Post-Feeding Physiology in Rhodnius Prolixus: the Possible Roles of Calcitonin-Like Diuretic Hormone (Or DH31) and Fglamide-Related Allatostatins

Post-feeding physiology in Rhodnius prolixus: The possible roles of calcitonin-like diuretic hormone (or DH31) and FGLamide-related allatostatins

Meet Zandawala

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Department of Cell and Systems Biology University of Toronto

roles of calcitonin-like diuretic hormone (or DH31) and FGLamide-related allatostatins

Meet Zandawala

Doctor of Philosophy

Department of Cell and Systems Biology University of Toronto

2014 Abstract

Unfed Rhodnius prolixus (Order: Hemiptera) are in a state of arrested development. Blood- gorging results in short-term physiological/endocrinological changes that are essential to the survival of the insect, and also triggers several crucial developmental processes such as molting, growth and reproduction. Two neuropeptides that potentially play a role in this post-feeding physiology are FGLamide-related allatostatins (FGLa/ASTs) and calcitonin-like diuretic hormone (CT/DH). In order to determine the roles of these neuropeptides, I cloned and characterized their cDNA sequences and those of their cognate G protein-coupled receptors. I showed that both these neuropeptides are predominantly expressed in the central nervous system and are capable of being released into the haemolymph as hormones following feeding. I have also characterized two functionally different receptors for Rhopr-CT/DH, for the first time in any insect, and one receptor for Rhopr-FGLa/ASTs. Spatial expression analyses of these receptor transcripts revealed a wide distribution of the receptors in tissues associated with post-feeding physiology. Examination of their physiological effects using various biological assays revealed that these two neuropeptides may play indirect roles, but not direct roles during the rapid post-

ii feeding diuresis and also may be involved with long term physiological changes such as metamorphosis and egg production.

iii

Acknowledgements

I would like to begin by thanking my supervisor, Dr. Ian Orchard, for allowing me to carry out research in his laboratory, starting as an inexperienced second-year undergraduate. I am grateful and fortunate to have had such a wonderful and supportive mentor like you for the past 8 years.

Thank you for always having faith in me and allowing me to pursue all the projects that I desired. I gladly move on to the next phase in my career knowing that I have learnt from the very best.

I am also thankful to my immediate mentor and role-model, Dr. Jean-Paul Paluzzi, for training and guidance. Your selfless and continuing support throughout the course of my research has been invaluable. I would also like to thank my committee members, Dr. Angela Lange and Dr.

Tim Westwood, for their advice and support. A special thanks to Dr. Lange for giving me the encouragement to always go a step further. I am very appreciative of all your input.

Sincere thanks to all my past and present lab mates, colleagues and friend at UTM who have helped me achieve my goals and made my time memorable. I couldn’t have asked for a better work environment and colleagues. You all know who you are!

Words cannot express how thankful I am to have such a supportive family. To my parents, thank you for all the sacrifices you have had to make to support my education. I wouldn’t be here if it wasn’t for your bravery to immigrate to Canada for the sake of my education. Dad, thank you for trying to learn about my research. Mom, even though it is hard for you to understand what it is that I do, I appreciate all your support and the wonderful meals you prepare when I come home from work.

Finally, I would like to thank my best friend and confidant, my girlfriend, Aruna. This thesis would not have been possible without your undying love and support. Thank you for always being there for me. You always knew how to lift me up when I had been down countless times due to failed experiments. Thank you for respecting my dedication to work and believing in my goals, more than myself. I am lucky to have you in my life!

Table of Contents Abstract ...... ii Acknowledgments ...... iv Table of Contents ...... vi Organization of the Thesis ...... x List of Figure and Tables ...... xii List of Appendices ...... xvi List of Abbreviations ...... xvii

Chapter 1: General Introduction ...... 1 Rhodnius prolixus ...... 2 Post-feeding physiology ...... 2 Neuropeptides ...... 4 Diuretic and anti-diuretic hormones ...... 6 Calcitonin-like diuretic hormone ...... 7 FGLamide-related allatostatins (FGLa/ASTs) ...... 9 G protein-coupled receptors (GPCRs) ...... 10 Objectives and organization of thesis ………………...... 11 References ...... 14

Chapter 2: Isolation and characterization of the cDNA encoding DH31 in the kissing bug, Rhodnius prolixus ...... 18 Abstract ...... 19 Introduction ...... 20 Material and methods ...... 24 Results ...... 31 Discussion ...... 53 References ...... 58 Acknowledgements ...... 63 Appendices ...... 64

Chapter 3: Isolation and functional characterization of calcitonin-like diuretic hormone receptors in Rhodnius prolixus ...... 68 Abstract ...... 69 Introduction ...... 70 Material and methods ...... 72 Results ...... 77 Discussion ...... 97 References ...... 101 Acknowledgements ...... 105 Appendices ...... 106

Chapter 4: Structure-activity relationships of two Rhodnius prolixus calcitonin-like diuretic hormone analogs ...... 118 Abstract ...... 119 Introduction ...... 120 Material and methods ...... 121 Results ...... 123 Discussion ...... 129 References ...... 131 Acknowledgements ...... 133

Chapter 5: Cloning of the cDNA, localization, and physiological effects of FGLamide- related allatostatins in the blood-gorging bug, Rhodnius prolixus...... 134 Abstract ...... 135 Introduction ...... 136 Material and methods ...... 138 Results ...... 144 Discussion ...... 173 References ...... 179 Acknowledgements ...... 185 Appendices ...... 186

vii

Chapter 6: Post-feeding physiology in Rhodnius prolixus: the possible role of FGLamide- related allatostatins ...... 193 Abstract ...... 194 Introduction ...... 195 Material and methods ...... 196 Results ...... 200 Discussion ...... 219 References ...... 223 Acknowledgements ...... 226

Chapter 7: Isolation and functional characterization of FGLamide-related allatostatin receptor in Rhodnius prolixus ...... 227 Abstract ...... 228 Introduction ...... 229 Material and methods ...... 230 Results ...... 233 Discussion ...... 250 References ...... 254 Acknowledgements ...... 258

Chapter 8: General Discussion ...... 259 Major conclusions ...... 260 Rhopr-CT/DH and its receptors ...... 260 FGLa/ASTs and their putative receptor ...... 268 Integrating the whole ...... 271 Future directions ...... 276 References ...... 279

Appendix A: Calcitonin-like diuretic hormone in insects ...... 282 Abstract ...... 283 Introduction ...... 284 Discovery ...... 285 mRNA and prepropeptide structures ...... 288 viii

Distribution ...... 295 Biological activity ...... 301 Receptors ...... 307 Conclusion and future directions ...... 313 References ...... 315 Acknowledgements ...... 323

Organization of the Thesis Chapter 1 provides a general introduction of my research topic.

Unless stated otherwise, I performed all the experiments and data analysis. Dr. Orchard provided invaluable input in the form of suggestions and comments for each manuscript. Copyright permission, if required, was granted from each of the publishers to reprint Chapters 1-6 and Appendix A.

Chapter 2 was published in Molecular and Cellular Endocrinology (Zandawala, M., Paluzzi, J.P., and Orchard, I. (2011) Mol. Cell. Endocrinol. 331(1): 79-88; doi: 10.1016/j.mce.2010.08.012). Dr. Paluzzi mentored me and taught me all the molecular techniques utilized in the manuscript.

Chapter 3 was published in PLoS One (Zandawala, M., Li, S., Hauser, F., Grimmelikhuijzen, C., and Orchard, I. (2013) PLoS One 8: e82466; doi: 10.1371/journal.pone.0082466). Dr Li taught me cell culture techniques and the functional receptor assay. Dr. Hauser and Dr. Grimmelikhuijzen provided the research space and reagents for the receptor-ligand functional analysis.

Chapter 4 will be published in Peptides (Zandawala, M., Poulos, C., and Orchard, I. (In press) doi: 10.1016/j.peptides.2014.03.019). Dr. Poulos synthesized the analogs that were tested in the assays.

Chapter 5 was published in Insect Biochemistry and Molecular Biology (Zandawala, M., Lytvyn, Y., Taiakina, D., and Orchard, I. (2012) Insect Biochem. Mol. Biol. 42(1): 10-21; doi: 10.1016/j.ibmb.2011.10.002). Yulia Lytvyn cloned the cDNA sequence and performed northern blot hybridization. Daria Taiakina performed the muscle contraction assays.

Chapter 6 was published in General and Comparative Endocrinology (Zandawala, M., and Orchard, I. (2013) Gen. Comp. Endocrinol. 194(1): 311-317; doi: 10.1016/j.ygcen.2013.10.005). Dr. Orchard performed the neurophysiology.

Chapter 7 will be submitted for publication in Molecular and Cellular Endocrinology.

Chapter 8 summarizes all the research chapters and provides a general discussion of the thesis.

Appendix A was published in Insect Biochemistry and Molecular Biology (Zandawala, M. (2012) Insect Biochem. Mol. Biol. 42(10): 816-825; doi: 10.1016/j.ibmb.2012.06.006).

List of Figures and Tables

Chapter 2: Isolation and characterization of the cDNA encoding DH31 in the kissing bug, Rhodnius prolixus ...... 18

Table 1. Structures of mature DH31 peptides (deduced or sequenced) from 16 species ...... 23

Figure 1. Northern blot analysis of Rhopr-DH31 in R. prolixus fifth-instar CNS ...... 34

Figure 2. Rhopr-DH31 sequence and structure of Rhopr-DH31 splice variants ...... 36

Figure 3. Multiple sequence alignment of the DH31 prepropeptide and its corresponding phylogenetic tree ...... 40

Figure 4. RT-PCR of Rhopr-DH31 in fifth-instar R. prolixus ...... 43

Figure 5. Relative expression of Rhopr-DH31 variants in fifth-instar CNS ...... 47

Figure 6. Fluorescence in situ hybridization (FISH) showing Rhopr-DH31 expression in CNS …. ……………………………………………………………………………………………………49

Figure 7. Diagrammatic representation of Rhopr-DH31 transcripts expression ...... 51

Table S1. Forward and reverse primers used to obtain the partial Rhopr-DH31 cDNA sequence ...... 64 Table S2. Forward and reverse primers used to perform the 5’ PCR RACE reactions ...... 65 Table S3. Forward and reverse primers used to perform the 3’ PCR RACE reactions ...... 66 Table S4. Forward and reverse primers used to obtain the 3’ UTR sequence ...... 67

Chapter 3: Isolation and functional characterization of calcitonin-like diuretic hormone receptors in Rhodnius prolixus ...... 68 Figure 1. cDNA sequences and the deduced amino acid sequences of Rhopr-CT/DH-Rs ...... 80 Figure 2. Rhopr-CT/DH-R1 and Rhopr-CT/DH-R2 splicing ...... 82 Figure 3. Functional assay of R. prolixus CT/DH receptor isoforms transiently expressed in HEK293/CNG cell lines ...... 85 Figure 4. Multiple sequence alignment of select insect CT/DH receptors ...... 88 Figure 5. A cladogram of family B1 GPCRs ...... 90 Figure 6. Spatial expression analysis of Rhopr-CT/DH-Rs in fifth instar R. prolixus ...... 93 Figure 7. Spatial expression analysis of Rhopr-CT/DH-Rs in R. prolixus adult reproductive tissues ...... 95 Table S1. Primers used to amplify the partial cDNA sequence for Rhopr-CT/DH-R1 and Rhopr- CT/DH-R2 ...... 106

xii

Table S2. Primers used to perform 5’ RACE PCR reactions ...... 107 Table S3. Primers used to perform 3’ RACE PCR reactions ...... 108 Table S4. Primers used to amplify the largest cDNA fragments ...... 109 Table S5. Primers used to amplify full ORF and introduce Kozak sequence ...... 110 Table S6. Primers used for qPCR reactions ...... 111 Figure S1. Rhopr-CT/DH-R1-A cDNA sequence and the deduced amino acid sequence ...... 112 Figure S2. Rhopr-CT/DH-R2-A cDNA sequence and the deduced amino acid sequence ...... 114 Figure S3. Kinetics of the bioluminescence responses of HEK/CNG and CHO/G16 cells expressing Rhopr-CT/DH-R1-B ...... 116

Chapter 4: Structure-activity relationships of two Rhodnius prolixus calcitonin-like diuretic hormone analogs ...... 118 Figure 1. Effect of Rhopr-CT/DH and its analogs on frequency of R. prolixus hindgut contractions ...... 125 Figure 2. Functional assay of R. prolixus CT/DH receptor isoforms expressed in CHO/G16 cells ...... 127

Chapter 5: Cloning of the cDNA, localization, and physiological effects of FGLamide- related allatostatins in the blood-gorging bug, Rhodnius prolixus ...... 134 Table 1. Rhopr-FGLa/AST predicted peptides ...... 147 Figure 1. Rhopr-FGLa/AST sequence and structure ...... 148 Figure 2. Multiple sequence alignment of the Rhopr-FGLa/AST prepropeptide from different insects ...... 150 Figure 3. An unrooted phylogram showing evolutionary relationships of FGLa/AST prepropeptides ...... 152 Figure 4. Expression analyses of Rhopr-FGLa/AST in fifth-instar R. prolixus ...... 156 Figure 5. Fluorescent in situ hybridization portraying Rhopr-FGLa/AST expression in fifth-instar R. prolixus dorsal CNS ...... 158 Figure 6. Diagrammatic representation of Rhopr-FGLa/AST expression in R. prolixus CNS ..160 Figure 7. Sample traces showing the effects of Rhopr-FGLa/AST-2 on spontaneous contractions of R. prolixus anterior midgut ...... 163 Figure 8. Dose-response curves for the effects of Rhopr-FGLa/ASTs on the frequency of anterior midgut contractions ...... 165 xiii

