General and Comparative Endocrinology 162 (2009) 105–112

Contents lists available at ScienceDirect

General and Comparative Endocrinology

journal homepage: www.elsevier .com/locate /ygcen

Biological activity of diuretic factors on the anterior midgutof the blood-feeding bug, Rhodnius prolixus

Victoria Te Brugge *, Juan P. Ianowski, Ian OrchardDepartment of Biology, University of Toronto Mississauga, 3359 Mississauga Road, Mississauga, Ont., Canada L5L-1C6

a r t i c l e i n f o a b s t r a c t

Article history:Received 2 December 2008Revised 28 January 2009Accepted 29 January 2009Available online 7 February 2009

Keywords:InsectCropTransportDiuresisDiuretic peptideSerotonin

0016-6480/$ - see front matter � 2009 Elsevier Inc. Adoi:10.1016/j.ygcen.2009.01.025

* Corresponding author. Fax: +1 905 828 3792.E-mail address: victoria.tebrugge@utoronto.ca (V.

Probing of a host and ingestion of a blood-meal in a fifth instar Rhodnius prolixus results in a cascade oftightly integrated events, including salivary gland secretion, plasticisation of the abdominal cuticle,increased ion and water movement across the anterior midgut (crop) and Malpighian tubules (which rap-idly produce urine) and the regular expulsion of urine from the hindgut. In this study we have focussedon the role of the anterior midgut during the rapid postprandial diuresis. The huge blood-meal is pumpedinto the anterior midgut, during feeding, then modified by diuresis and stored until it is digested. Changesin the anterior midgut activity are rapid. Within minutes of the commencement of feeding there is anincrease in the frequency of anterior midgut contractions and diuresis begins with the movement of saltand water across the epithelium of the anterior midgut into the haemolymph.

While serotonin, a diuretic hormone in R. prolixus, is known to play a role in the physiological activity ofthe anterior midgut, we were interested in exploring further the role of serotonin, and other diuretic pep-tides. We have tested the activity of several peptides, including R. prolixus calcitonin-like diuretic hor-mone (Rhopr-DH31), corticotropin-releasing factor (CRF)-like peptide from Zootermopsis nevadensis DH(Zoone-DH) and a kinin from Leucophaea maderae, Leucokinin 1 (LK1). These peptides families are knownto be present in the central nervous system of R. prolixus, are putative neurohormones released into thehaemolymph after the start of feeding, and have been shown to have activity on a variety of tissuesinvolved in post-feeding diuresis. We show here that both serotonin and Zoone-DH increase the cAMPcontent of the anterior midgut and that serotonin, Zoone-DH and cAMP analogues increase absorptionof water from the anterior midgut, increase the short circuit current and voltage, while decreasing theresistance across the epithelium. While LK1 and Rhopr-DH31 do not significantly increase absorption,or short circuit current, LK1 does significantly decrease the resistance and transepithelial voltage ofthe anterior midgut epithelium. All of the factors studied increase the frequency of contractions of theanterior midgut.

� 2009 Elsevier Inc. All rights reserved.

1. Introduction

As Rhodnius prolixus prepare to ingest a blood meal, they probethe host to find a suitable site. Once a site is located with their pro-boscis inserted in the host, they start to inject saliva and to feed.The ingested blood is taken into the anterior midgut for storage,and diuresis begins. Diuresis involves the movement of salt andwater across the epithelium of the anterior midgut into the haemo-lymph, and from the haemolymph across the cells of the upperMalpighian tubule into the lumen of the tubule. Urine flows alongthe tubule to the lower Malpighian tubule where K+ is reabsorbed(Collier and O’Donnell, 1997), and this modified urine continuesinto the hindgut where it is expelled at regular intervals. Unlike

ll rights reserved.

Te Brugge).

some other insects (see Phillips et al., 1998) the hindgut in R. pro-lixus does not participate in modifying the urine.

Although diuresis begins with the movement of water and saltacross the epithelium of the anterior midgut, much of the researchon diuresis to date has focussed on the Malpighian tubules. Studiesfocussing on the anterior midgut and diuresis were reported byFarmer et al. (1981) and Barrett (1982). Along with the transportof water and salt, which concentrates the ingested blood, the ante-rior midgut increases its frequency of muscle contraction rapidlyafter the start of feeding (Maddrell, 1964). This increase in the rateof contraction is believed to contribute to the mixing of ingestedblood inside the anterior midgut and haemolymph outside theanterior midgut, thus minimizing unstirred layers. At the sametime, these contractions increase the circulation of hormones with-in the haemolymph. R. prolixus tubules have been shown to havelarge unstirred layers surrounding the tubule in vitro, whilein vivo these unstirred layers are significantly reduced (Collier

106 V. Te Brugge et al. / General and Comparative Endocrinology 162 (2009) 105–112

and O’Donnell, 1997). The reduction of the unstirred layers in vivois potentially due to contractions of the anterior midgut and hind-gut and subsequent mixing of the haemolymph.

