Gene 232 (1999) 2534 www.elsevier.

com/locate/gene

Gene structure and polymorphism of an invertebrate

nitric oxide synthase gene
S. Luckhart a, *, R. Rosenberg b
a Virginia Tech, Department of Biochemistry, 124 Engel Hall, Blacksburg, VA 24061, USA
b US Army Medical Research Unit , Unit 64109, APO AE 09831-4019, Nairobi, Kenya
Received 14 December 1998; received in revised form 4 March 1999; accepted 20 March 1999; Received by K. Gardiner

Abstract

Nitric oxide synthases (NOSs) are ubiquitous in living organisms. However, little is known about the evolution of this large
gene family. The first inducible NOS to be described from an invertebrate regulates malaria parasite (Plasmodium spp.) development
in the mosquito Anopheles stephensi. This single copy gene shows the highest homology to the vertebrate neuronal isoforms,
followed by decreasing homology to endothelial and inducible isoforms. The open reading frame of 1247 amino acids is encoded
by 19 exons, which span #33 kilobases. More than 50% of the mosquito exons, distributed around the putative heme, calmodulin,
and FAD/NADPH cofactor-binding domains, are conserved with those of the three human genes. Repetitive elements identified
within the larger introns include a polymorphic dinucleotide repeat, two tandem repeats, and a putative miniature inverted repeat
transposable element. Sequence analysis and primer extension indicate that the upstream promoter is TATA-less with multiple
transcription start sites within #250 base pairs of the initiation methionine. Transcription factor binding sites in the 5-flanking
sequence demonstrate a bipartite distribution of lipopolysaccharide- and inflammatory cytokine-responsive elements that is
strikingly similar to that described for vertebrate inducible NOS gene promoters. 1999 Elsevier Science B.V. All rights reserved.

Keywords: Anopheles; Insect immunity; Malaria; Mosquito

1. Introduction neurotransmission, vasodilation, and antipathogen and

Nitric oxide (NO), produced during the conversion
of -arginine to -citrulline by NO synthase (NOS), is
both a free radical biological messenger and a killing
molecule. In humans, three NOS isoforms are encoded
by genes on different chromosomes and function in
Abbreviations: CaM, calmodulin; C/EBPb, CCAAT enhancer bind-
ing protein b; EDTA, ethylenediaminetetraacetic acid; FAD PPi,
flavin-adenine dinucleotide pyrophosphate; FISH, fluorescence in-situ
hybridization; FMN, flavin mononucleotide; IL6-RE, interleukin
6-response element; IRF-1, interferon response factor-1; kb, kilobase;
LPS, lipopolysaccharide; MITE, miniature inverted-repeat transpos-
able element; NADPH, nicotinamide-adenine dinucleotide phosphate;
NF-IL6, nuclear factor-interleukin 6; NF-kB, nuclear factor-kB; NO,
nitric oxide; NOS, nitric oxide synthase; Oct, octamer binding protein;
RACE, rapid amplification of cDNA ends; RTPCR, reverse tran-
scriptasepolymerase chain reaction; SR, straight repeat; SSC, 0.15 M
NaCl, 0.015 M sodium citrate, pH 7.0; STAT, signal transducer and
activator of transcription; TBE, 89 mM Tris, 89 mM boric acid,
2.6 mM EDTA buffer; TIR, tandem inverted repeat; TNF, tumor
necrosis factor; TR, tandem repeat; UTR, untranslated region.
* Corresponding author. Tel.: +1-540-231-5729;
fax: +1-540-231-9070.
E-mail address: luckhart@vt.edu (S. Luckhart)
antitumor responses. Nitric oxide production, however,
is ubiquitous in living organisms, having been detected
in bacteria (Chen and Rosazza, 1995), plants
(Delledonne et al., 1998) and invertebrates (Regulski
and Tully, 1995; Yuda et al., 1996; Luckhart et al., 1998).
Early work with insects demonstrated that NOS
expression is associated with neuronal signalling
( Elphick et al., 1993) and production of NO-loaded
salivary gland proteins that facilitate blood-feeding by
hematophagous insects (Ribeiro and Nussenzveig,
1993), whereas biochemical evidence suggested that
some insects also produced NO in response to bacterial
infection (Choi et al., 1995). Recently, the first molecular
evidence of an immune-inducible NOS was reported
from insects: NOS expression and NO production in
the mosquito Anopheles stephensi limit malaria parasite
development in the insect host (Luckhart et al., 1998).
Taken together, these studies demonstrate that insects
and vertebrates possess a similarly broad range of
NO-based physiologies, an observation that is perhaps
less surprising after an examination of the significant
homology between insect and vertebrate NOS genes.

