vaidyanathan jme 05

VECTOR/PATHOGEN/HOST INTERACTION, TRANSMISSION

Isolation of a Myoinhibitory Peptide from Leishmania major(Kinetoplastida: Trypanosomatidae) and Its Function in theVector Sand Fly Phlebotomus papatasi (Diptera: Psychodidae)

RAJEEV VAIDYANATHAN1, 2

J. Med. Entomol. 42(2): 142Ð152 (2005)

ABSTRACT Protozoan parasites in the genus Leishmania are ingested by sand ßies with blood andmultiply in the gut until they are transmitted to a vertebrate host when the sand ßy blood feeds again.Infections of the enzootic vector Phlebotomus papatasi Scopoli result in distended midguts with nospontaneous gut contractions. Using a P. papatasi hindgut contraction bioassay, a paralytic factorsensitive to trypsin, chymotrypsin, proteinase-K, and heating at 56�C was detected in crude lysates ofLeishmania major promastigotes. Application of parasite lysate to isolated hindguts resulted in re-versible, dose-dependent inhibition of spontaneous contractions. Mean volume of isolated midgutsand hindguts increased by 50Ð60% after application of L. major lysate. L. major paralytic factor waspuriÞed 104-fold over the total protein preparation and yielded a hydrophobic 12-kDa peptide.Myoinhibitory activity eluted as a single peak in reverse phase-high-pressure liquid chromatography.Tandem mass spectrometry resulted in 15 amino acid sequences, three of them sharing 45Ð73%homology with short hypothetical gene products of undeÞned function from Pseudomonas, Halo-bacterium, andDrosophila. This unique protozoan peptide mimics the function of endogenous insectneuropeptides that control visceral muscle contractions. By this novel mechanism, parasites persist inthe expanded, relaxed midgut after blood meal and peritrophic matrix digestion. This allows time fordevelopment and migration of infective forms, facilitating sand ßy vector competence and parasitetransmission.

KEYWORDS Leishmania major, Phlebotomus papatasi, vector competence, myoinhibition, hindgut

PROTOZOA IN THE GENUS Leishmania (Kinetoplastida:Trypanosomatidae) are dimorphic parasites that al-ternate between a ßagellated, extracellular promasti-gote stage in the gut of a sand ßy vector and anintracellular amastigote stage within the reticuloen-dothelial cells of a mammal host (Peters and Killick-Kendrick 1987, Dedet et al. 1999). Leishmania para-sites cause a spectrum of diseases in humans, includingthe clinically distinct forms of visceral, cutaneous, andmucocutaneous leishmaniasis. Phlebotomus papatasiScopoli (Diptera: Psychodidae) is the most importantvector of cutaneous leishmaniasis in Israel, maintain-ing enzootic transmission of Leishmania major amongfat sand rats, Psammomys obesus, and acting as solevector of L. major to humans (Schlein et al. 1982,Janini et al. 1995).

Female sand ßies ingest blood that includes mac-rophages containing Leishmania amastigotes whenfeeding on an infected vertebrate host. In the midgut,the blood is encased in a peritrophic matrix (PM),a semipermeable chitinous sac produced by the gut

epithelium(Blackburnet al. 1988).Within thePM, theamastigotes divide repeatedly and transform into auniform population of promastigotes. The PM disin-tegrates at the end of blood digestion, and free para-sites can be voided. Rapid loss of infection by excre-tion indicates that susceptible parasites are not killedoutright but rather voided with the PM and blood mealremnants. Parasites might persist in the gut by an-choring ßagella between midgut microvilli or by con-formational changes in surface glycoconjugates (War-burg et al. 1989; Pimenta et al. 1992; for summary, seeSacks and Kamhawi 2001). However, a large popula-tion of unattached parasites persists in the gut lumen(Walters et al. 1987, 1989), and it has not been ex-plained why these parasites are not excreted. Freepromastigotes transform into infective metacyclicforms, which migrate forward and are transmittedwhen the sand ßy blood feeds again.

Sand ßies heavily infected with L. major had dis-tended midguts with no tonus and no peristalsis. Im-mobility of the gut of infected ßies seemed to be amechanism by which parasites avoided expulsion, pre-sumably by factors produced by L. major. The iden-tiÞcation of this factor and deÞnition of its functionwere the purpose of this study.

1 Department of Parasitology, Hebrew University of Jerusalem,Hadassah Medical School, Ein Kerem, Jerusalem 91120 Israel.

2 Current address: Department of Entomology, University of Cal-ifornia, Davis, CA 95616.

0022-2585/05/0142Ð0152$04.00/0 � 2005 Entomological Society of America

Materials and Methods

Parasite Preparation. L. majorMHOM/IL/86/Blum(Jordan Valley strain) was obtained from the WorldHealth Organization Leishmania Reference Center,Department of Parasitology, Hebrew University,Jerusalem, Israel. Reagents and protease inhibitorswere purchased from Sigma (Rehovot, Israel), unlessotherwise speciÞed. Parasites were grown in Dulbec-coÕs modiÞed EagleÕs medium (Biological Industries,Beit Haemek, Israel) with high glucose content, 10%heat-inactivated fetal calf serum, 4 mM L-glutamine,2 mM adenosine, and 2% (vol:vol) Þlter-sterilizedhuman urine. Cultures were maintained at 28�C andpassaged every 4 d.

Two other kinetoplastid parasites were used as con-trols. Herpetomonas muscarum, an obligate parasite ofhouse ßies, was cultured identically to the L. majorcultures. Crithidia fasciculata, an obligate gut parasiteof mosquitoes, was cultured in brain-heart infusion at28�C and passaged every day.

