plTplporblilhlaPliateprsitc

br

Ep

E

Journal of Invertebrate Pathology 73, 289–293 (1999)Article ID jipa.1998.4836, available online at http://www.idealibrary.com on

Rhodnius prolixus Infected with Trypanosoma rangeli:In Vivo and in Vitro Experiments

S. A. O. Gomes, D. Feder, N. E. S. Thomas, E. S. Garcia, and P. Azambuja1

Departamento de Bioquımica e Biologia Molecular, Instituto Oswaldo Cruz, Fundacao Oswaldo Cruz, Avenue Brasil 4365,CEP 21045-900, Rio de Janeiro, RJ, Brazil

Received April 20, 1998; accepted November 2, 1998

metld(itcSt

smhaw1msbooalslwlt(rtpdt4Tb(

Studies were carried out on the activation of therophenoloxidase (proPO) in adults of Rhodnius pro-ixus infected by short and long epimastigote forms ofrypanosoma rangeli. The in vitro activation of theroPO cascade using L-DOPA as substrate was very

ow in the absence of fat body extract, hemolymph, andarasites. On the other hand, a higher PO activity wasbserved when short, but not long, epimastigotes of T.angeli were incubated with fresh hemolymph, fatody extract, and L-DOPA. Supernatant from lysed

ong epimastigotes increased the PO activity at levelsdentical to those observed with supernatants fromysed short epimastigotes. Similarly, the PO activity ofemolymph obtained from inoculated insects with

ong epimastigotes of T. rangeli showed a very lowctivity when incubated with L-DOPA compared to theO activity of hemolymph taken from insects inocu-

ated with short epimastigotes of T. rangeli. Controlnsects inoculated with sterile PBS showed no POctivity. These data indicate the presence of (a) fac-or(s) in the hemolymph as well as in the fat bodyxtract that may be released (or induced) by theresence of short epimastigotes of T. rangeli and whichesults in the activation of the R. prolixus proPOystem. The implications of these findings are discussedn relation to the development of T. rangeli and its abilityo overcome the proPO system, survive, and successfullyolonize the hemolymph of R. prolixus. r 1999 Academic Press

Key Words: prophenoloxidase; phenoloxidase; fatody; hemolymph; Rhodnius prolixus; Trypanosomaangeli.

INTRODUCTION

Phenoloxidase (o-diphenol, oxygen oxidoreductaseC 1.10.3.1) is an enzyme involved in the oxidation ofhenols to toxic quinones, which is responsible for the

1 To whom correspondence should be addressed. Fax: 21 590 3495.

i-mail: Azambuja@gene.dbbm.fiocruz.br.

289

elanization processes in insects that result in thencapsulation of pathogens as well as cuticle sclerotiza-ion and wound repair (Andersen, 1985). The propheno-oxidade (proPO), an inactive precursor of phenoloxi-ase (PO) found in the plasma fraction of hemolymphPye, 1974) or in the hemocytes (Leonard et al., 1985) ofnvertebrates, can be activated by proteases, such asrypsin and chymotrypsin, bacteria, and fungi, or by theirell wall components (Ashida et al., 1983; Aso et al., 1985).ince the oxidation of phenols to toxic products can killhe insects, this process must be transient.

It is well known that in Triatominae vectors, Trypano-oma rangeli, after being ingested as trypomastigotes,ultiply as epimastigotes in the midgut, invade the

emolymph and hemocytes to continue their growth,nd complete their development in the salivary glandhere metacyclogenesis takes place (Hoare, 1972; Cuba,975). Invertebrate host and parasite parameters mayodulate the hemocoelic T. rangeli invasion and its

urvival in the vector. For example, no correlationetween the digestive proteases and the developmentf the parasites in the gut of Rhodnius prolixus wasbserved (Bauer, 1981). A similar insensitivity to prote-ses was described for Trypanosoma cruzi in R. pro-ixus (Garcia and Gilliam, 1980). Hecker et al. (1990)howed that T. rangeli passes through the gut epithe-ium to the hemocoel by using an intracellular routeithin a parasitophorous vacuole. Once in the hemo-

