evidence for a new sibling species of anopheles minimus from the ryukyu archipelago, japan

16 Pages·2004·4.97 MB·English

Journal of the American Mosquito Contol Association, 17(2):9g_113,2OO1
Copyright O 2001 by the American Mosquito Control Association, Inc.
EVIDENCE FOR A NEW SIBLING SPECIES OF ANOPHELES MINIMUS
FROM THE RYUKYU ARCHIPELAGO, JAPAN
PRADYA SOMBOON,' CATHERINE WALTON,' ROSIE G. SHARPE,' YUKIKO HIGA,3 NOBUKO TUNO.3
YOSHIO TSUDA3 AND MASAHIRO TAKAGI3
ABSTRACT, The Anopheles minimus complex is known to comprise at least 2 sibling species (A and C) in
Thailand and Vietnam. This study investigated the specific status of An. minimus on Ishigaki Island, the Ryukyu
Archipelago, Japan using morphological and genetic analyses. Morphological studies revealed that almost all
(99.5Ea) of the adult mosquitoes are characterized by the humeral pale spot on the costa of their wings, a
character that partially differentiates species A and C elsewhere. A high frequency (81.4Vo) have a pale fringe
spot at the tip of vein lA, a character rarely observed in other An. minimus populations. Significant seasonal
variation in the size of wild An. minimus mosquitoes on the island was observed, with the largest size in the
winter. Scanning micrographs of the cibarial armature of females from Ishigaki lsland revealed that over 90Va
had cone filaments clearly differing in shape from those of species A or C. The Giemsa-stained metaphase
karyotypes of larval brain cells were somewhat similar to those of species A, with a few exceptions, bur were
very different from those reported for species C. Crossing experiments between species A (CM strain) from
Thailand and the progeny of An. minimus from Ishigaki Island (ISG strain) revealed postzygotic genetic incom-
patibility, although no prezygotic isolation. Hybrid progeny were only obtained from CM female x ISG male.
F, hybrid progeny were not obtained, since the hybrid males were sterile or almost sterile with atrophied testes
or abnormal spermatozoa, although the polytene chromosomes of hybrid larvae showed synapsis. The hybrid
females backcrossed with either CM or ISG males laid eggs with significantly lowered fertility and viability.
The sequence for the D3 region of the 28S gene of ribosomal DNA of the ISG strain differed from those of
species A and C. In addition, sequence data from Vietnamese mosquitoes suggest that the An. minimus complex
may contain additional species. The morphological, cytogenetic, molecular, and hybridization evidence together
suggest the existence of another sibling species of the An. minimus complex on Ishigaki Island, which is pro-
visionally designated An. minimus species E.
KEY WORDS Anopheles minimus, Japan, Thailand, morphology, DNA, hybridization, species complex
INTRODUCTION al. 1999). According to these reports, An. minimus
species A is probably the most widespread species
Anopheles minimus Theobald is one of the major
of the complex in the Oriental Region and has been
vectors of malaria throughout the Oriental Region
suggested to be Theobald's species (Harrison et al.
(Reid 1968, Rao 1984), apart from Sri Lanka, most
l99O). Species A in India, however, appears to be
of Malaysia, Singapore, Brunei, the Philippines,
different from that found in Thailand with respect
and Indonesia, where it is considered to be absent
to host feeding preference, resting behavior, spo-
(Harrison et al. 1990). An extensive study by Har-
rozoite positivity, and the response to vector control
rison (1980) provided useful information on intra-
using insecticide-treated bed nets (Jana-Kara et al.
and interspeciflc morphological variation of Az.