The integration of events associated with feeding and diuresisin R. prolixus is controlled and coordinated by a combination ofneurohormones, neurotransmitters and neuromodulators. Seroto-nin is one of these coordinating factors (see Orchard, 2006). Sero-tonin is present in the central nervous system (CNS), salivaryglands and anterior midgut of R. prolixus (Lange et al., 1988; Orch-ard et al., 1988; Miggiani et al., 1999). Serotonin plasticises thecuticle (Reynolds, 1974; Orchard et al., 1988), increases Malpighiantubule secretion (Maddrell et al., 1969, 1971), increases the rate ofcontraction of the dorsal vessel (Chiang et al., 1992), and increasesion and water transport across the anterior midgut (Farmer et al.,1981). Serotonin acts through a cAMP-dependent mechanism inR. prolixus anterior midgut, hindgut, Malpighian tubules and epi-dermis (Barrett and Orchard, 1990; Barrett et al., 1993), althoughthe activation of other pathways is also possible. Similarly, pepti-dergic diuretic hormones in R. prolixus and in other insects havebeen shown to have diuretic activity, moving water and salt acrossepithelia including Malpighian tubules, and to have myotropicactivity on a variety of tissues such as Malpighian tubules, foregutand hindgut (Te Brugge et al., 1999, 2001, 2002a,b; Blake et al.,1996; Coast, 1998; Coast et al., 2005).

Recently we identified and sequenced the calcitonin-like diure-tic hormone, Rhopr-DH31, from R. prolixus (Te Brugge et al., 2008)and found it to be identical in amino acid sequence to the Diplop-tera punctata DH31 (Furuya et al., 2000). Previously, we have dem-onstrated the presence of CRF-like, DH31-like and kinin-likeimmunoreactivity throughout the CNS of R. prolixus and found evi-dence that these peptides are released as neurohormones after thestart of feeding. In addition, we have observed DH31-like, kinin-likeand CRF-like immunoreactive processes over the posterior midgutand hindgut (Te Brugge et al., 1999, 2001, 2005), as well as kinin-like immunoreactive endocrine cells in the posterior midgut (TeBrugge et al., 2001). The presence of these peptidergic diuretic fac-tors in neurohaemal sites and posterior midgut and hindgut sug-gests the possibility that these peptides could potentially play aphysiological role on many tissues, including the anterior midgut,during feeding, and therefore might be involved in the post-feed-ing diuresis. In this study we investigate the biological activity ofCRF-like, Rhopr-DH31 and kinin peptides on the anterior midgutof fifth instar R. prolixus.

2. Materials and methods

2.1. Insects

Fifth instars of R. prolixus were taken from a long standing col-ony maintained at 25 �C under high humidity. The insects wereunfed for 5–7 weeks and had previously been fed on rabbit’s bloodas fourth instars.

2.2. Peptides and chemicals

The native R. prolixus CRF-like and kinin-like peptides have notbeen identified or sequenced. In previous studies on the Malpi-ghian tubules of R. prolixus the CRF-like peptide from Zootermopsisnevadensis DH (Zoone-DH) and the kinin-like peptide from Leucop-haea maderae, Leucokinin 1 (LK1) have been utilized. We have usedthe same peptides in the present study of the anterior midgut. TheCRF-like peptide, Zoone-DH, and R. prolixus diuretic hormone 31(Rhopr-DH31) were a gift from Dr. D.A. Schooley. Leucokinin 1was obtained from Bachem (Torrance, CA, USA) while serotoninand all other chemicals were obtained from Sigma–Aldrich (Miss-

issauga, Ontario, Canada) unless otherwise noted. Insect tissueswere dissected and experiments were preformed utilizing salinecomposed of NaCl 129, KCl 8.6, MgCl2 8.5, CaCl2 2, NaHCO3 10.2,NaH2PO4 4.3, Glucose 20, and HEPES 8.6 mM (pH 7).

2.3. Phalloidin staining

Anterior midguts were dissected from fifth instar R. prolixus un-der saline. Tissues were fixed in 2% paraformaldehyde overnight at4 �C then washed in phosphate buffered saline (PBS). They wereincubated in 4% Triton X 100 with 10% normal sheep serum forone hour at room temperature. The tissues were then transferredinto PBS containing Phalloidin conjugated to Cy3 at a concentrationof 1:1000 for 0.5 h at room temperature, followed by a wash in PBSfor 18 h at 4 �C. Preparations were mounted on slides in glyceroland viewed on a confocal microscope (Zeiss, LSM 510).

2.4. Cyclic AMP quantification

Anterior midguts were dissected from fifth instars under saline.The tissues were then transferred to a microfuge tube containing10�3 M of the phosphodiesterase inhibitor, 3-isobutyl-1-methyl-xanthine (IBMX) and either saline, Zoone-DH, Rhopr-DH31, LK1,or serotonin, final volume 50 ll. Tissues were incubated for10 min and the reaction was stopped with 250 ll of boiling0.05 M sodium acetate. Samples were placed in a boiling waterbath for 5 min, then frozen at �20 �C until assayed. To assay thecAMP content of the anterior midgut, the samples were thawed,sonicated and centrifuged at 8800g for 10 min and the supernatantdecanted. The cAMP in the supernatant was measured using aradioimmunoassay kit (Mandel/NEN, Guelph, ON, Canada) withmodifications as described in Lange and Orchard (1986). The re-sults are expressed as mean ± SE (n = 5).

2.5. Anterior midgut absorption assay: water transport

Anterior midguts were removed from fifth instar R. prolixus andwashed for 5 min in saline. The posterior end of the anterior mid-gut was ligated with a silk thread, and the anterior end was cutopen and the anterior midgut contents were flushed out with sal-ine. Twenty to thirty microlitres of saline was introduced into thelumen of the anterior midgut and then the anterior end was alsoligated with a silk thread. The anterior midgut was then gentlyblotted and weighed on a Mettler AE 240 balance. Once weighed,the tissues were placed in a microfuge tube with 100 ll of salineor saline containing serotonin or peptide solution. The tissues wereincubated for 30 min at room temperature (23 ± 0.5 �C) then gentlyremoved, blotted and weighed. The difference in weight, initialminus final, was calculated. A rate of absorption (ll/min) was thencalculated assuming a specific gravity of 1. The results are ex-pressed as mean ± SE (n = 5–10).