Figure 9. Sample traces showing the effects of Rhopr-FGLa/AST-3 on leucokinin 1 (LK1)- induced contractions of R. prolixus hindgut ...... 167 Figure 10. Dose-response curves for the effects of Rhopr-FGLa/ASTs on the amplitude of LK1- induced contractions of R. prolixus hindgut ...... 169 Figure 11. Dose-response curves for the effects of Rhopr-FGLa/ASTs on the frequency of LK1- induced phasic contractions of R. prolixus hindgut ...... 171 Table S1. Primers used to sequence Rhopr-FGLa/AST ORF ...... 186 Table S2. Primers used to perform 3’ RACE reactions ...... 187 Table S3. Primers used to perform 5’ RACE reactions ...... 188 Table S4. Primers used to amplify the complete Rhopr-FGLa/AST cDNA sequence ...... 189 Table S5. Primers used to amplify Rhopr-β-actin (housekeeping gene) in RT-PCR reactions …... …………………………………………………………………………………………………..190 Table S6. Primers used to synthesize RNA probe for fluorescent in situ hybridization ...... 191 Table S7. Summary of FGLa/AST prepropeptides used for phylogenetic analysis ...... 192

Chapter 6: Post-feeding physiology in Rhodnius prolixus: the possible role of FGLamide- related allatostatins ...... 193 Figure 1. FGLa/AST-like and serotonin-like double-label immunohistochemistry ...... 201 Figure 2. Immunohistochemistry, Lucifer yellow injection and physiology of a DUM neuron ..... …………………………………………………………………………………………………..204 Figure 3. FGLa/AST-like immunoreactivity in the DUM neurons in the posterior MTGM of unfed and fed fifth-instar R. prolixus ...... 206 Figure 4. FGLa/AST-like immunoreactivity in the abdominal nerves of unfed and fed fifth-instar R. prolixus ...... 208 Figure 5. FGLa/AST-like immunoreactivity in the gut of unfed and fed fifth-instar R. prolixus ... …………………………………………………………………………………………………..210 Figure 6. Effects of Rhopr-FGLa/ASTs on serotonin-stimulated anterior midgut absorption ..213 Figure 7. Effects of Rhopr-FGLa/ASTs on serotonin-stimulated Malpighian tubule secretion ...... …………………………………………………………………………………………………..215 Figure 8. Effects of Rhopr-FGLa/ASTs on Rhopr-kinin-2 (RK2)-induced contractions of R. prolixus hindgut ...... 217

xiv

Chapter 7: Isolation and functional characterization of FGLamide-related allatostatin receptor in Rhodnius prolixus...... 227 Figure 1. Rhopr-FGLa/AST-R cDNA sequence and the deduced amino acid sequence ...... 234 Figure 2. Rhopr-FGLa/AST-R gene structure ...... 236 Figure 3. Multiple sequence alignment of select insect FGLa/AST receptors ...... 239 Figure 4. A phylogram of FGLa/AST and galanin GPCRs ...... 241 Figure 5. Spatial expression analysis of Rhopr-FGLa/AST-R in fifth instar R. prolixus ...... 244 Figure 6. Spatial expression analysis of Rhopr-FGLa/AST-R in R. prolixus adult reproductive tissues ...... 246 Figure 7. Spatial expression analysis of Rhopr-FGLa/AST-R in R. prolixus corpora cardiaca/corpora allata complex ...... 248

Chapter 8: General Discussion……………………………………………………………….259 Figure 1. Functional assay to investigate any possible interaction between Rhopr-CT/DH-R1 and Rhopr-CT/DH-R2 ...... 265 Figure 2. Schematic of the R. prolixus alimentary canal summarizing the physiological roles played by various diuretic and anti-diuretic hormones ...... 272 Figure 3. A model summarizing the potential physiological roles of Rhopr-FGLa/ASTs and Rhopr-CT/DH ...... 274

Appendix A: Calcitonin-like diuretic hormone in insects ………………………………….282 Table S1. Structures of mature CT/DHs (deduced or sequenced) ...... 287 Table S2. Summary of CT/DH transcripts in insects ...... 290 Table S3. Summary of CT/DH effects on Malpighian tubule fluid secretion ...... 304 Table S4. Studies examining synergistic or additive effects between CT/DHs and other diuretic factors on Malpighian tubule secretion ...... 305 Figure S1. Molecular organization of insect CT/DHs based on BLAST analysis and gene structure prediction ...... 291 Figure S2. CT/DH prepropeptides in insects ...... 293 Figure S3. Rhopr-CT/DH expression pattern determined using fluorescent in situ hybridization in fifth instar R. prolixus CNS ...... 297 Figure S4. Phylogenetic tree of insect GPCRs, belonging to family B ...... 310

List of Appendices Appendix A: Calcitonin-like diuretic hormone in insects ……………………………….…282

xvi

List of Abbreviations Note: Most of the abbreviations utilized in this thesis are included in this list.

5-HT – 5-hydroxytryptamine (or serotonin) ADH – anti-diuretic hormone ANOVA – Analysis of variance ASTs – Allatostatins ATP – Adenosine triphosphate BLAST – Basic Local Alignment Search Tool BSA – Bovine serum albumin CA – Corpus allatum cAMP – Adenosine 3’,5’-cyclic monophosphate CC – Corpus cardiacum cGMP – Guanosine 3’,5’-cyclic monophosphate CGRP – Calcitonin gene-related peptide CHO – Chinese hamster ovary CLR or CRLR – Calcitonin receptor-like receptor CNG – Cyclic nucleotide-gated channel CNS – Central nervous system CRF/DH – Corticotropin releasing-factor-related diuretic hormone CRF/DH-R – Corticotropin releasing-factor-related diuretic hormone receptor CT – Calcitonin CTR – Calcitonin receptor CT/DH or CLDH – Calcitonin-like diuretic hormone CT/DH-R – Calcitonin-like diuretic hormone receptor DH – Diuretic hormone

DH31 – Diuretic hormone 31 DIG - Digoxigenin DMEM/F12 – Dulbecco’s modified eagle medium nutrient mixture F12-Ham DUM – Dorsal unpaired median ETHR – Ecdysis-triggering hormone receptor FBS – Fetal bovine serum

xvii

FGLa/AST – FGLamide-related allatostatins FGLa/AST-R – FGLa/AST receptor FISH – Fluorescent in situ hybridization GPCR – G protein-coupled receptor HEK293 – Human embryonic kidney 293 JH – Juvenile hormone LK1 – Leucokinin 1 MALDI-TOF – Matrix assisted laser desorption ionization time-of-flight MIP – Myoinhibitory peptide MNSC – Medial neurosecretory cells MT – Malpighian tubule MTGM – Mesothoracic ganglionic mass NCCII – Nervi corpori cardiaci II NGS – Normal goat serum NJ – Neighbor-joining ORF – Open reading frame PBS – Phosphate-buffered saline PBT – PBS with 0.1% Tween-20 PBTB – PBT containing 1% (w/v) blocking reagent PCR – Polymerase chain reaction PDF – Pigment-dispersing factor PDFR – Pigment-dispersing factor receptor PG – Prothoracic gland (including associated fat body) PISCF/AST – PISCF-related allatostatin PRO – Prothoracic ganglion PTHR – Parathyroid hormone receptor PTTH – Prothoracicotropic hormone PVC – Posterior ventral cells qPCR – Quantitative PCR RACE – Rapid amplification of cDNA ends RAMP – Receptor activity modifying proteins RCP – Receptor component protein RIA – Radioimmunoassay xviii

RNAi – RNA interference RT-PCR – Reverse transcriptase polymerase chain reaction SG – Salivary gland SOG – Suboesophageal ganglion UTR – Untranslated region

xix

Chapter 1: General Introduction

Rhodnius prolixus

Rhodnius prolixus (common name: the kissing bug) is a member of the Reduviidae family

(Order: Hemiptera) and is native to Central and South America. This hemimetabolous insect has five nymphal instars and an adult stage; all the developmental stages and both sexes of R. prolixus are obligate blood feeders. Unfed R. prolixus are arrested in their development, whereas blood-gorging triggers several crucial developmental processes such as molting, growth and reproduction. This trait, where the life cycle and basic biology of an insect is so highly regulated by a blood meal, is one of the main reasons why R. prolixus was used by Sir Vincent

Wigglesworth to perform pioneering studies on metamorphosis in the early 20th century

(Wigglesworth, 1934b). However, more recently, it has garnered more attention due to its medical and economic importance. Rhodnius prolixus is one of the main vectors of the protozoan parasite, Trypanosoma cruzi, which causes Chagas disease in humans. According to the most- recent estimates, approximately 7-8 million people are infected with Chagas disease worldwide, with about 50000 new cases diagnosed each year (see Lima et al., 2010). There is currently no vaccine for Chagas disease and no treatment for the chronic phase of the infection despite the disease being discovered over a century ago. This has led the World Health Organization to classify Chagas disease amongst the most neglected tropical diseases. The recent completion of the R. prolixus genome sequencing, however, will undoubtedly aid in gaining knowledge of this disease-vector and assist in controlling disease transmission.

Post-feeding physiology

Rhodnius prolixus have evolved to be so efficient at hematophagy that the fifth-instar nymphs can consume a blood meal that can be as large as 10-12 times their unfed body weight. This blood meal results in short-term physiological/endocrinological changes that are essential to the 2 survival of the insect. The huge blood meal not only results in osmotic stress, but also makes the insect relatively immobile and thus prone to predation. In order to counteract this osmotic stress and reduce the threat of predation, R. prolixus must rapidly eliminate excess water and NaCl to preserve the volume and ionic balance of its haemolymph. In the process, it is also able to concentrate the nutritious component of the meal, the red blood cells. This post-feeding diuresis commences immediately after feeding and continues at a high rate for up to 3-4 hours during which time it secretes urine at a rate of 400nL/min (see Orchard, 2009). It is able to do so with the aid of four Malpighian tubules (MTs) whose epithelia possess the fastest secreting cells known (Maddrell, 1991). It is for this reason that R. prolixus MTs serve as a great model for the study of ion transporters in insects (see O'Donnell et al., 2003).

There are three main steps involved in the rapid post-feeding diuresis: (1) absorption of water and NaCl across the anterior midgut into the haemolymph, (2) secretion of water, NaCl and KCl from the haemolymph into the lumen of distal MTs and (3) reabsorption of KCl from the proximal end of MTs into the haemolymph (see Orchard, 2009). Eventually, the primary urine that enters the R. prolixus hindgut is similar to the plasma of the blood meal in terms of its ionic and osmotic composition. This urine, along with any digested waste products from the previous meal, are then expelled via the contraction of hindgut muscles. It is along with this urine that R. prolixus transmits the parasite, T. cruzi. The parasite enters the circulation of the human host mostly through the bite wound that R. prolixus makes when it initiates a new blood meal. Thus, it is of some medical importance to understand the post-feeding physiology of R. prolixus in order to control diuresis and prevent the spread of Chagas disease.

As previously mentioned, the blood meal does not only alter the short-term physiology but it also results in long term physiological changes which include both molting and egg production. The input to trigger both of these processes is believed to originate from the stretching of the anterior midgut induced by the blood meal (see Davey, 2007). This signal may be detected by stretch and/or pressure receptors and relayed to the brain via the nervous system. Feeding, therefore, has a direct effect on egg production such that the volume of the blood meal consumed influences the number of eggs produced. Moreover, activation of the stretch receptors following blood-gorging triggers the release of prothoracicotropic hormone (PTTH) from brain neurosecretory cells

(Wigglesworth, 1934a). PTTH then acts on the prothoracic glands to trigger ecdysteroid secretion and, consequently, affects ecdysteroid-dependent developmental events such as molting (Steel and Vafopoulou, 1997); however, not much is known about the downstream neurochemicals that influence egg production and molting following feeding in R. prolixus.

Neuropeptides

The nervous systems of present day animals utilize a wide variety of neurochemicals for neuronal communication. These include amino acids (e.g. glutamate), biogenic amines (e.g. octopamine), acetylcholine, and neuropeptides (e.g. galanin, FMRFamide and tachykinin) amongst others. The first nervous systems are thought to have evolved in the common ancestor of cnidarians (basal animal group), more than 700 million years ago (see Grimmelikhuijzen and

Hauser, 2012). Sequenced genomes of some of these basal animals such as Hydra magnipapillata show that these ancient nervous systems mainly utilized neuropeptides for communication (Chapman et al., 2010). Most animals emerging after cnidarians either belong to

Protostomia (comprising most invertebrates) or Deuterostomia (comprising all vertebrates and

4 few invertebrates). Many mammalian neuropeptide signaling systems have orthologs in

Protostomia (see Grimmelikhuijzen and Hauser, 2012).

Neuropeptides are small protein-like molecules that are produced in the central nervous system

(CNS), although they are not exclusive to the CNS and some neuropeptides are made by peripheral neurosecretory and endocrine cells. Neuropeptides are produced as part of larger precursor proteins which include the signal peptide, mature peptide(s), and, in most cases, one or more precursor (also known as spacer or flanking) peptides. To yield a mature peptide, the signal peptide is first cleaved off, after which it undergoes proteolytic processing at mainly basic amino acid residues (see Veenstra, 2000). Some neuropeptides may also undergo further modifications such as the amidation of C-terminal glycine before they become biologically active.