1995, Somboon et al. 1995). Species C was first
minimus and its close relatives. More recent studies,
reported in Thailand (Sucharit et al. 1988) and is
which employed primarily enzyme electrophoresis,
common in Kanchanaburi Province in sympatry
have shown that An. minimus is in fact a species
with species A, but absent or rare in other provinces
complex consisting of at least 2 closely related spe-
(Green et al. 1990, Sharpe et al. 1999). Recently,
cies (Sucharit et al. 1988, Green et al. l99O). Other
'Form based on enzyme electrophoresis, species C has
putative members of the complex, B' on
Hainan Island, China (Yu Yuan 1987), and a pos- been reported in Vietnam (Van Bortel et al. 1999),
sible species D in Thailand (Baimai 1989), have where it occurs in sympatry with species A in vary-
been proposed, but sufficient information to support ing proportions depending on locality, host prefer-
their species status is not yet available. ences, and season. The vector status of Az. minimus
Anopheles minimus species A is the predominant species C in transmitting malaria has not been de-
species of the complex in Thailand (Green et al. termined. Little is known about the distribution of
1990). It has also been reported in India (see Sub- these 2 species and any other sibling species in the
barao 1998 for review) and Vietnam (Van Bortel et An. minimus complex in other countries, including
Japan.
Anopheles minimus was the primary vector of
rDepartment of Parasitology, Faculty of Medicine, falciparum malaria in the southern islands of the
Chiang Mai University, Chiang Mai 50200, Thailand. Ryukyu Archipelago, Japan, particularly after
'(cid:0)Ecology and Evolution Group, School of Biology,
World War II until 1962 when transmission was
University of Leeds, Leeds LS2 9JI United Kingdom.
3 Department of Medical Entomology, Institute of Trop- completely disrupted (WHO 1966). The density of
ical Medicine, Nagasaki University, Nagasaki 852-8523, An. minimus was markedly reduced because of the
Japan. effect of DDT residual spraying during antimalaria
JUNB2 001 ANopnatns un,ttuus Rnou JepaN 99
control operations. However, recent surveys have tempts were made to find morphological characters
found that the species has become widely prevalent of the various stages that might help to differentiate
on Ishigaki and Iriomote Islands, along with, to a the specimens from the known sibling species. The
much lesser extent, Miyako Island (Toma and Mi- size of adult mosquitoes was determined by mea-
yagi 1986; Toma et al. 1996a,1996b). suring the length of wings, usually the right wing,
The discovery of sibling species in the An. min- from the axillary incision to the wing tip, excluding
imus cornplex elsewhere prompted us to investigate the fringe, using an ocular micrometer. Anopheles
the specific status of An. minimus in the Ryukyu minimus from the collection made in October 1998
Archipelago. Previously, Kanda et al. (1984) re- were retained in colony and designated as ISG
ported cytogenetic and hybridization studies among strain, and their progeny were used for hybridiza-
a strain of An. minimas from Ishigaki Island and 2 tion and other studies. An artiflcial mating tech-
strains (KCH-I and KCH-2) from Kanchanaburi nique (Ow Yang et al. 1963) was necessary to
Province, Thailand. However, because rearing con- maintain the colony. The morphological terminol-
ditions caused high mortality in the immature stag- ogy used in the report follows Harbach and Knight
es (even of the control crosses), no clear conclu- (1980).
sions could be made from the crosses. Nonetheless, A stenogamous strain of An. minimus (CM
the polytene chromosomes of the 3 strains were all strain) from northern Thailand (Somboon and Su-
similar in their banding patterns, and complete syn- wonkerd 1997) was taken to Nagasaki for crossing
apsis was observed in the hybrids of alt crosses. and other experiments. There was no genetic in-
The KCH-1 strain is most probably An. minimus compatibility when crossed with several strains of
species A (as most individuals lacked the humeral An. minimus s.l. in northern Thailand. The CM
pale spot), whereas the KCH-2 strain was most strain was confirmed as species A by enzyme elec-
probably An. minimus species C (as all individuals trophoresis (K. Sawabe, personal communication)
possessed the humeral pale spot). The suggestion and DNA analysis (see results). The insectarium
that species A and C could be distinguished on the was maintained at 27"C and'7O7o relative humidity
basis of the presence or absence of humeral pale (RH) with a photoperiod of l6:8 h L:D. The rearing
spots was first put forward by Sucharit et al. (1988). methods followed Kanda (1979).