ymph, T. rangeli may be recognized and activated byhe defense system of the vector. Gregorio and Ratcliffe1991) showed that the susceptibility of R. prolixus to T.angeli infection, at least in part, might be related tohe suppression of the activation of proPO in theresence of this flagellate. Recently, Mello et al. (1995)emonstrated that the rate of T. rangeli development inhe hemolymph of R. prolixus linearly increased withindays after infection. Numerous short epimastigotes of. rangeli were present until day 2 in the hemolymph,ut after this time long epimastigotes were dominantMello et al., 1995). Interestingly, the PO activity

nduced by the infection with T. rangeli presented a

0022-2011/99 $30.00Copyright r 1999 by Academic Press

All rights of reproduction in any form reserved.

pea

vbseegeres

I

cpwff

T

d2srtpttntesca(sw

H

psdapswR

Caasbc

P

µ2arvb1maiaL

sa(EarttlTrtmpelraatt

P

wC(

P

v

290 GOMES ET AL.

eak of activity on day 1 after infection, just when shortpimastigotes predominate in the hemolymph (Mello etl., 1995).The objective of the present study was to describe in

ivo and in vitro experiments in the hematophagousug, R. prolixus, infected with T. rangeli. Our resultshowed that short epimastigotes of T. rangeli werefficient in activating the proPO system while longpimastigotes were not. In addition, this paper alsoives an account of studies which support the hypoth-sis that the fat body extract may have some factor(s)elated to the activation of the proPO system. Thesexperiments shed light on the question of how T. rangeliurvive and develop in the hemolymph of the vector.

MATERIALS AND METHODS

nsects and Feeding

All experiments were undertaken with randomlyhosen adult R. prolixus, reared and maintained asreviously described (Garcia et al., 1984). The insectsere starved for 15 days after the last ecdysis and then

ed on citrated human blood through a membraneeeder (Garcia et al., 1984).

rypanosoma rangeli Infection

T. rangeli strain H14 (supplied by Dr. Maria Auxilia-ora de Sousa, FIOCRUZ, Brazil) was maintained at8°C in BHI (brain heart infusion, DIFCO) mediumupplemented with 20% heat-inactivated fetal calf se-um. The grown epimastigotes were taken from cul-ures and washed three times in 0.14 M NaCl in 0.01 Mhosphate buffer (PBS), pH 7.2. Short and long epimas-igotes were checked by Giemsa staining. Short epimas-igotes were obtained from fresh cultures in the expo-ential growth phase (until day 7 of cultivation 99% ofhe parasites were short epimastigotes), while the longpimastigotes, predominantly (.98%) found in thetationary growth phase, were taken after 12 days ofulture. Insects were inoculated into the thorax, 5 daysfter feeding, with 1 µL of a trypanosome suspension1 3 106 cells/mL) in PBS using a 10-µL Hamiltonyringe connected to a fine-glass needle. Control insectsere inoculated with 1 µL of sterile PBS.

emolymph and Abdominal Fat Body Collections

Hemolymph was carefully collected with micropi-ettes at different intervals after inoculation of para-ites by cutting off the insect legs and immediatelyiluted in insect isotonic buffer (0.01 M sodium cacodyl-te, 0.01 M CaCl2, 280 mM sucrose, pH 7.4) in aroportion of 1 part hemolymph per 9 parts bufferolution. For fat body extract preparation, R. prolixusere laterally cut and opened by dorsal dissection in

hodnius saline (0.1 M NaCl, 25 mM KCl, and 10 mM f

aCl2), and the digestive tract, Malphigian tubules,nd reproductive organs were removed. The ventralbdominal fat body was then washed in Rhodniusaline and homogenized in insect isotonic cacodylateuffer, and after centrifugation the soluble fraction wasollected.

