain reaction (PCR) in 1,105 mosquitoes collected from 16 locations in Guatemala and in single collections from
Mexico, Honduras, Nicaragua, Costa Rica, Panama, Colombia, and Venezuela. The PCR products were tested for
variation using single strand conformation polymorphism analysis and 45 haplotypes were detected. Haplotype frequencies did not vary between coasts in Guatemala. Populations within ;200 km of one another were panmictic.
However, at distances . 200 km, FST and geographic distances were correlated suggesting that populations are isolated
by distance.
these reasons, mitochondrial DNA has been used in studies
of gene flow and phylogenetic relationships among and within the four species of the A. quadrimaculatus complex,16
among populations of A. aquasalis, A. rangeli, A. trinkae,
and A. nuneztovari,17–19 and among and within the African
malaria vectors A. gambiae and A. arabiensis.20–22
Following earlier studies of mitochondrial DNA in our
laboratory,23 we used single strand conformation polymorphism (SSCP) analysis24 as a quick, sensitive, and inexpensive means to screen for variation among mitochondrial
genes amplified from individual mosquitoes. We then sequence the most common haplotypes to test the sensitivity
and reproducibility of the SSCP technique and to gather data
with which to assess phylogenetic relationships among haplotypes. We examine patterns of gene flow between northern
(Atlantic) and southern (Pacific) populations of A. albimanus
in Guatemala and among single populations from Mexico,
Honduras, Nicaragua, Costa Rica, Panama, Colombia, and
Venezuela. As in our earlier survey, mosquitoes were sampled in a nested spatial design with frequencies partitioned
among individuals within a sample, among samples in a region, among regions distributed in either the northern or
southern parts of Guatemala, and among South and Central
American countries.
Anopheles albimanus is one of the major vectors of malaria in Central America, the Caribbean, and coastal regions
of northern South America. It feeds more readily on animals
than humans,1–3 and is generally exophilic in host-seeking
and feeding behavior.4–6 Sporozoite rates are generally low,
with the greatest reported rate at only 1.6% (1 of 63).7 However, these behavioral characteristics and vector competence
estimates vary greatly among A. albimanus populations.8 We
have been testing for the presence of cryptic species within
A. albimanus because of the strong historical precedence for
the existence of cryptic, isomorphic Anopheles species.
In an earlier study of Guatemalan A. albimanus populations, we observed a high degree of variation both in the
frequency and copy number of intergenic spacer (IGS)
length variants among individual A. albimanus.9 Regardless
of the collection site, most of the variation either in frequency or intensity of IGS variants was detected among
mosquitoes within a population. The next largest component
of genetic differentiation was detected between populations
on the Pacific and Atlantic coasts of Central America. In
contrast, populations from the Atlantic coast of Central
America and from South America had similar IGS patterns.
We suggested that the many mountain ranges that run
through northern Central America could form barriers to
gene flow among Atlantic and Pacific coast populations.
However, extreme local variation in rDNA IGS has been
found in many population studies of other insect species and
is consistent with a model whereby molecular drive rapidly
gives rise to unique spacer patterns in populations and then
drives them to high frequencies.10–13 In the case of A. albimanus, molecular drive probably occurs through unequal
crossing-over and gene conversion within and among the
100-basepair (bp) repeat units constituting the IGS.14
In the present paper, we test the inferences arising from
our earlier rDNA study using an independent marker: variation in the mitochondrial genome (mtDNA). This type of
DNA is maternally inherited and does not recombine.15
When sequence data are collected, these properties allow for
phylogenetic analysis of maternal lineages. In addition, patterns of variation in mtDNA haplotype frequencies can be
used to estimate rates of gene flow among populations. For
MATERIALS AND METHODS
Mosquito collections. The locations, sample sizes, and
collectors at each site are listed in Table 1. Mosquitoes were
collected in cattle corrals and brought live to the laboratory
in cardboard cartons. There they were killed by freezing,
identified to species,25 and DNA was extracted. Mosquitoes
that were not used immediately for DNA extraction were
stored in 70% ethanol for later use. The geographic locations
of all sampling sites are shown in Figure 1.