0378-1119/99/$ see front matter 1999 Elsevier Science B.V. All rights reserved.
PII: S0 3 7 8 -1 1 1 9 ( 9 9 ) 0 0 12 1 - 3
26 S. Luckhart, R. Rosenberg / Gene 232 (1999) 2534
BLAST homology analysis (Altschul et al., 1990)
indicates that the open reading frames of the four known
insect NOS genes from Drosophila melanogaster
(Regulski and Tully, 1995), Rhodnius prolixus ( Yuda
et al., 1996), Manduca sexta (Nighorn et al., 1998), and
A. stephensi (Luckhart et al., 1998) show the highest
homology to neuronal NOS (NOS1; a mean of 49%
identity), followed by decreasing homology to endothe-
lial NOS (NOS3; a mean of 47% identity) and inducible
NOS genes (NOS2; a mean of 44% identity). This
pattern of homology may be the result of sampling error
(i.e. this is a coincidental finding from a small number
of insect genes), or it may reflect an evolutionary pattern
of NOS gene development. A striking conservation of
exon/intron structure across human NOS1, NOS3, and
NOS2 genes suggests that these genes are derived from
a single ancestral gene (Hall et al., 1994). The insect
coding sequence homologies suggest a link to a neuronal-
type ancestral form. However, a comparison of the
genomic organization across species is an essential com-
plement to coding sequence analysis, providing insights
into the time scale of gene evolution relative to species
divergence and the mechanisms of gene evolution, i.e.
loss versus acquisition of introns over time.
There were two major aims of the current study. One
aim was to characterize the first invertebrate NOS
genomic structure to provide an insight into NOS gene
evolution. A second aim was to characterize gene ele-
ments that may be associated with NOS biology in A.
stephensi. Because immune-related NOS expression in
vertebrates is driven by lipopolysaccharide (LPS ) and a
suite of inflammatory cytokines, the 5-flanking region
of A. stephensi NOS (AsNOS ) was analyzed as a prelude
to investigating potentially related regulatory signals for
inducible expression in mosquitoes. Published reports
have also indicated that human NOS polymorphisms
can be correlated with several human disease states,
including the prevalence and severity of malaria infection
(Anstey et al., 1996; Kun et al., 1998). This prompted
a search for repetitive elements and a preliminary analy-
sis of polymorphism in AsNOS as a foundation for
studying the correlation between various Anopheles poly-
morphs and malaria parasite infection in natural popula-
tions of mosquitoes.
2. Materials and methods
2.1. Genomic Southern and chromosome fluorescence
in-situ hybridization (FISH)
Aliquots of 17 mg of A. stephensi genomic DNA and
20 mg of Plasmodium falciparum DNA were digested
overnight with BamHI, HindIII, or PstI, electrophoresed
through 0.8% agarose, and transferred to a Magna-
Graph (Micron Separations) nylon membrane. The
membrane was screened at high stringency with an
[a-32P]dATP (Dupont New England Nuclear, specific
activity 3000 Ci/mmol ) random prime-labeled cDNA
fragment characterized from earlier reverse tran-
scriptasepolymerase chain reaction (RTPCR) assays
(Luckhart et al., 1998). In semi-quantitative RTPCR
assays of malaria parasite-infected and uninfected mos-
quitoes, greater quantities of this 3-NOS-derived frag-
ment were consistently amplified from Plasmodium-
infected mosquitoes, indicating that the fragment was
derived from an inducible gene (Luckhart et al., 1998).
Polytene chromosome squashes from A. stephensi were
prepared as described previously (Crampton et al.,
1997). Biotinylation of the NOS RTPCR fragment
(Luckhart et al., 1998; amino acids 10841182) and
FISH were performed according to the manufacturers
instructions (BioPrime, Life Technologies).
2.2. Isolation of genomic clones
Recombinant bacteriophage clones were isolated by
high-stringency plaque hybridization from an A. ste-
phensi EMBL3 genomic library (Clontech). Four DNA
fragments were used to individually screen 5105 clones;
these were derived from the NOS RTPCR assay
described in 2.1 (amino acids 10841182), 3- and
5-rapid amplification of cDNA ends (RACE; amino
acids 12131247 and 21242) and internal PCR ampli-
fication from a 5-RACE clone (amino acids 134234).
DNA probes were random prime-labeled using either
[a-32P]dATP or digoxigenin-dUTP (Boehringer
Mannheim); the hybridization and wash conditions were
performed using standard procedures (Crampton et al.,
1997) or according to the manufacturers instructions.
2.3 DNA sequencing
Restriction fragments from bacteriophage clones were
subcloned into pBluescript II KS (+) (Stratagene) or
pZErO-1 (Invitrogen), and PCR fragments amplified
from genomic DNA were subcloned into pCR 2.1-TOPO
( Invitrogen). All clones were sequenced using the
PRISM Dye Terminator Cycle Sequencing kit (Perkin
Elmer-Applied Biosystems) according to the manufac-
turers instructions and analyzed on an Applied
Biosystems automated DNA sequencer. Genomic DNA
sequences were analyzed using Genetics Computer
Group (GCG) sequence analysis software (Devereux
et al., 1984), Transcription Factor Search (Heinemeyer
et al., 1998), MatInspector v2.2 (Quandt et al., 1995),
Signal Scan (Prestridge, 1991), and GenWeb RepEater
(Compugen) analysis programs.
2.3. Rapid amplification of cDNA ends (RACE) and
primer extension
5-RACE and primer extension were used to confirm
5-coding sequence and identify the transcription start
site, and 3-RACE was used to confirm the 3-coding
