Báo cáo Y học: Characterization of the lectin from females of Phlebotomus duboscqi sand flies doc

Characterization of the lectin from females of
Phlebotomus
duboscqi
sand flies
Petr Volf
1
, Sona Skarupova
´
1
and Petr Man
2,3
1
Department of Parasitology and
2
Department of Biochemistry, Charles University, Prague, Czech Republic;
3
Institute of
Microbiology, Academy of Sciences of the Czech Republic, Prague, Czech Republic
Lectin from females of the important sand fly vector,
Phlebotomus duboscqi (Diptera: Psychodidae), was isolated
by immunoaffinity chromatography using a minicolumn
with immobilized anti-lectin immunoglobulins. Carbohy-
drate-binding specificity of active fractions corresponded to
that of midgut and salivary gland lysates. Haemagglutina-
tion was inhibited by
D
-glucosamine,
D
-galactosamine and
D
-mannosamine. The homogeneity and molecular mass of

the purified lectin was examined by SDS/PAGE in both
reducing and nonreducing conditions. The active fractions
showed one band strongly stained by Coomassie blue or
silver nitrate; the molecular mass of the lectin was 42 kDa
under nonreducing and 44 kDa under reducing conditions.
SDS/PAGE of active fractions from the gel filtration
revealed four to six protein bands, but the 42/44-kDa protein
present in all active fractions was the only component
reacting with specific antibodies in Western blots. Local-
ization of the lectin in the gut of females was studied using
indirect immunofluorescence on sections. The positive
reaction of specific antibodies was localized in the lumen and
along the microvillar surfaces of epithelial cells. The lectin
was partially sequenced and characterized by MS. Peptide
maps were obtained by MALDI-TOF MS, and several
sequence tags were identified from tandem mass spectra on
an ion trap. These sequences displayed high similarity to
salivary protein precursors previously identified in a cDNA
library of the sand flies Phlebotomus papatasi and Lutzomyia
longipalpis. Two main hypotheses on the role of female lectin
in Leishmania development are discussed.
Keywords: immunoaffinity chromatography; lectin; Phle-
botomus duboscqi;sandfly.
Females of the sand fly genera Phlebotomus and Lutzomyia
are insect vectors of Leishmania parasites, causative agents
of a wide spectrum of human diseases, ranging from self-
healing cutaneous lesions (e.g. Leishmania major)to
progressive and fatal systemic involvement (e.g. Leishmania
donovani). The vector part of the life cycle is crucial for
Leishmania circulationinnature;Leishmania develop and

multiply in the midgut of female sand flies and are
transmitted by bite to mammalian hosts. Identification of
molecular interactions at the sand fly–Leishmania interface
is fundamental to any study of vector competence; the
mechanisms responsible for controlling sand fly susceptibi-
lity to Leishmania infections, however, are not fully
understood.
The interplay between the parasite and the vector appears
to include a number of potential barriers to complete
parasite development. Midgut digestive enzymes may
inhibit the early phase of development [1–3], peritrophic
matrix behaves as a physical barrier to parasite migration
[2,4,5], and putative receptors specific to the parasite
glycoconjugate lipophosphoglycan (LPG) seem to be
involved in species-specific binding of parasites to the
epithelium of the sand fly midgut [6,7]. Another intrinsic
factor of the vector that might be involved in sand
fly–Leishmania interaction is the lectin activity present in
the sand fly midgut.
In insects, lectins act as effector, receptor and regulatory
molecules in the processes of self/nonself recognition and
innate immunity, cell adhesion and tissue differentiation.
They also play a regulatory role in pathogen–vector
interactions (for reviews, see [8,9]). In Reduviid bugs or
Glossina flies, they are involved in the establishment and
maturation of trypanosomatid infections (for reviews, see
[10,11]).
In sand flies, lectin activity has been demonstrated in
lysates of various tissues, including head, gut, ovaries,
haemolymph [12,13] and salivary glands [14]; the same

sugar-binding specificity of activities found in different
tissues suggested the presence of the same lectin molecule.
The lectin activity is sex-dependent, and high activities were
found exclusively in females [15]. In vitro, midgut lysates of
female sand flies agglutinated Leishmania promastigotes
[12,16], but experiments on inhibition of lectin activity
in vivo did not clarify the role of this molecule in the
Leishmania life cycle [17].
The main aim of this work was to purify and characterize
the lectin from females of Phlebotomus duboscqi,an
important vector of L. major in Subsaharian Africa. The
small size of sand flies required a simple, preferentially one-
step purification technique. Preliminary experiments
showed that affinity chromatography is not suitable because
of the low affinity of the lectin for simple carbohydrates, and
immunoaffinity chromatography was therefore used.
Correspondence to P. Volf, Department of Parasitology, Vinicna 7,
128 44 Prague 2, Czech Republic.
Fax: + 420 2 24919704, Tel.: + 420 2 21953196,
E-mail:
Abbreviation: LPG, lipophosphoglycan.
(Received 22 June 2002, revised 16 September 2002,
accepted 5 November 2002)
Eur. J. Biochem. 269, 6294–6301 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.03349.x
MATERIALS AND METHODS
Sand flies
A colony of P. duboscqi (Senegal strain, obtained from
R. Killick-Kendrick, Imperial College at Silwood Park,
Ascot, Surrey, UK) was reared under standard conditions
at 25–26 °C and 14 h : 10 h light/dark photoperiod.