Extraction of DNA. The DNA was extracted from mosquitoes using a protocol provided by Dr. Patricia Romans
(University of Toronto, Toronto, Ontario, Canada). Mosquitoes were placed individually in 1.5-ml tubes and 25 ml of
grinding buffer (0.1 M NaCl, 0.2 M sucrose, 0.1 M TrisHCl, 0.05 M EDTA, and 0.5% sodium dodecyl sulfate
230
mtDNA VARIATION IN A. ALBIMANUS
TABLE 1
Locations, dates, collectors, and sample sizes of Anopheles albimanus collections*
Country
Region
Date
Guatemala (N. Padilla, C.C. de Rosales,
P. Peralta, J. Garcia)
Northern Guatemala
Champona
III-6 (12), VII-6 (32), IX-5 (43)
Cristina VII-4
Nahua
III-9 (9), VII-3 (29), IX-6 (43)
San Luis Peten VII-10
San Luis Buena Vista
VII-10 (43), IX-4 (44)
Southern Guatemala
Chano
II-7 (26), VI-6 (15), VIII-3 (13)
Cuto
II-6 (42), VI-5 (26), IV-5 (89)
Iglesias VI-5
Lauro
VI-7 (4), IV-4 (19), II-9 (31)
Mango
III-15 (22), II-28 (33), III-7-96 (37)
Ruperto
II-8 (16), VI-7 (16), IV-3 (9)
Tallado
VI-9 (12), II-7 (40), VIII-2 (38), IV-6 (9)
Eastern Guatemala
ElMotor II-20-96
ElZapote IX-13
Vega IX-12
Puente Blanco
VII-11 (46), II-21-96 (46), IX-11 (45)
Mexico (E. Loyola, D. Bown, M. H. Rodriguez)
Honduras (H. Cosenza)
Nicaragua (D. Bown, P. R. Mendoza)
Costa Rica (T. Solano)
LP
EA
El Paso
Panama (M. Nelson, A. Kuhar, A. Blas)
Colombia (V. P. Howley)
Venezuela (Y. Rubio, R. Zimmerman)
Study total
N
1,105
338
87
40
81
43
87
528
54
157
31
54
92
41
99
239
45
26
31
137
11
17
25
64
24
15
25
36
12
12
1,282
* Unless otherwise indicated, all collections were made in 1995. Sample sizes for specific
dates appear in parentheses.
[SDS], pH of the entire mixture adjusted to 9.2) was added.
The mosquito was homogenized with a Kontes pestle (Kimble/Kontes, Vineland, NJ) and the pestle was washed with
an additional 25 ml of grinding buffer. The contents of each
tube were briefly centrifuged and heated in a water bath at
658C for 30 min. As tubes were removed from the bath, 7
ml of 8 M potassium acetate were added, the tubes were
mixed by tapping and placed on ice for 30 min. Each tube
was centrifuged at 17,000 3 g for 15 min to pellet debris
and precipitated SDS. The supernatant was removed to a
second 1.5-ml tube and 100 ml of 100% ethanol was added
to precipitate nucleic acids. This mixture was divided in two
tubes, one for use in the present study and the other stored
for future use. A 50-ml aliquot was centrifuged for 15 min
at 17,000 3 g and the pellet was washed in 70% ethanol
followed by a second wash in 100% ethanol. The pellet was
dried at room temperature for 12 hr and resuspended in 100
231
ml of filtered, autoclaved TE buffer (10 mM Tris-HCl, 1 mM
EDTA, pH 8.0).
Gene amplification with the polymerase chain reaction.
The ND5 gene was amplified using primers ND5P1 (59TWG CSC CTA ATC CKG CTA TA-39) and ND5M2 (59YTW GGA TGA GAT GGS TTA GG-39) where Y 5 pyrimidine, R 5 purine, S 5 C or G, K 5 G or T, and W 5
A or T. Amplification was done in an MJ Research (Watertown, MA) thermocycler in which tubes were heated to 958C
for 5 min and held at 808C for 20 min while 1 unit of Taq
polymerase was added to each tube. Tubes were then subjected to10 cycles of 928C for 1 min, 488C for 1 min, and
728C for 1.5 min. This was followed by 32 cycles of 928C
for 1 min, 548C for 35 sec, and 728C for 1.5 min. This was
followed by a final extension at 728C for 7 min.
The amplified regions correspond with nucleotides 7,282–
7,671 in A. quadrimaculatus (GenBank #L04272) and nucleotides 7,169–7,558 in A. gambiae (GenBank # L20934).
The 16S gene was amplified using primers 16Sbr (59-CCG
GTY TGA ACT CAG ATC AYG T-39) and 16Sar (59-CRC
CTG TTT AYC AAA AAC AT-39) and amplification proceeded with the same thermal cycling conditions listed
above. These correspond with nucleotides 12,784–13,235 in
A. quadrimaculatus and nucleotides 12,611–13,058 in A.
gambiae.
Acquisition of DNA sequence data. The nucleotide sequences of the 15 ND5 haplotypes most common in Guatemala were sequenced. The 16S gene was amplified and
sequenced in the same 15 individuals. Amplified products
were sequenced along both strands using double-stranded
sequencing with the ABI 3948 Nucleic Acid Synthesis and
Purification System (Perkin-Elmer, Norwalk, CT) at Macromolecular Resources at Colorado State University.