Neuropeptides are very diverse and as such grouped into various families based on their sequence or functional similarity. Neuropeptides control several essential physiological processes including feeding, reproduction and metamorphosis amongst others, and they do so by either functioning as neurotransmitters, neurohormones or neuromodulators. Their mode of action affects the privacy (i.e. localized action), speed of delivery, and duration of the message, thereby allowing the CNS to achieve flexibility in messaging. The neurotransmitters that are released at synapses are highly localized and private and both fast acting and short lasting.

Neurohormones, on the other hand, are released into the blood or haemolymph and thus less localized and private, slow-acting and have longer lasting effects. These two classes represent the two extremes; neurochemicals working in between these extremes are loosely referred to as neuromodulators. Hence, neuropeptides can be used to intricately control and coordinate a

5 variety of physiological and endocrinological events that are associated with a common behavior such as feeding.

Diuretic and anti-diuretic hormones

Urine production (or diuresis) in insects is controlled by various hormones and involves a variety of tissues, with MTs, analogous to the vertebrate kidney, being central in most insects. Insects undergo diuresis at various times during their life cycle; they undergo diuresis during ecdysis, in preparation for flight to reduce weight, during flight to get rid of waste products, and following feeding to get rid of excess water and salts. The rapid post-feeding diuresis that takes place following blood-gorging in R. prolixus is of interest here. Based on the research in various species, it is now evident that a multitude of hormones (neuropeptides and biogenic amines) regulate MT secretion (see Schooley et al., 2012). Diuretic hormones stimulate secretion whereas antidiuretic hormones either inhibit MT secretion or stimulate fluid reabsorption from the hindgut. In R. prolixus, the more distal ends of MTs are responsible for secretion of water and ions while the proximal end is where the reabsorption of KCl takes place (see Orchard, 2009).

Moreover, no evidence exists that point towards any reabsorption taking place in R. prolixus hindgut following post-feeding diuresis. Consequently, hormones that are known to promote fluid reabsorption in the hindgut of other insects such as neuroparsins, ion transport peptide and chloride transport stimulating hormone, have not received much attention in R. prolixus (see

Schooley et al., 2012).

There are four main families of hormones that control diuresis in insects. These include the corticotropin releasing-factor (CRF)-related diuretic hormones (CRF/DHs), kinins, CAPA or

CAP2b, and calcitonin-like diuretic hormones (CT/DHs) (see Schooley et al., 2012). Biogenic

6 amines serotonin (5-hydroxytryptamine, 5-HT) and tyramine are also diuretic in R. prolixus and

D. melanogaster, respectively. In most insects, these four peptidergic hormone families are diuretic; however, in R. prolixus, the situation is more complex. Only Rhopr-CRF/DH and serotonin act as “true” diuretic hormones (able to stimulate water and ion absorption across the anterior midgut, and secretion across the MTs), with both hormones evidently working via the intracellular messenger, cAMP, and both eliciting maximum secretion (see Orchard, 2009).

Moreover, a synergistic effect on secretion by isolated MTs is evident when serotonin is tested in conjunction with Rhopr-CRF/DH in an in vitro fluid secretion assay (commonly known as the

Ramsay assay), despite both hormones working via cAMP (Paluzzi et al., 2012). However, only serotonin and not Rhopr-CRF/DH is able to stimulate reabsorption across the proximal end of

MTs (see Orchard, 2009). Rhopr-kinins have no diuretic activity while Rhopr-CT/DH’s role in diuresis is questionable to say the least (see below for more details). Surprisingly, Rhopr-CAPA-

2 is an antidiuretic hormone in R. prolixus that is able to inhibit secretion stimulated by serotonin but not by Rhopr-CRF/DH (see Paluzzi, 2012). To make the situation more interesting, Rhopr-

CAPA-2 inhibits the synergistic effect between serotonin and Rhopr-CRF/DH (Paluzzi et al.,

2012). As mentioned, the interaction of these hormones has only been determined in vitro and the contributions of all these hormones towards diuresis in R. prolixus, in vivo, have yet to be determined.

Calcitonin-like diuretic hormone (CT/DH)

The existence of insect hormones related to the vertebrate calcitonin (CT) has been known for quite some time. Their presence was first examined in the corpus cardiacum and corpus allatum of the insect Leucophaea maderae via immunohistochemistry using antisera raised against mammalian CT (Hansen et al., 1982). Similar immunohistochemical analyses were also

7 performed in the tobacco hornworm moth, Manduca sexta, and the Colorado potato beetle,

Leptinotarsa decemlineata (El-Salhy et al., 1983; Veenstra et al., 1985). However, it wasn’t until

2000 that the first representative of the CT/DH family was isolated and functionally characterized in the Pacific beetle cockroach Diploptera punctata (Furuya et al., 2000). This peptide was originally referred to as diuretic hormone 31 (DH31), owing to the fact that it increased fluid secretion in MTs of various insects and contained 31 amino acids (Furuya et al.,

2000; Te Brugge et al., 2005). Although it has low sequence identity to the vertebrate CT, it has the conserved C-terminal Glycine-X-Proline-NH2 and they are similar in length (Furuya et al.,

2000). These are both important features in terms of peptide bioactivity as has been shown using vertebrate CTs (see Andreotti et al., 2006). Residues 9-19 of salmon calcitonin are involved in forming a stable α-helix which interacts with the C-terminus (Amodeo et al., 1999). Furthermore, the α-helical region has been shown to directly interact with the N-terminus of the receptor and varying the length of the helix causes a reduction in the peptide’s bioactivity (Andreotti et al.,

2006; Stroop et al., 1996). Since their isolation from D. punctata, CT/DHs have been identified and cloned in several insect species and their role as diuretic hormones examined in few insects.

Due to minimal sequence similarity between vertebrate CTs and insect CT/DHs, as well as their inability to act as “true” diuretic hormones (they are mostly only able to stimulate MT secretion) in most of the species examined so far, one might question the appropriateness of the name attributed to this peptide family. Due to the functional similarity between the vertebrate calcitonin receptors and insect CT/DH receptor(s), the CT-like nature of this peptide family remains assured. However, their role as diuretic hormones in insects is still questionable.

Note: The above excerpt was extracted from the review article “Calcitonin-like diuretic hormones in insects” authored by myself (see: Appendix A). The entire article is not presented here as it contains some findings from the later chapters.

FGLamide-related allatostatins (FGLa/ASTs)

Allatostatins (ASTs) are insect neuropeptides that were first identified based on their ability to inhibit juvenile hormone (JH) biosynthesis by the corpora allata (CA) (see Bendena and Tobe,

2012). FGLa/ASTs, related to the vertebrate galanin, represent one of three families of ASTs in insects that are characterized by their conserved C-terminus sequences. All three types of ASTs are present in most insects but only one type of AST inhibits JH synthesis, if at all, in any given insect. FGLa/ASTs have been shown to possess allatostatic activity in cockroaches, crickets, termites and locusts; however, whether they play such a role in R. prolixus is still unclear (see

Bendena and Tobe, 2012). FGLa/ASTs have now been shown to possess numerous other physiological roles, ranging from inhibition of visceral muscle contractions to influencing processes associated with reproduction (see Bendena and Tobe, 2012).

Several lines of evidence in R. prolixus and other insects point to the fact that FGLa/ASTs may be involved in post-feeding physiology in R. prolixus. The direct involvement of FGLa/ASTs in feeding has already been shown in D. melanogaster and Blattela germanica, where activation of a subset of FGLa/AST neurons and injection of FGLa/ASTs inhibited feeding behavior, respectively (Aguilar et al., 2003; Hergarden et al., 2012). Consistent with these results,

FGLa/ASTs also influence foraging in D. melanogaster (Wang et al., 2012). Other results based on immunohistochemical analyses and in vitro biological assays also indirectly imply a role for

FGLa/ASTs in feeding-related physiological events. For instance, FGLa/AST-like

9 immunoreactivity has been shown to be associated with the foregut, midgut, and hindgut, as well as open-type midgut endocrine cells in a variety of insects, including R. prolixus (see Audsley and Weaver, 2009; Sarkar et al., 2003). Hence FGLa/ASTs may modulate gut motility, digestion of food (via their effect on digestive enzymes secretion) and/or absorption of ions. It is hypothesized that FGLa/ASTs may also be released from midgut endocrine cells for actions on nearby tissues such as the MTs (see Veenstra, 2009). Lastly, but importantly, FGLa/AST-like immunoreactivity is associated with 5 dorsal unpaired median (DUM) neurons in the mesothoracic ganglionic mass (MTGM) of R. prolixus (Sarkar et al., 2003). These 5 DUM neurons are distinct from 5 DUM neurons in the MTGM expressing Rhopr-CT/DH and serotonin

(Orchard et al., 1989; Te Brugge et al., 2005), and responsible for the regulation of diuresis (see

Orchard, 2009). The location of the 5 Rhopr-FGLa/AST expressing DUM neurons alongside the

5 serotonergic DUM neurons in the MTGM presents the interesting possibility of neural integration of the two groups. Hence Rhopr-FGLa/ASTs may be involved in feeding-related activities, possibly counteracting the stimulatory effects of Rhopr-CT/DH and serotonin.

Consequently, it may be able to inhibit MT secretion and thus represent a class of antidiuretic hormones in insects.

G protein-coupled receptors (GPCRs)

Most neuropeptides mediate their effects by binding to GPCRs, which represent the largest family of cell membrane receptors. GPCRs can be further subdivided into three main families: class-A (also known as rhodopsin-like receptors), class-B (also known as secretin-like receptors) and class-C (which includes the metabotropic glutamate receptors) (see Venkatakrishnan et al.,

2013). Thus FGLa/ASTs and CT/DH mediate their effects by binding to class-A and class-B

GPCRs, respectively. All GPCRs are characterized by seven transmembrane domains, an

10 extracellular N-terminus and an intracellular C-terminus. The different families also have features that are restricted to them. For example, the N-terminal domain of class-B GPCRs contain six conserved cysteine residues that form three disulfide bridges (see Couvineau and

Laburthe, 2012). GPCRs sense molecules/ligands outside the cell through the N-terminus and/or one or more extracellular loops. This causes a conformational change in the receptor and activates the heterotrimeric G-protein (composed of alpha, beta and gamma subunits) that is associated with its C-terminus and/or intracellular loops. Eventually, one or more intracellular signaling pathways are activated depending on which type of G-protein alpha subunit the receptor couples with. For instance, class-B GPCRs usually couple with Gs alpha subunits that activate the cAMP pathway (see Couvineau and Laburthe, 2012). Since GPCRs regulate various vital processes in all animals, targeting these receptors using biostable, species-specific peptide mimetics that alter the normal physiology of insects is a promising avenue to control insect populations.

Objectives and organization of thesis

Recent advances in genome sequencing and bioinformatics has made it possible to predict novel neuropeptides, in silico. However, most, if not all, currently known insect neuropeptides were first isolated and biochemically purified based on them having an action in a particular bioassay.

Hence, FGLa/ASTs were first isolated based on their ability to inhibit JH biosynthesis in D. punctata (Woodhead et al., 1989). However, as is the case with many neuropeptides, they may not have the same effect in all insects. FGLa/ASTs and CT/DH are two such neuropeptides whose roles in R. prolixus are still unclear. Hence this thesis aims to understand the roles that these neuropeptides play, and in the process clarify their contribution, if any, towards post-

11 feeding physiology in R. prolixus. The central hypothesis tested in this thesis is that CT/DH and

FGLa/ASTs are present in R. prolixus and are involved in post-feeding physiology.

In Chapter 2 of this thesis, I utilized an array of molecular biology techniques to isolate cDNA sequences of three splice variants encoding Rhopr-CT/DH and to characterize their expression pattern using reverse transcriptase-PCR (RT-PCR) and fluorescent in situ hybridization (FISH).

This work shows that Rhopr-CT/DH is expressed throughout the CNS, including some neurosecretory cells, suggesting its action as a neurohormone on peripheral targets.

To identify the target tissues for CT/DH, in Chapter 3, receptors for Rhopr-CT/DH were isolated, functionally characterized, and their transcript expression profiles determined. The results of these studies show that CT/DH mediates its effects via two receptors in most insects and also identifies novel target tissues for this neuropeptide in R. prolixus.

Chapter 4 is an extension of this work, where effects of two Rhopr-CT/DH analogs (full-length form and N-terminal truncated form) were tested on hindgut contractions and in a heterologous receptor expression system. This work indicates that the peptide length is critical for receptor activation and hence truncated analogs fail to elicit any response.

In Chapter 5, using an approach similar to the one used in Chapter 2, a cDNA sequence encoding

Rhopr-FGLa/ASTs was cloned and characterized. Moreover, biological assays were also used to determine their physiological effects, in vitro. These results show that Rhopr-FGLa/AST is expressed throughout the CNS as well as in peripheral tissues. A role for Rhopr-FGLa/ASTs in

12 post-feeding physiology is also implicated since they dose-dependently inhibit contractions of the dorsal vessel, anterior midgut and hindgut.

Post-feeding physiological roles for Rhopr-FGLa/ASTs were further examined in Chapter 6. The release and distribution pattern of FGLa/ASTs into the haemolymph were examined post-feeding using immunohistochemistry and their effects tested on absorption across the anterior midgut and

MT secretion. Results show that although Rhopr-FGLa/ASTs are released into the haemolymph when diuresis needs to be halted, they are unable to inhibit anterior midgut absorption or MT secretion.