rophenoloxidase-Activating Assay

Unless otherwise stated, for in vitro experiments, 10L of diluted hemolymph taken 5 days after feeding (ca.0 µg protein) were pre-incubated for 2 h with anliquot of 20 µg protein of fat body extract and 1 µL of T.angeli suspension (1 3 106 cells/mL), and the finalolume was adjusted to 45 µL with isotonic cacodylateuffer. The reactions were initiated by the addition of5 µL of DOPA solution and allowed to proceed for 30in at room temperature. For in vivo experiments

liquots of 10 µL of hemolymph taken from inoculatednsects, diluted to 10% in isotonic cacodylate buffer, anddded to 35 µL of buffer, were assayed with 15 µL of-DOPA (Sigma Chem. Co.) saturated solution as de-cribed before (Azambuja et al., 1991a). The absorbancet 492 nm was measured in an ELISA plate readerModel Anthos Labtec HT2 from Dely Instruments,ngland). The enzyme unit (U) was expressed asbsorbance at 492 nm 3 100 for 30 min incubation atoom temperature. Previous experiments have shownhat the PO activity was minimal when protein concen-ration of fat body extract or hemolymph samples wereess than 10 µg, pH below 6.0, and in absence of CaCl2.o study the interference of long epimastigotes of T.angeli in the proPO activation induced by short epimas-igotes, 0.5 µL of short epimastigotes suspension wasixed with 0.5 µL of long epimastigotes suspension and

re-incubated with fresh hemolymph and fat bodyxtract for 2 h as described above. In some experiments,ysed suspensions of short and long epimastigotes of T.angeli (1 3 106 cells/mL) were obtained by freezingnd thawing the suspension of washed parasites, andfter centrifugation at 10,000g for 15 min, superna-ants were collected and used for in vitro activation ofhe proPO system.

rotein Quantification

Proteins of hemolymph and fat body extract samplesere quantified with protein dosage kit (Sigma Chem.o.) using bovine serum albumin (BSA) as standard

Lowry et al., 1951).

RESULTS

reliminary in Vitro Experiments and Precautions

To obtain clear and reproducible results on the initro assays we studied the importance of hemolymph,

at body extract, and the time of pre-incubation on the

pssrbdealshfsthwibesewt

D

PwopFns

L

Twossaastu

P

seicaTdPotreoa7

agbiL

a

oT(ppll

291Rhodnius prolixus INFECTED WITH Trypanosoma rangeli

roPO activation. The results, summarized in Fig. 1,how the PO activity for each group tested. It demon-trates that addition of short or long epimastigotes of T.angeli in a medium containing hemolymph and fatody extract activated the proPO system, showing aifferent kinetics of melanin formation. In fact, shortpimastigotes of T. rangeli were able to activate the POctivity after 2 h of pre-incubation 17-fold more thanong epimastigotes. The preliminary observationshowed that the time of pre-incubation, the presence ofemolymph, and the fat body extract and parasiteorms used modified the in vitro activation of the proPOystem. The lowest PO activities were observed whenhe pre-incubation media with parasites containedemolymph or fat body separately. These PO activitiesere similar to the control group, i.e., with the pre-

ncubation medium containing hemolymph and fatody extract but not parasites. Therefore, for furtherxperiments, we decided that the assay of PO activityhould be carried out using hemolymph with fat bodyxtract and adding different preparations of parasitesith a pre-incubation period of 2 h at room tempera-

ure.

ifferent Epimastigote Forms of T. rangeliand the proPO Cascade

Short epimastigotes of T. rangeli induced 64 units ofO activity (Fig. 2). A similar experiment carried outith long epimastigotes of T. rangeli was able to inducenly 8 units. Control group assayed in the absence ofarasites produced 7 units of proPO activation (Fig. 2).igure 2 also demonstrates that long epimastigotes didot interfere with the melanin formation induced byhort epimastigotes.