Subsequently, however, Green et al. (1990) and Van Cibarial aftnature: Freshly killed adult females
Bortel et al. (1999) did not find this character to be were stored in TOVoe thanol until dissected. The ci-
completely diagnostic. This report presents the re- barial armature was dissected from heads in a drop
sults of morphological and genetic studies of An. of distilled water under a dissecting microscope,
minimus on Ishigaki Island that provide evidence covered with a coverslip and observed under a
that it is another species in the An. minimus com- compound microscope (400X magnification). For
plex. scanning electron microscopy, the dissected cibarial
armatures were dehydrated through a graded etha-
MATERIALS AND METHODS nol series and mounted on stubs. After being sput-
ter-coated with gold, the specimens were scanned
Ishigaki Island: The island (24'26'N, I24"ll,E\ in a JEOL scanning electron microscope (JSM-
is about 222 km2 and located approximately 400 km 84OAN; JEOL Ltd., Akishima, Japan).
southwest of the main island of Okinawa, Japan. A few An. minimus species A and C females
The island's climate is classifled as subtropical with from Thailand (Green et al. 1990, Sharpe et al.
average minimum-maximum temperatures for the 1999) and Vietnam (Van Bortel et al. 1999) iden-
past 22 years (1968-1990) in August (summer), tified previously by these authors using enzyme
October (autumn), and February (winter) of 27-32, markers or a DNA analysis were obtained as dry
22-27, and l6-21'C. respectively. More informa- specimens. In order to dissect these specimens, it
tion regarding the island, as well as the distribution was necessary to place the head-thoracic portions
of An. minimrrs, was provided by Toma and Miyagi in I mI of distilled water and incubate them at 4.C
(1986) and Toma er al. (1996a). for l-2 days. After incubation, the supernatant was
Mosquitoes: Fourth-stage larvae and pupae of replaced with TOVo ethanol. The specimens were
An. minimus were collected from Nishihama stream then dissected as above.
on the north side of the island in August and Oc- Metaphase karyotypes: The brain ganglia of the
tober 1998 and February 1999. Adults and imma- 4th-stage larvae were examined for metaphase kar-
ture stages were also collected from the satne a.rea yotypes, using a modification of the techniques de-
in August 1999. The specimens were kept in the scribed by French et al. (1962) and Baimai (197j).
field at room temperature (about 25.C) for a few Briefly, instead of placing the dissected brain gan-
days. The larvae were then transferred to an insec- glia in colcemid solution, the larvae were placed in
tary in Nagasaki and reared in stream water from O.lVo colchicine for 2 h and transferred to a drop
the collection site with the addition of larval food of 17o trisodium citrate for dissection of ttre brain.
until they pupated. Emerged adults with associated Fixing and staining methods then followed Baimai
larval and pupal exuviae were identified using kevs 0977).
in Tanaka et al. (1979) and Harrison (19g0-). Ar- Crossing experiments: The ISG strain was
r00 JounNer-o F THEA MERrc,q,MNo seuno CoNrnol AssocrerroN Vol. 17,N o.2
crossed with the CM strain to determine genetic tions, on an Applied Biosystems Model 373 auto-
compatibility. Virgin females separated at the pupal mated sequencer. All products were sequenced in
stage were placed in screened cups provided with both directions, and no ambiguities were observed.