 S. Luckhart, R. Rosenberg / Gene 232 (1999) 2534
sequence and to determine the site of poly(A) addition.
For 5-RACE, three nested gene-specific primers
designed against the region encoding amino acids 235
260 were used for cDNA synthesis and two rounds
of PCR amplification. A total of 500 ng of mRNA
from P. falciparum-infected A. stephensi (9 days post-
infection) were reverse-transcribed with Superscript II
(Life Technologies), treated with RNase H (Life
Technologies) and RNase A/T1 cocktail (Ambion), puri-
fied with GlassMax (Life Technologies), and tailed with
dCTP using terminal transferase (Life Technologies).
An anchored poly(G) upstream primer and nested gene-
specific primer were used in the first round of amplifica-
tion, and anchor-complement and gene-specific primers
modified for uracil DNA glycosylase capture-cloning
(Life Technologies) were used in a second round of
amplification. For 3-RACE, two nested gene-specific
primers were selected from the region of amino acids
11681186. cDNA was reverse-transcribed with
SuperscriptII (Life Technologies) using an anchored
poly( T ) primer. First-round PCR was performed with
upstream gene-specific and anchor-complement primers,
and second-round PCR was performed with nested gene-
specific and anchor-complement primers modified for
capture-cloning (Life Technologies). For primer exten-
sion, two 30-mers ( EXT1 and EXT2) were selected
within 100150 bp of 5-RACE termination sites.
Reactions with EXT1 and EXT2 were performed in
tandem and replicated with 15-mg aliquots of mRNA
from P. berghei-infected A. stephensi (2 days post-infec-
tion). Because AsNOS expression is inducible above a
background of apparently constitutive expression
(Luckhart et al., 1998), an aliquot of rRNA remaining
after mRNA isolation was used as a control. EXT1 and
EXT2 were labeled with T4 kinase (Life Technologies)
to a high specific activity with [c-32P]dATP (Dupont
New England Nuclear, specific activity 5000 Ci/mmol )
and purified using NucTrap (Stratagene) push columns.
Primer extension reactions were analyzed on 9% (wt/vol )
acrylamide gels (1:30 bis-acrylamide:acrylamide) with
7 M urea in 1 TBE buffer.
2.4. Microsatellite analysis
A preliminary investigation was designed to assess
NOS polymorphism in field-collected A. stephensi and
to serve as the basis for future studies. A single dinucleo-
tide repeat in intron 2 was randomly selected for PCR
analysis using the forward primer 5-CAGACGATGCA-
CGCTTTC-3 and the reverse primer 5-CTTTGG-
CAGGATGGATTC-3. Laboratory-reared A. stephensi
( Walter Reed Army Institute of Research) were used to
develop the assay and included as controls in all amplifi-
cations from field-collected mosquitoes. Wild A. ste-
phensi were collected in New Delhi, India, and a rural
Myanmar site (Oktwin Township) and kindly provided
27
by Dr C. Prasitissuk, World Health Organization SE
Asia Regional Entomologist.
3. Results
3.1. Isolation and organization of the A. stephensi NOS
gene
Genomic Southern hybridization indicated that the
NOS gene fragment amplified in semi-quantitative NOS
RT-PCR assays of Plasmodium-infected A. stephensi
(Luckhart et al., 1998) was derived from a mosquito
gene that is likely present as a single copy in the genome
( Fig. 1A). Chromosome FISH with the same gene frag-
ment showed a single point of hybridization in three
polytene squashes (Fig. 1B); localization to a specific
chromosome was not possible because limited informa-
tion is available on chromosome banding patterns in
this species.
The AsNOS gene was constructed from EMBL3
genomic DNA library clones (l55; Fig. 2A), library
subclones (l42-3.8, l42-2.3, l16-4.0, l2-7.0; Fig. 2A)
and two PCR products (PCR1 and PCR2; Fig. 2A)
amplified from genomic DNA. Ambiguous junctions
were confirmed by comparing PCR product sizes esti-
mated from known sequence with PCR products ampli-
fied from genomic DNA.
Alignments of AsNOS genomic with Drosophila NOS
cDNA sequence (Regulski and Tully, 1995) and of
AsNOS genomic with 5- and 3-RACE sequences were
used to identify an open reading frame of 1247 amino
acids that is encoded by 19 exons spread over approxi-
mately 33 kb of genomic DNA (Fig. 2A, Table 1). The
nucleotide sequence is available through Genbank under
Accession Nos AF130124 AF130134. The theoretical
molecular weight of the encoded protein was calculated
to be 141 636 Da, which is intermediate to the predicted
molecular weights of NOS3 (Gnanapandithen et al.,
1996; 132 880 Da) and NOS1 (Hall et al., 1994;
160 787 Da) and to the predicted molecular weights of
Rhodnius NOS ( Yuda et al., 1996; 132 331 Da) and
Drosophila NOS (Regulski and Tully, 1995; 151 938 Da).
The exon size ranged from 92 to 432 bp, whereas introns
ranged from 62 bp to over 7 kb ( Table 1). All introns,
except 2 and 3, were completely sequenced. The first
3.6 kb and last 11 bp were sequenced from intron 2, and
the first 280 bp and last 200 bp were sequenced from
intron 3. Exon/intron boundaries conformed to the
GT/AG splice donor/acceptor rule, with the associated
sequences showing strong homology to those reported
for exon/intron boundaries in the human NOS1 and
NOS3 genes (not shown; Marsden et al., 1993; Hall
et al., 1994). Cofactor binding sites exhibited a sequen-
tial arrangement that is common to the NOS gene
family, with the calmodulin binding site split between
exons 9 and 10 and the remaining sites encoded by
28 S. Luckhart, R. Rosenberg / Gene 232 (1999) 2534