Adults were maintained on 50% sucrose and females
bloodfed on anaesthetized mice once a week. Dissected
midguts, salivary glands or whole bodies of females were
homogenized with a Teflon homogeniser in Eppendorf
tubes on ice in Tris/NaCl buffer (20 m
M
Tris/HCl,
pH 7.6, 150 m
M
NaCl). Previous experiments showed
that the addition of calcium and other bivalent cations is
not required. Samples were centrifuged at 10 000 g for
10 min at 4 °C, and supernatants were collected for
subsequent assay. Protein concentration was determined
by the Bradford assay (Bio-Rad kit) using BSA (Sigma)
as a standard.
Haemagglutination assay
Sand fly tissue lysates and fractions obtained from chro-
matography assays were tested for haemagglutination
activity in Tris/NaCl buffer on 96-well U-bottomed micro-
titration plates, as described previously [13,18]. Briefly,
samples (50 lL) were serially diluted twofold, and an equal
volume of 2% (v/v) suspension of washed rabbit erythro-
cytes was added. The plates were incubated for 1 h at room
temperature, the haemagglutination titre being defined as
the reciprocal of the highest dilution showing visual
agglutination of erythrocytes. In controls, the lysates or
fractions were replaced by Tris/NaCl buffer only.
Assay of haemagglutination inhibition
Inhibition tests with carbohydrates were performed in

microtitration plates as described elsewhere [13,14]. The
carbohydrate-binding specificity of the lectin activity was
known from previous experiments. Therefore, five
noninhibitory monosaccharides,
D
-glucose,
D
-galactose,
D
-mannose, N-acetyl-
D
-glucosamine and N-acetyl-
D
-gal-
actosamine, and three inhibitory ones,
D
-glucosamine,
D
-galactosamine and
D
-mannosamine, were chosen to
compare the binding specificity of gut lysates and active
fraction from chromatography techniques. Twofold dilu-
tions of carbohydrates were prepared in 50 lLTris/NaCl
buffer and mixed with an equal volume of lysate or
chromatography fractions adjusted to contain  1.5
haemagglutination units. Then, an equal volume of 2%
suspension of rabbit erythrocytes was added to each well.
The minimum concentration of inhibitors required to
block haemagglutination was determined after 2 h incu-

bation at room temperature.
Anti-lectin immunoglobulins
Antibodies against haemagglutinin of P. duboscqi females
were raised in rabbits (female; Great Chinchila;  4 kg) as
described by Yeaton [19]. The rabbit was bled from the ear,
andwashednativeerythrocyteswereadjustedto2%
suspension in sterile Tris/NaCl buffer. Then 15 mL eryth-
rocyte suspension was agglutinated by filtered gut lysates of
P. duboscqi females for 1 h. Then, agglutinated erythrocytes
were washed three times in sterile Tris/NaCl buffer by
centrifugation at 750 g for 15 min, the pellet was resus-
pended in 2 mL incomplete Freund’s adjuvant (Difco,
Detroit, MI, USA) and injected subcutaneously into the
same rabbit. Four immunizations at 2-week intervals were
followed after 2 months by an intravenous booster (without
adjuvant). Immune sera were obtained 1 week after the final
booster. IgG fractions of the sera were isolated by rivanol
(2-ethoxy-6,9-diaminoacridine lactate hydrate) and ammo-
nium sulfate [20]. Purified IgG samples were stored in
aliquots at )70 °C and used for Western blotting and
immunoaffinity chromatography.
Localization of the lectin in midgut tissue
Previous studies detected haemagglutination activity in both
midgut epithelium and the midgut content of females [15].
In this work, indirect immunofluorescence with anti-haem-
agglutinin IgG was used for more precise localization of the
lectin in the midgut tissue. P. duboscqi females (4–6-days-
old) were fixed in 70% ethanol and embedded in LR White
resin according to instructions of the manufacturer (Poly-
sciences, Warrington, Lancs, UK). Parasagittal sections,

1–2 lm thick, obtained with an Ultracut E (Reichert Jung,
Wien, Austria), were incubated overnight at 4 °C with Tris/
NaCl buffer containing 0.1% (v/v) Tween 20 (Tris/NaCl/
Tween) and 5% (w/v) BSA to prevent nonspecific binding
of serum to hydrophobic epitopes of the section. Then, the
sections were incubated with immune rabbit serum diluted
in Tris/NaCl/Tween, washed, and incubated with fluoresc-
ein isothiocyanate-conjugated swine anti-rabbit immuno-
globulins (Sevac, Prague, Czech Republic) diluted in Tris/
NaCl/Tween. In control sections, the preimmune serum
from the same rabbit was used or the serum incubation step
was omitted (control of unspecific binding of the conjugate).
Both incubations with sera and conjugate were performed
in a moist chamber for 45 min at 37 °C. Sections stained
with Evans blue were photographed using a Jenalumar
(Karl Zeiss-Jena) fluorescent microscope.
Purification of the lectin
Medium-pressure liquid chromatography system BioLogic
(Bio-Rad) and two different methods, gel filtration and
immunoaffinity chromatography, were used for purification
of the lectin. Samples were prepared from batches of about
500 P. duboscqi females, 3–8 days old, which had never had
a blood meal. Females were homogenized in 500 lLTris/
NaCl buffer as described above, and supernatant containing
 1mgÆmL
)1
protein was filtered using 0.45 lmMicrocon
filters (Amicon) before loading on the chromatography
columns.
Preliminary experiments with three different gel-filtration