3Vo sugar solution and offered a bloodmeal when 5 The sequences were aligned using the PILEUP pro-
days old. After the females took 1 bloodmeal, re- gram of the Genetics Computer Group (Program
ciprocal crosses using the forced mating technique Manual for the Wisconsin Package Version 8, 1994,
noted above were made between the virgin females 575 Science Drive, Madison, WI). The PHYLIP 3.4
and males of both strains. Following mating, each (Felsenstein 1993) programs, DNAPARS, SE-
female was isolated in an oviposition vial. Eggs QBOOT, CONSENSE, DNADIST, and NEIGH-
were counted and left until they hatched. Following BOR were used for phylogeny reconstruction, and
oviposition, females were dissected to check for TREEVIEW (Page 1996) was used to visualize
spermatozoa in their spermatheca, and eggs from phylogenetic trees.
uninseminated females were excluded. Some of the
females were checked after taking another blood-
RESULTS
meal and laying another batch of eggs. The larvae
reared from these egg batches were used to exam- Morphological characterlsflcs.' Some of the
ine polytene chromosomes of the salivary glands wing characters on An. minimus collected from Ish-
(Kanda 1979). Newly hatched larvae (normally 2 igaki Island and randomly selected progeny of the
days after oviposition) from each egg batch were ISG strain are summarized in Table l. All except a
counted and placed in rearing trays until they pu- small proportion of the mosquitoes collected in the
pated. Egg batches with no or little hatching were winter (February 1999) had the humeral pale spot
allowed to stand for another 3 days, after which on both wings. In addition, the pale fringe spot at
they were examined for embryonation. Pupae were the tip of vein 1A was present in high proportion.
removed daily, sexed, and placed separately in cups Collections in February 1999 failed to obtain
until the adults emerged. The F, hybrid adults that high numbers. Nonetheless, it is interesting that the
emerged were counted and their morphological wild 4th-stage larvae, as well as the emerged mos-
characteristics were noted. Their fertility and via- quitoes, were significantly larger than those col-
bility were observed by further crosses among the lected in other seasons (Table 1), suggesting a sea-
hybrids and backcrosses with the parental colonies. sonal effect on mosquito development. Six feral
The testes and ovaries of the hybrids were also dis- females previously collected from human bait in the
sected to check fertility. The crosses were made in same area in February 1997 were also large (mean
the same rooms housing the colonies, and the test wing length 3.13 mm, range 2.92-3.35 mm). The
specimens were kept in these rooms under identical mean wing length of the feral females collected in
laboratory conditions to those of the colonies. August 1999 (Table 1) was not significantly differ-
To observe if there is preferential mating behav- ent from that of An. minimus s.l. females collected
ior in the laboratory, a crossing experiment was from human bait in Chiang Mai, Thailand, in June
carried out in which virgin females or males of the L999 in the early rainy season (mean 2.46 mm,
CM strain were confined with the opposite sex of nnge 2.17-2.70 mm; t = O.79, df 155, P : O.43).
the ISG strain in a 30-cm cage. In each cage, 100 In the laboratory, the ISG females tended to be
pairs of l-4-day-old adults were released, and l0 larger than the CM females, as investigated in a
days after their release, the insemination rate was rearing experiment using similar conditions (mean
determined by checking spermathecae for the pres- wing length of ISG females 2.72 rr.m, range 2.49-
ence of sperm. As the control, 1@ pairs of the CM 2.97 mm1' mean of CM females 2.55 mm, range
and ISG mosquitoes were released in separate cag- 2.31-2.69 mm; t : 5.39, df 45, P < 0.0001).
es, and their insemination rate was compared. The eggs of both the CM and ISG strains resem-
DNA sequencing.' Genomic DNA was extracted bled the general descriptions of An. minimus s.l.