Late log-phase cultures of the three parasites athigh density (107Ð108 parasites/ml) were spun at2,000 � g for 10 min at 8�C and washed twice withAedes aegypti L. buffered saline (ABS, Þnal concen-tration 0.6 mM MgCl2, 4.0 mM KCl, 1.8 mM NaHCO3,150 mM NaCl, 25 mM HEPES-NaOH, 1.7 mM CaCl2,pH 7.4). A protease inhibitor cocktail was added towet parasite volume to inhibit autolysis (Þnal concen-tration 1.0 mM AEBSF, 0.5 mM EDTA, 65 �M bestatin,7 �M E-64, 0.5 �M leupeptin, 0.15 �M aprotinin). Tolyse parasites, cell pellets were snap-frozen in liquidN2 and thawed three times at 30�C. Samples werechecked under phase-contrast microscope to verifyparasite lysis. Crude homogenates were frozen in liq-uid N2 and stored at �70�C until use. Concentrationsof lysate proteins were assayed by the Bradfordmethod (Bradford 1976).

Crude homogenates were thawed on ice and spunat 12,000 � g for 30 min to precipitate particulates.Both precipitate and supernatant were tested in thesand ßy hindgut bioassay (below).Sand Flies, Hindgut Bioassay, and Gut DistensionMeasurements.TheP. papatasi colony originated withßies from Kfar Adumim, 10 km east of Jerusalem. Thecolony was maintained according to Modi and Tesh(1983), and insectary conditions were 26 � 1�C, 80%RH,andaphotoperiodof17:7(L:D)h.Two- to6-d-oldsugar-fed sand ßies were used for all experiments.

Whole guts from male and female sand ßies weredissected into a watch glass with 90Ð99 �l of oxygen-ated ABS warmed to 30�C and allowed to recover untilhindgut contractions stabilized. Hindgut contractionswere counted for 5-min increments. Samples of par-asite lysate proteins were added 5 min after the hind-gut stabilized and contracted regularly. Preparationswere mixed with a pipette to distribute proteins. Oneunit of paralytic activity is 1% inhibition of P. papatasihindgut contractions relative to the initial 5-min pe-riod before addition of proteins. SpeciÞc activity isdeÞned as units of paralytic activity per milligram oflysate proteins. To test whether inhibition was revers-

ible, the treated gut preparations were rinsed twice inABS, and returned to 100 �l of warm, oxygenated ABS.P. papatasimidguts and hindguts were measured on

an Olympus BH phase-contrast microscope. Wholeguts from unfed ßies were dissected into 98 �l ofoxygenated ABS warmed to 30�C in glass-coveredwatch glasses. Width and length of the midgut and thehindgut were measured, 6.4 �g of L. major proteinswas added (Þnal concentration 64 �g/ml), and theguts were kept in a humid chamber until they weremeasured again at 5 and 30 min. Midgut and hindgutvolumes were estimated as cylinders for the calcula-tions, as widest cross-sectional area multiplied bylength. Ten ßies (Þve males, Þve females) were used.An equal concentration of H. muscarum lysate pro-teins was used as a control.

SigniÞcant differences in speciÞc activity in thesand ßy hindgut bioassay were tested using a two-sample t-test assuming equal variances. Gut volumecalculations were tested for signiÞcant differences us-ing the MannÐWhitney U test (Daniel 1991).Protease and Heat Inactivation of the ParalyticFactor. A 1-ml sample of L. major lysate proteins(10.55 mg/ml) was prepared according to the above-mentionedprotocol.To test for thermal inactivationofparalytic activity, lysate was heated at 56�C in a waterbath, gently agitated, and aliquots were removed at0, 30, 60, 90, 120, 150, and 180 min. Each aliquot wasfrozen in liquid N2 and stored at �70�C until tested onthe sand ßy hindgut bioassay.L. major promastigote lysates were treated with

trypsin, chymotrypsin, and proteinase-K to test sen-sitivity of the paralytic factor to proteases. Aliquotsof parasite proteins (3 mg/ml) were treated 1:1 with1 mg/ml trypsin and chymotrypsin, or 10:1 with0.5 mg/ml proteinase-K, heated for 30 min at 37�C,and reactions halted with proteinase inhibitors, asdescribed above. Lysate samples treated with pro-teases were frozen in liquid N2 and stored at �70�Cuntil tested on the sand ßy hindgut bioassay.Purification of the Paralytic Factor. In brief, to pu-

rify the paralytic factor to apparent homogeneity,cell proteins were precipitated with 85% (NH4)2SO4

at 0�C and subjected to two steps of hydrophobicchromatography, size exclusion chromatography,and Þnally reverse phase-high-pressure liquid chro-matography (RP-HPLC). Chromatography, centrifu-gation, and dialysis were performed at 0Ð4�C (Web-ster and Prado 1970). RP-HPLC fractions were testedfor paralytic activity by using the sand ßy hindgutbioassay. Active fractions were submitted to elec-trospray ionization and tandem mass spectrometryanalysis to determine amino acid sequence (Nassel1999, Mann et al. 2001). These sequences, identiÞedfrom peptide fragmentation data after mass spectrom-etry, were matched to protein sequence databases.

Fraction 1 lysate proteins from freeze/thawed par-asites, cultured and harvested as described above werebulk precipitated with 85% (NH4)2SO4, pH adjustedto 7.2 with NH4OH at 0�C (Englard and Seifter 1990).Samples were centrifuged 5 � 103 g for 30 min, and thesupernatant was discarded. Pellets were pooled, im-

March 2005 VAIDYANATHAN: L. major MYOINHIBITORY PEPTIDE 143

mediately frozen in liquid N2, and stored at �70�C(fraction 2). Aliquots of fraction 1, fraction 2, andsupernatant were assayed using the sand ßy hindgut.Fraction 2 and supernatant were Þrst dialyzed over-night against 4.0 liters of ABSm (ABS modiÞed to50 mM NaCl) to remove (NH4)2SO4.