columns showed Superose 12 to be the most suitable one;
400 lL filtered supernatant was applied to the column
(1 · 40 cm), pre-equilibrated with Tris/NaCl buffer.
Elution was carried out with the same buffer at a flow rate
of 0.4 mLÆmin
)1
. Fractions were examined for haemagglu-
tination activity and active fractions were checked for
binding specificity using selected carbohydrates (see above).
Then the fractions were concentrated (to  0.5 mgÆmL
)1
)
Ó FEBS 2002 Characterization of lectin from sand fly females (Eur. J. Biochem. 269) 6295
using centifugation on Microcon YM-10 filters (Amicon),
and protein composition was determined by electrophoresis.
For immunoaffinity chromatography, purified anti-lectin
IgG was immobilized on CNBr-activated Sepharose. Sam-
ples with high agglutinating activity from gel filtration or
about 500 lL filtered supernatant from homogenized
femaleswereloadedontoacolumn(4mL)equilibratedwith
Tris/NaCl buffer. After extensive washing with Tris/NaCl
buffer (flow rate 0.5 mLÆmin
)1
for 70 min), the immunospe-
cific bound protein was eluted with a linear pH gradient of
citrate buffer (50 m
M
citrate, 100 m
M
NaCl, pH 2.6). The

eluted fractions were adjusted to pH 7.5 with 1
M
Tris, tested
for haemagglutinating activity and carbohydrate-binding
specificity, concentrated, and analysed electrophoretically.
Electrophoresis and Western blots
Supernatants of tissue lysates or concentrated fractions
obtained by chromatography were boiled for 3 min in
sample buffer with or without 2% (v/v) 2-mercaptoethanol
andloadedontoanSDS/10%polyacrylamidegel(thick-
ness 0.75 mm). Separations were carried out at a constant
200 V for 50 min using Mini-Protean II apparatus (Bio-
Rad). Gels were stained with Coomassie Brilliant blue
R-250 or silver nitrate.
Proteins separated by SDS/PAGE were transferred to
nitrocellulose membrane (0.2 lm; Serva) using a Semiphor
unit (Hoefer Scientific Instruments). Blotting was performed
for90minat1.5mAÆcm
)2
at room temperature. The blot
was rinsed in Tris/NaCl/Tween, stained for proteins with
1% (w/v) Ponceau red, and incubated for 2 h in Tris/NaCl/
Tween with 5% (w/v) skimmed milk (Oxoid, UK). The
incubation with rabbit immunoglobulins diluted 1 : 200 in
Tris/NaCl/Tween (2 h at room temperature) was followed
by repeated rinsing in Tris/NaCl/Tween and then by 1 h
incubation with swine anti-rabbit immunoglobulins conju-
gated with horseradish peroxidase (Sevac; diluted 1 : 1000
in Tris/NaCl/Tween). The peroxidase reaction product was
developed in 4-chloro-1-naphthol solution.

In-gel digestion and esterification
For MS analysis and protein microsequencing, the active
fraction from immunoaffinity chromatography was used for
electrophoresis on a 12% (w/v) gel. A Coomassie-stained
spot was excised from the gel and cut into small pieces. The
gel was washed with water. The wash solution was
discarded and replaced with 100 m
M
ethylmorpholine
acetate buffer, pH 8.5, in 50% acetonitrile. After complete
gel destaining in a sonication bath, the gel pieces were
washed with water, shrunk by dehydration in acetonitrile,
reswelled in water, and dehydrated again by addition of
acetonitrile. The supernatant was removed and the gel was
partly dried in a vacuum centrifuge. The gel pieces were then
swollen in a digestion buffer containing 50 m
M
ehylmorph-
oline acetate, pH 8.0, 1 m
M
CaCl
2
, 10% (v/v) acetonitrile
and sequencing grade trypsin (trypsin to protein ratio
1 : 75). After overnight digestion (shaking at 37 °C), the
resulting peptides were extracted from the gel by increasing
the acetonitrile concentration to 50% and by addition of
trifluoroacetic acid to a final concentration of 1%. Subse-
quently, the tubes were sonicated for 15 min. The liquid
phase with the extracted peptides was divided into two