FIG. 1. Temporal course of in vitro activation of proPO system ofdults of Rhodnius prolixus by epimastigotes of Trypanosoma ran-eli. Pre-incubation was performed for 2 h with hemolymph and fatody extract in the presence of short (d) or long (j) epimastigotes orn the absence of parasites (control, n). After the pre-incubation time-DOPA was added as a substrate and the absorbance was measured

Pt 492 nm 30 min later.

ysis of Epimastigotes and proPO Cascade

The capacity of lysed short and long epimastigotes of. rangeli to induce the activation of the proPO systemas tested. As expected, the control sample presentednly background levels of activation of the proPOystem (Fig. 2). However, the lysed supernatants ofhort and long epimastigotes were able to drasticallyctivate the proPO system, attaining 82 and 76 units ofctivation, respectively. When supernatant from lysedhort epimastigotes was mixed with lysed long epimas-igotes the production of melanin reached about 92nits under the assayed condition (Fig. 2).

henoloxidase Activity in Vivo

We then investigated the activation of the proPOystem in the hemolymph of R. prolixus inoculated withpimastigotes of T. rangeli 5 days after feeding. Twomportant results were observed when the data fromontrol and experimental groups are compared. First, POctivity in the group inoculated with short epimastigotes of. rangeli was activated by the flagellate inoculation andetected a few hours after inoculation, reaching 44 units ofO activation after 6 h, and continued increasing to a peakf 63 units of activation 24 h after infection (Fig. 3). Afterhat, PO activity decreased to 22 units in 48 h andemained stable later. Second, the inoculation of longpimastigotes of T. rangeli increased the PO activitynly to 19 units within the first 6 h, reached 5 units ofctivation in 24 h, and remained stable at 12 units until2 h. Finally, control insects inoculated with sterile

FIG. 2. In vitro activation of the proPO system in the hemolymphf adults of Rhodnius prolixus by short and/or long epimastigotes ofrypanosoma rangeli. Live (C) or supernatant from lysed parasites

h) were pre-incubated for 2 h in the presence of fresh hemolymphlus fat body extract before the addition of L-DOPA. Control waserformed with hemolymph plus fat body extract in the absence ofive and lysed parasites (j). Each point represents a pool of hemo-ymph (n 5 5 insects). The experiment was repeated four times.

BS had only background levels of PO activity (Fig. 3).

sRplaptaa1iposRHphrbahtHtattrwili

awteTcp

mh1mcGbhp(ahrametgpwafssmuMswsisiwpScmletHgtpa

oelme

292 GOMES ET AL.

DISCUSSION

The regulation of the proPO/PO pathway has beentudied in some hematophagous insects. For example,owley et al. (1986) demonstrated that in simuliids theathway for melanin formation exists within the hemo-ymph and that the serine-proteases are involved in thectivation of the proPO in vivo. Recent work on theroPO/PO activation in larval Ae. aegypti suggestedhat this proenzyme is present in these mosquitoes,ctivated in the presence of a bivalent cation (Ca21),nd a limited proteolysis gave rise to PO (Ashida et al.,990). It was demonstrated that serine proteases arenvolved in this activation (Ashida et al., 1990). In R.rolixus, the activation of the proPO pathway wasbserved in insects inoculated with bacteria or trypano-omatids (Azambuja et al., 1986, 1989; Gregorio andatcliffe, 1991; Mello et al., 1995; Feder et al., 1996).owever, nothing is known about the localization ofroPO activating factors in this insect. Insects mustave a process by which the substrate proPO is sepa-ated from its activator until needed. Of course, fatody and hemolymph are important clear targets fornalysis. It is also known that in other insect species,aemocytes/hemolymph are the source of proPO prioro the formation of the PO active enzyme (Ham, 1992).owever, in the present study we could not determine if

his system is confined to the hemocytes or is present assoluble fraction in the noncellular hemolymph, since

he filtration of hemolymph gave us no clear evidence ofhe proPO source (not shown). Azambuja et al. (1991)eported that the preparation of cell-free hemolymphithout incidental activation of the hemocytes (lysis!)