2.6. Ussing chamber assay: ion transport

Anterior midguts were removed under saline from fifth instar R.prolixus. The anterior midguts were cut longitudinally to create aflat sheet of tissue and washed with saline. The tissues weremounted in one half of a 500 ll Ussing chamber (4 mm diameteropening) then clamped in place with the second half (500 ll)chamber. The chamber had four ports for the electrodes and fourports for the introduction of fluid (two on the top of each chamber).The chamber was connected to a World Precision Instruments(WPI, New Haven, CT, USA), EVC 4000 amplifier with an EVC 3pre-amplifier via two voltage and two current calomel electrodeswith 4% agar in 150 mM NaCl bridges. The data were collectedusing an 8 channel Powerlab IDE and recorded with Chart version

Fig. 1. Phalloidin staining on the anterior midgut. Filled white arrow ()) indicateslongitudinal muscle fibre. Open white arrow (*) indicates circular muscle fibre.Scale bar: 50 lm.

V. Te Brugge et al. / General and Comparative Endocrinology 162 (2009) 105–112 107

4. The tissue was mounted in place and the chambers were eachfilled with 500 ll saline. Air was slowly bubbled into the lumenside of the chamber. Assays were run at room temperature. The po-tential difference (PD) was recorded for the tissue. Only tissueswith initial PD greater than 4 mV were used for the study. The tis-sues were voltage clamped at 0 with a 10 mV, 200 ms pulse every6 s. The tissues were allowed to equilibrate for 20 min. Anteriormidgut tissues had a PD that was lumen negative. Serotonin orpeptide was added to the basolateral side of the preparation to afinal concentration of 10�6 M. The tissue was washed by gentlyremoving the contents of the chamber on the basolateral side ofthe tissue and adding saline, 3–4 times, then refilling the chamber.Short circuit current (Isc, lA/cm2) was measured at 0 voltage, andresistance (Xcm2) and transepithelial voltage (Vte, mV) were cal-culated (n = 5–11).

2.7. Anterior midgut contraction assay

Anterior midgut contraction assays were conducted on isolatedfifth instar anterior midguts maintained under saline. The prepara-tion consisted of an isolated anterior midgut secured with minutinpins through the oesophagus and posterior midgut to a Sylgard-coated dish. Contractions were monitored using an impedancemonitor (UFI, Morro Bay, California, USA), with electrodes placedeither side of the posterior portion of the anterior midgut. Contrac-tions were monitored on a chart recorder (Linear 1200). Tissueswere equilibrated in 50 ll of saline for 20 min, then the salinewas removed and replaced with either saline or test solutions con-taining serotonin or peptide. Frequency of contractions (contrac-tions/min, mean ± SE, n = 5–10) and percent increase over salinecontrols were calculated.

2.8. Statistics

The results are expressed as mean ± SE. Differences between themeans were determined using paired or unpaired (as appropriate)Students t tests. The data were considered statistically different atP < 0.05.

Fig. 2. The cAMP content of anterior midgut tissues incubated for 10 min in thepresence of IBMX with serotonin, Zoone-DH and Rhopr-DH31 is significantlyincreased over saline controls (Student’s t test, P < 0.05). Incubation of the anteriormidgut with the kinin peptide, LK1, does not increase the cAMP content of theanterior midgut. Mean ± SE (n = 5). Asterisk denotes significantly different fromsaline control.

3. Results

3.1. Phalloidin staining of the anterior midgut

Phalloidin binds to actin fibres in muscle, and the stainingshown in Fig. 1 indicates the pattern of muscle fibres over the fifthinstar anterior midgut. Both circular and longitudinal muscle fibresare observed, with the circular muscles overlain with the longitu-dinal muscle fibres. These fibres are evenly distributed over thebulk of the anterior midgut, but are more densely distributed atthe narrowed ends, anteriorly at the oesophagus and posteriorlyat the junction with the posterior midgut.

3.2. Cyclic AMP quantification

Cyclic AMP content of R. prolixus fifth instar anterior midgut issignificantly increased (Student’s t test, P < 0.05) in the presenceof serotonin, Zoone-DH, and Rhopr-DH31 when tested at 10�6 M(Fig. 2). The kinin, LK1, does not increase the cAMP content ofthe anterior midgut over saline controls.

3.3. Anterior midgut absorption assay: water transport

The anterior midgut absorption assay is a sensitive assay tomonitor the movement of water across the epithelium of the ante-rior midgut. In these assays, anterior midguts incubated with sero-

tonin increase their rate of absorption in a dose-dependent manner(Fig. 3A) with a threshold between 10�10 and 10�9 M and maximaat about 10�7 M. The rate of absorption is also significantly in-creased (Student’s t test, P < 0.05) by Zoone-DH (10�6 M) but notby Rhopr-DH31 or LK1 (Fig. 3B). The cyclic AMP analogues, dibutrylcAMP and 8-bromo cAMP (5 � 10�3 M), also significantly increasethe rate of absorption across the anterior midgut epithelium com-pared to saline controls (Fig. 3B).