Statistical analysis of haplotype frequencies. Variation
in haplotype frequencies was examined using the analysis of
molecular variance (AMOVA) procedure on Arlequin 1.1.26
The AMOVA procedure was initially performed on collections across seasons to test for evidence of seasonal changes
in haplotype frequencies. This procedure was next used to
partition variation among northern, southern, and eastern
portions of Guatemala. Lastly, AMOVA partitioned variation
among Central and South American countries. The significance of the variance components associated with each level
of genetic structure was tested using non-parametric permutation tests.26 Arlequin 1.1 was also used to compute FST,
a standardized measure of variation in haplotype frequencies,27 among all populations and between all population
pairs.
Pairwise FST values were transformed to FST/(1 2 FST) and
regressed on pairwise geographic distances among populations to determine if geographic distance among populations
serves as a barrier to gene flow.28 Geographic distances were
obtained by the Geographic Information System (GIS) using
program ATLAS-GIS 3.0 (Environmental Systems Research
Institute, Inc., Redlands, CA). The collection sites were located in the map of each location in a department or country
using the coordinates of each site. Linear distances were
measured between every two sites. Distances were taken
from a planimetric map, but altitudes were not considered.
This regression was repeated using a natural logarithm transformation of geographic distance.29 Transformations, regres-
232
DE MERIDA AND OTHERS
FIGURE 1.
Map of Latin America showing the locations of Anopheles albimanus collections.
sion analyses, and Mantel’s test30 were performed using a
FORTRAN program MANTEL (available from William C.
Black). The reciprocal of the slope estimated by this regression provides an estimate of the average effective population
size.29 Pairwise transformed FST values were entered into a
distance matrix and subject to cluster analysis using unweighted pair group method using averages (UPGMA) analysis31 in the NEIGHBOR procedure in PHYLIP3.5C.32
The 390 bp of the ND5 gene were manually aligned according to codons. No gaps were required to obtain an optimal alignment. The ;450 bp of the 16S gene were manually aligned among haplotypes by insertion of 3 gaps to
yield an aligned sequence of 454 bp. The aligned ND5 and
16S datasets are available by e-mail at wcb4@lamar.
colostate.edu. Phylogenetic relationships among haplotypes
were estimated with PAUP4B41 using maximum likelihood,33 maximum parsimony, and distance/neighbor joining
analyses.34,35 Analyses of the two genes were initially performed separately. The ND5 and 16S sequences were then
appended for each haplotype and the congruence of the datasets was tested using the partition homogeneity test.36 The
individual gene phylogenies were congruent and the homogeneity test confirmed that phylogenies arising from the two
genes were not significantly different. The remainder of the
analyses was therefore performed on the combined (844 bp)
dataset.
mtDNA VARIATION IN A. ALBIMANUS
For each collection, the nucleotide sequence and the frequency of each haplotype were entered into DnaSP.37 This
analysis could not be completed for all individuals in this
study because we only sequenced the 15 most common haplotypes in Guatemala. Nevertheless, 90% (305 of 338), 95%
(502 of 528), and 97% (232 of 239) of individuals collected
in northern, southern, and eastern Guatemala, respectively,
were used. For each collection, we estimated the number of
polymorphic sites, the average number of nucleotide differences (k) (equation A3),38 the nucleotide diversity (p) (equation 10.5),39 and the nucleotide diversity with the Jukes and
Cantor correction (p2) (equation 10.19 and 5.3).39 Between
each pair of collections, we estimated the number of polymorphic sites, k, p1, and the average number of nucleotide
substitutions per nucleotide site between populations with
the Jukes and Cantor correction (DXY (JC)) (equation
10.20).39 Also calculated were the number of net nucleotide
substitutions per site between populations with the Jukes and
Cantor correction Da (JC) (equation 10.21).39 Sequence data
were used to estimate pair-wise estimates of NST.40 As above,
pairwise NST values were transformed to NST/(1 2 NST) and
regressed on pairwise geographic distances and on a natural
logarithm transformation of geographic distance.29
RESULTS
Haplotype frequencies. A total of 45 different ND5 haplotypes were detected among the 1,282 A. albimanus examined in this study. The locations, dates, collectors and
sample sizes at each sites are listed in Table 1. Haplotype
frequencies in each collection are available by e-mail at
[email protected]. The relative frequencies of all 45
haplotypes are displayed graphically in Figure 2. Haplotype
number designations correspond to the order in which the
haplotypes were discovered. Once all populations were surveyed, all unique haplotypes were compared on a single gel.