from individual adult mosquitoes using either a (Reid 1968). Most eggs had a complete deck, but
phenol-chloroform method (Sambrook et al. 1989) approximately half of the broods had some individ-
or a salting out method (Sunnucks and Hales 1996). uals with an incomplete deck. Such individuals
A product of approximately 370 base pair (bp) of were usually rare (lOVo of the brood), but occa-
the D3 region of the 28S gene of ribosomal DNA sionally more common (up to 897o in the CM and
(rDNA) was amplified by PCR using primers D3a 657o in the ISG strains). The adults reared from
(5' GACCCGTCTTGAAACACGGA 3') and D3b eggs with an incomplete deck usually produced
(5' TCGGAAGGAACCAGCTACTA 3'). Amplifi- eggs with a complete deck. The length of 653 CM
cation was performed according to Sharpe et al. and 755 ISG eggs randomly selected from the col-
(1999), although the hot start procedure was found onies varied from 0.383 to 0.508 mm (mean 0.421
to be optional. The products were purified on spin mm. SD 0.025) and 0.393 to 0.548 mm (mean
columns (Promega, Madison, WI) and sequenced O.452 mm, SD 0.031), respectively. The 4th-stage
using the PCR primers and TaqFS dye-terminator larvae of the 2 strains also resembled the general
fluorescent chemistry (Applied Biosystems, Foster descriptions of An. minimas s.l. (Reid 1968, Har-
City, CA), according to the manufacturer's instruc- rison 1980), including the anterior tergal plate on
Juvs 2001 ANopnuns umwus pnov hpnN 101
€o tr- il-VII, seta 0-IV V and the number of teeth of
9 n Q \ \ ;
(rnn -SNbro^o: o^ :-ord q tteheet hd oarts tohme ebnatusme) . H(uoswueavlleyr , 9s, eitnac 1lu-dI ining IS2G mlainrvuatee
(89/100) was fully palmate with leaflets with dis-
r
tinct shoulders, whereas in CM larvae with distinct
o FT o shoulders they were found in a much lower pro-
I
\ 9 q c n n $ poftion (7/100). Over 100 pupal exuviae ofthe CM
.Io r0O) o(.lo €oOl i:^Fi- F- 'J and ISG strains were examined, but no significant
H o ; e character differences were detected, including setae
o
7-VI. -VII. and O-I[-V[.
cr)
Cibarial arrnature: A total of 64 CM and 66 ISG
c.l females were examined. Figs. 1-6 show scanning
d c:
c.l micrographs of the cibarial armature. The appear-
PFi qne909Is ance of the armature under a light microscope
btb-R OO\F- 6v)\) (40OX) is shown in Figs. 7-8. The CM females
< : have cone filaments that are wide at the base and
? *
a0 n change abruptly to form a fine needle at the apex
al (in about tL to h of the filaments), producing a lan-
cetlike shape (Figs. 1-2). The ISG females typically
CO o
.H N . vlnnQn; have cone filaments that are relatively narrow and
o ra 8 F E ^ 3 F ; g gradually reduced to pointed ends, producing a
U) E thornlike shape (Figs. 4-5). None of the CM fe-
o H
males examined had thornlike filaments. However,
tr- a few ISG females (6166 or 9Vo) had the tip of fil-
qqqoecc9I.i E ii aments similar to the CM type. In addition, varia-
tion in the apex of the filaments was observed in
o\ O$O\O\Oi^i
O O\ oO r F-:l A both strains where a few filaments (up to 7, usually
: >
bb 1-3) had bifurcated or more rarely fimbriated ends
n
{oYc .si Eo C(FMig s.s tr3a,i n6 )(. 34T/h6i4s voar ri5a3nVt oi)s thmaonr ein cothmem IoSnG inst rtahine
\ E o n .a (19/66 or 29Vo). ^fhe shape of the cone filaments,
N
t a ca; including the occurrence of bifid or fimbriated ends,
lU=(vh- ,9 \ q q e \ \ F I can also be distinguished under the light micro-
€ iOt\ ncr\ ]oc'.\1 oi^\Oi S L) scope (Figs. 7-8).