Fraction 2 samples were thawed on ice, diluted withABSm to 1.5 M (NH4)2SO4, and centrifuged 5 � 103

g for 30 min to remove insoluble particulates. In total,2,878 mg of fraction 2 was loaded onto a Phenyl Sepha-rose 6 Fast Flow column (6.5 by 23 cm) (AmershamBiosciences Inc., Piscataway, NJ), equilibrated with1.5 M (NH4)2SO4 in ABSm (Kennedy 1990). The col-umn was eluted stepwise with the same buffer con-taining 1.0, 0.75, 0.70, 0.65, 0.60, 0.55, 0.50, 0.25, and0.0 M (NH4)2SO4. Aliquots of 100 �l were dialyzedagainst 4.0 liters of ABSm overnight and bioassayed.Pooled active fractions were deÞned as fraction 3.

Fraction 3 (40 mg) was loaded onto a FractogelEMD Propyl Sepharose column (1.5 by 9 cm) (Merck,Whitehouse Station, NJ), equilibrated with 1.5 M(NH4)2SO4 inABSm.Samplewaseluted stepwisewiththe same buffer containing 1.25, 1.00, 0.75, 0.50, 0.25,and 0.0 M (NH4)2SO4 (Kennedy 1990). Aliquots of50 �l were dialyzed against 4.0 liters of ABSm over-night and tested on the sand ßy hindgut bioassay.Active fractions from Propyl Sepharose chromatogra-phy were deÞned as fraction 4. Before proceeding tosize exclusion chromatography, fraction 4 was con-centrated on a 1-ml bed volume Phenyl Sepharose 6Fast Flow column, equilibrated with 1.5 M (NH4)2SO4

in ABSm. The sample was eluted with the same bufferat 1.5 and 0.0 M (NH4)2SO4, dialyzed, and bioassayed.This concentrated sample was deÞned as fraction 5.

Protein recovery was conÞrmed by Bradford assay(Bradford 1976), and fraction 5 was Þltered (0.22 �m)and loaded onto a Merck Fractogel EMD BioSEC (S)size exclusion chromatography column, adapted forHPLC(Merck,Darmstadt,Germany).ChromatographywasperformedusingaVarian5000liquidchromatographand Rheodyne solvent delivery module (Cotati, CA).FractionsweredetectedwithaVarianR1Ð3UVdetectorset at 280 nm. The following molecular weight markerswere run: �-amylase (200 kDa), bovine serum albumin(66 kDa), chymotrypsinogen (25 kDa), and aprotinin(6.5 kDa). The column was equilibrated and washedwith ABS at a 0.5 ml/min ßow rate (Fischer 1980, Stell-wagen 1990). Ninety 0.8-ml fractions were collected onice, and 10-�l aliquots were bioassayed. Fractions withparalytic activity were pooled and deÞned as fraction 6.

Fraction 6 was evaporated to dryness under vac-uum, reconstituted in HPLC-grade triple-distilledH2O, and Þltered (0.22 �m) before loading onto aVydac 214 TP RP-C4HPLC column (0.46 by 25 cm).Equipment was the same as for size exclusion chro-matography, with the exception of UV detection at220 nm. Column elution was performed with gradientsof solution A (5% acetonitrile, 95% H2O, and 0.1%trißuoroacetic acid [TFA]) at 30 min, eluted using alinear gradient with solution B (95% acetonitrile,5% H2O, and 0.1% TFA) at a ßow rate of 0.5 ml/minfor 60 min. Ninety 0.5-ml fractions were collected

manually and immediately frozen in liquid N2; 2-�laliquots were evaporated to dryness before reconsti-tution in triple-distilled H2O and tested on sand ßyhindgut (Veelaert et al. 1996, Duve et al. 1999).Mass Spectrometric Identification of the ParalyticFactor. Active Vydac fractions were combined andvacuum-evaporated to �50 �l. A further 25 �l of25mMNH4HCO3 wasaddedatpH8.0 toact as abufferduring trypsin addition. This 75 �l was divided intothree aliquots. The Þrst aliquot was digested overnightat 37�C with 10 �l of trypsin (0.1 �g/10 �MNH4HCO3). It was then eluted using a ZipTip C18 tip(Millipore Corporation, Billerica, MA) and 75%CH3CN, 1% formic acid (Merck). Part of this samplewas used as load, and the remaining part was digestedwith Asp-N (see protocol below).

The second aliquot was digested overnight at 37�Cwith 10 �l of 0.04 �g of Asp-N (Roche Diagnostics,Mannheim, Germany) in 25 mM NH4HCO3 at pH 8.0.After enzymatic digestion, the sample was eluted us-ing a ZipTip C18 tip and 75% CH3CN, 1% formic acid.

The third aliquot underwent reduction alkylation.It was treated with 10 �l of 45 mM dithiothreitol for30 min at 60�C and 10 �l of 100 mM iodoacetamide for30 min at 25�C. It was eluted with a ZipTip C18 tipand 75% CH3CN, 1% formic acid. After alkylationreduction, the sample was divided into two fractions.The Þrst was treated with 10 �l of 0.04 �g of Asp-N in25 mM NH4HCO3 at pH 8.0 overnight at 37�C andeluted with a ZipTip C18 tip and 75% CH3CN, 1%formic acid. The second was treated with 10 �l oftrypsin (0.1 �g/10 �M NH4HCO3) overnight at 37�Cand eluted with a ZipTip C18 tip and 75% CH3CN, 1%formic acid.

A load of 3 �l of each eluted sample was injectedthrough a Long NanoES spray capillary (Protana Cor-poration, Toronto, Canada) to an electrospray ioniza-tion quadropole time of ßight (ESI QTOF2) (Micro-mass London, London, United Kingdom) to ionizeprotein fragments and to detect the mass-to-chargeratio (m/z) of ionized peptides (Fenn et al. 1989).Capillary voltage was 1,200 V, cone voltage was 25 V,collision energy was 10.