s rather difficult, despite the use of special anticoagu-ant solutions. Thus, care has to be taken in interpret-

FIG. 3. In vivo activation of the proPO system in the hemolymphf Rhodnius prolixus adults by inoculation with short (d) or long (j)pimastigotes of Trypanosoma rangeli or PBS control (m). Hemo-ymphs from inoculated insects were incubated with L-DOPA for 30

in. Each point represents a pool of hemolymph (n 5 5 insects). Thexperiment was repeated four times.

ng results demonstrating hemolymph-derived proPO d

ctivity. However, the activation of the proPO systemas dependent on the presence of fat body extract in

he reaction medium. Neither hemolymph nor fat bodyxtract alone was able to activate the proPO cascate.herefore, hemolymph and fat body extract taken fromontrol insects were important as a source of the proPOathway and other activating factor(s).The importance of the vector immune system in theodulation of the parasite–insect vector interaction

as been recognized (Molyneux et al., 1986; Kaaya,989). The proPO system has been implicated in hu-oral immunity, non-self recognition, and cell–cell

ooperation in insects (Ratcliffe et al., 1985). Recently,regorio and Ratcliffe (1991) showed that the suscepti-ility of R. prolixus to T. rangeli hemocoel infection mayave been related to the suppression of the activation ofroPO in the presence of this flagellate. Mello et al.1995) demonstrated that T. rangeli triggered proPOctivation in vivo with a maximal level at 24 h. From 24

on, while PO activity falls the parasite numberapidly increases (Mello et al., 1995). These authorslso demonstrated that short epimastigotes rapidlyultiply in the hemolymph and transform into long

pimastigotes of T. rangeli. This fact conflicts withhose of Anez (1983), who described that long epimasti-otes of T. rangeli multiplied in the hemolymph of R.rolixus, but fully agrees with Tobie’s observation,hich revealed that short epimastigotes of T. rangelippear soon in the hemolymph, and then they trans-orm in long epimastigotes that invade hemocytes andalivary glands (Tobie, 1970). These contraditory re-ults emphasize the difficulty in reproducing experi-ents in which factors such as insect or parasite strain

sed can greatly influence the data. Nevertheless,ello et al. (1995), Anez (1983), and Tobie (1970) did not

eparate the epimastigotes into short/long forms. Here,e describe for the first time that the inoculation of

hort and long epimastigotes of T. rangeli strain H14nto the hemolymph induced different kinetics of proPOystem activation: short epimastigotes were competentn activating the PO activity and long epimastigotesere not. Similar findings on the activation of theroPO system were observed in an in vitro system.imilarly, preliminary results from our laboratory indi-ate that short epimastigotes of T. rangeli strain H14ultiply fast in the hemolymph and transform into

ong epimastigotes (Gomes et al., 1997). Also, longpimastigotes of T. rangeli H14, when inoculated intohe hemolymph, disappear within 2 days of inoculation.owever, there was an increase of the percentage of

iant hemocytes (Azambuja et al., 1991b), suggestinghat long epimastigotes penetrated into the hemolym-hatic cells as observed by Tobie (1970) and Gomes etl. (1997).In conclusion, these results provide the first basis for

elineating the in vivo mechanism that is involved in

ttfspoederpfiggmsaeR

DaABco

A

A

A

A

A

A

A

A

B

C

F

G

G

G

G

H

H

H

K

L

L

M

M

P

R

R

T

293Rhodnius prolixus INFECTED WITH Trypanosoma rangeli

he establishment of T. rangeli infection in the vector:he passage of short to long epimastigotes is importantor the development of intrahemocoelic stages of para-ites and to overcome the proPO system. Based on theresent model, we postulate that differential activationf the proPO system of R. prolixus by short and longpimastigotes of T. rangeli appears to be responsible forifferences in the dynamics of infection. This hypoth-sis assumes that some factor(s) in the membrane oreleased by the short epimastigotes activate(s) theroPO cascade. Evidence for this conclusion comesrom the fact that long epimastigotes did not block thencrease of PO activity induced by the short epimasti-otes of T. rangeli and lysed parasites of both epimasti-otes similarly activated the formation of melanin. Thisodel will allow an investigation into differences in the

urface components of the short and long epimastigotesnd/or their secretions which permit the T. rangeli tostablish the infection in the hemolymph of the vector,. prolixus.