B q0r n cil r'rnTzhse s paercmieastu Are aonf d 1C4, arneds p1e2c tfiveemlya,l esth aotf hAand. bmeienn-
oi identified by other authors (i.e., Green et al. 1990,
b0 ao I Sharpe et al. 1999, Van Bortel et al. 1999) were
a) cr.:) uts examined. Several wild-caught An. minimus s.l. fe-
b0 xXr*, NI r F A 6 males from Thailand and a few from Yunnan,
B oio*eo{irOi,ori)+FfR o otr southern China, were also examined. All exhibited
<hh€9 OO\O\ F..\O:) e.2 ; 5 lancetlike cone filaments similar to the CM type.
s .9EH Some, particularly those of Green et al. (1990),
v
o 6i : a = showed bifid or fimbriated ends, as found in several
I oiP CM females.
L v = ,e -a Metaphase karyotype: More than 4O larvae of
oBo N3':AxN0aUe= A X9 !doixatHPLrod= l>!Yi€Bia;ts=r Ho:, emmaeoctshao pmohfea tshneeu kmCabMrye ort ayn(p2den sI S. : GT h6se)t,yr a ciwonness rweis etuirnnegi f eoxormaf m I iinnpe acdihr rfooor-f
o heteromorphic sex chromosomes and 2 pairs of au-
q2 ; ! x - g x x
0) ad =b\ = U E tosomes similar to other members in the Myzomyia
s6 -t, s 6 > fi9 d F ; Series reported by Baimai et al. (1996a). However,
e € E 5
q a a x!?H€oF oo@ some variation was observed in the sex chromo-
9 l d! !x" 9- F c.2 somes and in the pericentric heterochromatin of au-
69 ll (.) iE!lY ![. Jt 'Tx4E.:' F F=E F.:Eo d t tosao)m Tehse aCs Mfo llsotwrasi.n (Figs. 9-16). Female larvae of
T ; E
- . , :IIl t'=i i aecq € Go. : goE0 io: c..9 .tr0.:' H! ttrhicis Xst racihnr oemxhoisboitmede s.s hTohrte anshdo./rotre lro nXg cshurbommeotsaocmene-
. , 1 s s E ; ' E E *9,6.0?,!F9^5, had a ratio of short to long arms of about 1:2 and
55e:s"f
F I the longer chromosome l:2.5 to 1:3 (Fig. 9), which
are comparable to X, and Xr, respectively, de-

JuNe 20Ol Atopauns ummus rnou hpaN 103
scribed by Baimai et al. (1996a). In addition, some F, hybrid males (5-7 days old) revealed that 85.47o
larvae exhibited X chromosomes having a smaller were completely sterile with atrophied testes (Fig.
"Ihe
heterochromatic area at approximately the middle 26). remaining had spermatozoa, but most
of the long arm (Figs. lO-12). This variation was were inactive and had an enlarged head (Fig. 27).
common in the short X chromosome, but rare in Their accessory glands looked normal, but the vas
the long. Because the meaning of this variation is efferens were very fragile, often causing the testes
not yet known, it is preliminarily called X," or Xr", to be detached during dissection. These males
depending on the length of the chromosomes. In failed to inseminate, although they succeeded in
male larvae, the X,, Xr, or X,. chromosomes were copulating with females when subjected to force
observed. There were short and long Y chromo- mating. Thus, no F, hybrids were obtained. The
somes (Figs. 13-16). The short chromosomes have ovaries of the hybrid females looked normal. When
a ratio of short to long arms of about l:3, compa- backcrossed to either CM or ISG males, they pro-
rable to the submetacentric Y, chromosome previ- duced eggs with low embryonation and hatching
ously reported by Baimai et al. (1996a). The long rates (Table 3). A high mortality was observed
chromosome has a ratio of l:4, being somewhat among lst-stage larvae. A total of l0 females and
subtelocentric (acrocentric) in appearance. Because 15 males were obtained from the backcrosses, but
this variation has not been reported previously, it is half of the females died within 5 days. Four of 8
designated as Y3 (Y, is described in An. minimus males dissected had spermatozoa, some of which
species C). In good preparations, the submetacen- had an enlarged head. Further crosses among the
tric autosomes usually exhibit a conspicuous block backcross progeny yielded a total of 146 eggs from
of pericentric heterochromatin in the short arm, 2 females, but the overall embryonation rate was
whereas the metacentric autosomes usually have a 32.97o and the hatching rate was 6.87o. Furlher
small amount of pericentric heterochromatin in crosses were not attempted.