Pumping the analyte at a low microliter per minuteßow rate at high voltage causes electrostatic disper-sion of micrometer-sized droplets that rapidly evap-orate and impart their charge onto the analyte mole-cules (Mann et al. 2001). Electrosprayed ions weredetected by tandem mass spectrometry (MS-MS)spectra, with identical capillary and cone voltages asdescribed above, but with collision energy between 30and 40 in the presence of argon (Hunt et al. 1986,Biemann and Scoble 1987).

The MS-MS spectra were matched against nonre-dundant database sequences; a sequence tag search(www.matrixscience.com) and a full sequence search(BLASTandFASTA)weredoneusing small identiÞedpeptide fragments (Altschul et al. 1997, GrifÞn andAebersold 2001) against all recorded proteins in thedatabase.

144 JOURNAL OF MEDICAL ENTOMOLOGY Vol. 42, no. 2

Results

Inhibition of P. papatasi Hindgut Contractions.Sand ßy hindguts in warmed ABS spontaneously con-tracted for 1Ð6 h. Midgut activity was irregular; thehindgutkept a steady rhythm.Activityofboth sectionswas stopped by addition of L. major lysate; however,only hindguts were suitable for measuring inhibitoryeffect. Contractions of the rectum were not affectedby addition of parasite lysate.

A dose of 50Ð70 �g/ml L. major lysate proteinsinhibited 60% of spontaneous P. papatasi hindgut con-tractions within 5 min of application (Fig. 1). Hind-guts exposed to parasite lysate for 20 min, rinsed inABS, and returned to fresh ABS resumed contracting

(Fig. 1). No effect was seen with equivalent concen-trations of lysates prepared fromH.muscarum or fromC. fasciculata (data not shown).

Increasing concentrations of L. major lysate pro-teins inhibited hindgut contractions in a dose-depen-dent manner (Fig. 2). An application of 12 or 24 �g/mlL.major lysateproteins inhibited spontaneoushindgutcontractions by 20 and 34%, respectively. Distendednodes formed along the length of the hindgut. Anapplication of 48 and 96 �g/ml lysate proteins de-creased hindgut contractions by 60 and 80%, respec-tively. Hindguts completely distended within secondsof a concentration of 48 �g protein/ml or more. Hind-gut lumen Þlled with liquid; gut and epithelial cellswere turgid.

Fig. 1. Inhibition of P. papatasi hindgut contractions with L. major lysate proteins (50Ð70 �g/ml). Contractions weremonitored for 5 min. Hindguts in the “return” group were treated with L. major lysate proteins for 20 min, rinsed in ABS,and immediately returned to fresh ABS to determine whether myoinhibition was reversible. Means and standard errors areshown for 36 trials.

Fig. 2. Inhibition of hindgut contractions by different concentrations of L. major lysate proteins within 5 min. Means andstandard errors are shown for Þve trials.


Distension of P. papatasi Midgut and Hindgut.Based on observations of distended hindguts afterapplication of L. major lysate proteins, midgut andhindgut distension was quantiÞed 5 and 30 min after

addition of 6.4 �g of protein (64 �g/ml) of L. majorand H. muscarum lysates (Fig. 3A and B).P. papatasi midgut volume increased after 5-min

incubation with 64 �g/ml L. major proteins, saline,

Fig. 3. Distension of P. papatasimidgut (A) and hindgut (B) volume 5 and 30 min after application of 64 �g/ml L. majoror H. muscarum lysate proteins. Means and standard errors are shown for 10 trials (Þve males, Þve females).

Fig. 4. Heat inactivation of L. major lysate proteins tested on P. papatasi hindgut assays.


and 64 �g/ml H. muscarum proteins by 19.5, 17.4, and12.2%, respectively. The increase due to L. majorproteins was not signiÞcantly different from the salinecontrol orH. muscarum proteins (P� 0.05). Howeverthe difference was signiÞcant (P� 0.01) after 30-minincubation with L. major proteins (48.7%) versustreatment with saline (19.2%) or H. muscarum pro-teins (7.8%) (Fig. 3A).

Average hindgut volume increased 5 min after ap-plication of 64 �g/ml L. major proteins, saline, and64 �g/ml H. muscarum proteins by 20.7, 7.3, and13.1%, respectively. After 30-min incubation, hindgutvolume increased by 57.1, 11.0, and 13.1%, respec-tively. The increase in hindgut volume incubated withL. major proteins was signiÞcantly greater than theother two applications (P� 0.01) at both 5 and 30 min(Fig. 3B).Protease andHeat Inactivation of the Paralytic Fac-tor. Heating of L. major lysate for 30 min at 56�Cresulted in a 42% loss of activity. Aliquots heated from60 to 180 min resulted in 65Ð70% loss of activity, butthere was no difference among samples heated 60 minor longer (Fig. 4).

Lysate samples treated with trypsin, chymotrypsin,or proteinase-K no longer inhibited spontaneous hind-gut contractions (data not shown). Based on these

results, the paralytic factor was considered to be aprotein.Purification and Identification of the Paralytic Pro-tein. A series of Þve chromatography columns wereused to purify the paralytic protein from L. majorlysate to apparent homogeneity. The puriÞcationscheme was performed four times; results for theÞnal trial are presented. Average values for activity(percentage of inhibition per milliliter), protein con-centration, and speciÞc activity are summarized inTable 1.

After initial puriÞcation steps, more than one(NH4)2SO4-precipitated fraction had myoinhibitoryactivity on spontaneous contractions in P. papatasihindgut. The fraction with highest speciÞc activitywas puriÞed. Elution conditions for the active frac-tions were: Phenyl Sepharose 6 Fast Flow, 0.65 M(NH4)2SO4; EMD Propyl Sepharose, 1.0 M (NH4)2SO4;Phenyl Sepharose 6 Fast ßow (to concentrate sam-ple), plain ABSm; Fractogel EMD BioSEC, 134.9-minelution time; and Vydac RP-C4HPLC, 33.5% acetoni-trile. SpeciÞc activity increased 104-fold from the orig-inal crude proteins to the Þltered load injected ontothe Vydac RP-HPLC column (Table 1).