Fig. 3. (A) Serotonin increases the rate of absorption across the epithelium of theanterior midgut in a dose-dependent manner. Mean ± SE (n = 5). (B) Serotonin,Zoone-DH and cAMP analogues (Db cAMP, 8-bromo cAMP) significantly increasethe rate of absorption across the anterior midgut compared to saline controls(Student’s t test, P < 0.05). Neither Rhopr-DH31 or LK1 significantly increase the rateof absorption. Mean ± SE (n = 5–10). Asterisk denotes significantly different fromsaline control.

108 V. Te Brugge et al. / General and Comparative Endocrinology 162 (2009) 105–112

3.4. Ussing chamber assay: ion transport

In order to explore the movement of ions across the epitheliumof the anterior midgut we have utilized a modified Ussing Cham-ber. In this assay, fifth instar anterior midguts are mounted inthe chamber as a flat sheet. The tissues are lumen negative andthe initial potential difference (PD) across the fifth instar anteriormidgut ranges from �4 to �35.8 mV with an average of�18.3 mV. Sample traces for these assays for each of the factorsare shown in Figs. 4–6. The means of measurements taken at thetime of maximum response were compared to each tissues salinecontrol using a paired Student’s t test, P < 0.05 (Fig. 7). Both seroto-nin (Figs. 4A–C and 7A–C) and Zoone-DH (Figs. 4D–F and 7A–C) in-crease the short circuit current (Isc, lA/cm2), transepithelialvoltage (Vte, mV), but decrease resistance (X cm2). The kinin-likepeptide, LK1, causes a small, but significant, decrease in the resis-tance and voltage across the epithelium of the anterior midgut.However, there is no significant change in the Isc (Figs. 5A–C and7A–C). Rhopr-DH31 does not significantly change the Isc, Vte orresistance (Figs. 5D–F and 7A–C). The cAMP analogue, 8-bromocAMP (Fig. 6A–C) increases Isc (38.6 ± 3.4 saline, 8-bromo cAMP121.9 ± 8.36 lA/cm2) and Vte (saline 7.9 ± 2.5; 8-bromo cAMP20.45 ± 3.5 mV) and decreases resistance (saline 206.3 ± 63.82; 8-bromo cAMP 175.4 ± 47.5 X cm2).

Both serotonin and Zoone-DH reach a similar peak value for Isc(Fig. 7A), however the serotonin response is slower (354.2 ± 22.8 sto reach the peak) relative to Zoone-DH (168.5 ± 8.0 s). The overallshape of the Isc curve is also variable for both serotonin and Zoone-DH; in some cases the curves have a distinct peak followed by adownward slope and a plateau similar to that seen for theZoone-DH example (Fig. 4D), while in others the curve reaches aplateau that is maintained (e.g. Fig. 4A) for the period of observa-tion (10 min).

3.5. Anterior midgut contraction assay

Contractions of the anterior midgut are complex, composed of aperistaltic wave (circular muscle contraction) which travels thelength of the anterior midgut, generally in an anterior to posteriormanner (although peristaltic waves initiated at the posterior endare also observed) and a longitudinal shortening of the anteriormidgut (longitudinal muscle contraction). Serotonin increases thefrequency of contractions of the anterior midgut in a dose-depen-dent manner (Fig. 8) with a threshold between 10�10 and 10�9 Mand maxima between 10�6 and 10�5 M. The peptides Zoone-DH,Rhopr-DH31, and LK1 significantly increase (paired Student’s t test,P < 0.05) the frequency of contractions of the anterior midgut oversaline controls, as does 8-bromo cAMP (Fig. 9).

4. Discussion

Post-feeding diuresis in R. prolixus is a highly coordinated eventin which the ingested blood is concentrated by the rapid move-ment of water and salt across the epithelium of the anterior midgutinto the surrounding haemolymph. In turn, the rapid activation ofthe Malpighian tubules results in the movement of water and saltfrom the haemolymph to the lumen of the upper Malpighian tu-bules and the reabsorption of K+ from the lower tubule. The urineempties into the hindgut. Contractions of the hindgut then lead toexpulsion of the urine through the anus. Failure to coordinate thesetissues would result in an ionic and volume inbalance in the hae-molymph. In this study, we have examined the activity of a varietyof diuretic factors on the anterior midgut; factors which have beenshown to have activity on the Malpighian tubules.

Structurally the anterior midgut is well suited for the efficientrapid movement of water and ions. It is a large ventricle composedof an epithelial layer, a single layer of cuboidal cells with a luminalbrush border and a highly folded basal-laminal membrane sur-rounded by a thick basal lamina (Billingsley, 1990), which is sur-rounded by bands of circular and longitudinal muscles revealedby Phalloidin staining. Maddrell (1964), has shown that contrac-tions of the anterior midgut increase within 30 s of the start offeeding. These contractions result in mixing of both the surround-ing haemolymph and of the ingested blood meal.

The diuretic factors tested in this study could reach the anteriormidgut in several ways, one of which is via the stomatogastric ner-vous system, which sends processes over the anterior portion ofthe anterior midgut. Several compounds have been shown to bepresent in the R. prolixus stomatogastric nervous system and tobe in the innervation to the anterior portion of the anterior midgut,including serotonin (Lange et al., 1988). In other insects, endocrinecells are found throughout the midgut (Zitnan et al., 1993; Veen-stra et al., 1995). Billingsley and Downe (1986) however, foundthat in R. prolixus, endocrine cells were only present in the poster-ior midgut (where the digestion of the blood meal occurs), and ourimmunohistochemistry studies confirm this (Te Brugge et al., 1999,2001; Tsang and Orchard, 1991).