Fig. 1. Genomic Southern hybridization and chromosome FISH of a NOS gene expressed in Plasmodium-infected Anopheles stephensi. (A) Genomic
Southern hybridization. Aliquots of genomic DNA from both species were digested overnight with BamHI (B), HindIII (H ), or PstI (P),
electrophoresed through 0.8% agarose in 89 mM Tris, 89 mM boric acid, 2.6 mM EDTA buffer, and transferred to a nylon membrane. Examination
of the agarose gel and membrane post-transfer indicated that all samples transferred efficiently. The membrane was screened at high stringency
with [a-32P]dATP-labelled NOS cDNA fragment RTPCR amplified from parasite-infected mosquitoes (Luckhart et al., 1998). The probe hybridized
only to single restriction fragments of A. stephensi DNA, indicating that this NOS is encoded by a single copy gene in the mosquito genome. (B)
Chromosome FISH. Polytene chromosome squashes from A. stephensi were prepared as described previously (Crampton et al., 1997). Chromosome
slides were heat-denatured at 90C for 12 min with 70 ml of 0.5 mg/ml of biotinylated NOS cDNA fragment (Luckhart et al., 1998) in 10% (wt/vol )
dextran sulfate and 1 hybridization solution, incubated overnight at 42C, washed in 0.2 sodium chloridesodium citrate (1 SSC=0.15 M
NaCl, 0.015 M sodium citrate, pH 7.0; 20 min at 42C, 20 min at room temperature), and covered with blocking solution for 15 min at room
temperature. Slides were then incubated with 100 ml of 4 mg/ml of BODIPYAFL-streptavidin in blocking solution for 30 min at 37C in the dark
and washed twice in Tris-buffered saline (25 mM Tris, 137 mM NaCl, 2.7 mM KCl, pH 7.0), 20 min each, at room temperature. The preps were
counterstained with 1 mg/ml of propidium iodide at room temperature for 5 min, drained, and coverslipped for microscopic examination. Three
squashes showed hybridization to a single, localized point (arrow).