This established identity among haplotypes that were initially incorrectly identified as unique and occasionally identified unique haplotypes that were initially thought to be
identical when run on different gels (these haplotypes are
marked with 9 or 0 superscripts). The process of rescreening
haplotype variation resulted in the identity of ;1% of haplotypes being redesignated in the final dataset. Sequence
analysis of the 15 most common haplotypes in Guatemala
confirmed that the SSCP technique is sensitive to single substitutions. The SSCP haplotypes 2 and 16 differed by a single A ↔ G transition as did haplotypes 8 and 1, while haplotypes 7 and 31 differed by a single C ↔ T transition.
Sequences of two mosquitoes with SSCP haplotype 1 were
identical and confirmed the reproducibility of the technique.
Identical sequences were also obtained from two mosquitoes
with haplotype 2.
Haplotype frequencies are similar among northern (Atlantic), eastern and southern (Pacific) Guatemalan collections
(Figure 2A). Aside from haplotypes 44 and 6 in Honduras,
populations from Mexico, Guatemala, Honduras, and Nicaragua also had similar haplotype frequencies (Figure 2B).
However, haplotype frequencies shift in collections from
Costa Rica and all collections to the south from Panama,
Colombia, and Venezuela (Figure 2C).
Haplotype frequencies were compared among seasonal
233
collections from the same locations in northern and southern
Guatemala collections using AMOVA. In both regions,
;99% of the variation in haplotype frequencies arose among
individuals in a collection and the remaining variation arose
among sites within a season. Negative variation was estimated among seasons indicating that there was more variation among sites within seasons than among seasons. There
was no variation in haplotype frequencies between wet and
dry season collections in either northern or southern Guatemala. Because haplotypes did not vary among season, all
seasonal collections were grouped together by location in all
subsequent analyses.
Haplotype frequencies were next compared among northern, southern, and eastern regions in Guatemala (Table 2A).
Within Guatemala, 97.5% of the variation in haplotype frequencies arose among individuals in a collection, while only
;1% of variation arose among collections in a region and
;1.5% of variation arose among regions. These analyses
agree with Figure 2A in indicating very little variation
among populations from northern, southern, and eastern
Guatemala.
Haplotype frequencies were also compared among countries (Table 2B). Most of the variation in haplotype frequencies (84.5%) arose among individuals in a collection and
only ;1.5% variation arose among collections in a country.
However, almost 14% of the variance in haplotypes arose
among countries, indicating (as shown in Figure 2C) that a
substantial proportion of variation occurs among Latin
American countries.
Genetic relationships among populations. The FST was
computed between all pairs of populations and transformed
as FST /(1- FST). This matrix of pairwise differences between
populations was subject to cluster analysis using UPGMA in
NEIGHBOR on PHYLIP3.5C (Figure 3). All northern Guatemala populations cluster together but also cluster with the
collections from Tallado and Ruperto from southern Guatemala and with the collection from Nicaragua. All southern
and eastern Guatemala populations cluster together with the
nearby Tapachula, Mexico collection. The collections from
Costa Rica cluster together as do the two collections from
South America. Panama differs from all populations in the
frequencies of several haplotypes (Figure 2C).
Pairwise transformed FST estimates were regressed against
geographic distances to test for a pattern of isolation by distance and there was a significant correlation between untransformed geographic distances and FST (Figure 4A and
Table 3). The average FST between pairs of populations tended to increase with the geographic distances between them.
Transformed FST estimates were also regressed against the
natural logarithm of the geographic distance to determine
both the geographic distance over which isolation occurs and
to estimate the average effective population size. This regression was performed on all collections (Figure 4B and
Table 3) and among only Guatemalan collections (Figure 4C
and Table 3). In all regression analyses, Mantel’s test indicated that the correlation was significantly greater than zero.
Among all collections, FST did not increase until geographic distances approached 150–250 km (e5–e5.5). However, even within Guatemala, the regression between FST and
geographic distance was significant (Figure 4C and Table 3).
The average effective population size was 68 individuals
234
DE MERIDA AND OTHERS
FIGURE 2. Frequencies of the 45 ND5 haplotypes in A, north, east, and south Guatemala; B, Guatemala, Mexico, Honduras, and Nicaragua;
and C, Costa Rica, Panama, Colombia, and Venezuela.
among all collections and 227 among collections within
Guatemala.
Phylogenetic relationships among haplotypes. The ND5
and 16S sequences of A. albimanus were easily manually
aligned with the homologous regions of A. gambiae and A.
quadrimaculatus. Phylogenetic analysis did not identify any
well-supported clades among A. albimanus haplotypes (Figure 5). Only two clades (H8/H18 and H7/H31) had slight
levels of bootstrap support. These haplotypes were not
unique to any region of Guatemala.