The diuretic factors could also reach the anterior midgut as neu-rohormones. The release of serotonin from neurohaemal areas ele-

Fig. 4. Ussing chamber assays for serotonin (A–C) and Zoone-DH (D–F) at a final concentration 10�6 M. (A) Short circuit current (Isc) sample trace recorded in saline and thenin the presence of serotonin (filled bar). (B) Resistance sample trace in saline and in serotonin. (C) Voltage (Vte) sample trace in saline and serotonin. (D) Isc sample tracerecorded in saline and in the presence of Zoone-DH (filled bar). (E) Resistance sample trace in saline and in Zoone-DH. (F) Voltage (Vte) sample trace in saline and in Zoone-DH.

Fig. 5. Ussing chamber assays for LK1 (A–C) and Rhopr-DH31 (D–F) at a final concentration of 10�6 M. (A) Short circuit current (Isc) sample trace recorded in saline and then inthe presence of LK1(filled bar). (B) Resistance sample trace in saline and LK1. (C) Voltage (Vte) sample trace in saline and then LK1. (D) Isc sample trace recorded in saline andthen Rhopr-DH31 (filled bar). (E) Resistance sample trace in saline and Rhopr-DH31. (F) Voltage (Vte) sample trace in saline and Rhopr-DH31.

V. Te Brugge et al. / General and Comparative Endocrinology 162 (2009) 105–112 109

vates the serotonin titre in the haemolymph within minutes of thestart of feeding, with basal levels increasing from 0.03 lM to about0.1 lM in fifth instar and adult R. prolixus (Lange et al., 1989; Bar-rett et al., 1993). These levels drop over the next 20 min to lessthan 10 nM. The elevated levels of serotonin in the haemolymphare sufficient to stimulate activity of both the anterior midgut

(Farmer et al., 1981) and the Malpighian tubules (Maddrell et al.,1969, 1971). The distribution of CRF-like, DH31-like and kinin-likeimmunoreactivity (Te Brugge et al., 1999, 2001, 2005) in neurohae-mal areas, as well as evidence for release of these peptides into thehaemolymph (Te Brugge et al., 2002a,b, 2005), suggests that thesepeptides are also neurohormones. Interestingly, of the diuretic fac-

Fig. 6. Ussing chamber assay sample trace for 8-bromo cAMP (A–C) at a finalconcentration of 2 � 10�3 M. (A) Short circuit current (Isc) sample trace recorded insaline then 8-bromo cAMP (filled). (B) Resistance sample trace in saline then 8-bromo cAMP. (C) Voltage (Vte) sample trace in saline then 8-bromo cAMP.

Fig. 7. Summary of measurements taken at the time of maximum response tovarious factors in the Ussing chamber assays. (A) Short circuit current (Isc);Serotonin and Zoone-DH significantly increase the short circuit current across theepithelium of the anterior midgut over saline controls. (B) Resistance (Ohms cm2);Serotonin and LK1 significantly decrease the resistance across the anterior midgutepithelium compared to saline controls. (C) Voltage (Vte); Serotonin and Zoone-DHsignificantly increase the voltage while LK1 significantly decreases the voltage oversaline controls. Mean ± SE, n = 5–11, (paired Student’s t test, P < 0.05). Asteriskdenotes significantly different from saline control.

110 V. Te Brugge et al. / General and Comparative Endocrinology 162 (2009) 105–112

tors tested, serotonin is the only one that is also present in pro-cesses over the anterior portion of the anterior midgut, potentiallydelivered directly to the target tissue.

Previously serotonin has been shown to cause a dose-depen-dent increase in the cAMP content of the anterior midgut of R. pro-lixus (Barrett et al., 1993). In the present study serotonin, Zoone-DH and Rhopr-DH31, but not LK1, increased the cAMP content ofthe anterior midgut. The result for both serotonin and Zoone-DHis consistent with their ability to increase cAMP content in R. pro-lixus Malpighian tubules (Te Brugge et al., 1999) and implies thatcAMP may play a role as a second messenger in the anterior midgut(see later).

The absorption assay is a novel assay developed with the R. pro-lixus anterior midgut. In the present study, serotonin, Zoone-DHand cAMP analogues, significantly increased the rate of absorptionacross the anterior midgut. Over all this assay gave very similar re-sults to those of Farmer et al. (1981), using a very different tech-nique, for maximum rates of absorption, although the presenttechnique may be more sensitive, with the serotonin dose-re-sponse curve shifted to the left. Interestingly, while Rhopr-DH31

significantly increased the cAMP content of the anterior midgut tis-sues, this was not accompanied by a significant increase in the rateof absorption across the anterior midgut epithelium, suggestingthe possibility that the increase in cAMP induced by Rhopr-DH31

might be related to its effect on muscle contraction (see later).The kinin, LK1, like Rhopr-DH31, was also inactive in stimulating

absorption across the anterior midgut. The results obtained inthe present study on anterior midgut are consistent with those ob-tained for these factors when assayed on Malpighian tubule secre-tion (Te Brugge et al., 2002b, 2005) suggesting that the anteriormidgut and Malpighian tubules are coordinated in similar ways.