tubes, and one was subjected to ethyl esterification in
ethanolic HCl prepared by mixing 1 mL ethanol with
160 lL acetyl chloride. The reaction was carried out for
2.5 h and stopped by drying in a SpeedVac concentrator.
The second part of the peptide mixture was dried in a
SpeedVac concentrator. Both samples were redissolved with
5 lL 50% (v/v) acetonitrile/1% (v/v) trifluoroacetic acid.
MALDI-TOF MS
A saturated solution of a-cyano-4-hydroxycinnamic acid
(Sigma) in aqueous 50% (v/v) acetonitrile/0.2% (v/v)
trifluoroacetic acid was used as a MALDI matrix. A 2-lL
volume of sample and 2 lL matrix solution were premixed
in a tube; 0.5 lL of the mixture was placed on the sample
targetandallowedtodryattheambienttemperature.
Positive ion MALDI mass spectra were measured on a
Bruker BIFLEX reflectron time-of-flight mass spectrometer
(Bruker-Franzen, Bremen, Germany) equipped with a
SCOUT 26 sample inlet, a gridless delayed extraction ion
source, and a nitrogen laser (337 nm) (Laser Science,
Cambridge, MA, USA). The ion acceleration voltage
was 19 kV, and the reflectron voltage was set at 20 kV.
Spectra were calibrated externally using the monoisotopic
[M + H]
+
ion of a-cyano-4-hydroxycinnamic acid and a
peptide standard (angiotensin II; Aldrich).
lHPLC-nano ESI ion trap MS
The tryptic peptides were loaded on to a homemade
capillary column (0.18 · 100 mm) packed with reversed-
phase resin (MAGIC C-18; 200 A

˚
;5lm; Michrom Bio-
Resources, Auburn, CA, USA) and separated using a
gradient from 5% (v/v) acetonitrile/0.5% (v/v) acetic acid to
35% (v/v) acetonitrile/0.5% (v/v) acetic acid for 50 min at a
flow rate of 2 lLÆmin
)1
. The column was connected directly
to an LCQ
DECA
ion trap mass spectrometer (ThermoQuest,
San Jose, CA, USA) equipped with a nanoelectrospray ion
source. The spray voltage was held at 1.6 kV and the tube
lens potential was )2 V. The heated capillary was kept at
175 °C with a voltage of 13 V. Full-scan spectra were
recorded in positive mode over the mass range 350–1300
atomic mass units. MS/MS data were automatically
acquiredonthemostintenseprecursorionineachfull-
scan spectrum. Acquired MS/MS spectra were interpreted
manually.
RESULTS
Western blots with female tissue
For both midgut lysate and salivary gland lysate, the purified
IgG fraction of the immune serum specifically recognized a
single protein band. The band represented a major salivary
protein and a minor midgut protein; its molecular mass
was 42 kDa under nonreducing and 44 kDa under reducing
conditions (Fig. 1). When the whole immune serum was
used, an additional protein band of molecular mass
 70 kDa was visualized in the midgut lysate (Fig. 1) but

not in salivary glands. Both preimmune serum and the
negative control without serum gave no reaction with
both antigens. A similar result was observed when midgut
lysate of the closely related species Phlebotomus papatasi
6296 P. Volf et al.(Eur. J. Biochem. 269) Ó FEBS 2002
was used: anti-haemagglutinin IgG specifically recognized
the 42–44-kDa region (data not shown).
Localization of the lectin
Anti-haemagglutinin IgG reacted with the content of the
midgut lumen and along the surfaces of midgut epithelial
cells. A positive reaction was observed in both thoracic and
abdominal parts of the midgut (Fig. 2). Antibody binding
was specific: no reaction was observed on control sections
incubated with preimmune sera or with fluorescein conju-
gate only.
Purification of the lectin by gel filtration
Gel filtration of whole body lysates on a Superose 12
column revealed about six protein peaks. Haemagglutina-
tion activity against rabbit erythrocytes was observed
between peaks 3 and 4, with a broad maximum in fractions
18–21 (Fig. 3A). The carbohydrate-binding specificity of the
active fractions was similar to that of midgut lysates.
Inhibition was achieved with
D
-glucosamine,
D
-galactosa-
mine (both at 20 m
M
final concentration) and

D
-mannosa-
mine (40 m
M
), whereas
D
-glucose,
D
-galactose,
D
-mannose,
N-acetyl-
D
-glucosamine and N-acetyl-
D
-galactosamine had
no inhibitory effect at 160 m
M
final concentration. The
active fractions were concentrated and submitted to SDS/
PAGE under reducing conditions; four to six protein bands
were detected in each fraction (Fig. 4). The 44-kDa protein
present in all active fractions was the only component that
reacted with anti-haemagglutinin immunoglobulins in
Western blotting. Antibodies from preimmune rabbit serum
gavenoreaction(Fig.4).
Isolation of the lectin by immunoaffinity
chromatography
Fractions with haemagglutinating activity (titres 1 : 8 and
1 : 16) against native rabbit erythrocytes were present in the

first peak eluted from the immunoaffinity column by low
pH (Fig. 3B). The homogeneity and molecular mass of the
purified lectin were examined by SDS/PAGE in both
reducing and nonreducing conditions. The active fractions
showed one band strongly stained with Coomassie blue or
silver nitrate; the molecular mass of the lectin was 42 kDa
under nonreducing and 44 kDa under reducing conditions
(Fig. 4). The second peak eluted from the column at low pH
had no haemagglutinating activity and contained a frag-
ment of IgG detached from the column (data not shown).
MS and data processing
In the first step, we analyzed a tryptic peptide mixture by
MALDI-TOF MS. Despite the fact that the spectrum
contained a considerable number of fully resolved peaks
(Fig. 5A), the approach of peptide mapping gave no
positive hit. In the second step, the peptide mixture was
analyzed by LC-MS/MS on an ion trap mass spectrometer.
In this experiment, we obtained several tandem mass spectra
of peptides, which were interpreted manually (Fig. 5B). The
sequences were read out from y-ion and b-ion series
according to known fragmentation mechanisms proposed
and described elsewhere [21]. We also measured the peptide
mixture after ethyl esterification and thus were able to assign
the number of acidic residues in each peptide. Because the
ion trap instrument does not allow detection of low-mass
and ammonium ions, we were not able to assign the
N-terminal di-residues accurately in all cases.
Fig. 2. Parasagittal section of the abdomen of P. duboscqi female under
the fluorescent microscope. Autofluorescence of the cuticular sclerit (sc)
surrounding thoracic muscles (mu). Specific reaction of the midgut