A preliminary estimation of the apparent nativeprotein mass could be deduced from the plotting of

Fig. 5. Elution sequence of molecular weight markers from Fractogel EMD BioSEC size exclusion column. Arrow denotespoint at which myoinhibitory activity eluted.

Table 1. Increase of myoinhibitory activity at different steps of purification of L. major proteins

FractionVol

(ml)Activity

(units/ml)Total

activity�Protein(mg/ml)

Totalprotein(mg)

SpeciÞcactivity(units/

mg)

1 Original lysate 3,100 1.36 � 104 4.21 � 107 3.8 11,780 3.58 � 103

2 AS precipitate 530 7.11 � 104 3.77 � 107 18.0 9540 3.95 � 103

3 Phenyl Sepharose 1,300 1.73 � 104 2.25 � 107 0.15 195 1.15 � 105

4 Propyl Sepharose 60 2.62 � 105 1.57 � 107 0.015 0.9 1.75 � 107

6 Size exclusion 2.1 8.00 � 104 1.68 � 105 0.005 0.0105 1.60 � 107

Final RP-HPLC 1.5 2.85 � 105 4.28 � 105 �0.005 �0.0075 �5.7 � 107

Data are shown for active fractions only. AS, ammonium sulfate. Note that fraction 5 was a concentration step and is not shown.


log molecular weight of the protein markers versustheir retention time on the preparative BioSEC col-umn (Fig. 5). The paralytic activity eluted with aretention time of 134.9 min, marked with an arrowbetween chymotrypsinogen and aprotinin, indicatingan apparent mass of �12 kDa. A more accurate ana-lytical measurement of the protein apparent mass hasyet to be conducted, based on both Stokes radii andsedimentation measurements.

Myoinhibitory activity in the RP-HPLC step wasdetected between 48 and 50 min (Fig. 6). The aceto-nitrile gradient increased from 5% at 30 min (time 0)to 95% at 90 min (that is, 60 min after the gradientbegan). Using the linear regression equation y 1.5x� 5, the calculated acetonitrile concentration atwhich myoinhibitory activity eluted is 33.5% (Fig. 7).

Active fractions from the RP-HPLC fractionationstep were subjected to enzymatic digestion, mass

Fig. 6. RP-HPLC fractionation on Vydac C4 column of a previous size exclusion chromatography fraction of �12 kDa.Myoinhibitory activity eluted between 48 and 50 min (denoted by arrow).

Fig. 7. Increasing acetonitrile gradient on RP-HPLC Vydac C4 column during isocratic elution with solution B (95%acetonitrile, 5% H2O, and 0.1% TFA). Arrow denotes point at which myoinhibitory activity eluted, 18Ð20 min after an initial30-min void volume (corresponding to activity at 48Ð50 min in Fig. 6).


spectrometry, and MS-MS analysis (Fenn et al. 1989).The m/z of ionized peptides was detected by aQTOF2 Micromass mass spectrometer. The active my-oinhibitory peptide was extremely difÞcult to frag-ment, despite sequential trials of enzymatic digestionwith and without reduction alkylation. Peaks in theMS spectra range of 1000 m/z (Fig. 8) are highlycharged, representing peptides �6,000 Da. Tandemmass spectrometry of 10 major peaks and dozens ofminor peaks yielded m/z values, molecular weights,and putative sequences for the myoinhibitory peptide.Currently, sequence data are being used as templatesfor RT-polymerase chain reaction and production of

recombinant peptide. Amino acid sequences may bepublished once a more complete picture is obtained.

All the peptide sequences obtained from tandemmass spectrometry were used to search for sequencehomologies in known proteins by using Mascot,BLAST, and FASTA search programs (Altschul et al.1997, GrifÞn and Aebersold 2001). By selecting entiredatabases, three organisms yielded sequences of fourto seven amino acids with �45% homology toL.majormyoinhibitory peptide and low probability that thesearch sequence was a random string (an expectedvalue, or E, � 1.0). A putative gene product of Pseudo-monas aeruginosa had a 53% homology with some

Fig. 8. Tandem mass spectra data from electrospray ionization of active fractions from RP-HPLC. The 650Ð700 range isexpanded to show major peaks used for peptide sequence analysis. Each peak represents one peptide fragment; multiplenumbers for one peak represent amino acid modiÞcations within that fragment.


identiÞed fragments and an E-value of 0.13 (Stover etal. 2000). One hypothetical protein of aHalobacteriumsp. had 73% homology with one identiÞed fragmentand an E-value of 0.35. Three putative gene prod-ucts in Drosophila melanogaster Meigen had 45% ho-mology with identiÞed fragments and E-values of0.84 (Adams et al. 2000). The functions for these pro-teins in P. aeruginosa, Halobacterium, and D. melano-gaster are unknown.

Discussion

Spontaneous contractions of P. papatasi gut prepa-rations were inhibited byL.major lysate proteins; theyresumed regular activity after rinsing (Fig. 1). Lysatesfrom kinetoplastid parasites of other Diptera did notaffect P. papatasi hindgut contractions. Inhibition ofmuscle activity was dose-dependent, although the ef-fect was not strictly linear (Fig. 2). Similar lysatepreparations signiÞcantly increased midgut and hind-gut volumes (Fig. 3). Paralytic activity in parasitelysate was reduced when lysates were heated (Fig. 4)and lost when lysates were treated with proteases,indicating that the paralytic factor is a protein.Leishmania promastigotes that are entirely in the

sand ßy gut beneÞt from decreased peristalsis andincreased gut volume. A distended gut retains moresand ßy food, providing nutrients for parasites andmore room for multiplication. Immobility of the ex-panded gut protects parasites from expulsion after PMdisintegration. This is coincident with development ofinfective forms, or metacyclogenesis (Sacks and Per-kins 1985). By inhibiting gut peristalsis and increasinggut volume, parasites condition the midgut for meta-cyclogenesis and transmission of infective forms.