The absorption assay essentially monitors the movement ofwater across the anterior midgut epithelium. To monitor themovement of ions across gut tissues, Ussing type chambers havebeen used in a variety of insects including Manduca sexta, Hyalo-phora cecropia, Bombyx mori and Schistocerca gregaria (Chamberlinand King, 1998; Lee et al., 1998; Phillips et al., 1998; Wood andMoreton, 1978; Zeiske et al., 2002). Ion transport has also beenstudied in Aedes aegypti midgut using a variety of techniques (Clarket al., 1999, 2000; Onken et al., 2004a,b). In R. prolixus the move-ment of water across the epithelium of the anterior midgut is be-

Fig. 8. Serotonin causes a dose-dependent increase in the frequency of contractionsof the anterior midgut. Values are expressed as a percent increase over salinecontrols. Mean ± SE (n = 5).

Fig. 9. Anterior midgut contraction assay (frequency of contraction). All of thepeptides and the cAMP analogue, 8-bromo cAMP, increase the frequency ofcontractions of the anterior midgut over saline controls (Student’s t test, P < 0.05).Mean ± SE (n = 5–10). Asterisk denotes significantly different from saline control.

V. Te Brugge et al. / General and Comparative Endocrinology 162 (2009) 105–112 111

lieved to be driven by the active transport of Na+ ions across thebasal surface of the anterior midgut (Farmer et al., 1981; Barrett,1982) by a Na+/K+ ATPase. The movement of Cl� and water arethought to be passive.

Previously, both serotonin and cAMP were shown to increasethe transport of NaCl across the anterior midgut (Farmer et al.,1981). In the present study, Zoone-DH, as well as serotonin and8-bromo cAMP, increased the Isc and Vte, and decreased the resis-tance across the anterior midgut. The traces obtained in the pres-ent study for serotonin and Zoone-DH are similar to thoseobtained for serotonin by Farmer et al. (1981). Rhopr-DH31 showedno activity on the current, voltage or resistance. Interestingly, LK1caused no significant change in current but did cause a small, butsignificant, change in the resistance and voltage across the anteriormidgut. The nature of this change will require further detailedstudy using the native kinin peptides, but the data suggest the pos-sibility of a change in passive or paracellular pathways in responseto kinin peptide. In A. aegypti tubules it has been shown that theaddition of leucokinin-VIII led to the immediate collapse of theVte and a 6-fold decrease in resistance (Pannabecker et al.,

1993), changing a moderately ‘tight’ epithelium to a ‘leaky’ epithe-lium. Whether this is a physiological effect in R. prolixus remains tobe seen.

The absorption and Ussing chamber experiments monitor the ef-fects of diuretic factors on epithelial transport. After the start of feed-ing, however, there is also an increase in the frequency of anteriormidgut contractions (Maddrell, 1964). This increase in the rate ofcontractions contributes to the mixing of ingested blood inside theanterior midgut and haemolymph outside the anterior midgut, thusminimizing unstirred layers and facilitating ion transport, while atthe same time increasing circulation of hormones within the haemo-lymph. This is important for efficient transport across the epitheliumof the anterior midgut and the Malpighian tubules.R. prolixus Malpi-ghian tubules apparently have no muscle fibres associated withthem, hence the tubules are not capable of independent movement.The reduction of the unstirred layers in vivo is potentially due to con-tractions of the anterior midgut and hindgut which would mix thehaemolymph (and gut contents) as well as move the associated Mal-pighian tubules through the haemolymph. Previously, we haveshown that some diuretic factors have myotropic activity on thehindgut (Te Brugge et al., 2002a,b, 2005) and salivary glands (Orch-ard and Te Brugge, 2002). In other insects, both CRF-like peptidesand serotonin have been shown to increase the frequency of contrac-tion of gut (Blake et al., 1996; Onken et al., 2004a). In this study, wedemonstrate that serotonin and all of the peptides tested also in-crease the frequency of contractions of the anterior midgut, and thatcAMP analogues mimic this effect. However, LK1 does not stimulatean increase in cAMP and so its mode of action in increasing fre-quency of contraction must be via a separate pathway.

Looking at the data combined, a model for coordinating anteriormidgut and Malpighian tubules is suggested. Serotonin and Zoone-DH are active on R. prolixus anterior midgut, elevating cAMP con-tent (present study; Barrett et al., 1993), increasing frequency ofcontractions (present study), and increasing ion and water move-ment (present study; Farmer et al., 1981). Similarly both serotoninand Zoone-DH act on the upper Malpighian tubule to increasecAMP content and rate of secretion. The anterior midgut and upperMalpighian tubules are coordinated by both of these diuretic hor-mones. Interestingly, however, the lower portion of the Malpighiantubules, where the reabsorption of K+ occurs, is stimulated by sero-tonin, but not by Zoone-DH (Donini et al., 2008).

The calcitonin-like peptide, Rhopr-DH31, produces a small butsignificant increase in the cAMP content of the anterior midgut,but has no effect in the absorption assays or in the Ussing chamberassays, suggesting that this peptide does not play a significant roleon its own in the movement of water and salts across the anteriormidgut epithelium. Similarly, this peptide has only a very small ef-fect on Malpighian tubule secretion (Te Brugge et al., 2005, 2008).Rhopr-DH31 does, however significantly increase the cAMP contentand frequency of contractions of the anterior midgut. The kinin-like peptide, LK1, increases the frequency of anterior midgut con-tractions, but it does so without an increase in cAMP content. Thispeptide also has no effect on the rate of absorption across the ante-rior midgut, a result that is consistent with the findings that it doesnot stimulate Malpighian tubule secretion (Te Brugge et al., 2001).Interestingly, although LK1 does not stimulate fluid absorption/secretion in anterior midgut or Malpighian tubules, it does de-crease the resistance and voltage across the anterior midgut. Thephysiological significance of this result is as yet unknown.