To further examine the potential physiological roles of FGLa/ASTs, in Chapter 7, I isolated and determined the expression pattern of a receptor related to insect FGLa/AST receptors. The expression pattern of the putative Rhopr-FGLa/AST receptor demonstrates that Rhopr-

FGLa/ASTs may be involved in long-term physiological changes (e.g. metamorphosis) that are triggered upon feeding.

As a result of this extensive study, this thesis elucidates the role of CT/DH and FGLa/ASTs in post-feeding physiology and sets the foundation for identifying other roles for these neuropeptides in R. prolixus.

References

Aguilar, R., Maestro, J.L., Vilaplana, L., Pascual, N., Piulachs, M.D. and Belles, X. (2003). Allatostatin gene expression in brain and midgut, and activity of synthetic allatostatins on feeding-related processes in the cockroach Blattella germanica. Regul Pept. 115, 171-7.

Amodeo, P., Motta, A., Strazzullo, G. and Castiglione Morelli, M.A. (1999). Conformational flexibility in calcitonin: the dynamic properties of human and salmon calcitonin in solution. J Biomol NMR. 13, 161-74.

Andreotti, G., Mendez, B.L., Amodeo, P., Morelli, M.A., Nakamuta, H. and Motta, A. (2006). Structural determinants of salmon calcitonin bioactivity: the role of the Leu-based amphipathic alpha-helix. J Biol Chem. 281, 24193-203.

Audsley, N. and Weaver, R.J. (2009). Neuropeptides associated with the regulation of feeding in insects. Gen Comp Endocrinol. 162, 93-104.

Bendena, W.G. and Tobe, S.S. (2012). Families of allatoregulator sequences: a 2011 perspective. Can J Zoolog. 90, 521-544.

Chapman, J.A., Kirkness, E.F., Simakov, O., Hampson, S.E., Mitros, T., Weinmaier, T., Rattei, T., Balasubramanian, P.G., Borman, J., Busam, D., Disbennett, K., Pfannkoch, C., Sumin, N., Sutton, G.G., Viswanathan, L.D., Walenz, B., Goodstein, D.M., Hellsten, U., Kawashima, T., Prochnik, S.E., Putnam, N.H., Shu, S., Blumberg, B., Dana, C.E., Gee, L., Kibler, D.F., Law, L., Lindgens, D., Martinez, D.E., Peng, J., Wigge, P.A., Bertulat, B., Guder, C., Nakamura, Y., Ozbek, S., Watanabe, H., Khalturin, K., Hemmrich, G., Franke, A., Augustin, R., Fraune, S., Hayakawa, E., Hayakawa, S., Hirose, M., Hwang, J.S., Ikeo, K., Nishimiya-Fujisawa, C., Ogura, A., Takahashi, T., Steinmetz, P.R., Zhang, X., Aufschnaiter, R., Eder, M.K., Gorny, A.K., Salvenmoser, W., Heimberg, A.M., Wheeler, B.M., Peterson, K.J., Bottger, A., Tischler, P., Wolf, A., Gojobori, T., Remington, K.A., Strausberg, R.L., Venter, J.C., Technau, U., Hobmayer, B., Bosch, T.C., Holstein, T.W., Fujisawa, T., Bode, H.R., David, C.N., Rokhsar, D.S. and Steele, R.E. (2010). The dynamic genome of Hydra. Nature. 464, 592-6.

Couvineau, A. and Laburthe, M. (2012). The family B1 GPCR: structural aspects and interaction with accessory proteins. Curr Drug Targets. 13, 103-15.

Davey, K. (2007). The interaction of feeding and mating in the hormonal control of egg production in Rhodnius prolixus. J Insect Physiol. 53, 208-15.

El-Salhy, M., Falkmer, S., Kramer, K.J. and Speirs, R.D. (1983). Immunohistochemical investigations of neuropeptides in the brain, corpora cardiaca, and corpora allata of an adult lepidopteran insect, Manduca sexta (L). Cell Tissue Res. 232, 295-317.

Furuya, K., Milchak, R.J., Schegg, K.M., Zhang, J., Tobe, S.S., Coast, G.M. and Schooley, D.A. (2000). Cockroach diuretic hormones: characterization of a calcitonin-like peptide in insects. Proc Natl Acad Sci USA. 97, 6469-74.

Grimmelikhuijzen, C.J. and Hauser, F. (2012). Mini-review: the evolution of neuropeptide signaling. Regul Pept. 177 Suppl, S6-9.

Hansen, B.L., Hansen, G.N. and Scharrer, B. (1982). Immunoreactive material resembling vertebrate neuropeptides in the corpus cardiacum and corpus allatum of the insect Leucophaea maderae. Cell Tissue Res. 225, 319-29.

Hergarden, A.C., Tayler, T.D. and Anderson, D.J. (2012). Allatostatin-A neurons inhibit feeding behavior in adult Drosophila. Proc Natl Acad Sci U S A. 109, 3967-72.

Lima, F.M., Oliveira, P., Mortara, R.A., Silveira, J.F. and Bahia, D. (2010). The challenge of Chagas' disease: has the human pathogen, Trypanosoma cruzi, learned how to modulate signaling events to subvert host cells? N Biotechnol. 27, 837-43.

Maddrell, S.H. (1991). The fastest fluid‐secreting cell known: The upper malpighian tubule of Rhodnius. BioEssays. 13, 357-362.

O'Donnell, M.J., Ianowski, J.P., Linton, S.M. and Rheault, M.R. (2003). Inorganic and organic anion transport by insect renal epithelia. Biochim Biophys Acta. 1618, 194-206.

Orchard, I. (2009). Peptides and serotonin control feeding-related events in Rhodnius prolixus. Front Biosci. 1, 250-62.

Orchard, I., Lange, A.B., Cook, H. and Ramirez, J.M. (1989). A subpopulation of dorsal unpaired median neurons in the blood-feeding insect Rhodnius prolixus displays serotonin-like immunoreactivity. J Comp Neurol. 289, 118-28.

Paluzzi, J.P. (2012). Anti-diuretic factors in insects: the role of CAPA peptides. Gen Comp Endocrinol. 176, 300-8.

Paluzzi, J.P., Naikkhwah, W. and O'Donnell, M.J. (2012). Natriuresis and diuretic hormone synergism in R. prolixus upper Malpighian tubules is inhibited by the anti-diuretic hormone, RhoprCAPA-alpha2. J Insect Physiol. 58, 534-42.

Sarkar, N.R., Tobe, S.S. and Orchard, I. (2003). The distribution and effects of Dippu- allatostatin-like peptides in the blood-feeding bug, Rhodnius prolixus. Peptides. 24, 1553-62.

Schooley, D.A., Horodyski, F.M. and Coast, G.M., 2012. 9 - Hormones Controlling Homeostasis in Insects, in: Lawrence, I.G. (Ed.), Insect Endocrinology. Academic Press, San Diego, pp. 366-429.

Steel, C.G. and Vafopoulou, X. (1997). Ecdysteroidogenic action of Bombyx prothoracicotropic hormone and bombyxin on the prothoracic glands of Rhodnius prolixus in vitro. J Insect Physiol. 43, 651-656.

Stroop, S.D., Nakamuta, H., Kuestner, R.E., Moore, E.E. and Epand, R.M. (1996). Determinants for calcitonin analog interaction with the calcitonin receptor N-terminus and transmembrane-loop regions. Endocrinology. 137, 4752-6.

Te Brugge, V.A., Lombardi, V.C., Schooley, D.A. and Orchard, I. (2005). Presence and activity of a Dippu-DH31-like peptide in the blood-feeding bug, Rhodnius prolixus. Peptides. 26, 29-42.

Veenstra, J.A. (2000). Mono- and dibasic proteolytic cleavage sites in insect neuroendocrine peptide precursors. Arch Insect Biochem Physiol. 43, 49-63.

Veenstra, J.A. (2009). Peptidergic paracrine and endocrine cells in the midgut of the fruit fly maggot. Cell Tissue Res. 336, 309-23.

Veenstra, J.A., Romberg-Privee, H.M., Schooneveld, H. and Polak, J.M. (1985). Immunocytochemical localization of peptidergic neurons and neurosecretory cells in the neuroendocrine system of the Colorado potato beetle with antisera to vertebrate regulatory peptides. Histochemistry. 82, 9-18.

Venkatakrishnan, A.J., Deupi, X., Lebon, G., Tate, C.G., Schertler, G.F. and Babu, M.M. (2013). Molecular signatures of G-protein-coupled receptors. Nature. 494, 185-94.

Wang, C., Chin-Sang, I. and Bendena, W.G. (2012). The FGLamide-allatostatins influence foraging behavior in Drosophila melanogaster. PLoS One. 7, e36059.

Wigglesworth, V. (1934a). Memoirs: The physiology of ecdysis in Rhodnius prolixus (Hemiptera). II. Factors controlling moulting and ‘metamorphosis’. Q J Microsc Sci. 2, 191-222.

Wigglesworth, V.B. (1934b). Factors controlling moulting and metamorphosis in an insect. Nature. 133, 725-726.

Woodhead, A.P., Stay, B., Seidel, S.L., Khan, M.A. and Tobe, S.S. (1989). Primary structure of four allatostatins: neuropeptide inhibitors of juvenile hormone synthesis. Proc Natl Acad Sci U S A. 86, 5997-6001.

Chapter 2: Isolation and characterization of the cDNA

encoding DH31 in the kissing bug, Rhodnius prolixus

The proceeding chapter was reprinted/reproduced with permission from Elsevier.

Full citation details:

Isolation and characterization of the cDNA encoding DH(31) in the kissing bug, Rhodnius prolixus.

Zandawala M, Paluzzi JP, Orchard I.

Mol Cell Endocrinol. 2011 Jan 1;331(1):79-88. doi: 10.1016/j.mce.2010.08.012.

Abstract

Rhodnius prolixus undergoes a period of rapid diuresis after ingesting large blood meals.

Neurohormones with either diuretic or anti-diuretic activity control diuresis by acting on several tissues including the Malpighian tubules. One of the neurohormones that potentially plays a role in diuresis is diuretic hormone 31 (DH31) which belongs to the insect calcitonin-like family of diuretic hormones. Here we determine the complete cDNA sequences of three Rhopr-DH31 splice variants (Rhopr-DH31-A, Rhopr-DH31-B and Rhopr-DH31-C) and characterize their expression in unfed fifth-instar R. prolixus. Reverse transcriptase-PCR demonstrates that Rhopr-

DH31 is predominantly expressed in the central nervous system (CNS) of unfed fifth instars.

However, the expression of the three splice variants differs with Rhopr-DH31-B expression being the highest followed by Rhopr-DH31-A and Rhopr-DH31-C, as determined using semi- quantitative Southern blot analysis. Fluorescent in situ hybridization reveals that Rhopr-DH31 is expressed in a variety of cells in the CNS, including some neurosecretory cells.

Introduction

Rhodnius prolixus is a blood-feeding hemipteran which is confined to drier savannah areas of

Central and South America (Dujardin et al., 1998; Feliciangeli et al., 2004; Monroy et al., 2003;

Ramsey et al., 2000). This insect is a principal vector of Chagas' disease, an incurable illness damaging the human heart and nervous system, caused by the parasite Trypanosoma cruzi

(Koberle, 1968). Rhodnius prolixus ingests large blood meals and then undergoes a period of rapid diuresis. Infection occurs when R. prolixus releases protozoans in urine that it deposits near the site of feeding (see Prata, 2001). Diuresis, the process of urine production, involves a variety of processes and tissues. These include ion and water movement across the epithelium of the anterior midgut and the Malpighian tubules, and muscle contractions of the anterior midgut, hindgut and dorsal vessel which facilitate mixing of the blood-meal, mixing of the haemolymph, and the expulsion of waste (Coast et al., 2005; Coast et al., 2001; Donini et al., 2008; Te Brugge et al., 2009; Te Brugge et al., 2005; Te Brugge et al., 2008). Neurohormones with either diuretic or anti-diuretic activity control the function of Malpighian tubules (see Coast et al., 2002; see

Schooley et al., 2005). Diuretic hormones (DHs) cause acceleration in primary urine production, whereas anti-diuretic hormones (ADHs) either stimulate fluid reabsorption in the hindgut or reduce Malpighian tubules secretion (see Coast et al., 2002). Insect DHs include serotonin (5- hydroxytryptamine) and various families of neuropeptides such as the calcitonin-like DHs

(CLDH), kinin-like DHs, corticotropin-releasing factor (CRF)-like DHs and CAPA-like DHs

(see Coast et al., 2002; Furuya et al., 2000; Kean et al., 2002; Maddrell et al., 1991; Te Brugge et al., 2009; Te Brugge et al., 2005; Te Brugge et al., 1999; Te Brugge et al., 2002).