Persistence of Leishmania promastigotes in the gutby insertion of ßagella between midgut microvilli hasbeen studied by light microscopy (Adler and Theodor1926) and later by electron microscopy (Killick-Ken-drick et al. 1974, Warburg et al. 1986). Using isolatedßagella, Warburg et al. (1989) found that a ßagellarsurface protein facilitated attachment to midgut epi-thelial cells. Extensive studies have shown that pro-cyclic promastigotes bind to gut epithelium by li-pophosphoglycan (LPG), the most abundant cellsurface glycoconjugate. Other studies have detaileddifferences in LPG structure and modiÞcations, cor-relation with vector competence and speciÞcity, andwork with genetically modiÞed parasites (Pimenta et al.1992; Sacks et al. 1994, 1995; Sacks and Kamhawi 2001).However, infective metacyclic parasites with modiÞedLPG and other morphotypes (Walters et al. 1987, 1989)remain free in the gut lumen. Relaxation of the midgutmay protect the population of unattached parasites thatinhabit the midgut after PM disruption.

In the anterior midgut of infected ßies, a gelatinousplug (Killick-Kendrick 1979) consisting of an elec-tron-dense, Þlamentous precipitate surrounds manyparasites (Walters et al. 1987, Lawyer et al. 1990). Thegel plug has a framework of parasite-derived mucin-like proteophosphoglycan that contains Leishmania-secreted acid phosphatase. The gel plug presumably

enhances parasite transmission by impeding the in-gestion of blood and causing repeated probing (Stier-hof et al. 1999). Differentiation of promastigotes intometacyclic forms also takes place in the plug (Rogerset al. 2002). Assembly of the plug is cumulative; with-out inhibition of gut peristalsis, gel components maybe expelled and plug assembly interrupted at an earlystage.

The factor responsible for inhibition of spontaneousgut contractions was determined to be a protein be-cause of its inactivation by heat and proteases. Basedon this observation, L. major lysate was precipitatedand fractionated to yield a 12-kDa peptide from size-exclusion chromatography. A single peak from RP-HPLC yielded a peptide with 104-fold speciÞc activityover the original crude protein extract. The peptide isnamed stambhanin, from the Sanskrit verb stambh,meaning “to hinder, suppress, paralyze, stupefy; tobecome stiff or rigid” (Apte 1963).

SpeciÞc activity increased 100-fold from the Phenylto Propyl Sepharose column. A 100-fold increase inspeciÞc activity is unlikely by simple peptide isolationand is most likely due to the removal of endogenousinhibitors for the myoinhibitory peptide. Its elutionorder in Phenyl and Propyl Sepharose chromatogra-phy indicates that stambhanin is highly hydrophobic.Myoinhibition was detected in more than one fractionof Phenyl and Propyl Sepharose chromatography.Several myoinhibiting peptides must be present inL. major for activity to be present in more than onefraction. PuriÞcation steps proceeded with fractionsof highest speciÞc activity and highest recovery(Table 1).

Several peaks in the 650Ð700 range from the massspectrometric data represent the same peptide frag-ment (Fig. 8). In these most likely sequences, hydro-phobic amino acids (Gly, Ala, Val, Leu, Ile, Met, Phe,Trp, and Pro) account for 64Ð73% of those identiÞed.This explains the hydrophobic elution order foractive fractions in Phenyl and Propyl Sepharose chro-matography. In addition, cysteine was detected in twosequence fragments, possibly stabilizing internal pep-tide conformation through disulÞde linkages. Cur-rently, only fragmented sequence data are available.Studies are focused on production of recombinantpeptide for further characterization.

A search of prokaryotic and eukaryotic genomesyielded no homology with any arthropod-borne path-ogen. Only three organisms expressed a protein with�45% homology, and even these were for single frag-ments. The three best-Þt sequences from P. aeruginosa,Halobacterium sp., and D. melanogaster were putativegene products with no known function. This meanseither the sequences obtained are highly variable, orstambhanin is a novel molecule that functions likeendogenous viscerotropic neuropeptides of insects.

The enteric nervous system in arthropods releasesendogenous neuropeptides and neurotransmittersthat regulate visceral muscle contraction (Coast andWebster 1998). The Þrst inhibitory neuropeptide iso-lated from an insect was the FMRF-amide-relatedpeptide leucomyosuppressin, which inhibits sponta-


neous contractions of the cockroach hindgut (Holmanet al. 1986). FLRF-amides of Locusta migratoria L.inhibit locust heart rhythm, reduce spontaneous ovi-duct contractions, and decrease amplitude of hindgutcontractions (Schoofs et al. 1993). Many of these pep-tides, such as myosuppressins and locustamyoinhibit-ing peptide, block voltage-gated and ligand-gatedCa2� channels in the plasma membrane (Wilcox andLange 1995, Orchard et al. 1997). Blocking Ca2� chan-nels and decreasing Ca2�-dependent action potentialsis a common mechanism shared by insect neuropep-tides and may explain how stambhanin operates.

There are many examples of parasites modifying hostphysiology to their own advantage. In this study, a pro-tozoan produces a peptide that mimics the function ofmyoinhibitory neuropeptides of insects. Amino acid se-quences for fragments of stambhanin were comparedwith proteins and protein fragments from several data-bases,but its sequencesdidnotcorrespondtoanyknownmyoinhibiting agent. It seems to be a new protein thatreversibly inhibits visceral muscle contraction. Myoin-hibition, in concert with modiÞcations in LPG, ßagellarbinding, and the gel plug, conditions the sand ßy gut fordevelopment of infective forms and facilitates transmis-sion of infective parasites.