Zoone-DH, LK1 and Rhopr-DH31, along with serotonin, haveactivity on several tissues that are associated with feeding andwith post-feeding diuresis, including salivary glands, anterior mid-gut, Malpighian tubules and hindgut. Their activities on both ante-rior midgut and Malpighian tubules illustrate a fine tuning andcoordinated control over the movement of ions and water acrossthese two epithelia, along with their other physiological roles.

112 V. Te Brugge et al. / General and Comparative Endocrinology 162 (2009) 105–112

Acknowledgments

The authors thank Roger Avery for technical assistance withsome of the contraction assays. This work was supported by theNatural Sciences and Engineering Research Council of Canada.

References

Barrett, F.M., 1982. Absorption of fluid from the anterior midgut in Rhodnius. J.Insect Physiol. 28, 335–341.

Barrett, F.M., Orchard, I., 1990. Serotonin-induced elevation if cAMP levels in theepidermis of the blood-sucking bug, Rhodnius prolixus. J. Insect Physiol. 36, 625–633.

Barrett, F.M., Te Brugge, V.A., Orchard, I., 1993. Characteristics of serotonin-inducedcyclic AMP elevation in the integument and anterior midgut of the blood-feeding bug, Rhodnius prolixus. J. Insect Physiol. 30, 581–587.

Billingsley, P.F., 1990. The midgut ultrastructure of hematophagous insects. Annu.Rev. Entomol. 35, 219–248.

Billingsley, P.F., Downe, A.E.R., 1986. Non-digestive cell types in the midgutepithelium of Rhodnius prolixus (Hemiptera: Reduviidae). J. Med. Entomol. 23,212–216.

Blake, P.D., Kay, I., Coast, G.M., 1996. Myotropic activity of Acheta diuretic peptideon the foregut of the house cricket, Acheta domesticus (L.). J. Insect Physiol. 42,1053–1059.

Chamberlin, M.E., King, M.E., 1998. Changes in midgut active ion transport andmetabolism during the fifth instar of the tobacco hornworm (Manduca sexta). J.Exp. Zool. A: Comp. Exp. Biol. 280, 135–141.

Chiang, R.G., Chiang, J.A., Davey, K.G., 1992. A sensory input inhibiting heart rate inan insect, Rhodnius prolixus. Experientia 48, 1122–1125.

Clark, T.M., Koch, A., Moffett, D.F., 1999. The anterior and posterior stomach regionsof larval Aedes aegypti midgut: regulation of ion transport and stimulation by 5-hydroxytryptamine. J. Exp. Biol. 202, 247–252.

Clark, T.M., Koch, A., Moffett, D.F., 2000. The electrical properties of the anteriorstomach of the larval mosquito (Aedes aegypti). J. Exp. Biol. 203, 1093–1101.

Coast, G.M., Garside, C.S., Webster, S.G., Schegg, K.M., Schooley, D.A., 2005. Mosquitonatriuretic peptide identified as a calcitonin-like diuretic hormone in Anophelesgambiae (Giles). J. Exp. Biol. 208, 3281–3291.

Coast, G.M., 1998. The influence of neuropeptides on Malpighian tubule writhingand its significance for excretion. Peptides 19, 469–480.

Collier, K.A., O’Donnell, M.J., 1997. Analysis of epithelial transport by measurementsof K+, Cl� and pH gradients in extracellular unstirred layers: ion secretion andreabsorption by Malpighian tubules of Rhodnius prolixus. J. Exp. Biol. 200, 1627–1638.

Donini, A., O’Donnell, M.J., Orchard, I., 2008. Differential actions of diuretic factorson the Malpighian tubules of Rhodnius prolixus. J. Exp. Biol. 211, 42–48.

Farmer, J.S., Maddrell, S.H.P., Spring, J.H., 1981. Absorption of fluid by the midgut ofRhodnius. J. Exp. Biol. 94, 301–316.

Furuya, K., Milchak, R.J., Schegg, K.M., Zhang, J., Tobe, S.S., Coast, G.M., Schooley, D.A.,2000. Cockroach diuretic hormones: characterization of a calcitonin-likepeptide in insects. Proc. Natl. Acad. Sci. USA 97, 6469–6474.

Lange, A.B., Orchard, I., 1986. Identified octopaminergic neurons modulatecontractions of locust visceral muscle via adenosine 30 ,50-monophosphate(cyclic AMP). Brain Res. 363, 340–349.

Lange, A.B., Orchard, I., Barrett, F.M., 1989. Changes in haemolymph serotonin levelsassociated with feeding in the blood-feeding bug Rhodnius prolixus. J. InsectPhysiol. 35, 393–399.

Lange, A.B., Orchard, I., Lloyd, R.J., 1988. Immunohistochemical and electrochemicaldetection of serotonin in the nervous system of the blood-sucking bug, Rhodniusprolixus. Arch. Insect Biochem. Physiol. 8, 187–201.

Lee, K.-Y., Horodyski, F., Chamberlin, M.E., 1998. Inhibition of midgut ion transportby allatotropin (Mas-AT) and Manduca FLRFamides in the tobacco hornwormManduca sexta. J. Exp. Biol. 201, 3067–3074.