A combined analysis of frequencies and nucleotide substitutions (Table 3) indicated very slight levels of variation
among haplotypes in Guatemala. The results of regression
analysis of NST on transformed and untransformed geographic distances yielded similar estimates to those obtained with
FST (Table 3). The similarity of NST and FST is not surprising
mtDNA VARIATION IN A. ALBIMANUS
235
TABLE 2
Analysis of molecular variance of geographic variation among
Anopheles albimanus collections
Source of
variation
Sum of
squares
df*
Variance
components
% variation
A, Variation among regions in Guatemala
Among north, east,
and south regions
Among collections
within regions
Within collections
Total
2
6.063
0.007†
1.67
13
7.987
0.003‡
0.86
1,089
1,104
425.430
439.480
0.391§
0.401
97.46
Fixation indices
F(regions)
F(Collections in
regions)
F(All collections)
0.017†
0.009‡
0.026§
B, Variation among countries
Among countries
Among collections
within countries
Within collections
Total
Fixation indices
F(Countries)
F(Collections in
countries)
F(All collections)
7
15
24.834
14.049
0.064§
0.008§
13.73
1.74
1,259
1,281
492.575
531.458
0.391§
0.463
84.53
0.137§
0.020§
0.155§
* df 5 degrees of freedom.
† Probability (random value # observed value) 5 0.00062 6 0.00008 (100,172 permutations).
‡ Probability (random value # observed value) 5 0.00253 6 0.00016 (100,172 permutations).
§ Probability (random value # observed value) 5 0.00000 6 0.00000 (100,172 permutations).
given the slight levels of nucleotide variation among haplotypes.
DISCUSSION
Mitochondrial haplotype frequencies were similar between northern and southern populations of A. albimanus in
Guatemala (Figure 2A and Table 2), and FST values were
small among populations from these regions (Table 3). These
results suggest that we overestimated the magnitude of the
barriers to gene flow between Atlantic and Pacific populations of A. albimanus in our earlier study of rDNA IGS
variation.9 However, the findings of the present study do not
entirely contradict the results of our earlier study. First, it is
possible that Atlantic and Pacific populations of A. albimanus are truly isolated and that mitochondrial haplotypes introgress through semi-permeable barriers to hybridization.41
Anopheles gambiae and A. arabiensis share 8 mitochondrial
haplotypes in common and a high proportion of shared polymorphisms.21 While these are recognized as valid species,
fertile female hybrids have been found between these species
in nature, albeit very rarely,42 and there is evidence from
different molecular genetic approaches that A. gambiae and
A. arabiensis are or recently were undergoing gene exchange
through introgressive hybridization.20–22,43, 44 If so, this predicts that we will find similar patterns of reproductive isolation in A. albimanus as those detected in the rDNA study
when analyzing single copy nuclear genes or microsatellites.
FIGURE 3. Unweighted pair group method using averages analysis of pair-wise FST/(1 2 FST) relationships between collections.
Those studies are in progress; however, this outcome is highly unlikely because rare hybridization events generate genetic bottlenecks that dramatically change the frequencies of
mitochondrial haplotypes. In contrast, mitochondrial haplotypes were extremely similar between northern and southern
populations of A. albimanus (Figure 2A).
A more likely explanation of the difference between this
study and the earlier rDNA study is that even slight amounts
of reproductive isolation between northern and southern
populations detected in the present study would be sufficient
to allow for concerted evolution to rapidly drive elements to
fixation between regions. Populations in the north and south
could acquire unique IGS patterns if molecular drive caused
IGS repeats to converge among members of a population
more rapidly than gene flow introduced new variants that
would disrupt IGS associations. Precisely the same discrepancy appeared in studies with rDNA IGS markers in A. gambiae and Aedes albopictus. Microgeographic variation in
rDNA intergenic spacers was found among populations of
A. gambiae in western Kenya and inferred limited dispersal
among mosquito populations from nearby villages.45 However, recent studies using mitochondrial markers,21 microsatellites,46,47 and reanalysis of allozyme data48 all indicate large
amounts of gene flow among populations distributed across
Africa. Similarly, variation in rDNA IGS was found among
A. albopictus collected in nearby locations within cities11
while gene flow estimates using allozymes markers suggested high rates of gene flow.49–51
An isolation by distance model detected in this study sug-
236
DE MERIDA AND OTHERS
FIGURE 4. Regression analysis of pairwise FST/(1 2 FST) estimates against pairwise untransformed (A), natural logarithms of geographic
distances among all sites (B), or among sites in Guatemala (C). * 5 times symbol (3); d 5 geographic distance; prob. 5 probability.
gests that gene flow among A. albimanus populations decreases with increasing geographic distances. Clusters of An.
albimanus within 200 km of one another appear to exchange
genes continuously. However, rates of gene flow decrease
among clusters of An. albimanus separated by distances .