individual exons ( Table 1). The putative translation
initiation site is in exon 1, with the sequence around
this site conforming with the consensus of Kozak
( Kozak, 1984; A at 3, G at +4). An in-frame transla-
tional stop signal was located in exon 19.
A comparison of the exon/intron structure of AsNOS
with the three human NOS genes (Marsden et al., 1993;
Chartrain et al., 1994; Hall et al., 1994) demonstrated
a high level of structural conservation (Table 1). In this
comparison, exons were considered essentially homolo-
gous if size differences were small and if relative positions
across species were conserved. For homologous exons,
most size differences ranged from one nucleotide to two
codons; the largest differences (three and seven codons)
were noted between mosquito exon 9 and the corre-
sponding human NOS1 and NOS3 exons ( Table 1).
Only mosquito exon 16 differed slightly from its human
counterparts in splice site location. When sequences
were aligned, mosquito exon 16 extended eight nucleo-
tides downstream of the corresponding human exon
boundaries. With the exception of intron 16, intron
splicing type was conserved when the corresponding
preceding exon was conserved. The relative frequencies
of AsNOS intron types were 67% (12/18) type 0, 6%
(1/18) type I, and 28% (5/18) type II; the same frequency
pattern (type 0 > type II > type I ) was reported for
human NOS1 and NOS3 genes (Marsden et al., 1993;
Hall et al., 1994).
Two exons in the mosquito gene appeared to be a
fusion of homologous human exons (Marsden et al.,
1993; Chartrain et al., 1994; Hall et al., 1994). AsNOS
exon 6 appeared to be be composed of human exons
9/10 (NOS1, NOS2) or 7/8 (NOS3), whereas exon 16,
although differing slightly in size and splice site at the
3-end, appeared to be homologous to human exons
23/24/25 (NOS1) or 21/22/23 (NOS3, NOS2). AsNOS
exon 6 preceded an intron with a splicing type conserved
with that of its human NOS gene counterparts; exon 16
did not conform to this pattern.
3.2. Determination of transcription initiation and
termination sites and sequence analysis of the upstream
region
A sequence analysis of four separate 5-RACE pro-
ducts suggested the presence of multiple transcription
 S. Luckhart, R. Rosenberg / Gene 232 (1999) 2534 29

Table 1
Features of A. stephensi NOS genomic structure and comparison with human NOS (NOS1, NOS3, NOS2)a exon/intron structure

Exon Features Intron Features Typeb Conserved?

Number Size (bp) Number Size (bp)

1 402 1 437 0 N
2 137 2 #7100 2 (dCdA) , SRc II N
n
3 (6,4,6d) 163 Heme 3 (6,4,6) #4000 0 Y
4 (7,5,7e) 92 4 (7,5,7) 1394 (dGdT ) II Y
n
5 (8,6,8) 142 5 (8,6,8) 118 0 Y
6 (9+10, 7+8, 9+10) 315 6 (10,8,10) 65 0 Y
7 182 7 73 II N
8 115 8 584 0 N
9 (13, 11, NCf)g 95 CaM 9 (13, 11, NC ) 5118 II Y
10 (14, 12, 13) 145 CaM 10 (14, 12, 13) 3444 TRh 0 Y
11 (15, 13, 14) 105 11 (15, 13, 14) 2699 MITEi, (dCdA) , TR 0 Y
n
12 408 FMN 12 63 II N
13 255 13 240 0 N
14 148 FAD PPi 14 63 I N
15 (22, 20, 20)j 170 15 73 0 Y
16 (23+24+25, 21+22+23, 432 FAD/NADPH 16 63 0 N
21+22+23)k
17 138 17 68 0 N
18 (27, 25, 25)l 192 NADPH Ade 18 (27, 25, 25) 436 0 Y
19 229 NADPH, 3-UTR

a NOS1 (Hall et al., 1994); NOS3 (Marsden et al., 1993); NOS2 (Chartrain et al., 1994).
b Type 0, splice junction occurs between codons; type 1, spice juction occurs after the first nucleotide of a codon; type II, spice junction occurs
after the second nucleotide of a junction.
c SR, straight repeats.
d NOS2 exon 7 is 91 base pairs (bp).
e NOS2 exon 8 is 143 bp
f NC, not conserved.
g Corresponding human exons are 86 bp (NOS1) and 74 bp (NOS3).
h TR, tandem repeats.
i MITE, miniature inverted-repeat transposable element (493 bp total size).
j Corresponding human exons are 170 bp (NOS1), 173 bp(NOS3) and 164 bp (NOS2).
k Corresponding human exons as indicated are 421 bp (NOS1), 421 bp(NOS3) and 418 bp (NOS2).
l Corresponding human exons are 195 bp.