ACKNOWLEDGMENTS

This investigation was supported by the Conselho Nacional deesenvolvimento Cientıfico e Tecnologico (CNPq), Programa de Apoio

` Pesquisa Estrategica em Saude (PAPES, FIOCRUZ), Escola Brasil–rgentina de Biotecnologia (Ministerio da Ciencia e Tecnologia,rasil), and Projeto CAPES/DAAD (ProBAL, Ministerio da Educa-

¸ao, Brasil). We thank Dr. C. Pirmez (FIOCRUZ) for critical readingf the manuscript.

REFERENCES

ndersen, S. O. 1985. Sclerotization and tanning of the cuticle. In‘‘Comprehensive Insect Physiology, Biochemistry and Pharmacol-ogy’’ (G. A. Kerkut and L. I. Gilbert, Eds.), Vol. 3, pp. 59–74.Pergamon, Oxford.

nez, N. 1983. Studies on Trypanosoma rangeli Tejera, 1920. VI.Developmental pattern in the hemolymph of Rhodnius prolixus.Mem. Inst. Oswaldo Cruz 74, 413–419.

shida, M., Ishizaki, Y., and Iwahana, H. 1983. Activation of prophe-noloxidase by bacterial cell wall or by 1,3-glucans in plasma of thesilkworm, Bombyx mori. Biochem. Biophys. Res. Commun. 113,562–568.

so, Y., Kramer, K. J., Hopkins, T. L., and Lockhart, G. L. 1985.Characterization of hemolymph protyrosinase and a cuticularactivator from Manduca sexta (L). Insect Biochem. 15, 9–17.

zambuja, P., Freitas, C. C., and Garcia, E. S. 1986. Evidence andpartial characterization of an inducible antibacterial factor in thehemolymph of Rhodnius prolixus. J. Insect Physiol. 32, 807–812.zambuja, P., Mello, C. B., and Garcia, E. S. 1989. Immunity toRhodnius prolixus: Inducible peptides against bacteria and trypano-somes. In ‘‘Host Regulated Developmental Mechanism in VectorArthropods’’ (D. Borovsky and A. Spielman, Eds.), pp. 270–276.Florida Univ. Press, Vero Beach, FL.

zambuja, P., Garcia, E. S., Ratcliffe, N. A., and Warthen, J. D., Jr.1991a. Immune-depression in Rhodnius prolixus induced by thegrowth inhibitor, azadirachtin. J. Insect Physiol. 37, 771–777.zambuja, P., Garcia, E. S., and Ratcliffe, N. A. 1991b. Aspects ofclassification of hemiptera hemocytes from six triatomine species.Mem. Inst. Oswaldo Cruz 86, 1–10.

auer, P. G. 1981. ‘‘Ultrastrukturelle und Physiologische Aspekte des

Mitteldarms von Rhodnius prolixus Stal (Insecta: Hemiptera).’’Dissertation, Universitat Basel.

uba, C. A. 1975. Estudo de uma cepa peruana de Trypanosomarangeli. VI. Observacoes sobre sua evolucao e morfogenese nahemocele e nas glandulas salivares de Rhodnius ecuatoriensis. Rev.Inst. Med. Trop. Sao Paulo 17, 283–297.