200 km. Furthermore, these changes are not continuous over
the geographic range of this study. We observed abrupt
changes in mitochondrial haplotype frequencies when comparing samples from Guatemala, Mexico, Honduras, and
Nicaragua with those from Costa Rica and Panama (Figure
2C). There are no obvious geographic barriers to account for
the abrupt disruption of gene flow between northern Central
American populations and Costa Rica and Panama populations. More intensive collections should be completed in
Costa Rica and Panama to confirm that this is not an artifact
of sampling.
Whether measured within Guatemala, or across the entire
geographic range of this study, the average effective size of
A. albimanus populations appeared quite small. This may
reflect a patchy distribution of the species along coastal areas
in association with cattle sheds. Generally speaking, A. albimanus is more prevalent during the rainy season. Its numbers are greatly reduced during the dry season and in some
areas they seem to disappear completely, only to reappear
237
mtDNA VARIATION IN A. ALBIMANUS
TABLE 3
Summary of variability, FST, and neighborhood size in north, south, and east sites
North
South
East
North vs. South
North vs. East
South vs. East
N
Polymorphic
sites
k
p1
p2
305
502
232
36
37
29
4.264
4.247
3.784
0.00507
0.00505
0.00450
0.00509
0.00508
0.00452
N
Polymorphic
sites
k
p1
DXY (JC)
Da (JC)
NST
807
537
734
37
36
37
4.305
4.114
4.101
0.00512
0.00489
0.00488
0.00522
0.00495
0.00480
0.00013
0.00014
0.00000
0.02492
0.02837
20.00016
Slope
Intercept
R2
Mantel
probability
4Dps2
Guatemala
FST/(1 2 FST) 5 ln (distance)
0.00440
0.00493
0.06
0.001
227 individuals
Central America (20 sites)
FST/(1 2 FST) 5 ln (distance)
NST/(1 2 NST) 5 ln (distance)
0.01462
0.01175
20.03146
20.02697
0.14
0.09
0.001
0.001
68 individuals
85 individuals
All sites
FST/(1 2 FST) 5 ln (distance)
0.05784
20.21638
0.37
0.001
17 individuals
rapidly when the rain begins. An overall correlation is observed between high and low abundance of A. albimanus in
the wet and dry season, respectively.
The breeding structure of A. albimanus populations has
also been studied in Colombia using isozymes.52 Northern
populations were 3–4 times more heterozygous than the
southern populations and formed a distinct cluster in a dendrogram using genetic distance analysis. These results are
consistent with the findings of the present study and suggest
that Colombian populations of A. albimanus may also be
isolated by distance.
The patterns of gene flow in A. albimanus are quite different from patterns identified in other Anopheles species
examined to date. The nucleotide diversity (p1) estimates
among mitochondrial haplotypes in A. aquasalis, A. rangeli,
A. trinkae, and A. nuneztovari were an order of magnitude
larger that those reported here.17–19 Furthermore, the nucle-
otide diversity estimates (p1), variance components, and FST
values among populations of A. nuneztovari were all at least
an order of magnitude larger that those reported here.19 However, despite finding a much higher level of variation among
haplotypes and populations, there was no significant correlation between genetic and geographic distances in A. nuneztovari.19 The nucleotide diversity (p1) estimates within
Anopheles gambiae (0.0038) and A. arabiensis (0.0023–
0.0051) populations were very similar to those found among
A. albimanus (Table 3). However, despite sampling over
greater geographic distances, there was no evidence for isolation by distance among populations in either species.21 It
thus appears that effective migration rates and effective population sizes may be smaller in A. albimanus than in the
other Anopheles species examined to date.
Acknowledgments: We thank Celia Cordon-Rosales, Juan Garcia,
Pedro Peralta, and Alfonzo Salam for their collaboration in the collection of mosquitoes. Lourdes Pineda and Estuardo Barrios helped
in measuring distances between collection sites. Dr. Jan Conn and
two anonymous reviewers greatly improved the manuscript.
Financial support: This project was supported by the UNDP/World
Bank/WHO Special Program for Research and Training in Tropical
Diseases (TDR), grant no. 940392.
Authors’ addresses: Ana Maria P. De Merida, Margarita Palmieri,
Maria Marta Yurrita, Alvaro Molina, and Eduviges Molina, Medical
Entomology Research and Training Unit, Universidad del Valle de
Guatemala, Guatemala City, Guatemala. William C. Black IV, Department of Microbiology, Colorado State University, Fort Collins,
CO 80523.
Reprint requests: Director (AMP), Medical Entomology Research
and Training Unit, USEMB/HHS/MERTUG Unit 3321, APO AA
Miami, FL 34024.
REFERENCES
FIGURE 5. Maximum likelihood tree showing phylogenetic relationships among individual haplotypes. Bootstrap support using
maximum parsimony analysis appears above each branch while
bootstrap support using Tamura-Nei genetic distance/neighbor joining appears below each branch.