One of the neurohormones that may play a role in diuresis in R. prolixus is referred to as diuretic hormone 31 (DH31), and this belongs to the CLDH family of insect peptides (Te Brugge et al.,

2005). The first CLDH was identified in the Pacific beetle cockroach, Diploptera punctata

(Furuya et al., 2000). Since then, CLDHs have been identified in several insects, as well as crustaceans and chelicerates (Table 1) (Christie, 2008; Christie et al., 2010; Coast et al., 2005;

Coast et al., 2001; Gard et al., 2009; Li et al., 2008; Schooley et al., 2005). Here, we annotated

CLDHs in the southern house mosquito, Culex quinquefasciatus and the honey bee varroa mite,

Varroa destructor following a BLAST search of their genome databases. CLDHs are 31 amino acids in length in arthropods, with the exception of Ixodes scapularis and V. destructor

(predicted) where they contain 34 amino acids. These peptides show a high degree of amino acid identity and are amidated at their carboxyl termini. Invertebrate CLDHs are less similar to vertebrate calcitonin but share the C-terminal Gly-X-Pro-NH2 (Furuya et al., 2000). CLDHs have been shown to stimulate fluid secretion by Malpighian tubules (Coast et al., 2001; Furuya et al.,

2000; Maddrell et al., 1991; Te Brugge et al., 2005; Te Brugge and Orchard, 2008; Te Brugge et al., 2002), to have potent natriuretic activity (Coast et al., 2005) and to increase dorsal vessel and hindgut contractility (Te Brugge et al., 2008). Moreover, Rhopr-DH31 causes an increase in cAMP concentration and contraction frequency in the anterior midgut (Te Brugge et al., 2009).

Since these processes are associated with diuresis, CLDHs might play an important role in post- feeding diuresis in R. prolixus.

In the present study, complete cDNA sequences of three Rhopr-DH31 splice variants (Rhopr-

DH31-A, Rhopr-DH31-B and Rhopr-DH31-C) were obtained, which encode a mature peptide that is 100% identical to that determined previously by MALDI-TOF mass spectrometry for Rhopr-

DH31 (Te Brugge et al., 2008). Northern blot hybridization was performed to confirm the approximate size of the DH31 gene (DH31) transcripts. Reverse transcriptase-PCR (RT-PCR) analysis also confirmed Rhopr-DH31 expression in the central nervous system (CNS). Semi-

21 quantitative analysis using Southern blot hybridization to determine the relative expression of the three splice variants within the CNS revealed that Rhopr-DH31-B expression was the highest followed by Rhopr-DH31-A and Rhopr-DH31-C in fifth-instar R. prolixus. Using Rhopr-DH31-A partial cDNA sequence to design a probe, fluorescent in situ hybridization (FISH) was performed to localize the cell-specific expression of Rhopr-DH31, revealing a number of cells distributed throughout various regions of the CNS.

Table 1: Structures of mature DH31 peptides (deduced or sequenced) from 16 species.

Species Peptide structure Reference Insects 1 Rhodnius prolixus GLDLGLSRGFSGSQAAKHLMGLAAANYAGGP-NH2 This study, Te Brugge et al., 2008 1 Diploptera punctata GLDLGLSRGFSGSQAAKHLMGLAAANYAGGP-NH2 Furuya et al., 2000 1 Apis mellifera GLDLGLSRGFSGSQAAKHLMGLAAANYAGGP-NH2 Schooley et al., 2005 2 Tribolium castaneum GLDLGLGRGFSGSQAAKHLMGLAAANFAGGP-NH2 Li et al., 2008 Bombyx mori AFDLGLGRGYSGALQAKHLMGLAAANFAGGP-NH2 Schooley et al., 2005 Drosophila melanogaster TVDFGLARGYSGTQEAKHRMGLAAANFAGGP-NH2 Coast et al., 2001 3 Anopheles gambiae TVDFGLSRGYSGAQEAKHRMAMAVANFAGGP-NH2 Coast et al., 2005 3 Aedes aegypti TVDFGLSRGYSGAQEAKHRMAMAVANFAGGP-NH2 Schooley et al., 2005 Nasonia vitripennis GLDLGLNRGFSGSQAAKHLMGLAAANYAGGP-NH2 Predicted Acyrthosiphon pisum GLDLGLSRGYSGTQAAKHLMGMAAANFAGGP-NH2 Predicted

Nilaparvata lugens GLDLGLSRGFSGSQAAKHLMGLAAANYAAGP-NH2 Predicted 3 Culex quinquefasciatus TVDFGLSRGYSGAQEAKHRMAMAVANFAGGP-NH2 This study, Predicted Crustaceans Daphnia pulex GVDFGLGRGYSGSQAAKHLMGLAAANYAIGP-NH2 Gard et al., 2009 2 Homarus americanus GLDLGLGRGFSGSQAAKHLMGLAAANFAGGP-NH2 Christie et al., 2010 Chelicerates Ixodes scapularis AGGLLDFGLSRGASGAEAAKARLGLKLANDPYGP-NH2 Christie et al., 2008 Varroa destructor SNGMLDFGLARGMSGVDAAKARLGLKYANDPYGP-NH2 This study, Predicted

Amino acids that are shared by all sequences are highlighted in gray. 1 Identical sequences. 2 Identical sequences. 3 Identical sequences.

Material and methods

Animals

Fifth-instar R. prolixus of both sexes were taken from a long standing colony at the University of

Toronto Mississauga. Insects were maintained at high relative humidity in incubators at 25°C and were fed on rabbits’ blood. All the tissues used were dissected from six weeks post-fed (as fourth-instars) insects in nuclease-free phosphate-buffered saline (PBS) (Sigma–Aldrich,

Oakville, ON, Canada) and were either used immediately or stored at -20°C in RNAlaterTM RNA stabilization reagent (Qiagen Inc., Mississauga, ON, Canada).

Screening of fifth-instar CNS cDNA library

Degenerate forward primers (dh31forward1, dh31forward2, dh31forward3 and dh31forward4)

(see Appendices: Table S1) were designed using the Rhopr-DH31 peptide sequence (Te Brugge et al., 2008). These primers were used along with plasmid reverse primers (DNR-LIB REV110 and DNR-LIB REV77) (see Appendices: Table S1) to obtain a partial Rhopr-DH31 cDNA sequence. Next, modified 5’ and 3’ Rapid Amplification of cDNA Ends (RACE) PCR reactions were performed. The partial Rhopr-DH31 cDNA sequence was used to design gene-specific forward and reverse primers for 3’ and 5’ RACE, respectively. For 5’ RACE, nested PCRs were performed using two plasmid forward primers (DNR-LIB FOR1 and DNR-LIB FOR2) and four gene-specific reverse primers (DH31 5race1, DH31 5race2, DH31 5race3 and DH31 5race4) (see

Appendices: Table S2). One plasmid reverse primer (pDNR-LIB 3 – 25 REV) and four gene- specific forward primers (FOR1DH31, FOR2DH31, FOR3DH31 and FOR4DH31) were used to perform the semi-nested PCRs for 3’ RACE (see Appendices: Table S3). The forward and reverse primers were used successively in order to selectively amplify the specific products. PCR product of each reaction was either column purified using EZ-10 Spin Column PCR Products

Purification Kit (Bio Basic Inc., Markham, ON, Canada) or gel extracted using EZ-10 Spin

Column DNA Gel Extraction Kit (Bio Basic Inc., Markham, ON, Canada) and used as a template for the subsequent reaction. Plasmid DNA isolated from a R. prolixus CNS cDNA library was used as the template for the above reactions (Paluzzi et al., 2008). All PCR reactions were performed using GeneAmp® PCR System 9700 (Applied Biosystems) Thermocycler.

Temperature-cycling profiles remained constant and were based on the following profile: initial denaturation at 95°C for 3 min, followed by 35 cycles of 94°C for 30 sec, 50°C-60°C (depending on the primers used) for 30 sec and 72°C for 1 min. A final 10 min extension at 72°C was also included. Products generated after the final PCR were gel extracted, cloned using the pGEM®-T

Easy vector (Promega Corporation, Madison, WI, USA), and the plasmid DNAs isolated from the overnight cell cultures and sequenced at the SickKids DNA Sequencing Facility (The Centre for Applied Genomics, Hospital for Sick Children, Toronto, ON, Canada).

Northern blot analysis revealed that the size of transcripts was comparatively larger (see below) than that obtained following the 5’ and 3’ RACE PCR reactions. In silico analyses confirmed that a considerable region at the 3’ end could not be cloned using the 3’ RACE reactions and thus the following approach was adopted to obtain the complete Rhopr-DH31 cDNA sequence for all three splice variants. Rhopr-DH31 3’ UTR sequence was predicted using the R. prolixus preliminary genome assembly and used to design forward and reverse gene-specific primers for

PCR (see Appendices: Table S4). CNS single-stranded cDNA was used as a template for these

PCR reactions (for cDNA synthesis, see below).

cDNA sequence analysis

The deduced Rhopr-DH31 prepropeptide sequences were examined for potential signal peptides using SignalP 3.0 (Bendtsen et al., 2004) and potential ubiquitination sites using UbPred

(Radivojac et al., 2010). The intron-exon boundaries of Rhopr-DH31 were determined using

BLAST (Altschul et al., 1990) search results obtained from the R. prolixus preliminary genome assembly (http://www.ncbi.nlm.nih.gov/sutils/genom_table.cgi?organism=insects – last accessed on April 12, 2010) and confirmed using Genie, online software for splice site prediction (Reese et al., 1997). The three Rhopr-DH31 variant prepropeptide sequences (products of transcript variant A - HM030716, transcript variant B - HM030715 and transcript variant C - HM030714) and their homologous sequences from other insect species, including Drosophila melanogaster transcript variant A (NM_078790.3), D. melanogaster transcript variant C (NM_164825.2),

Bombyx mori (NP_001124379.1), Anopheles gambiae (EAA01397.3 / XP_321755.3),

Acyrthosiphon pisum (XP_001945901.1), Aedes aegypti (EAT40182.1), Nilaparvata lugens

(DB826761) and Nasonia vitripennis (XP_001599948.1), were aligned using ClustalW

(http://www.ch.embnet.org/software/ClustalW.html - last accessed on April 12, 2010) and the figure presented was obtained using the BOXSHADE 3.21 server

(http://www.ch.embnet.org/software/BOX_form.html - last accessed on April 12, 2010). DH31 sequence from Homarus americanus (GQ290461.1) was also included in the alignment and utilized as an outgroup for the phylogeny construction. D. melanogaster transcript variant D

(NM_001169442.1) and Tribolium castanuem (EEZ99367.1) sequences were not included in the alignment as the prepropeptide sequence did not contain the mature peptide and the sequence was truncated at the 5’ end, respectively. The alignment obtained upon performing the multiple sequence alignment was exported and used to produce a phylogenetic tree using MEGA 4.02

(Tamura et al., 2007). The rooted phylogram was obtained using neighbour-joining analysis and presented along with bootstrap values based on 1000 replicates.

Northern blot hybridization

Total RNA was purified from CNS using Trizol® reagent (Life Technologies Corporation,

Carlsbad, CA, USA) followed by mRNA extraction using PolyATtract® mRNA isolation systems III and IV (Promega Corporation, Madison, WI, USA). Approximately 300ng CNS mRNA was denatured at 75°C for 5 min followed by immediate incubation on ice. To enable size determination, a 250ng sample of RiboRulerTM High Range RNA Ladder (Fermentas

Canada Inc., Burlington, ON, Canada) was prepared in a similar manner. Samples of CNS mRNA and RNA ladder were then separated on a 1% agarose gel containing formaldehyde at

70V for 140 min. In order to ensure that RNA samples were separated sufficiently, the gel was quickly observed under a UV transilluminator. Next, excess formaldehyde was rinsed off the gel by incubating the gel in diethyl pyrocarbonate-treated double-distilled water for 1 hour, changing the water every 20 min. The RNA was then transferred overnight in 20X Saline-Sodium Citrate

(Fisher Scientific Ltd., Ottawa, ON, Canada) to a positively charged nylon membrane (Roche,

Mannheim, Germany) through downward capillary transfer and immobilized using a UV crosslinker at a setting of 30,000μJ/cm2. Hybridization was then performed using the DIG

Northern Starter Kit (Roche, Mannheim, Germany). The protocol supplied by the manufacturer was followed with some minor modifications. Specifically, hybridization was performed using digoxigenin (DIG)-labeled RNA probe. Plasmid DNA containing a 445bp partial Rhopr-DH31-A cDNA fragment was linearized using NcoI and then purified using Wizard® SV Gel and PCR

Clean-Up System (Promega Corporation, Madison, WI, USA). The linearized recombinant plasmid was used as template to synthesize the DIG-labeled RNA probe via in vitro transcription using the DIG RNA labeling kit SP6/T7 (Roche Applied Science, Mannheim, Germany). SP6

RNA polymerase was used to synthesize this anti-sense probe. Following RNA probe synthesis,

27 the template DNA was removed by incubating the probe with deoxyribonuclease I for 15 min at

37°C. During prehybridization, the membrane was incubated in pre-warmed hybridization solution at 68°C for 30 min. The prehybridization solution was then replaced with hybridization solution containing probe (50ng/mL) and the membrane incubated at 68°C for 16 to 18 hours.

The blot was then washed under stringency conditions and used for immunological detection.

Exposures were made at room temperature for various times ranging from 1 min to 20 min using

Bioflex Scientific Imaging Films (Clonex Corporation, Markham, ON, Canada). The blot was stripped off the probe and reprobed using DIG-labeled probe for the RNA ladder to determine the size of the observed fragment.