lumen (lu) and microvillar layer of the midgut epithelium (ep) with
purified anti-lectin immunoglobulins. Ft, Fat body.
Fig. 1. SDS/PAGE and Western blotting of lysates from salivary
glands and midgut of P. duboscqi females. Protein (1–3 lgperlane)was
loaded and samples run as described in Materials and methods. Gels
werestainedwithsilvernitrate,andreactiononWesternblotswas
visualized with 4-chloro-1-naphthol solution. SDS/PAGE: lane 1,
protein markers (BenchMark Protein Ladder; Gibco); lane 2, salivary
gland lysate under reducing conditions; lane 3, the same salivary gland
lysate sample under nonreducing conditions; lane 4, midgut lysate
under nonreducing conditions. Western blotting (nonreduced sam-
ples): lane 5, reaction of midgut lysate with immune (+) and preim-
mune (–) serum; lane 6, reaction of midgut lysate with purified
immunoglobulins from immune (+) and preimmune (–) sera.
Ó FEBS 2002 Characterization of lectin from sand fly females (Eur. J. Biochem. 269) 6297
The sequences obtained are summarized in Table 1.
Searches were carried out against a nonredundant protein
database by using
MS
-
BLAST
(l-heidelberg.
de/Blast2/msblast.html). High similarity was found to a
42-kDa salivary protein from P. papatasi (SwissProt
number Q95WD9).
DISCUSSION
Lectin from P. duboscqi females was purified and charac-
terized by liquid chromatography and SDS/PAGE as a 42–
44-kDa protein. Inhibition tests with carbohydrates gave
identical results in purified fractions and crude midgut

lysates. This confirmed that the purified lectin corresponds
to the haemagglutinin present in various sand fly tissues,
including the midgut and salivary glands. Similar electro-
phoretic migration of the molecule in reducing and nonre-
ducing conditions implies a monomer structure. Most insect
lectins characterized to date contain polypeptide chains
linked by disulfide bridges, and their activity is Ca
2+
dependent [22,23].
In bloodsucking Diptera, namely tsetse flies and mosqui-
toes, lectins have been purified from the haemolymph by
various chromatographic techniques, including affinity
chromatography [23,24]. In midgut tissue, chromatographic
isolation has been less successful and therefore erythrocytes
have frequently been used as affinity ligands. In the
mosquito Anopheles gambiae, Mohamed and Ingram [22]
identified a 65-kDa lectin band using adsorption of midgut
extracts with human erythrocytes. In tsetse flies, Grubhoffer
et al. [25] detected two lectin bands of molecular mass 27
and 29 kDa in Glossina tachinoides midgut using Western
blots with anti-haemagglutinin immunoglobulins raised by
the technique of Yeaton [19]. In the gut tissue of another
tsetse fly, Glossina longipennis,Osiret al.[26]purifieda
protein with two subunits of 27 and 33 kDa; the larger was
proposed to be an agglutinin with glucosamine-binding
lectin activity, while the smaller showed trypsin activity.
The lectin from P. duboscqi females was partially
sequenced and characterized by MS. Peptide maps were
obtained by MALDI-TOF, and several tandem mass
spectra were observed using an ion trap. Several sequence

tags were identified from the tandem mass spectra. These
sequences displayed a high similarity to salivary protein
Fig. 4. SDS/PAGE and Western blots of the purified female sand fly
lectin. The haemagglutinating fractions from gel-filtration and immu-
noaffinity chromatography were concentrated, loaded on the gel, and
run under nonreducing conditions as described in Materials and
methods. The gel was stained with silver nitrate, and reaction on
Western blots was visualized with 4-chloro-1-naphthol solution. Lane
1, protein markers (Bio-Rad); lane 2, active fraction (no. 20) from
Superose 12; lane 3, Western blot of fraction 20 with preimmune (–)
and immune anti-lectin serum (+); lane 4, active fraction (no.18) from
immunoaffinity chromatography.
Fig. 3. Purification of P. duboscqi lectin by gel filtration (A) and
immunoaffinity chromatography (B). (A) Supernatant from 500 females
(500 lL)wasfilteredandloadedontoSuperose12column
(1 · 40 cm), pre-equilibrated with Tris/NaCl buffer. Elution was car-
ried out with the same buffer (flow rate 0.4 mLÆmin
)1
). Fractions were
examined for haemagglutinating activity as described in Materials and
methods. (B) Filtered supernatant from 500 females was loaded on to a
minicolumn (4 mL) with anti-lectin immunoglobulins immobilized on
CNBr-activated Sepharose. After the column had been washed with
Tris/NaCl buffer (buffer A) the immunospecific bound protein was
eluted by a linear pH gradient of buffer B (citrate buffer; 50 m
M
citrate,
100 m
M
NaCl, pH 2.6). The eluted fractions were adjusted to pH 7.5