Maddrell, S.H.P., 1964. Excretion in the blood-sucking bug, Rhodnius prolixus Stål. llThe normal course of diuresis and the effect of temperature. J. Exp. Biol. 41,163–176.

Maddrell, S.H.P., Pilcher, D.E.M., Gardiner, B.O.C., 1971. Pharmacology of theMalpighian tubules of Rhodnius and Carausius: the structure-activityrelationship of tryptamine analogues and the role of cyclic AMP. J. Exp. Biol.54, 779–804.

Maddrell, S.H.P., Pilcher, D.E.M., Gardiner, B.O.C., 1969. Stimulatory effect of 5-Hydroxytryptamine (serotonin) on the secretion by Malpighian tubules ofinsects. Nature 222, 784–785.

Miggiani, L., Orchard, I., Te Brugge, V., 1999. The distribution and function ofserotonin in the large milkweed bug, Oncopeltus fasciatus: A comparative studywith the blood feeding bug, Rhodnius prolixus. J. Insect Physiol. 45, 1029–1036.

Onken, H., Moffett, S.B., Moffett, D.F., 2004a. The anterior stomach of larvalmosquitoes (Aedes aegypti): effects of neuropeptides on transepithelial iontransport and muscular motility. J Exp. Biol. 207, 3731–3739.

Onken, H., Moffett, S.B., Moffett, D.F., 2004b. The transepithelial voltage of theisolated anterior stomach of mosquito larvae (Aedes aegypti): pharmacologicalcharacterization of the serotonin-stimulated cells. J. Exp. Biol. 207, 1779–1787.

Orchard, I., 2006. Serotonin: a coordinator of feeding-related physiological events inthe blood-gorging bug, Rhodnius prolixus. Comp. Biochem. Physiol. A Mol. Integr.Physiol. 144, 316–324.

Orchard, I., Lange, A.B., Barrett, F.M., 1988. Serotonergic supply to the epidermis ofRhodnius prolixus: evidence for serotonin as the plasticising factor. J. InsectPhysiol. 34, 873–879.

Orchard, I., Te Brugge, V.A., 2002. Contractions associated with the salivary glands ofthe blood-feeding bug, Rhodnius prolixus: evidence for both a neural andneurohormonal coordination. Peptides 23, 693–700.

Pannabecker, T.L., Hayes, T.K., Beyenbach, K.W., 1993. Regulation of epithelial shuntconductance by the peptide leucokinin. J. Mem. Biol. 132, 63–76.

Phillips, J.E., Wiens, C., Audsley, N., Jeffs, L., Bilgen, T., Meredith, J., 1998. Nature andcontrol of chloride transport in insect absorptive epithelia. J. Exp. Zool. A: Comp.Exp. Biol. 275, 292–299.

Reynolds, S.E., 1974. Pharmacological induction of plasticization in the abdominalcuticle of Rhodnius. J. Exp. Biol. 61, 705–718.

Te Brugge, V.A., Miksys, S.M., Coast, G.M., Schooley, D.A., Orchard, I., 1999. Thedistribution of a CRF-like diuretic peptide in the blood-feeding bug, Rhodniusprolixus. J. Exp. Biol. 202, 2017–2027.

Te Brugge, V.A., Nässel, D.R., Coast, G.M., Schooley, D.A., Orchard, I., 2001. Thedistribution of a kinin-like peptide and its co-localization with a CRF-likepeptide in the blood-feeding bug, Rhodnius prolixus. Peptides 22, 161–173.

Te Brugge, V.A., Orchard, I., 2002a. Evidence for CRF-like and kinin-like peptides asneurohormones in the blood-feeding bug, Rhodnius prolixus. Peptides 23, 1967–1979.

Te Brugge, V.A., Schooley, D.A., Orchard, I., 2002b. The biological activity of diureticfactors in Rhodnius prolixus. Peptides 23, 671–681.

Te Brugge, V.A., Lombardi, V.C., Schooley, D.A., Orchard, I., 2005. Presence andactivity of a Dippu-DH31-like peptide in the blood-feeding bug, Rhodniusprolixus. Peptides 26, 29–42.

Te Brugge, V.A., Schooley, D.A., Orchard, I., 2008. Amino acid sequence andbiological activity of a calcitonin-like diuretic hormone (DH31) from Rhodniusprolixus. J. Exp. Biol. 211, 382–390.

Tsang, P.W., Orchard, I., 1991. Distribution of FMRFamide-related peptides in theblood-feeding bug, Rhodnius prolixus. J. Comp. Neurol. 311, 17–32.

Veenstra, J.A., Lau, G.W., Agricola, H.J., Pezel, D.H., 1995. Immunohistologicallocalisation of regulatory peptides in the midgut of the female mosquito Aedesaegypti. Histochem. Cell Biol. 104, 337–347.

Wood, J.L., Moreton, R.B., 1978. Refinements in the short-circuit technique and itsapplication to active potassium transport across the Cecropia midgut. J. Exp. Biol.77, 123–146.

Zeiske, W., Meyer, H., Wieczorek, H., 2002. Insect midgut K+ secretion: concertedrun-down of apical/basolateral transporters with extra-/intracellular acidity. J.Exp. Biol. 205, 463–474.

Zitnan, D., Šauman, I., Sehnal, F., 1993. Peptidergic innervation and endocrine cellsof insect midgut. Arch. Insect Biochem. Physiol. 22, 113–132.