start sites (R; Fig. 2B) within 250 bp of the initiation
methionine codon; two products terminated at the same
nucleotide (2R; Fig. 2B). Extension reactions with
EXT1 and EXT2 ( Fig. 2B) were performed in tandem
and replicated with 15-mg aliquots of mRNA from P.
berghei-infected A. stephensi (2 days post-infection).
Multiple labeled extension products were produced from
mRNA, but not rRNA, in replicated assays with EXT1
(P1; Fig. 2B) and EXT2 (P2; Fig. 2B). No predominant
start site was identified, suggesting that AsNOS tran-
scription proceeds from multiple initiation sites.
Inspection of the sequence downstream from an inverted
CCAAT box (101 to 105) did not reveal a consensus
TATAA motif. Multiple transcription start sites have
also been described for the TATA-less human NOS3
promoter (Marsden et al., 1993).
The upstream 2.3 kb of AsNOS showed a bipartite
distribution of putative inflammatory cytokine- and
LPS-responsive elements (Fig. 2B) that is remininscent
of the murine NOS2 promoter (Lowenstein et al., 1993;
Xie et al., 1993). Putative binding sites were identified
for four CCAAT enhancer binding protein b (C/EBPb)-
like elements, nuclear factor-interleukin 6 (NF-IL6),
vertebrate and invertebrate signal transducer and activa-
tor of transcription (STAT )-like elements, an IL6
response element ( IL6-RE), octamer binding protein
(Oct), a tumor necrosis factor response element
( TNF-RE ), activator protein 1 (AP-1), two interferon
regulatory factor 1 elements (IRF-1), GAAANN and
GATA factors, and nuclear factor kappa B (NF-kB).
Exon 19 did not contain the consensus polyadenyla-
tion signal AATAAA, but did contain an alternate
AAGAAA signal. Poly(A) addition occurred after the
C residue of a CA pair, 25 bp downstream of the
alternate signal, in three 3-RACE clones.
3.3. Repetitive elements in the A. stephensi NOS gene
and polymorphism of a (dCdA) element in intron 2
n
A variety of repetitive elements, including four dinu-
cleotide repeats, a pair of straight repeats, two tandem
30 S. Luckhart, R. Rosenberg / Gene 232 (1999) 2534
 S. Luckhart, R. Rosenberg / Gene 232 (1999) 2534 31

repeats and a MITE-like (miniature inverted-repeat

transposable element; Wessler et al., 1995) sequence,
were identified within introns 2, 4, 10, and 11 (Table 1)
of A. stephensi NOS. Genomic sequences flanking these
repeats have been submitted to GenBank and are avail-
able from the authors. In the human NOS1 gene,
polymorphic (dCdA) repeats were identified in the
n Anopheles stephensi NOS gene. DNA was isolated from single mosqui-
5-flanking region, in intron 2, and in exon 29 (Hall
et al., 1994). In the human NOS3 gene, tandem and
(dCdA) repeats were identified in introns 4 and 13,
n primer, an initial denaturation of 12 min at 94C, and 35 cycles of 30 s
respectively (Marsden et al., 1993), whereas a polymor-
phic pentanucleotide [(dC ) (dT ) ] repeat was identified
2 3n
in the upstream region of the human NOS2 gene ( Xu
et al., 1997).
The straight repeat pair in AsNOS intron 2 was not
associated with contiguous direct repeats and flanked
approximately 1 kb of sequence with no coding poten-
tial, indicating that this was not a retrotransposon.
Conversely, the MITE-like sequence in intron 11 pos-
sessed characteristics common to this family of transpos-
able elements ( Wessler et al., 1995): small size (492 bp),
presence of terminal inverted repeats (29-bp TIRs),
A+T richness (73% A+T ), presence of A+T-rich
target site duplications (5-TAA, 3-TAA), and potential
to form stable secondary structures (DG=86.5
90.4 kcal/mol, calculated using the default settings and
energy parameters of mfold server). MITE-like
sequences have recently been identified in the mosquito
Aedes aegypti ( Tu, 1997). These elements are often
highly duplicated and show a sequence homology that
segregates them into subfamilies. Additional copies of
this sequence are needed to confirm the presence of
MITE-like elements in A. stephensi.
Microsatellite analysis indicated that laboratory-
reared A. stephensi were homozygous for (dCdA) in
7
intron 2 ( Fig. 3). We detected seven alleles in mosquitoes
collected from rural Myanmar and New Delhi; with one
exception (the smallest allele), the alleles formed a ladder
separated by 2-bp steps. Four alleles were common to
Fig. 3. Characterization of (dCdA) microsatellite in intron 2 of the
n
toes and dissolved in 20 ml of 10 mM TrisCl, 1 mM EDTA buffer,
pH 8. Amplification conditions for 2 ml of DNA included 1.25 units of
AmpliTaq Gold (Perkin Elmer), 2.5 mM MgCl , 0.5 mM of each
2
at 94C, 30 s at 49C, and 30 s at 72C. Amplification products and a
10-bp DNA ladder (Life Technologies) were visualized with SybrGreen
II (Molecular Probes) following separation on 6% (wt/vol ) acrylamide
gels (1:30 bis-acrylamide:acrylamide) with 7 M urea in 1 TBE buffer.
Gels were photographed using a DC120 digital zoom camera ( Kodak);
Kodak Digital Science 1D image analysis software was used to deter-
mine sizes of the amplification products. Lane 1, PCR amplification
of genomic DNA from laboratory-reared A. stephensi. These mosqui-
toes were uniformly homozygous for (dCdA) and were used as con-
7
trols in all subsequent amplifications. Lanes 210, typical PCR
amplification of genomic DNA from individual A. stephensi field-col-
lected from Oktwin Township, Myanmar. Differences in PCR product
sizes represent unique alleles at this locus.
both groups; allele frequencies, however, showed marked
differences (data not shown) that likely reflect the geo-
graphic separation of the collecting sites. In mosquitoes
collected from New Delhi, where malaria transmission
is sporadic and A. stephensi is a primary vector (Sharma
et al., 1993), the distribution of six alleles among 32
individuals produced a heterozygosity or diversity index
of 0.80. The distribution of five alleles among 20 A.
stephensi collected from a rural Myanmar site charac-
terized by hyperendemic malaria ( Tun-Lin et al., 1995)
produced a heterozygosity index of 0.70; A. stephensi is
not a primary vector of malaria in this area ( Tun-Lin
et al., 1995). Polymorphism in the smallest repeat size
of AsNOS from two relatively small samples suggests
that a high level of AsNOS diversity is present in A.
stephensi populations.