1. Bruce-Chwatt L, Garrett-Jones C, Weitz B, 1966. Ten years’
study (1955–64) of host selection by anopheline mosquitoes.
Bull World Health Organ 35: 405–439.
2. Garrett-Jones C, 1964. The human blood index of malaria vec-
238
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
DE MERIDA AND OTHERS
tors in relation to epidemiological assessment. Bull World
Health Organ 30: 241–261.
Garrett-Jones C, Boreham P, Pant C, 1980. Feeding habits of
anophelines (Diptera: Culicidae) in 1971–1978, with reference to the human blood index: a review. Bull Entomol Res
70: 165–185.
Elliot R, 1968. Studies on man-vector contact in some malarious
areas in Colombia. Bull World Health Organ 38: 239–253.
Elliot R, 1972. The influence of vector behavior on malaria
transmission. Am J Trop Med Hyg 21: 755–763.
Hobbs J, Sexton J, St. Jean Y, Jacques J, 1986. The biting and
resting behavior of Anopheles albimanus in northern Haiti. J
Am Mosq Control Assoc 2: 150–153.
Warren McW, Mason J, Hobbs J, 1975. Natural infections of
Anopheles albimanus with Plasmodium in a small malaria focus. Am J Trop Med Hyg 24: 545–546.
Faran M, 1980. A revision of the Albimanus section of the Nyssorhynchus of Anopheles. Contrib Am Entomol Inst 15: 1–
215.
de Merida AM, de Mata MP, Molina E, Porter CH, Black WC,
1995. Variation in ribosomal DNA intergenic spacers among
populations of Anopheles albimanus in South and Central
America. Am J Trop Med Hyg 53: 469–477.
Dover G, 1986. Molecular drive in multigene families: how biological novelties arise spread and are assimilated. Curr
Trends Genet 8: 159–165.
Black IV WC, MacLain DK, Rai KS, 1989. Patterns of variation
in the rDNA cistron within and among populations of Aedes
albopictus (Skuse). Genetics 121: 539–550.
Tabachnick WJ, Black WC, 1996. Population genetics of disease vectors. Beaty BJ, Marquardt WC, eds. Biology of Disease Vectors. Boulder, CO: University Press of Colorado,
417–437.
Onyabe DY, Conn JE, 1999. Intragenomic heterogeneity of a
ribosomal DNA spacer (ITS2) varies regionally in the neotropical malaria vector Anopheles nuneztovari (Diptera Culicidae). Insect Mol Biol: (in press).
Beach RF, Mills D, Collins FH, 1989. Structure of ribosomal
DNA in Anopheles albimanus (Diptera: Culicidae). Ann Entomol Soc Am 81: 641–648.
Avise JC, 1994. Molecular Markers, Natural History and Evolution. New York: Chapman and Hall.
Mitchell SE, Narang SK, Cockburn AF, Seawright JA, Goldenthal M, 1992. Mitochondrial and ribosomal DNA variation
among members of the Anopheles quadrimaculatus (Diptera:
Culicidae) species complex. Genome 35: 939–950.
Conn JE, Cockburn AF, Mitchell SE, 1993. Population differentiation of the malaria vector Anopheles aquasalis. J Hered
84: 248–253.
Conn JE, Mitchell SE, Cockburn AF, 1997. Mitochondrial DNA
variation within and between two species of neotropical
anopheline mosquitoes (Diptera: Culicidae). J Hered 88: 98–
107.
Conn JE, Mitchell SE, Cockburn AF, 1998. Mitochondrial DNA
analysis of the neotropical malaria vector Anopheles nuneztovari. Genome 41: 313–327.
Besansky NJ, Powell JR, Caccone A, Hamm DM, Scott JA,
Collins FH, 1994. Molecular phylogeny of the Anopheles
gambiae complex suggests genetic introgression between
principal malaria vectors. Proc Natl Acad Sci USA 91: 6885–
6888.
Besansky NJ, Lehmann JA, Fahey GT, Fontenille D, Braack
LEO, Hawley WA, Collins FH, 1997. Patterns of mitochondrial DNA variation within and between African malaria vectors, Anopheles gambiae and An. arabiensis, suggest extensive gene flow. Genetics 147: 1817–1828.
Caccone A, Garcia BA, Powell JR, 1996. Evolution of the mitochondrial DNA control region in the Anopheles gambiae
complex. Insect Mol Biol 5: 51–59.
Norris DE, Klompen JSH, Keirans JE, Black WC, 1996. Population genetics of Ixodes scapularis (Acari: Ixodidae) based
on mitochondrial 16S and 12S genes. J Med Entomol 33: 78–
89.
Orita M, Iwahana H, Kanazawa H, Hayashi K, Sekiya T, 1989.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
Detection of polymorphisms of human DNA by gel electrophoresis as single-strand conformation polymorphisms. Proc
Natl Acad Sci USA 86: 2766–2770.