eder, D., Mello, C. B., Garcia, E. S., and Azambuja, P. 1996. Immuneresponses in Rhodnius prolixus: Influence of nutrition and ecdy-sone. J. Insect Physiol. 43, 513–519.arcia, E. S., Azambuja, P., and Contreras, V. T. 1984. Large-scalerearing of Rhodnius prolixus and production of metacyclic trypomas-tigotes of Trypanosoma cruzi. In ‘‘Genes and Antigens of Parasites,a Laboratory Manual’’ (C. M. Mrel, Ed.), pp. 44–47. FundacaoOswaldo Cruz, World Health Organization, Rio de Janeiro.arcia, E. S., and Gilliam, F. C. 1980. Trypanosoma cruzi develop-ment is independent of protein digestion of Rhodnius prolixus. J.Parasitol. 66, 1052–1053.omes, S. A. O., Feder, D., Thomas, N., Garcia, E. S., and Azambuja,P. 1997. Studies of Trypanosoma rangeli of development in Rhod-nius prolixus: Oral infection with different forms of parasites.Mem. Inst. Oswaldo Cruz 92(Suppl. I), 289.regorio, E. A., and Ratcliffe, N. A. 1991. The prophenoloxidasesystem in vitro interaction of Trypanosoma rangeli with Rhodniusprolixus and Triatoma infestans hemolymph. Parasite Immunol.13, 551–564.am, P. J. 1992. Immunity in haematophagous vectors of parasiteinfection. In ‘‘Advances in Disease Vector Research,’’ Vol. 9, pp.101–149. Springer-Verlag, New York.ecker, H., Schwarzenbach, M., and Rudin, W. 1990. Developmentand interactions of Trypanosoma rangeli in and with reduviid bugRhodnius prolixus. Parasitol. Res. 76, 311–318.oare, C. A. 1972. ‘‘The Trypanosomes of Mammals. A ZoologicalMonograph.’’ Blackwell Sci., Oxford–Edinburgh.aaya, G. P. 1989. A review of the progress in recent years onresearch and understanding of immunity in insect vectors ofhuman and animal diseases. Insect Sci. Appl. 10, 751–769.

eonard, C., Soderhall, K., and Ratcliffe, N. A. 1985. Studies onprophenol-oxidase activity in Blaberus cranifer haemocytes. InsectBiochem. 15, 803–810.

owry, O. H., Rosebrough, H. J., Farr, A. L., and Randall, R. J. 1951.Protein measurement with the Folin phenol reagent. J. Biol. Chem.193, 265–275.ello, C. B., Garcia, E. S., Ratcliffe, N. A., and Azambuja, P. 1995.Trypanosoma cruzi and Trypanosoma rangeli: Interplay with hemo-lymph components of Rhodnius prolixus. J. Invertebr. Pathol. 65,261–268.olyneux, D. H., Takle, G., Ibrahim, E. A., and Ingram, G. A. 1986.Insect immunity in Trypanosomatidae. In ‘‘Immune Mechanismsin Invertebrate Vectors’’ (A. M. Lackie, Ed.), pp. 117–144. Claren-don, Oxford.

ye, A. E. 1974. Microbial activation of prophenoloxidase fromimmune insect larvae. Nature (London) 251, 610–613.atcliffe, N. A., Rowley, A. F., Fitzgerald, S. W., and Rhodes, C. P.1985. Invertebrate immunity: Basic concepts and recent advances.Int. Rev. Cytol. 97, 183–250.

owley, A. F., Ratcliffe, N. A., Leonard, C. M., Richard, E. H., andRenwrantz, L. 1986. Humoral recognition factors in insects, withparticular reference to agglutinins and the prophenoloxidase sys-tem. In ‘‘Hemocytic and Humoral Immunity in Arthropods’’ (A. P.Gupta, Ed.), pp. 381–406. Wiley, New York.

obie, E. J. 1970. Observations on the development of Trypanosomarangeli in the haemocoel of Rhodnius prolixus. J. Invertebr. Pathol.15, 118–125.