both arms. More than 20 F, hybrid larvae from 5 of 7 cross-
b) The ISG strain (Figs. 17-24). Short and long es (CM X ISG F,, Table 2) were examined for poly-
submetacentric X chromosomes were observed in tene chromosomes. No asynapsis was observed
female larvae. Based on the ratio of short and long (Fig. 28).
arms, they are comparable to the X, and X, chro- Study of mating behavior showed that the adults
mosomes mentioned above. In addition, some lar- of the 2 strains mate readily in 30-cm cages. The
vae exhibited short X chromosomes with a lesser insemination rates of ISG female (Fu) x CM male
amount of heterochromatin or a small euchromatin and CM female X ISG male (Fu) crosses were
block near the distal end of the long arm (Figs. 17- 8l.97o (5O/61) and 8O.3Vo (53166), respectively.
18, 22). This type of chromosome is preliminarily These were as high as the insemination rate ob-
called X,o. In male larvae, X,, Xr, or X,o chromo- served in the CM control cage (86.OVo, 74/86), blt
somes were observed. There were 2 forms of Y higher than was observed in the ISG control cage
chromosome comparable to the Y, or.Y. as found (56.1Vo,23/41).
in the CM strain (Figs. 20-24). The autosomes look DNA sequence analysis: The D3 region of the
similar to those of the CM strain. rDNA has been used as the basis for a species iden-
Figure 25 shows a diagrammatic representation tification method for An. minimus species A and C
of Giemsa-stained mitotic karyotypes of the CM from Thailand and other members of the Minimus
and ISG strains observed in this study. Group (Sharpe et al. 1999) and is therefore poten-
Crossing experiments: Crosses carried out be- tially informative of the species status of the ISG
tween the CM and ISG strains resulted in hybrid strain. D3 rDNA sequence data was obtained from
progeny from only the CM females X ISG males the ISG strain; the CM strain; Anopheles flavirostris
(Table 2). There was a high rate of adult emer- (Ludlow) from Lombok Island, Indonesia; and An.
gence, apart from I cross that showed a rate ofonly minimus from Hoa Binh Province, Vietnam (kindly
ll.37o (ll/97). The sex ratio of each cross was l: provided by Wim Van Bortel), which was included
1 (all P > 0.05). Morphologically, hybrids having for comparison (Table 4). Figure 29 shows a phy-
a humeral pale spot on at least 1 wing were present logenetic tree of this data in the context of previous
in all crosses, but in varying proportions ranging sequence data from the An. minimus group, using
from lO-7OVo (overall 43Va) in the females and 10- An. flavirostris and An. aconitus Doenitz as out-
63Vo (overall 327o) in the males. Dissection of 82 groups. Both parsimony and distance-based meth-
(-
Figs. 1-8. The cibarial armature of CM and ISG females of Anopheles minimus examined under the scanning
electron microscope (1-6) and light microscope (7-8). 1, 2. Anterior and posterior aspects, respectively, ofCM females.
3. Anterior aspect of CM female showing a few filaments with bifid ends. 4, 5. Ant".io. and posterior aspects,
respectively, of ISG females. 6. Anterior aspect of ISG female showing cone filaments with bifid ends. 7-g. Anterior
view of CM and ISG females, respectively; the bifurcated ends of cone filaments are indicated by arrows. Note that
the scanning micrographs are present in different magnifications. cn, cone; Rd, rod.