with 1
M
Tris and tested for haemagglutinating activity as described in
Materials and methods.
6298 P. Volf et al.(Eur. J. Biochem. 269) Ó FEBS 2002
precursors found in the cDNA library of the closely related
species P. papatasi [27]. The coded proteins, named PpSP42
(Q95WD9) and PpSP44 (Q95WD8) and a similar Yellow
protein from salivary glands of another sand fly Lutzomyia
longipalpis showed motifs of the major royal jelly proteins of
honeybee (Apis mellifera) and Yellow protein of Drosophila
[27]. The biological role of these proteins remains unknown;
the major royal jelly proteins are believed to play a major
role in nutrition because of their high essential amino-acid
content [28]. Interestingly, in sand flies these 42–44-kDa
salivary proteins represent the main immunogens strongly
reacting with antibodies from hosts repeatedly bitten by
sand flies [29].
In the gut tissue of females, the lectin is present free in the
lumen of thoracic and abdominal parts of the midgut and
along the microvillar surface of midgut epithelium. These
observations confirmed previous results obtained by haem-
agglutination tests. Volf and Killick-Kendrick [15] showed
that high haemagglutination activity was present in both
parts of the midgut. In unfed females, the activity was
almost equally distributed between the epithelium and the
midgut content, whereas in fed females the activity titres
were elevated in the lumen, and most of the activity was
detected in the peritrophic space surrounded by peritrophic
matrix. Part of the midgut lectin activity may originate from

saliva swallowed during the feeding of the fly. However, the
midgut activity peaked not immediately after the blood
meal but 48 h later [15], suggesting that most of the lectin
present in midgut lumen is secreted by midgut epithelium
and passes through peritrophic matrix during blood meal
digestion. However, the site of synthesis of sand fly lectin
is not necessarily limited to salivary glands and midgut.
Biosynthesis of insect lectins takes place mainly in the fat
body or haemocytes [30,31]. In sand flies, various levels of
the lectin activity were found in different tissues, including
the ovaries and haemolymph [13], and hybridization in situ
will be required to identify lectin expression sites.
Two main hypotheses may be considered for the role of
sand fly lectins in Leishmania development: they could be
involved in Leishmania attachment to sand fly midgut or
they could serve as inhibitors of Leishmania development.
The ability of Leishmania promastigotestoattachtothe
midgut epithelium of female sand flies is a critical compo-
nent of vectorial competence. There is a close evolutionary
fit between sand fly vectors and Leishmania parasites in
some Old World leishmaniases: P. papatasi and Phleboto-
mus sergenti are susceptible only to L. major and Leishma-
nia tropica, respectively. The failure of other parasite species
to develop in these sand flies coincided with a time of
defecation of the blood meal remnants and is correlated
with the ability of promastigotes to attach to the sand fly
midgut by this time (for a review, see [32]). The attachment
is controlled by polymorphic, species-specific structures on
the parasite LPG [6,7] and a strong species-specific vector
competence of P. papatasi and P. sergenti is explained by

the presence of specific LPG-binding receptors on midgut
epithelium [32].
Midgut lectin of P. papatasi binds LPG of L. major [13],
and part of the activity is associated with the surface of the
midgut epithelium (see above). However, it is unlikely that it
is involved in the attachment or is identical with the LPG
receptor. Lectin activity with the same sugar-binding
specificity was present in all Phlebotomus and Lutzomyia
species studied [13,18], and the same is true for 42–44-kDa
Table 1. Sequences obtained from tandem mass spectra using lHPLC-
nanoESIiontrapMS.Comparison of data with the similar sequences
from salivary protein of P. papatasi. Peptides were separated on a
reversed-phase capillary column and analyzed on an ion trap mass
spectrometer equipped with a nanoelectrospray ion source (details are
given in Materials and methods). Acquired MS/MS spectra were
interpreted as depicted in Fig. 5. Numbers assign positions in the
polypeptide chain; (I/L) indicates that leucine or isoleucine is present in
this position (isobaric amino acids). Other characters in parentheses
may be in reverse order.
42-kDa salivary protein precursor
of P. papatasi (Q95WD9)
Sequences obtained from
P. duboscqi females
59-MLFFGIPR-67 M(I/L)FFG(I/L)PR
71-VPITFAQLSTR-81 VP(I/L)TVAQ(I/L)STR
90-NPPLDK-95 DPPLDK
167-NPLGYGGFAVDVVNPK-182 TP(I/L)GYGGFAVD
VVNPK
238-FKAGIFGIALGDR-250 (LE)TG(I/L)FG(I/L)
A(I/L)GDR