RT-PCR tissue profiling

Insect tissues were dissected and pooled into four different groups: 1) CNS 2) salivary glands 3) oesophagus, anterior midgut, posterior midgut, hindgut and Malpighian tubules 4) dorsal vessel, fat bodies, trachea, abdominal nerves, immature testes, immature ovaries, dorsal diaphragm and ventral diaphragm. Following the manufacturer supplied protocols, total RNA was isolated from these tissues using SV Total RNA isolation system (Promega Corporation, Madison, WI, USA) and used for cDNA synthesis using an oligo(dT) primer supplied with iScriptTM Select cDNA

Synthesis Kit (Bio-Rad Laboratories Ltd., Mississauga, ON, Canada). An aliquot of this single- stranded cDNA was used as a template to perform a subsequent PCR. The following temperature-cycling profile was used for PCR to amplify a Rhopr-DH31 fragment: initial denaturation at 95°C for 3 min, followed by 35 cycles of 94°C for 30 sec, 60°C for 30 sec, 72° for 1 min and a final 10 min extension at 72°C. The forward primer (FOR1DH31) had the 5'

CACTATATCGCCGGCAGTC 3' sequence and the reverse primer (DH315race4) had the 5'

CAAGATCCAAACCACGCTTC 3' sequence. The predicted size of the PCR product would be

28 between 317bp and 428bp depending on the type of variant expressed. For a positive control to test the quality of the template cDNA, a 229bp ribosomal protein 49 (RP49) fragment was amplified. The forward primer (rp49-qPCR-for) had the 5' GTGAAACTCAGGAGAAATTGGC

3' sequence and the reverse primer (rp49-rev2) had the 5' AGGACACACCATGCGCTATC 3' sequence. This procedure was repeated using three independent biological replicates.

Southern blot hybridization

RNA was isolated from CNS tissue and subsequently used to synthesize cDNA as mentioned earlier (see above). An aliquot of this single-stranded cDNA was used as a template for PCR, identical to the one described earlier for Rhopr-DH31 RT-PCR (see above). It was assumed that the target amplification was equally efficient for the three transcript variants since the primers used were designed over regions common to all three variants (i.e. the transcript variant region was located within the amplified region and not over the primer annealing sites). Several different PCR reactions were carried out with the number of cycles ranging from 26 to 35 since no amplification products were observed in PCR reactions of less than 26 cycles. This was carried out to ensure that the relative abundance of each transcript was consistent over the exponential amplification phase of the PCR reaction and not an end-point inhibition artifact. The

PCR product was purified as described earlier (see above) and 100ng of this purified PCR product was separated on a 2% agarose gel for 2 hours at 170V. The DNA was denatured, neutralized and transferred to a positively charged nylon membrane (Roche, Mannheim,

Germany) through downward capillary transfer. The DNA was fixed onto the membrane using a

UV crosslinker at a setting of 7,500μJ/cm2. Hybridization was performed using the Amersham

Gene Images AlkPhos Direct Labelling and Detection System (GE Healthcare, Piscataway, NJ,

USA) as described previously (Paluzzi et al., 2008). The manufacturer protocol was followed

29 with some minor modifications. In particular, hybridization was performed using an alkaline phosphatase-labeled probe, which was a 317bp long fragment amplified from Rhopr-DH31-A clone using the same primers as above for PCR. The probe was stored at -20°C in 50% glycerol until used for hybridization. The membrane was incubated in hybridization solution containing probe (10ng/mL) at 60°C. After overnight incubation for 22 to 24 hours, the membrane was washed in the primary wash buffer, in which the blocking solution was replaced by 0.1% (w/v)

BSA and 0.1% (w/v) milk powder. The blot was subsequently treated following manufacturer recommendations. Lastly, the image was captured using a STORM 840 Phosphorimager (GE

Healthcare, Piscataway, NJ, USA) at several time points after adding the ECF substrate to generate signal, and was analyzed using ImageQuant TL software (GE Healthcare, Piscataway,

NJ, USA). The Southern blot analysis of the RT-PCR products was repeated using three independent biological replicates to obtain average relative expression of the three transcripts.

To confirm that the average relative expression of the three transcripts differed significantly, two-tailed t-test (P < 0.05) was performed on arcsine transformed data.

Fluorescent in situ hybridization (FISH)

Rhopr-DH31 expression localization in fifth-instar R. prolixus CNS was determined using the protocols described earlier (Paluzzi and Orchard, 2010; Paluzzi et al., 2008), with the following modifications. Digoxigenin (DIG)-labeled anti-sense RNA was synthesized as mentioned earlier for northern blot hybridization (see above). Similarly, to obtain the sense-probe for the control experiment, the plasmid DNA was linearized using NdeI and T7 polymerase was used instead to label the RNA probe via in vitro transcription. The probe was treated with deoxyribonuclease I for 15 min at 37°C to remove the template DNA and then stored at -20°C for up to several weeks. Following the endogenous peroxidase quenching step, the tissues were incubated in 4%

Triton X-100 (Sigma–Aldrich, Oakville, ON, Canada) in PBS with 0.1% Tween-20 (PBT) and the tissues incubated for 45 min at room temperature. Pre-hybridization was performed at 56°C for 1 hour. Approximately 0.5-1ng probe (amount estimated using gel electrophoresis) per 1μL hybridization solution was used to perform hybridization. Following the primary antibody incubation, tissues were washed six times for 10 min each with PBT containing 1% (w/v) blocking reagent (PBTB) supplied with TSATM kit #24 (Molecular Probes Inc., Eugene, OR,

USA). Following the incubation with streptavidin horseradish peroxidase, the tissues were washed six times for 10 min each with PBTB. This was followed by one wash in PBT and two washes in PBS for 5 min each. Following the incubation with tyramide substrate, the tissues were washed five times for 10 min each with PBS and then left overnight (12 to 14 hours) in

PBS at 4°C before mounting.

Results

Three DH31 splice variants exist in R. prolixus

Three Rhopr-DH31 splice variants are present in fifth-instar R. prolixus, referred to as Rhopr-

DH31-A, Rhopr-DH31-B and Rhopr-DH31-C. The Rhopr-DH31-A cDNA sequence obtained after performing the initial 5’ and 3’ RACE was used to synthesize a labeled probe. Northern blot hybridization was then performed using this probe to determine the size of the Rhopr-DH31 transcripts. The expected size of the transcripts was approximately 550bp to 450bp. However, the Rhopr-DH31 transcripts were estimated to be just less than 1.5kb long (Figure 1), which was considerably larger than the cDNA sequences that had been determined. Hence, only partial cDNA sequences for the three variants had been obtained after the 5’ and 3’ PCR RACE reactions. Therefore a modified approach, as described earlier, was adopted to obtain the complete Rhopr-DH31 cDNA sequence (Figure 2A). Rhopr-DH31-A variant was found to be 1374

31 bp long while Rhopr-DH31-B and Rhopr-DH31-C were found to be 1431 bp and 1485 bp long, respectively. All three Rhopr-DH31 variants have the same 5’ and 3’ UTRs which are 91bp and at least 953bp long, respectively. Within the 3’ UTR of all the variants is a poly(A) tail which is at least 16 nucleotides long. All the variants contain a single open reading frame (ORF) but their lengths vary. Rhopr-DH31-A, Rhopr-DH31-B and Rhopr-DH31-C have an ORF of 330bp, 387bp and 441bp, respectively, which encode apparent prepropeptides of 109, 128 and 146 amino acids, respectively. A highly predicted signal peptide is present in the ORF with the most likely cleavage site occurring between alanine residue at position 23 and serine at position 24.

Moreover, the lysine residue at position 68 is predicted to undergo ubiquitination. All three variants produce the same mature peptide consisting of 31 amino acids and located between amino acid positions 108 and 140 (Figure 2A). Within the prepropeptide sequence, the sequence encoding the mature peptide is flanked by lysine and arginine dibasic amino acid residues at the

N terminus and a glycine and four arginine residues at the C terminus. The dibasic amino acid residues at the N terminus and the arginine residues at the C terminus are required for post- translational proteolytic processing following signal peptide cleavage while the glycine residue is required for amidation by peptidyl-glycine α-amidating monooxygenase (see Coast et al., 2002).

The mature peptide sequence predicted here is 100% identical to the sequence determined using

MALDI-TOF mass spectrometry analysis of the CNS (Te Brugge et al., 2008). Polyadenylation signal consensus sequence (AATAAA) is usually present in the 3’ UTR of a gene. This sequence marks the binding site for proteins that cleave the 3’ end of the RNA after which the poly(A) tail is added (see Zhao et al., 1999). Interestingly, 9 such consensus sequences are present in the 3’

UTR of Rhopr-DH31 sequence. After completing the PCRs to clone the Rhopr-DH31 3’ UTR, two sequences of varying lengths were obtained - a larger sequence (presented in Figure 2A) and a

32 shorter sequence. The shorter sequence was a truncated version of the larger sequence and had a poly(A) tail after nucleotide position 1385 (Figure 2A).

Molecular organization of Rhopr-DH31 was determined based on BLAST search and in silico analysis using intron prediction (Figure 2B). The gene comprises six exons (sequentially referred to as 1 – 6) which are separated by five introns. The lengths of the exons are 89bp, 93bp, 112bp,

57bp, 54bp and 1064bp. Rhopr-DH31-C is composed of exons 1 to 6, Rhopr-DH31-B is comprised of exons 1 to 4 and 6, while Rhopr-DH31-A is made up of exons 1 to 3 and 6. The size of the introns could not be determined since the R. prolixus genome assembly is incomplete. The inclusion of exon 4 in Rhopr-DH31-B and Rhopr-DH31-C causes a change in the resulting amino acid sequence and results in a lysine residue at position 68 instead of an asparagine residue at the same position in Rhopr-DH31-A. Similarly, the presence of exon 5 in Rhopr-DH31-C results in an arginine residue at position 87 instead of a serine residue found in Rhopr-DH31-B.

Figure 1: Northern blot analysis of Rhopr-DH31 in R. prolixus fifth-instar CNS. Hybridization was performed with a probe synthesized using Rhopr-DH31-A variant. The position of the RNA molecular weight markers (MW) is indicated.

Figure 2: Rhopr-DH31 sequence and structure of Rhopr-DH31 splice variants. (A) Nucleotide sequence of Rhopr-DH31-C cDNA and deduced amino acid sequence of the peptide precursor.

The numbering for each sequence is shown at right. The initial methionine start codon has been capitalized. The amino acid sequence for the mature DH31 peptide is shown in bold. The dibasic amino acid pair at the N terminus and the arginine residues at the C terminus that are required for post-translational cleavage are shaded in gray. The glycine residue required for amidation is boxed while the lysine residue predicted to undergo ubiquitination is double underlined. The nucleotides highlighted in black indicate exon-exon boundaries. Potential polyadenylylation signals are underlined and the partial poly(A) tail is dash underlined. The most likely site for cleavage of a signal sequence is indicated by an arrow. (B) Molecular organization of Rhopr-

DH31 based on the BLAST analysis and in silico analysis using intron prediction. The boxes represent exons and the dashes represent introns.

DH31 sequences in other insects

DH31 is found in several insect species; cDNAs encoding DH31 prepropeptides have been cloned and sequenced for several species and the sequence predicted for several others. We examined these prepropeptide sequences and aligned them to observe the conservation across various species (Figure 3A). DH31 sequence from the crustacean, Homarus americanus was also included in the alignment. 50% majority rule was implemented to highlight the identical and similar amino acids across sequences. The alignment shows that the mature peptide is very well conserved across all insects. Interestingly, prepropeptide sequences encoded by Rhopr-DH31-B and Rhopr-DH31-C are considerably longer than other sequences. These two sequences contain a prepropeptide sequence region that is absent in others. This sequence region is present just before the dibasic cleavage site for the mature peptide.

Phylogenetic relationship was inferred using a total of 12 DH31 prepropeptide sequences from 8 insect species and 1 crustacean species (H. americanus), which was utilized as an outgroup

(Figure 3B). Hemimetabolous insect sequences, comprising hemipterans (R. prolixus, A. pisum and N. lugens), form a monophyletic group since it contains all the hemipteran sequences and their common ancestor. DH31 is present in all four major orders of holometabolous insects but only three of them were included in our analysis. Holometabolous insect sequences, comprising dipterans (D. melanogaster, A. gambiae, A. aegypti), hymenopterans (N. vitripennis), and lepidopterans (B. mori), form a paraphyletic group since it contains all the insect sequences except for the hemipteran sequences. This observation differs from that of insect phylogeny determined using nuclear gene sequences (Savard et al., 2006), where the holometabolous insects form one monophyletic group and hemimetabolous insects form another monophyletic group.

Bootstrap support is fairly low within the clade comprising the hemipteran sequences (>66) but

38 very high within the clade comprising only the dipteran sequences (>96). Moreover, the bootstrap support for the relationships between the different orders is variable, with values ranging from 88 to those less than 50.

Figure 3: Multiple sequence alignment of the DH31 prepropeptide from different insect orders and a crustacean and its corresponding phylogenetic tree. (A) ClustalX alignment of the DH31 prepropeptide from the parasitic wasp, N. vitripennis (XP_001599948.1), the kissing bug, R. prolixus (transcript variant A - HM030716, transcript variant B - HM030715 and transcript variant C - HM030714), the brown planthopper, N. lugens (DB826761), the pea aphid, A. pisum

(XP_001945901.1), the silk worm, B. mori (NP_001124379.1), the fruit fly, D. melanogaster

(transcript variant A - NM_078790.3 and transcript variant C - NM_164825.2), the malaria mosquito, A. gambiae (EAA01397.3 / XP_321755.3) and the yellow fever mosquito, A. aegypti

(EAT40182.1). The sequence from the red lobster, H. americanus (GQ290461.1) is also included in the alignment. Following the 50% majority rule, identical amino acids are highlighted in black and similar amino acids are highlighted in gray. (B) A phylogram showing the evolutionary relationships of the 12 taxa inferred using neighbour-joining analysis. H. americanus taxon is utilized as an outgroup. Boot strap values (only those above 50) obtained on performing the boot strap test with 1000 replicates are indicated where the branches split. The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree.