Fig. 2. Genomic structure and 5-flanking region of the Anopheles stephensi NOS (AsNOS) gene. (A) Genomic structure. EMBL3 clones and PCR
products that comprise the sequence are indicated below the structure; size in kilobases of EMBL3 subclones completely or partially sequenced
are indicated as numbers following dashes. Exons are indicated as numbered shaded boxes. Exon 1 contains the deduced translational start site,
and exon 19 contains the deduced in-frame translational stop signal. Introns are indicated as lines; intron 2 is not drawn to scale, as indicated by
a slash. Repetitive elements found in introns 2, 4, 10 and 11 are indicated; element size reflects that detected in laboratory-reared A. stephensi.
(dCdA) or (dGdT ) , simple dinucleotide repeats; SR, straight repeats; TR, tandem repeats; MITE, miniature inverted repeat transposable
n n
element-like sequence. (B) 5-flanking region. The exon 1 coding sequence begins MetAlaAspThr Primers for primer extension determination
of the transcription start site ( EXT1 and EXT2) are indicated; termination sites for these primers (P) correspond numerically (i.e. P1 is a termination
site for EXT1). Termination sites for 5-RACE clones are shown as R; two clones terminated at a single nucleotide labeled 2R. A putative
inverted CCAAT box is indicated; no consensus TATAA sequence was identified. Analysis of the upstream sequence identified binding sites for a
variety of transcription factors associated with LPS- and cytokine-inducible expression, including C/EBPb (CCAAT/enhancer binding protein b),
NF-IL6 (nuclear factor IL6), vSTAT (vertebrate single transducer and activator of transcription), iSTAT (invertebrate STAT ), IL6-RE (IL-6
responsive element), Oct (octamer binding protein), TNF-RE (tumor necrosis factor responsive element), AP-1 (activator protein-1), IRF-1
(interferon regulatory factor-1), GATA and NF-kB-like (nuclear factor kappa B). Several GAAANN motifs are also indicated. The relatedness of
the underlined AsNOS sequence to a known or consensus binding sequence is indicated as a ratio; orientation of the element is indicated as +/.
The nucleotide sequence(s) reported in this study are available through Genbank under Accession Nos AF130124AF130134.
32 S. Luckhart, R. Rosenberg / Gene 232 (1999) 2534
4. Discussion
Genomic Southern hybridization and chromosome
FISH indicate that AsNOS is encoded by the A. stephensi
genome as a single copy gene. It is not currently known
whether A. stephensi possesses additional NOS genes.
Data from the four known insect genes, however, permit
us to speculate on the diversity of NOS genes in A.
stephensi and other insects. The pattern of coding
sequence homology to vertebrate NOS genes
(NOS1&NOS3&NOS2) of the four known insect genes
suggests that a single insect NOS gene type is associated
with diverse physiologies. This hypothesis is not without
precedent: the functions of human NOS1 and NOS2
may be broadened through heterodimerization of pep-
tides from alternative transcripts produced in response
to cytokines or tissue-specific signals ( Eissa et al., 1996;
Wang et al., 1997). Preliminary Northern blot data (not
shown) indicate that both constitutive and Plasmodium-
inducible transcripts are produced from AsNOS;
multiple NOS transcripts are also produced in
Drosophila (Regulski and Tully, 1995) and Rhodnius
( Yuda et al., 1996). Further characterization of insect
NOS genes is necessary to determine whether diverse
physiologies are associated with structural diversity of
a single gene type in these organisms.
The genomic organization of AsNOS also reflects the
homology pattern NOS1&NOS3&NOS2. Of the con-
served exons, ten of ten are homologous to correspond-
ing NOS1 and NOS3 exons, with sequence divergence
from AsNOS being greater for NOS3 than for NOS1
exons. In contrast, nine of ten conserved AsNOS exons
are homologous to the corresponding NOS2 exons.
Although the human exons 5 (NOS1, NOS2) and 3
(NOS3) do not correspond closely to AsNOS exon 2,
the more highly conserved COOH terminal amino acids
conform to the overall pattern of homology. Among
the last 22 amino acids encoded by the human exons,
the number conserved with AsNOS ranged from 14/22
for NOS1 exon 5, 11/22 for NOS3 exon 3, and 8/22 for
NOS2 exon 5.
Structural conservation of AsNOS exons 35, 911,
15 and 18 suggests that these exons predate the evolu-
tionary separation of insects and chordates, whereas the
apparent homology of exons 6 and 16 to small groupings
of the corresponding human exons suggests that some
intron insertion events have occurred relatively recently
in NOS gene evolution. Intron insertion into AsNOS
exon 16 would result in the segregation of the FAD-
and NADPH-binding functional domains that is seen
in the human NOS genes. This change could result in a
greater flexibility in duplicating and recombining these
domains in higher organisms, but it does not necessarily
indicate that small exons distinguish NOS genes in
higher organisms. It is perhaps more likely that early
NOS genes were more interrupted with introns, based
on the introns-early hypothesis (Poole et al., 1998),
and that intron removal and reinsertion have occurred
many times over the course of NOS gene evolution.
Additional genomic structures from divergent species
are needed to construct a hypothetical tree for NOS
gene evolution.
The 5-flanking region of AsNOS shows a bipartite
distribution of putative LPS- and cytokine-responsive
transcription factor binding sites that is reminiscent of
the murine NOS2 promoter (Lowenstein et al., 1993;
Xie et al., 1993). The presence of such an array suggests
that homologs of vertebrate cytokines and the transcrip-
tion factors that are responsive to them are present in
A. stephensi. Several studies have demonstrated that
invertebrates produce proteins similar to IL1, IL2, IL6,
and TNF-a (Ottaviani et al., 1993; Beck and Habicht,
1994). A number of insect immune gene promoters
possess sequences homologous to NF-IL6, GAANN,
GATA, and interferon-responsive elements that are
essential to promoter activity (Georgel et al., 1995;
Kadalayil et al., 1997). An IRF-1-like protein binds to
one of these elements in immune-stimulated Drosophila
(Georgel et al., 1995). Together, these data strongly
suggest that functional characterization of the AsNOS
promoter will likely identify additional conserved
inflammatory elements and pathways.
A growing number of studies have demonstrated
significant correlations between repeat element poly-
morphs and human disease states. Human NOS2 poly-
morphism, although not repeat-based, was associated
with malaria severity and time to re-infection (Anstey
et al., 1996; Kun et al., 1998). These studies suggested
that Anopheles NOS polymorphism, while useful for
population analyses, may also be correlated with malaria
parasite infection risk in field-collected mosquitoes. In
the present study, one dinucleotide repeat demonstrated
a substantial degree of polymorphism and allele frequen-
cies that reflected the collection sites. Although infection
data were not collected for the mosquitoes analyzed,
this polymorphism and previously observed differences
in A. stephensi involvement in malaria transmission at
the two sites are intriguing. The MITE-like sequence
indicates potential for further diversity if actively
transposing MITEs are present in A. stephensi.
Movement of MITEs into and out of the A+T-rich
sequence of AsNOS 5- and 3-flanking regions and
introns could also influence gene expression, mRNA
processing, and NO synthesis in response to malaria
parasite development. A more complete analysis of
Anopheles NOS polymorphism and malaria parasite
infection, therefore, may identify significant predictive
correlations for transmission risk assessment in the field.