295-TEAIALAYDPETK-307 TEA(I/L)A(I/L)AYDPETK
Fig. 5. MS of purified lectin of P. duboscqi females. (A) MALDI-TOF
mass spectrum of a tryptic peptide mixture after in-gel digestion. Peaks
labelled with an asterisk represent peptides successfully sequenced by
lHPLC-nano ESI MS. (B) Sequencing by lHPLC-nano ESI MS,
example of the peptide 1184.7.
Ó FEBS 2002 Characterization of lectin from sand fly females (Eur. J. Biochem. 269) 6299
protein precursors found by Valenzuela et al. [27]. There-
fore, the lectin cannot serve as the species-specific receptor
responsible for different vectorial competence of various
sand fly species.
The second hypothesis is based on similarity to the
Glossina–Trypanosoma system, where the lectin activity of
the vector was proposed to prevent establishment of
parasites in the ectoperitrophic space [33] and trigger cell-
suicide pathways in trypanosomes, analogous to apoptosis
in metazoa (for a review, see [34]). In addition, Glossina
lectins were reported to play a dual role, not only to kill
parasites but also to provide a signal for the maturation of
established ones [35]. At present, we cannot exclude the
possibility that sand fly lectin may affect Leishmania
development by similar mechanisms. Purification of the
lectin by immunoaffinity chromatography promotes further
study of the role of this molecule in sand fly–Leishmania
interaction.
ACKNOWLEDGEMENTS
We thank Professor R. Killick-Kendrick for the P. duboscqi colony and
help during sand fly research, and Professor L. Grubhoffer for long-
term support of parasite–vector studies. We are also grateful to Dr K.
Bezous

ˇ
ka,DrI.Hrdy´ and R. S
ˇ
uta
´
k for advice on lectins and
chromatography techniques and Vera Volfova
´
for sand fly dissections.
This study was supported by the Ministry of Education (projects MSM
113100001 and 113100004) and the Grant Agency of the Czech
Republic (project 206/03/0325).
REFERENCES
1. Borovsky, D. & Schlein, Y. (1987) Trypsin and chymotrypsin-like
enzymes of the sand fly Phlebotomus papatasi infected with
Leishmania and their possible role in vector competence. Med. Vet.
Entomol. 1, 235–242.
2. Pimenta, P.F.P., Modi, G.B., Pereira, S.T., Shahabuddin, M. &
Sacks, D.L. (1997) A novel role for the peritrophic matrix in
protecting Leishmania from the hydrolytic activities of the sand fly
midgut. Parasitology 115, 359–369.
3. Schlein, Y. & Jacobson, R.L. (1998) Resistence of Phlebotomus
papatasi to infection with Leishmania donovani is modulated by
components of the infective bloodmeal. Parasitology 117,467–
473.
4. Schlein, Y., Jacobson, R.L. & Shlomai, J. (1991) Chitinase
secreted by Leishmania functions in the sandfly vector. Proc. R.
Soc. Lond. B 245, 121–126.
5. Walters, L.L., Irons, K.P., Modi, G.B. & Tesh, R.B. (1992)
Refractory barriers in the sand fly Phlebotomus papatasi (Diptera:

Psychodidae) to infection with Leishmania panamensis. Am. J.
Trop.Med.Hyg.46, 211–228.
6. Pimenta, P.F.P., Saraiva, E.M.B., Rowton, E., Modi, G.B.,
Garraway, L.A., Beverley, S.M., Turco, S.J. & Sacks, D.L. (1994)
Evidence that the vectorial competence of phlebotomine sand flies
for different species of Leishmania is controlled by structural
polymorphisms in the surface lipophosphoglycan. Proc. Natl
Acad. Sci. USA 91, 9155–9159.
7. Sacks, D.L., Modi, G., Rowton, E., Spath, G., Epstein, L., Turco,
S.J. & Beverley, S.M. (2000) The role of phosphoglycans in
Leishmania–sand fly interactions. Proc. Natl Acad. Sci. USA 97,
406–411.
8. Ratcliffe, N.A. & Rowley, A.F. (1987) Insect responses to para-
sites and other pathogens. In Immune Responses in Parasitic
Infections (Soulsby, E.J.L., ed.), Vol. 4, pp. 271–332. CRC Press,
Boca Raton, FL, USA.
9. Yoshino, T.P. & Vasta, G.R. (1996) Parasite–invertebrate host
immune interactions. In Advances in Comparative and Environ-
mental Physiology (Cooper, E.L., ed.), Vol. 24, pp. 125–167.
Springer-Verlag, Berlin, Germany.
10. Ingram, G.A. & Molyneux, D.H. (1991) Insect lectins: role in
parasite–vector interactions. Lectin Rev. 1, 103–127.
11. Grubhoffer, L., Hypsa, V. & Volf, P. (1997) Lectins (hemagglu-
tinins) in the gut of the important disease vectors. Parasite 4,203–
216.
12. Wallbanks, K.R., Ingram, G.A. & Molyneux, D.H. (1986) The
agglutination of erythrocytes and Leishmania parasites by sandfly
gut extracts: evidence for lectin activity. Trop. Med. Parasitol. 37,
409–413.
13. Pala