Spatial expression of Rhopr-DH31 in fifth-instars

Although Rhopr-DH31-like immunoreactivity is present in the CNS, salivary glands, hindgut, anterior dorsal vessel and dorsal diaphragm of fifth-instar R. prolixus, immunoreactive cell bodies are present only in the CNS (Te Brugge et al., 2005; Te Brugge et al., 2008). To determine the spatial expression pattern of Rhopr-DH31, RT-PCR tissue profiling was performed

(n = 3 independent trials). RT-PCR results demonstrate that Rhopr-DH31 is most likely expressed only within the CNS (Figure 4) in unfed fifth instars. The expected size of the PCR product for

Rhopr-DH31 was between 317bp and 428bp depending on the type of variant being expressed in the tissue. All three transcript variants were observed in CNS, confirming that all are expressed in CNS of fifth-instar R. prolixus. In contrast, no PCR product was detected in other tissues in all three trials, except for salivary glands, where a very faint band corresponding to the expected product size for Rhopr-DH31-A was observed in one of the three trials (not shown). A 229bp

Rhopr-RP49 fragment was amplified as a positive control to assess the quality of the template cDNA. Bands of high intensity and of expected size were observed for all groups of tissues.

Hence the difference in RT-PCR result observed across the various groups of tissues is due to a difference in Rhopr-DH31 expression and not due to the template used for the reaction.

Figure 4: RT-PCR of Rhopr-DH31 in fifth-instar R. prolixus. Tissues were dissected from at least

10 insects and pooled together into four categories: 1) CNS, 2) salivary glands, 3) oesophagus, midgut, hindgut and Malpighian tubules, and 4) remaining tissues (trachea, abdominal nerves, immature testes and ovaries, dorsal and ventral diaphragm and fat bodies). PCR product is strongly observed in the CNS. Rhopr-RP49 was also amplified as a positive control to test the quality of the template cDNA. (n = 3 biological replicates).

Relative expression of the three Rhopr-DH31 variants

To determine the relative expression of the three variants, semi-quantitative Southern blot analysis was performed (Figure 5A). Following a 35 cycle PCR amplification, Rhopr-DH31-A variant had an average relative expression of 31.5%, Rhopr-DH31-B variant of 52.6% and Rhopr-

DH31-C of 15.9% (n = 3 independent trials) (Figure 5B). Two-tailed t-test shows that the relative expressions of the three transcript variants are significantly different from each other. In addition, a number of different cycles (26, 28, 30, 32 and 35 cycles) were chosen for analysis

(not shown) to ensure that the results were consistent over the exponential amplification phase of

PCR and not an artifact of end-point product inhibition.

Rhopr-DH31 expression in fifth-instar CNS

The cell-specific expression of Rhopr-DH31 in fifth-instar R. prolixus was localized using digoxigenin-labeled anti-sense RNA probe (Figure 6A-E). Rhopr-DH31 expression is observed in approximately 46 cells of the CNS. No staining is observed in these cells in the control preparations with the digoxigenin-labeled sense RNA probe (not shown). On examining Rhopr-

DH31 expression in all the components of the CNS which includes the brain, suboesophageal ganglion (SOG), prothoracic ganglion (PRO) and mesothoracic ganglionic mass (MTGM), it can be seen that the gene expression localization is basically organized into bilaterally symmetrical groups or pairs of cells. Within the dorsal brain, 18 cells containing the Rhopr-DH31 transcript(s) are faintly to moderately stained (Figure 7). Rhopr-DH31 expression is also observed in 6 moderately-stained cells in the ventral SOG, and in 4 bilaterally-paired cells in the ventral PRO

(Figure 7). Rhopr-DH31 expression is observed in 16 cells within the ventral MTGM (Figure 7), with a group of 6 strongly-stained cells (Figure 6D, 6E and 7) localized along the midline in the posterior MTGM; these cells stain the strongest and are observed in almost all the preparations.

However, these cells were variable in number, with few a preparations staining either 5 or 7 cells. Lastly, Rhopr-DH31 expression is observed in a pair of cells in the dorsal MTGM (Figure

7); these cells were very faintly stained and were observed in a few preparations only.

Figure 5: Relative expression of Rhopr-DH31 variants in fifth-instar CNS. (A) Southern blot analysis of RT-PCR amplified Rhopr-DH31 variants. (B) Semi-quantitative Southern blot analysis of Rhopr-DH31 showing the average relative expression of the three variants. The average expression is based on three independent trials. Bars indicate the standard error. Two tailed t-test (P < 0.05) comparison of arcsine transformed data indicate that the average relative expression of each variant is significantly different from others (denoted by letters a, b and c).

Figure 6: Fluorescence in situ hybridization (FISH) showing Rhopr-DH31 expression in whole mount preparations of (A) dorsal brain, (B) ventral sub-oesophageal ganglion, (C) ventral prothoracic ganglion, (D) ventral mesothoracic ganglionic mass (MTGM) and (E) higher magnification of ventral MTGM of fifth-instar R. prolixus. Where it is unclear, cell-specific staining is indicated by arrows. Scale bars: A, B, C = 50μm, D =100μm and E =20μm.

Figure 7: Diagrammatic representation of Rhopr-DH31 transcripts expression (A) ventral aspect of the CNS (brain, suboesophageal ganglion (SOG), prothoracic ganglion (PRO), and mesothoracic ganglionic mass (MTGM). (B) Dorsal aspect of the CNS. Map based on 20 preparations. Scale bar: 200μm.

Discussion

We have isolated and characterized a calcitonin-like DH31 from R. prolixus. This is the first study to identify and isolate three DH31 splice variants in an arthropod species. A modified approach was adopted to obtain the complete cDNA sequences for the three variants.

Surprisingly, the 3’ UTR contains 9 polyadenylation signal consensus sequences. It is fairly unlikely that all these 9 consensus sequences are functional and are most likely present due to the

3’ UTR being highly AT-rich. However, PCRs to clone the Rhopr-DH31 3’ UTR resulted in two sequences of varying lengths - a larger sequence (presented in Figure 2A) and a shorter sequence. Cleavage of the transcript occurs approximately 10 to 30 nucleotides downstream of the polyadenylation signal consensus sequence (see Zhao et al., 1999). There is a polyadenylation consensus sequence 19 nucleotides upstream of the larger sequence and there are two consensus sequences between 16 and 27 nucleotides upstream of the shorter sequence.

Also, the sequence preceding the poly(A) tail of this shorter sequence consists of 6 adenosine nucleotides. Therefore, the shorter version of the sequence could be another result of unspecific binding of the oligo(dT) primer or the two sequences of varying lengths could be a result of alternative polyadenylation, similar to the mammalian calcitonin and calcitonin gene-related peptide (CGRP) gene processing (Amara et al., 1982; Sabate et al., 1985). The mammalian calcitonin gene undergoes alternative splicing to generate two products – calcitonin and CGRP, both of which have a different poly(A) site. It is unclear if the three Rhopr-DH31 transcripts have different poly(A) sites or if any possible effects result from the alternate use of poly(A) site in R. prolixus.

Recently, Ons et al. submitted R. prolixus CLDH ORF sequences of two isoforms (isoform A -

GQ856316 and isoform B - GQ856317) to the GenBank database. Isoform A is identical to

Rhopr-DH31-C and isoform B is identical to Rhopr-DH31-A. However, the Rhopr-DH31-A and

Rhopr-DH31-C sequences presented here also include the additional 3’ UTR and 5’ UTR sequences. These isoforms are indeed splice variants as confirmed by our analyses. Moreover,

BLAST search results indicate that Rhopr-DH31 may have undergone partial gene duplication

(Hu and Worton, 1992). Specifically, Rhopr-DH31-C cDNA sequence was used as a query to perform a BLAST search of the R. prolixus preliminary genome assembly. Interestingly, 2 unique highly-similar hits, each located on a different supercontig, were returned for exons 5 and

6, suggesting that at least a portion of the gene is duplicated. Closer examination of the three transcript variants illustrates that residues present in the Rhopr-DH31-C variant are missing in

Rhopr-DH31-A and Rhopr-DH31-B. A gap is also present in the multiple sequence alignment of the DH31 prepropeptide sequences from R. prolixus and other invertebrate species. The gap is present just before the sequence coding for the mature peptide, and is due to Rhopr-DH31-B and

Rhopr-DH31-C variants being longer than other sequences. Comparing this result with the molecular organization of Rhopr-DH31 shows that exons 4 and 5 are responsible for the gap observed in the alignment. Phylogenetic analysis of DH31 prepropeptide sequences indicates that

Rhopr-DH31-A is more closely related to the N. lugens sequence compared to the other two variants. This suggests that Rhopr-DH31-A variant, which lacks exons 4 and 5, is the ancestral version of the gene, and the Rhopr-DH31-B and Rhopr-DH31-C variants possibly result from

DNA insertion caused by an event such as horizontal gene transfer or through DNA transposons

(McClintock, 1950; Syvanen, 1985).

Tissue-specific expression of the three Rhopr-DH31 splice variants could explain their presence in R. prolixus, especially since each variant is predicted to generate the same mature Rhopr-DH31 peptide through post-translational modification and processing (see Veenstra, 2000). Although

Rhopr-DH31 variants are most likely expressed only within the CNS, it does not eliminate the possibility of non-neuronal expression of this gene as is found in D. melanogaster midgut

(Veenstra, 2009). The presence of a very faint product corresponding to Rhopr-DH31-A in salivary glands for only one of the three RT-PCR trials (not shown) suggests that Rhopr-DH31-A expression in salivary glands is very low compared to that of CNS. Likewise, one or more

Rhopr-DH31 splice variants could be expressed in other tissues but their expression is too low to be detected. Furthermore, it is important to keep note of the fact that we are only examining one particular stage of R. prolixus development (6 weeks post-fed fifth-instars). Therefore Rhopr-

DH31 spatial expression pattern could be different when examined at other life stages.

The three Rhopr-DH31 variants significantly differ in their relative expression within the CNS.

Since the 5’ end of all the three variants is similar, they probably share the same promoter. Thus, the difference in relative expression of the three splice variants could be due to the variants being expressed in a cell-specific manner within the CNS. This could be beneficial in regulating the amount of mature peptide synthesized in different neurons since increased amounts of mRNA would most likely result in increased amounts of protein. Moreover, the difference in relative expression could be due to the difference in transcripts generated from alternative splicing. As mentioned earlier, exon 4, present in Rhopr-DH31-B and exons 4 and 5, present in Rhopr-DH31-C result in these transcripts being longer than their homologs from other insect species. The propeptide region coded by exons 4 and 5 is not similar to any known biologically active peptides (as determined by BLAST search). However, specific lysine residues within these peptide sequences could undergo various post-translational modifications such as ubiquitination or SUMOylation which would affect peptide stability (see Haglund and Dikic, 2005; see Hay,

2005; see Hochstrasser, 1995). Specifically, the presence of exon 4 in Rhopr-DH31-B and Rhopr-

DH31-C results in a lysine residue at position 68 instead of asparagine residue at the same position in Rhopr-DH31-A, which could be critical since this lysine is predicted to be ubiquitinated (Radivojac et al., 2010).

FISH was performed to localize Rhopr-DH31 expression within the CNS of fifth-instar R. prolixus. Comparing these results with the immunohistochemical localization of Rhopr-DH31- like peptide (Te Brugge et al., 2005), additional cells stain with the immunohistochemical procedure than are observed with FISH. Specifically, Rhopr-DH31-like immunoreactivity is observed in cells in the ventral brain, dorsal SOG and dorsal PRO where Rhopr-DH31 expression is not observed. Rhopr-DH31 expression was also not observed in dorsal unpaired median (DUM) neurons in the MTGM, although it must be noted that these stain quite weakly with immunohistochemistry. Rhopr-DH31 could yet still be expressed in DUM neurons; however, its expression appears to be too low to be detected using the approach and stage used here in this study. In terms of the similarities between the immunohistochemical localization and FISH, most of the cell-specific staining observed with FISH is also observed using immunohistochemistry.

In particular, on the posterior ventral surface of MTGM, 6 strongly immunoreactive bilaterally- paired cells also demonstrate Rhopr-DH31 expression. These cells are very strongly stained and observed in almost all the preparations. However, the numbers of these cells varied as some preparations contained either 5 or 7 cells instead of 6. In addition, consistent with the immunohistochemical localization, Rhopr-DH31 expression is also localized to lateral and medial neurosecretory cells in the dorsal brain. This provides evidence that DH31 produced by these cells might be released into the haemolymph via the corpus cardiacum for action as a neurohormone.

DH31 is present in several arthropod species. Phylogenetic analysis of DH31 prepropeptide sequences indicates that the R. prolixus sequence is most similar to that of N. lugens. It is interesting that holometabolous insect CLDH sequences form a paraphyletic group rather than a monophyletic group as inferred using nuclear gene sequences (Savard et al., 2006). All the dipteran sequences and hemipteran sequences form distinct clades, each with strong bootstrap support. Moreover, the predicted Rhopr-DH31 mature peptide is 100% identical to Dippu-DH31 and Apime-DH31 peptides. Within the DH31 sequence proper, 13 out of the 31 (or 34) amino acid residues are identical in sequences from all the species observed. Specifically, the C-terminal domain of the peptide sequence shows high identity across various species, which follow varying developmental patterns and utilize different feeding strategies. The high conservation observed for this peptide across various species suggests that apart from being involved in processes associated with diuresis, it could play other crucial roles in arthropods which are still unclear.