Acknowledgements
Oligonucleotide primer synthesis and DNA sequen-
cing were performed in core facilities at Walter Reed
 S. Luckhart, R. Rosenberg / Gene 232 (1999) 2534
Army Institute of Research ( Washington, DC ) and at
the Uniformed Services University of the Health
Sciences (Bethesda, MD). This work was supported by
grants from the National Institutes of Health, the United
Nations Development Program/World Bank/World
Health Organization Special Program for Research and
Training in Tropical Diseases, Uniformed Services
University of the Health Sciences, and the Department
of Defense.
References
Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990.
Basic local alignment search tool. J. Mol. Biol. 215, 403410.
Anstey, N.M., Weinberg, J.B., Hassanali, M.Y., Mwaikambo, E.D.,
Manyenga, D., Misukonis, M.A., Arnelle, D.R., Hollis, D.,
Mcdonald, M.I., Granger, D.L., 1996. Nitric oxide in Tanzanian
children with malaria: inverse relationship between malaria severity
and nitric oxide production/nitric oxide synthase type 2 expression.
J. Exp. Med. 184, 557567.
Beck, G., Habicht, G.S., 1994. Invertebrate cytokines. Ann. NY Acad.
Sci. 712, 206212.
Chartrain, N.A., Geller, D.A., Koty, P.P, Sitrin, N.F., Nussler, A.K.,
Hoffman, E.P., Billiar, T.R., Hutchinson, N.I., Mudgett, J.S., 1994.
Molecular cloning, structure and chromosomal localization of the
human inducible nitric oxide synthase gene. J. Biol. Chem. 269,
67656772.
Chen, Y., Rosazza, J.P.N., 1995. Purification and characterization of
nitric oxide synthase (NOSNoc) from a Nocardia species. J. Bacte-
riol. 177, 51225128.
Choi, S.K., Choi, H.K., Kadono-Okuda, K., Taniai, K., Kato, Y.,
Yamamoto, M., Chowdhury, S., Xu, J., Miyanoshita, A., Debnath,
N.C., Asaoka, A., Yamakawa, M., 1995. Occurrence of novel types
of nitric oxide synthase in the silkworm, Bombyx mori. Biochem.
Biophys. Res. Commun. 207, 452459.
Crampton, J.M., Beard, C.B., Louis, C., 1997. The Molecular Biology
of Insect Vectors of Disease: a Methods Manual. Chapman & Hall,
New York.
Delledonne, M., Xia, Y., Dixon, R.A., Lamb, C., 1998. Nitric oxide
functions as a signal in plant disease resistance. Nature 394,
585588.
Devereux, J., Haeberli, P., Smithies, O., 1984. A comprehensive set of
sequence analysis programs for the VAX. Nucleic Acids Res. 12,
387395.
Eissa, N.T., Strauss, A.J., Haggerty, C.M., Choo, E.K., Chu, S.C.,
Moss, J., 1996. Alternative splicing of human inducible nitric oxide
synthase mRNA, tissue-specific regulation and induction by cytok-
ines. J. Biol. Chem. 271, 2718427187.
Elphick, M.R., Green, I.C., OShea, M., 1993. Nitric oxide synthesis
and action in an invertebrate brain. Soc. Neurosci. Abstr. 19
(483), 4
Georgel, P., Kappler, C., Langley, E., Gross, I., Nicolas, E., Reichhart,
J.-M., Hoffmann, J.A., 1995. Drosophila immunity. A sequence
homologous to mammalian interferon consensus response element
enhances the activity of the diptericin promoter. Nucleic Acids Res.
23, 11401145.
Gnanapandithen, K., Chen, Z., Kau, C.L., Gorczynski, R.M., Mars-
den, P.A., 1996. Cloning and characterization of murine endothelial
constitutive nitric oxide synthase. Biochim. Biophys. Acta 1308,
103106.
Hall, A.V., Antoniou, H., Wang, Y., Cheung, A.H., Arbus, A.M.,
Olson, S.L., Lu, W.C., Kau, C., Marsden, P.A., 1994. Structural
33
organization of the human neuronal nitric oxide synthase gene
(NOS1). J. Biol. Chem. 269, 3308233090.
Heinemeyer, T., Wingender, E., Reuter, I., Hermjakob, H., Kel, A.E.,
Kel, O.V., Ignatieva, E.V., Ananko, E.A., Podkolodnaya, O.A.,
Kolpakov, F.A., Podkolodny, N.L., Kolchanov, N.A., 1998. Data-
bases on transcriptional regulation: TRANSFAC, TRRD, and
COMPEL. Nucleic Acids Res. 26, 364370.
Kadalayil, L., 1997. Adjacent GATA and kappa B-like motifs regulate
the expression of a Drosophila immune gene. Nucleic Acids Res.
25, 12331239.
Kozak, M., 1984. Compilation and analysis of sequences upstream
from the translational start site in eukaryotic mRNAs. Nucleic
Acids Res. 12, 857872.
Kun, J.F.J., Mordmuller, B., Lell, B., Lehman, L.G., Luckner, D.,
Kremsner, P.G., 1998. Polymorphism in promoter region of induc-
ible nitric oxide synthase gene and protection against malaria.
Lancet 351, 265266.
Lowenstein, C.J., Alley, E.W., Raval, P., Snowman, A.M., Snyder,
S.H., Russell, S.W., Murphy, W.J., 1993. Macrophage nitric oxide
synthase gene: two upstream regions mediate induction by
interferon gamma and lipopolysaccharide. Proc. Natl. Acad. Sci.
USA 90, 97309734.
Luckhart, S., Vodovotz, Y., Cui, L., Rosenberg, R., 1998. The
mosquito Anopheles stephensi limits malaria parasite development
with inducible synthesis of nitric oxide. Proc. Natl. Acad. Sci. USA
95, 57005705.
Marsden, P.A., Heng, H.H.Q, Scherer, S.W., Stewart, R.J., Hall, A.V.,
Shi, X.-M., Tsui, L.-C., Schappert, K.T., 1993. Structure and chro-
mosomal localization of the human constitutive endothelial nitric
oxide synthase gene. J. Biol. Chem. 268, 1747817488.
Nighorn, A., Gibson, N.J., Rivers, D.M., Hildebrand, J.G., Morton,
D.B., 1998. The nitric oxide-cGMP pathway may mediate com-
munication between sensory afferents and projection neurons in
the antennal lobe of Manduca sexta. J. Neurosci. 18, 72447255.
Ottaviani, E., Franchini, A., Fontanili, P., 1993. Presence of cytokine-
like molecules in molluscan hemocytes. Biochem. Biophys. Res.
Commun. 195, 984988.
Poole, A.M., Jeffares, D.C., Penny, D., 1998. The path from the RNA
world. J. Mol. Evol. 46, 117.
Prestridge, D.S., 1991. SIGNAL SCAN: a computer program that
scans DNA sequence for eukaryotic transcriptional elements. Curr.
Adv. Biol. Sci. 7, 203206.
Quandt, K., Frech, K., Haras, H., Wingender, E., Werner, T., 1995.
MatInd and MatInspector: new fast and versatile tools for detec-
tion of consensus matches in nucleotide sequence data. Nucleic
Acids Res. 23, 48784884.
Regulski, M., Tully, T., 1995. Characterization of dNOS: a Drosophila
Ca2+/calmodulin-dependent nitric oxide synthase. Proc. Natl.
Acad. Sci. USA 92, 90729076.
Ribeiro, J.M.C., Nussenzveig, R.H., 1993. Nitric oxide synthase activ-
ity from a hematophagous insect salivary gland. FEBS Lett. 330,
165168.
Sharma, S.N., Subbarao, S.K., Choudhury, D.S., Pandey, K.C., 1993.
Role of An. culicifacies and An. stephensi in malaria transmission
in urban Delhi. Ind. J. Malariol. 30, 155168.
Tu 1997. Three novel families of miniature inverted-repeat transpos-
able elements are associated with genes of the yellow fever
mosquito, Aedes aegypti. Proc. Natl. Acad. Sci. USA 94,
74757480.
Tun-Lin, W., Thu, M.M., Than, S.M., Mya, M.M., 1995. Hyperen-
demic malaria in a forested, hilly Myanmar village. J. Am. Mosq.
Control Assoc. 11, 401407.
Wang, Y., Goligorsky, M.S., Lin, M., Wilcox, J., Marsden, P.A., 1997.
A novel, testis-specific mRNA transcript encoding an
NH2-terminal truncated nitric oxide synthase. J. Biol. Chem. 272,
1139211401.
Wessler, S.R., Bureau, T.E., White, S.E., 1995. LTR-retrotransposons
34 S. Luckhart, R. Rosenberg / Gene 232 (1999) 2534

and MITEs: important players in the evolution of plant genomes.
Curr. Opin. Genet. Dev. 5, 814821.
Xie, Q.-W., Whisnant, R., Nathan, C., 1993. Promoter of the mouse
gene encoding calcium-independent nitric oxide synthase confers
inducibility by interferon gamma and lipopolysaccharide. J. Exp.
Med. 177, 17791784.
Xu, W., Liu, L., Emson, P.C., Harrington, C.R., Charles, I.G., 1997.
Evolution of a homopurine-homopyrimidine pentanucleotide
repeat sequence upstream of the human inducible nitric oxide syn-
thase gene. Gene 204, 165170.
Yuda, M., Hirai, M., Miura, K., Matsumura, H., Ando, K., Chinzei,
Y., 1996. cDNA cloning, expression and characterization of a nitric
oxide synthase from the salivary glands of the blood-sucking insect,
Rhodnius prolixus. Eur. J. Biochem. 242, 807812.