´
nova
´
,L.&Volf,P.(1997)Carbohydrate-bindingspecificities
and physico-chemical properties of lectins in various tissue of
phlebotominae sandflies. Folia Parasitol. 44, 71–76.
14. Volf, P., Tesarova
´
,P.&Nohy´ nkova
´
, E. (2000) Salivary proteins
and glycoproteins in phlebotominae sandflies of various species,
sex and age. Med. Vet. Entomol. 14, 251–256.
15. Volf, P. & Killick-Kendrick, R. (1996) Post-engorgement
dynamics of haemagglutination activity in the midgut of six
species of phlebotominae sandflies. Med. Vet. Entomol. 10,
247–250.
16. Svobodova
´
, M., Volf, P. & Killick-Kendrick, R. (1996) Aggluti-
nation of Leishmania promastigotes by midgut lectins of phlebo-
tominae sandflies. Ann. Trop. Med. Parasitol. 90, 329–336.
17. Volf, P., Svobodova
´
,M.&Dvora
´
kova
´
, E. (2001) Bloodmeal
digestion and Leishmania major infections in Phlebotomus

duboscqi: effect of carbohydrates inhibiting midgut lectin activity.
Med. Vet. Entomol. 15, 281–286.
18. Volf, P., Killick-Kendrick, R., Bates, P. & Molyneux, D.H. (1994)
Comparison of the haemagglutination activities in gut and head
extracts of various species and geographical populations of phle-
botomine sandflies. Ann. Trop. Med. Parasitol. 88, 337–340.
19. Yeaton, R.W. (1986) Occurence of non-lymphatic hemagglutinins
in arthropods and their possible functions. In Hemocytic and
Humoral Immunity in Arthropods (Gupta, A.P., ed.), pp. 505–515.
Wiley Interscience, New York, USA.
20. Steinbuch, M. (1981) Protein fractionation by ammonium sul-
phate, rivanol and caprylic acid precipitation. In Methods
in Plasma Protein Fractionation (Curling, J.M., ed.), pp. 34–56.
Academic Press, London.
21. Biemann, K. (1990) Sequencing of peptides by tandem mass
spectrometry and high-energy collision-induced dissociation. In
Methods in Enzymology: Mass Spectrometry (McCloskey, J.A.,
ed.), Vol. 193, pp. 455–479. Academic Press, San Diego, USA.
22. Mohamed, H.A. & Ingram, G.A. (1994) Effects of physico-che-
mical treatments on hemagglutination activity of Anopheles gam-
biae hemolymph and midgut extract. Med. Vet. Entomol. 8, 8–14.
23. Chen, C. & Billingsley, P.F. (1999) Detection and characterization
of a mannan-binding lectin from mosquito, Anopheles stephensi
(Liston). Eur. J. Biochem. 263, 360–366.
24. Ingram, G.A. & Molyneux, D.H. (1990) Lectins (haemaggluti-
nins) in the haemolymph of Glossina fuscipes fuscipes:isolation,
partial characterization, selected physico-chemical properties and
carbohydrate-binding specificities. Insect Biochem. 20, 13–27.
25. Grubhoffer, L., Muska, M. & Volf, P. (1994) Midgut hemagglu-
tinins in five species of tsetse flies (Glossina spp.): two different

lectin systems in the midgut of Glossina tachinoides. Folia Para-
sitol. 41, 229–232.
26. Osir, E.O., Abubakar, L. & Imbuga, M.O. (1995) Purification and
characterization of a midgut lectin-trypsin complex from the tsetse
fly Glossina longipennis. Parasite Res. 81, 276–281.
27. Valenzuela, J.G., Belkaid, Y., Garfield, M.K., Mendez, S.,
Kamhawi, S., Rowton, E.D., Sacks, D.L. & Ribeiro, J.M.C.
(2001) Toward a defined anti-Leishmania vaccine targeting vector
6300 P. Volf et al.(Eur. J. Biochem. 269) Ó FEBS 2002
antigens: characterization of a protective salivary protein. J. Exp.
Med. 194, 331–342.
28. Albert, S., Klaudiny, J. & Simuth, J. (1999) Molecular
characterization of MRJP3, highly polymorphic protein of hon-
eybee (Apis mellifera) royal jelly. Insect Biochem. Mol. Biol. 29,
427–434.
29. Volf, P. & Rohousova
´
, I. (2001) Species-specific antigens in sali-
vary glands of phlebotomine sandflies. Parasitology 122, 37–41.
30. Stiles, B., Bradley, R.S., Stuart, G.S. & Hapner, K.D. (1988)
Site of synthesis of the haemolymph agglutinin of Melanoplus
differentialis (Acrididae: Orthoptera). J. Insect Physiol. 34,
1077–1085.
31. Gupta, A.P. (1985) Cellular elements in the hemolymph. In
Comprehensive Insect Physiology, Biochemistry and Pharmacology
(Kerkut, G.A. & Gilbert, L.I., eds), Vol. 3, pp. 401–451. Pergamon
Press, Oxford, UK.
32. Sacks, D.L. (2001) Leishmania–sand fly interactions controling
species-specific vector competence. Cellular Microbiol. 3, 1–9.
33. Maudlin, I. & Welburn, S.C. (1987) Lectin mediated establish-

ment of midgut infections of Trypanosoma congolense and Try-
panosoma brucei in Glossina morsitans. Trop. Med. Parasitol. 38,
167–170.
34. Welburn, S.C., Barcinski, M.A. & Williams, G.T. (1997) Pro-
grammed cell death in Trypanosomatids. Parasitol. Today 13, 22–
26.
35. Welburn, S.C. & Maudlin, I. (1989) Lectin signalling of matura-
tion of T. congolense infections in tsetse. Med. Vet. Entomol. 3,
141–145.
Ó FEBS 2002 Characterization of lectin from sand fly females (Eur. J. Biochem. 269) 6301