JoURNALo p rrrp AvgnrcnN Mosquro CoNTRoLA ssocrnrroru Vor. 17, No. 2
tln
CM strain ISGs train
I T tl
tEl[tl ffiltt
llll
- t
Il lll Xt XsaX 2 X2oYj Ys lI tll Xt XrUX z Yr Y3
Ftg. 25. Diagrammatic representation and comparison of Giemsa-stained metaphase karyotypes of CM and ISG
strains of Anopheles minimus. Only I set of autosomes II and III is presented. Variable heteroihromatic portion is
depicted in black or shaded. The centromeres are indicated by constrictions of each chromosome. Chromosome lengths,
arm ratios, and heterochromatic portions are shown in proportion.
ods gave a tree of consistent topology. All the An. species A from Thailand and prior samples from
minimus s.l. sequences form a well-supported Vietnam (Sharpe et al. 1999). The implications of
clade, to which An. flavirostris is external. The the 3 novel sequences observed (1 from the Japa-
bootstrap values within the An. minimus s.l. clade nese ISG strain and 2 from Vietnamese mosquitoes)
reflect the limited amount of phylogenetic infor- are considered further in the discussion.
mation in the sample (because of the close rela-
tionships between An. minimus species) rather than
DISCUSSION
indicating a conflict between characters. Within the
An. minimus s.l. clade, the sequences fall into 2 In an attempt to determine the specific status of
clades, one of which contains An. minimrzs species An. minimus ISG, we have made extensive genetic
A and the other An. minimus species C. and morphological comparisons with species of the
Individuals of the CM strain had a sequence An. minimus complex and carried out hybridization
identical to that of An. minimas species A from experiments with An. minimus species A. In agree-
Thailand and Vietnam reported previously (Sharpe ment with previous observations (Tanaka et al.
et al. 1999), conflrming that this strain was An. min- 7979), we found that An. minimus mosquitoes in
imus species A (Fig. 29). Bott' individuals of the Ishigaki Island and its neighboring islands are often
ISG strain had the same sequence as each other, but characterized by the presence of both a humeral
this differed from all other sequences. One of the pale spot (86-lOOVo of individuals) and a pale
Vietnamese samples had the same sequence as An. fringe spot at the tip of vein lA (67-92Vo of indi-
minimus species C from Thailand. The other ZYiet- viduals) on the wing of the adult. The humeral pale
namese specimens fell within the clade containing spot is also common in An. minimus species C and
An. minimus species A. These sequences not only in the closely related species Anopheles aconitus,
differed from each otheq but also from An. minimus Anopheles jeyporiensis James, and Anopheles pam-
Table 2. Results from interstrain crosses between Anopheles minimus from Thailand (CM strain) and Japan
(ISG strain).1
No. of
Eggs
eggs Embry- Hatch-
Pupation rate3 Emergence rate3
No. of Average/ exa- onation ing
Crosses2 broods Total brood mined rate rates Female Male Total Female Male Total
Interstrain crosses
CM X ISG FI 7 729 104.1 n.d. n.d. 94.8 44.7 45.0 89.7 35.6 35.8 7r .4
ISG FI X CM 7 530 75.7 276 27.2 0
ISG F3 x CM 3 330 I10.0 301 26.2 0
'757
ISG F5 X CM 12 757 63.1 15.9 o.44
Control crosses
CMXCM 5 458 91.6 n.d. n.d. 97.8 39.9 41.5 81.4 38.6 37.r 75.7
ISG FI X ISG FI 3 251 83.7 n.d. n.d. 92.8 45.8 43.8 89.6 40.2 38.2 78.4
ISG F5 X ISG F5 5 342 68.4 n.d. n.d. 9O.4 n.d. n.d. n.d. n.d. n.d. n.d.
I n.d., not determined.
'(cid:0) All crosses are female x male.
3 Each rate was calculated by (total number of individuals that reached each developmental stage)/(total number of eggs) x 100
a All larvae died shortly after hatching.