Darsie Jr RF, Ward RA, 1981. Identification and geographical
distribution of the mosquitoes of North America, north of
Mexico. Mosq Syst 1 (suppl): 1–313.
Excoffier L, Smouse PE, Quattro JM, 1992. Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction
data. Genetics 131: 471–491.
Weir BS, 1996. Genetic Data Analysis II. Sunderland, MA: Sinauer Associates, Inc.
Slatkin M, 1993. Isolation by distance in equilibrium and non
equilibrium populations Evolution 47: 264–279.
Rousset F, 1997. Genetic differentiation and estimation of gene
flow from F-statistics under isolation by distance. Genetics
145: 1219–1228.
Mantel N, 1967. The detection of disease clustering and a generalized regression approach. Cancer Res. 27: 209–220.
Sokal RR, Sneath PHA, 1963. Principles of Numerical Taxonomy. San Francisco: W. H. Freeman and Co.
Felsenstein J, 1993. PHYLIP: Phylogeny Inference Package.
Version 3.5C. Seattle, WA: University of Washington.
Hasegawa M, Kishino H, Yano T, 1985. Dating of the humanape splitting by a molecular clock of mitochondrial DNA. J
Mol Evol 22: 160–174.
Tamura K, Nei M, 1993. Estimation of the number of nucleotide
substitutions in the control region of the mitochondrial DNA
in humans and chimpanzees. Mol Biol Evol 10: 512–526.
Saitou N, Nei M, 1987. The neighbor-joining method: a new
method for reconstructing phylogenetic trees. Mol Biol Evol
4: 406–425.
Farris JS, Kallersjo M, Kluge AG, Bult C, 1995. Testing significance of incongruence. Cladistics 10: 315–319.
Rozas J, Rozas R, 1997. DnaSP version 2.0: a novel software
package for extensive molecular population genetics analysis.
Comput Applic Biosci 13: 307–311.
Tajima F, 1983. Evolutionary relationship of DNA sequences in
finite populations. Genetics 105: 437–460.
Nei, M. 1987. Molecular Evolutionary Genetics. New York: Columbia University Press.
Lynch M, Crease TR, 1990. The analysis of population survey
data on DNA sequence variation. Mol Biol Evol 7: 377–394.
Arnold ML, 1997. Natural Hybridization and Evolution. New
York: Oxford University Press.
Coluzzi M, Sabatini A, Petrarca V, di Deco MA, 1979. Chromosomal differentiation and adaptation to human environments in the Anopheles gambiae complex. Trans R Soc Trop
Med Hyg 73: 483–497.
Garcia BA, Caccone A, Mathiopoulos KD, Powell JR, 1996.
Inversion monophyly in African anopheline malaria vectors.
Genetics 143: 1313–1320.
Caccone A, Min GS, Powell JR, 1998. Multiple origins of cytologically identical chromosome inversions in the Anopheles
gambiae complex. Genetics 150: 807–814.
McLain DK, Collins FH, Brandling-Bennett AD, Were JBO,
1989. Microgeographic variation in rDNA intergenic spacers
of Anopheles gambiae in western Kenya. Heredity 62: 257–
264.
Lehmann T, Hawley WA, Kamau L, Fontenille D, Simard F,
Collins FH, 1996. Genetic differentiation of Anopheles gambiae populations from east and west Africa: comparison of
microsatellite and allozyme data. Heredity 77: 192–208.
Kamau L, Lehmann T, Hawley WA, Orago ASS, Collins FH,
1998. Microgeographic genetic variation of Anopheles gambiae mosquitoes from Asembo Bay, western Kenya: a comparison with Kilifi in coastal Kenya. Am J Trop Med Hyg 58:
64–69.
Miles SJ, 1978. Enzyme variation in the Anopheles gambiae
Giles group of species (Diptera: Culicidae). Bull Entomol Res
68: 85–96.
Black WC, Ferrari JA, Rai KS, Sprenger DA, 1988. Breeding
structure of a colonizing species: Aedes albopictus in the
United States. Heredity 60: 173–181.
mtDNA VARIATION IN A. ALBIMANUS
50. Black WC, Hawley WA, Rai KS, Craig GB, 1988. Breeding
structure of a colonizing species: Aedes albopictus in Malaysia. Heredity 61: 439–446.
51. Kambhampati S, Black WC, Rai KS, Sprenger D, 1990. Temporal variation in the genetic structure of a colonizing species:
239
Aedes albopictus (Skuse) in the United States. Heredity 64:
281–287.
52. Narang SK, Seawright JA, Suarez MF, 1991. Genetic structure
of natural populations of Anopheles albimanus in Colombia.
J Am Mosq Assoc 7: 337–345.