Acknowledgments

This work was carried out under the guidance of Y. Schlein,J. Shlomai, and A. Warburg (Department of Parasitology,Hebrew University of Jerusalem, Hadassah Medical School,Israel). Leishmania samples were provided by L. Schnur.R. L. Jacobson helped with parasite cultures. O. Moshel (Blet-terman Research Laboratory for Macromolecules and MassSpectrometry, Hebrew University of Jerusalem) performed themass spectrometric analysis.

References Cited

Adams, M. D., S. E. Celniker, R. A. Holt, C. A. Evans,J. D. Gocayne, P. G. Amanatides, S. E. Scherer, P. W. Li,R. A. Hoskins, R. F. Galle, et al. 2000. The genome se-quence ofDrosophila melanogaster. Science (Wash. DC)287: 2185Ð2195.

Adler, S., and O. Theodor. 1926. Further observations onthe transmission of cutaneous leishmaniasis to man fromPhlebotomuspapatasii.Ann. Trop. Med. Parasitol. 20: 175Ð190.

Altschul, S. F., T. L. Madden, A. A. Schaffer, J. Zhang,Z. Zhang, W. Miller, and D. J. Lipman. 1997. GappedBLAST and PSI-BLAST: a new generation of proteindatabase search programs. Nucleic Acids Res. 25: 3389Ð3402.

Apte, V. S. 1963. The studentÕs Sanskrit-English dictionary.Motilal Banarsidass, Delhi, India.

Biemann, K., and A. Scoble. 1987. Characterization by tan-dem mass spectrometry of structural modiÞcations inproteins. Science (Wash. DC) 237: 992Ð998.

Blackburn, K., K. R. Wallbanks, D. H. Molyneux, andD. R. Lavin. 1988. The peritrophic membrane of thefemale sandßy Phlebotomus papatasi. Ann. Trop. Med.Parasitol. 82: 613Ð619.

Bradford, M. M. 1976. A rapid and sensitive method forthe quantitation of microgram quantities of protein using

the principle of protein dye-binding. Anal. Biochem. 72:248Ð254.

Coast, G. M., and S. G. Webster [eds]. 1998. Recent ad-vances in arthropod endocrinology. SEB Seminar Series.Cambridge University Press, Cambridge, United King-dom.

Daniel, W. W. 1991. Biostatistics: a foundation for analysisin the health sciences. Wiley, New York.

Dedet, J. P., F. Pratlong, G. Lanotte, and C. Ravel. 1999.Cutaneous leishmaniasis: the parasite. Clin. Dermatol. 17:261Ð268.

Duve, H., P. D. East, and A. Thorpe. 1999. Regulation oflepidopteran foregut movement by allatostatins and al-latotropin from the frontal ganglion. J. Comp. Neurol. 413:405Ð416.

Englard, S., and S. Seifter. 1990. Precipitation techniques,pp. 285Ð300. InM. P. Deutscher [ed.], Methods in enzy-mology, vol. 182. Guide to protein puriÞcation. Academic,San Diego, CA.

Fenn, J. B., M. Mann, C. K. Meng, S. F. Wong, and C. M.Whitehouse. 1989. Electrospray ionization for massspectrometry of large biomolecules. Science (Wash. DC)246: 64Ð71.

Fischer, L. 1980. Gel Þltration chromatography. Elsevier/North-Holland Biomedical Press, Amsterdam, The Neth-erlands.

Griffin, T. J., and R. Aebersold. 2001. Advances in pro-teome analysis by mass spectrometry. J. Biol. Chem. 276:45497Ð45500.

Holman, G. M., B. J. Cook, and R. J. Nachman. 1986. Isola-tion, primary structure and synthesis of leucomyosup-pressin, an insect neuropeptide that inhibits spontaneouscontractions of the cockroach hindgut. Comp. Biochem.Physiol. 85C: 329Ð333.

Hunt, D. F., J. Shabanowitz, J. R. Yates, N.-Z. Zhu, andD.H. Russell. 1986. Protein sequencing by tandem massspectrometry. Proc. Natl. Acad. Sci. U.S.A. 83: 6233Ð6237.

Janini, R., E. Saliba, S. Khoury, O. Oumeish, S. Adwan, andS. Kamhawi. 1995. Incrimination of Phlebotomus pa-patasi as vector of Leishmania major in the southernJordan Valley. Med. Vet. Entomol. 9: 420Ð422.

Kennedy, R. 1990. Hydrophobic chromatography, pp. 339Ð342. In M. P. Deutscher [ed.], Methods in enzymology,vol. 182. Guide to protein puriÞcation., ed. Academic,San Diego, CA.

Killick-Kendrick, R. 1979. Recent advances and outstand-ing problems in the biology of phlebotomine sandßies. Areview. Acta Trop. 35: 297Ð313.

Killick-Kendrick, R., D. H. Molyneux, and R. W. Ashford.1974. Ultrastructural observations on the attachment ofLeishmania in the sandßy. Trans. R. Soc. Trop. Med. Hyg.68: 269Ð276.

Lawyer, P. G., P. M. Ngumbi, C. O. Anjili, S. O. Odongo,Y. B. Mebrahtu, J. I. Githure, D. K. Koech, and C. R.Roberts. 1990. Development of Leishmania major inPhlebotomus duboscqi and Sergentomyia schwetzi(Diptera: Psychodidae). Am. J. Trop. Med. Hyg. 43: 31Ð43.

Mann,M.,R.C.Hendrickson, andA. Pandey. 2001. Analysisof proteins and proteomes by mass spectrometry. Annu.Rev. Biochem. 70: 437Ð473.

Modi, G., and R. Tesh. 1983. A simple technique for massrearing Lutzomyia longipalpis and Phlebotomus papatasi(Diptera: Psychodidae) in the laboratory. J. Med.Entomol. 20: 568Ð569.

Nassel, D. R. 1999. Tachykinin-related peptides in inverte-brates: a review. Peptides 20: 141Ð158.


Orchard, I., B. C.Donly,M. Fuse, A. B. Lange, S. S. Tobe, andW. G. Bendena. 1997. FMRFamide-related peptides ininsects, with emphasis on the myosuppressins. Ann. N.Y.Acad. Sci. 814: 307Ð309.

Peters,W., and R. Killick-Kendrick, eds. 1987. The leishman-iases in biology and medicine, vols. 1 and 2. Academic,New York.

Pimenta, P.F.P., S. J. Turco, M. J. McConville, P. G. Lawyer,P. V. Perkins, and D. L. Sacks. 1992. Stage-speciÞc ad-hesion of Leishmania promastigotes to the sand ßy mid-gut. Science (Wash. DC) 256: 1812Ð1815.

Rogers, M. E., M. L. Chance, and P. A. Bates. 2002. The roleof promastigote secretory gel in the origin and transmis-sion of the infective stage of Leishmania mexicana by thesandßy Lutzomyia longipalpis. Parasitology 124: 495Ð507.

Sacks, D. L., and S. Kamhawi. 2001. Molecular aspects ofparasite-vector and vector-host interactions in leishman-iasis. Annu. Rev. Microbiol. 55: 453Ð483.

Sacks, D. L., and P. V. Perkins. 1985. Development of in-fective stage Leishmania promastigotes within phleboto-mine sand ßies. Am. J. Trop. Med. Hyg. 34: 456Ð459.

Sacks, D. L., P. F. Pimenta, M. J. McConville, P. Schneider,and S. J. Turco. 1995. Stage-speciÞc binding of Leishma-nia donovani to the sand ßy vector midgut is regulatedby conformational changes in the abundant surfacelipophosphoglycan. J. Exp. Med. 181: 685Ð697.

Sacks, D. L., E. M. Saraiva, E. Rowton, S. J. Turco, andP. F. Pimenta. 1994. The role of the lipophosphoglycanof Leishmania in vector competence. Parasitology 108:S55ÐS62.

Schlein, Y., A. Warburg, L. F. Schnur, and A. E. Gunders.1982. Leishmaniasis in the Jordan Valley II. Sandßies andtransmission in the central endemic area. Trans. R. Soc.Trop. Med. Hyg. 76: 582Ð586.

Schoofs, L., J. Vanden Broeck, and A. De Loof. 1993. Themyotropic peptides of Locusta migratoria: structures,distribution, functions and receptors. Insect Biochem.Mol. Biol. 23: 859Ð881.

Stellwagen, E. 1990. Gel Þltration, pp. 317Ð328. In M. P.Deutscher [ed.], Methods in enzymology, vol. 182. Guideto protein puriÞcation. Academic, San Diego, CA.

Stierhof, Y.-D., P. A. Bates, R. L. Jacobson, M. E. Rogers,Y. Schlein, E. Handman, and T. Ilg. 1999. Filamentousproteophosphoglycan secreted byLeishmania promastig-otes forms gel-like three-dimensional networks that ob-struct the digestive tract of infected sandßy vectors. Eur.J. Cell Biol. 78: 675Ð689.

Stover, C. K., X. Q. Pham, A. L. Erwin, S. D. Mizoguchi,P. Warrener, M. J. Hickey, F. S. Brinkman, W. O.Hufnagle, D. J. Kowalik, M. Lagrou, et al. 2000. Com-plete genome sequence of Pseudomonas aeruginosaPA01,an opportunistic pathogen. Nature (Lond.) 406: 947Ð948.

Veelaert, D., B. Devreese, L. Schoofs, J. Van Beeumen,J. Vanden Broeck, S. S. Tobe, and A. De Loof. 1996.Isolation and characterization of eight myoinhibitingpeptides from the desert locust Schistocerca gregaria: newmembers of the cockroach allatostatin family. Mol. Cell.Endocrinol. 122: 183Ð190.

Walters, L. L., G. B. Modi, G. L. Chaplin, and R. B. Tesh.1989. Ultrastructural development of Leishmania chagasiin its vector, Lutzomyia longipalpis (Diptera: Psychodi-dae). Am. J. Trop. Med. Hyg. 41: 295Ð317.

Walters, L. L., G. B. Modi, R. B. Tesh, and T. Burrage. 1987.Host-parasite relationship of Leishmania mexicana mexi-cana and Lutzomyia abonnenci (Diptera: Psychodidae).Am. J. Trop. Med. Hyg. 36: 294Ð314.

Warburg, A., G. S. Hamada, Y. Schlein, and D. Shire. 1986.Scanning electron microscopy of Leishmania major inPhlebotomus papatasi. Z. Parasitenkd 72: 423Ð431.

Warburg, A., R. B. Tesh, and D. McMahon-Pratt. 1989.Studies on the attachment of Leishmania ßagella to sandßy midgut epithelium. J. Protozool. 36: 613Ð617.

Webster, M. E., and E. S. Prado. 1970. Glandular kallikreinsfrom horse and human urine and from hog pancreas,pp. 681Ð699. In G. E. Perlman and L. Lorand [eds.],Methods in enzymology, vol. XIX. Proteolytic enzymes.Academic, New York.

Wilcox, C. L., and A. B. Lange. 1995. Role of extracellularand intracellular calcium on proctolin-induced contrac-tions in an insect visceral muscle. Regul. Pept. 56: 49Ð59.

Received 8 June 2004; accepted 1 November 2004.


vaidyanathan jme 05

Documents