In conclusion, the complete cDNA sequences of three Rhopr-DH31 splice variants have been obtained in R. prolixus. Although it is evident from in situ hybridization that Rhopr-DH31 is expressed in the CNS, it is still unclear which cells within the CNS express the different transcripts. It is evident that CLDHs are multi-functional peptides and have a wide array of functions in arthropods including increasing Malpighian tubules secretion, and hindgut and dorsal vessel contractility.

References

Altschul, S.F., Gish, W., Miller, W., Myers, E.W. and Lipman, D.J. (1990). Basic local alignment search tool. J Mol Biol. 215, 403-10.

Amara, S.G., Jonas, V., Rosenfeld, M.G., Ong, E.S. and Evans, R.M. (1982). Alternative RNA processing in calcitonin gene expression generates mRNAs encoding different polypeptide products. Nature. 298, 240-4.

Bendtsen, J.D., Nielsen, H., von Heijne, G. and Brunak, S. (2004). Improved prediction of signal peptides: SignalP 3.0. J Mol Biol. 340, 783-95.

Christie, A.E. (2008). Neuropeptide discovery in Ixodoidea: an in silico investigation using publicly accessible expressed sequence tags. Gen Comp Endocrinol. 157, 174-85.

Christie, A.E., Stevens, J.S., Bowers, M.R., Chapline, M.C., Jensen, D.A., Schegg, K.M., Goldwaser, J., Kwiatkowski, M.A., Pleasant, T.K., Jr., Shoenfeld, L., Tempest, L.K., Williams, C.R., Wiwatpanit, T., Smith, C.M., Beale, K.M., Towle, D.W., Schooley, D.A. and Dickinson, P.S. (2010). Identification of a calcitonin-like diuretic hormone that functions as an intrinsic modulator of the American lobster, Homarus americanus, cardiac neuromuscular system. J Exp Biol. 213, 118-27.

Coast, G.M., Garside, C.S., Webster, S.G., Schegg, K.M. and Schooley, D.A. (2005). Mosquito natriuretic peptide identified as a calcitonin-like diuretic hormone in Anopheles gambiae (Giles). J Exp Biol. 208, 3281-91.

Coast, G.M., Orchard, I., Phillips, J.E. and Schooley, D.A., 2002. Insect diuretic and antidiuretic hormones, in: Evans, P.D. (Ed.), Adv Insect Physiol. Academic Press Ltd., London, pp. 279-409.

Coast, G.M., Webster, S.G., Schegg, K.M., Tobe, S.S. and Schooley, D.A. (2001). The Drosophila melanogaster homologue of an insect calcitonin-like diuretic peptide stimulates V- ATPase activity in fruit fly Malpighian tubules. J Exp Biol. 204, 1795-1804.

Donini, A., O'Donnell, M.J. and Orchard, I. (2008). Differential actions of diuretic factors on the Malpighian tubules of Rhodnius prolixus. J Exp Biol. 211, 42-8.

Dujardin, J.P., Munoz, M., Chavez, T., Ponce, C., Moreno, J. and Schofield, C.J. (1998). The origin of Rhodnius prolixus in Central America. Med Vet Entomol. 12, 113-115.

Feliciangeli, M.D., Carrasco, H., Patterson, J.S., Suarez, B., Martinez, C. and Medina, M. (2004). Mixed domestic infestation by Rhodnius prolixus Stal, 1859 and Panstrongylus geniculatus Latreille, 1811, vector incrimination, and seroprevalence for Trypanosoma cruzi among inhabitants in El Guamito, Lara State, Venezuela. Am J Trop Med Hyg. 71, 501-505.

Gard, A.L., Lenz, P.H., Shaw, J.R. and Christie, A.E. (2009). Identification of putative peptide paracrines/hormones in the water flea Daphnia pulex (Crustacea; Branchiopoda; Cladocera) using transcriptomics and immunohistochemistry. Gen Comp Endocrinol. 160, 271- 87.

Haglund, K. and Dikic, I. (2005). Ubiquitylation and cell signaling. EMBO J. 24, 3353-9.

Hay, R.T. (2005). SUMO: a history of modification. Mol Cell. 18, 1-12.

Hochstrasser, M. (1995). Ubiquitin, proteasomes, and the regulation of intracellular protein degradation. Curr Opin Cell Biol. 7, 215-23.

Hu, X. and Worton, R.G. (1992). Partial gene duplication as a cause of human disease. Hum Mutat. 1, 3-12.

Kean, L., Cazenave, W., Costes, L., Broderick, K.E., Graham, S., Pollock, V.P., Davies, S.A., Veenstra, J.A. and Dow, J.A. (2002). Two nitridergic peptides are encoded by the gene capability in Drosophila melanogaster. Am J Physiol Regul Integr Comp Physiol. 282, R1297- 307.

Koberle, F. (1968). Chagas' disease and Chagas' syndromes: the pathology of American trypanosomiasis. Adv Parasitol. 6, 63-116. 59

Li, B., Predel, R., Neupert, S., Hauser, F., Tanaka, Y., Cazzamali, G., Williamson, M., Arakane, Y., Verleyen, P., Schoofs, L., Schachtner, J., Grimmelikhuijzen, C.J. and Park, Y. (2008). Genomics, transcriptomics, and peptidomics of neuropeptides and protein hormones in the red flour beetle Tribolium castaneum. Genome Res. 18, 113-122.

Maddrell, S.H., Herman, W.S., Mooney, R.L. and Overton, J.A. (1991). 5- Hydroxytryptamine: a second diuretic hormone in Rhodnius prolixus. J Exp Biol. 156, 557-566.

McClintock, B. (1950). The origin and behavior of mutable loci in maize. Proc Natl Acad Sci U S A. 36, 344-55.

Monroy, C., Rodas, A., Mejia, M., Rosales, R. and Tabaru, Y. (2003). Epidemiology of Chagas disease in Guatemala: infection rate of Triatoma dimidiata, Triatoma nitida and Rhodnius prolixus (Hemiptera, Reduviidae) with Trypanosoma cruzi and Trypanosoma rangeli (Kinetoplastida, Trypanosomatidae). Mem Inst Oswaldo Cruz. 98, 305-310.

Paluzzi, J.P. and Orchard, I. (2010). A second gene encodes the anti-diuretic hormone in the insect, Rhodnius prolixus. Mol Cell Endocrinol. 317, 53-63.

Paluzzi, J.P., Russell, W.K., Nachman, R.J. and Orchard, I. (2008). Isolation, cloning, and expression mapping of a gene encoding an antidiuretic hormone and other CAPA-related peptides in the disease vector, Rhodnius prolixus. Endocrinology. 149, 4638-4646.

Prata, A. (2001). Clinical and epidemiological aspects of Chagas disease. Lancet Infect Dis. 1, 92-100.

Radivojac, P., Vacic, V., Haynes, C., Cocklin, R.R., Mohan, A., Heyen, J.W., Goebl, M.G. and Iakoucheva, L.M. (2010). Identification, analysis, and prediction of protein ubiquitination sites. Proteins. 78, 365-80.

Ramsey, J.M., Ordonez, R., Cruz-Celis, A., Alvear, A.L., Chavez, V., Lopez, R., Pintor, J.R., Gama, F. and Carrillo, S. (2000). Distribution of domestic triatominae and stratification of Chagas Disease transmission in Oaxaca, Mexico. Med Vet Entomol. 14, 19-30.

Reese, M.G., Eeckman, F.H., Kulp, D. and Haussler, D. (1997). Improved splice site detection in Genie. J Comput Biol. 4, 311-323.

Sabate, M.I., Stolarsky, L.S., Polak, J.M., Bloom, S.R., Varndell, I.M., Ghatei, M.A., Evans, R.M. and Rosenfeld, M.G. (1985). Regulation of neuroendocrine gene expression by alternative RNA processing. Colocalization of calcitonin and calcitonin gene-related peptide in thyroid C- cells. J Biol Chem. 260, 2589-92.

Savard, J., Tautz, D., Richards, S., Weinstock, G.M., Gibbs, R.A., Werren, J.H., Tettelin, H. and Lercher, M.J. (2006). Phylogenomic analysis reveals bees and wasps (Hymenoptera) at the base of the radiation of Holometabolous insects. Genome Res. 16, 1334-8.

Schooley, D.A., Horodyski, F.M. and Coast, G.M., 2005. Hormones controlling homeostasis in insects: Endocrinology, in: Gilbert, L.I. et al. (Eds.), Comprehensive Molecular Insect Science. Elsevier, pp. 493-550.

Syvanen, M. (1985). Cross-species gene transfer; implications for a new theory of evolution. J Theor Biol. 112, 333-43.

Tamura, K., Dudley, J., Nei, M. and Kumar, S. (2007). MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Mol Biol Evol. 24, 1596-9.

Te Brugge, V.A., Ianowski, J.P. and Orchard, I. (2009). Biological activity of diuretic factors on the anterior midgut of the blood-feeding bug, Rhodnius prolixus. Gen Comp Endocrinol. 162, 105-112.

Te Brugge, V.A., Lombardi, V.C., Schooley, D.A. and Orchard, I. (2005). Presence and activity of a Dippu-DH31-like peptide in the blood-feeding bug, Rhodnius prolixus. Peptides. 26, 29-42.

Te Brugge, V.A., Miksys, S.M., Coast, G.M., Schooley, D.A. and Orchard, I. (1999). The distribution of a CRF-like diuretic peptide in the blood-feeding bug Rhodnius prolixus. J Exp Biol. 202, 2017-2027.

Te Brugge, V.A. and Orchard, I. (2008). Distribution and activity of a Dippu DH31-like peptide in the large milkweed bug Oncopeltus fasciatus. Peptides. 29, 206-13.

Te Brugge, V.A., Schooley, D.A. and Orchard, I. (2002). The biological activity of diuretic factors in Rhodnius prolixus. Peptides. 23, 671-681.

Te Brugge, V.A., Schooley, D.A. and Orchard, I. (2008). Amino acid sequence and biological activity of a calcitonin-like diuretic hormone (DH31) from Rhodnius prolixus. J Exp Biol. 211, 382-90.

Veenstra, J.A. (2000). Mono- and dibasic proteolytic cleavage sites in insect neuroendocrine peptide precursors. Arch Insect Biochem Physiol. 43, 49-63.

Veenstra, J.A. (2009). Peptidergic paracrine and endocrine cells in the midgut of the fruit fly maggot. Cell Tissue Res. 336, 309-323.

Zhao, J., Hyman, L. and Moore, C. (1999). Formation of mRNA 3' ends in eukaryotes: mechanism, regulation, and interrelationships with other steps in mRNA synthesis. Microbiol Mol Biol Rev. 63, 405-45.

Acknowledgements

We would like to thank Victoria Te Brugge for sharing the DH31 peptide sequence prior to its publication. This research was funded through an NSERC Discovery Grant to I.O.

Appendices

Table S1: Forward and reverse primers used to obtain the partial Rhopr-DH31 cDNA sequence.

The locations of the primers along the gene are also indicated and are based upon the nucleotide sequence numbering presented in Figure 2.

Primer Sequence Tm Location

(°C)

Degenerate forward primers

 dh31forward1 GCNGCNAAYTAYGCNGG 54.1 485-501

 dh31forward2 GCNAAYTAYGCNGGNGG 53.8 488-504

 dh31forward3 TAYGCNGGNGGNCCRGG 60.7 494-510

 dh31forward4 TAYGCNGGNGGNCCYGG 60.7 494-510

Plasmid reverse primers

 DNR-LIB REV110 GCACTAGTCATACCAGGA 53.0 -

 DNR-LIB REV77 GCTATGACCATGTTCACTTACCT 60.8 -

Table S2: Forward and reverse primers used to perform the 5’ PCR RACE reactions. The locations of the primers along the gene are also indicated and are based upon the nucleotide sequence numbering presented in Figure 2.

Primer Sequence Tm Location

(°C)

Plasmid forward primers

 DNR-LIB FOR1 GTGGATAACCGTATTACCGCC 63.9 -

 DNR-LIB FOR2 ACGGTACCGGACATATGCC 64.5 -

Gene-specific reverse primers

 DH31 5race1 CGAAGAGAAAGGAATTTTAGGCG 65.6 534-556

 DH31 5race2 GCTTGGCGCCTCCTTC 64.9 513-528

 DH31 5race3 TGCCTGGGCCTCCTG 65.2 498-512

 DH31 5race4 CAAGATCCAAACCACGCTTC 64.5 409-428

Table S3: Forward and reverse primers used to perform the 3’ PCR RACE reactions. The locations of the primers along the gene are also indicated and are based upon the nucleotide sequence numbering presented in Figure 2.

Primer Sequence Tm Location

(°C)

Gene-specific forward primers

 FOR1DH31 CACTATATCGCCGGCAGTC 63.2 1-19

 FOR2DH31 CCGAGCCCTACTAACTGCATC 64.7 39-59

 FOR3DH31 TCGAAGCGTGGTTTGGAT 64.3 407-424

 FOR4DH31 TGGATCACAGGCTGCC 62.5 448-463

Plasmid reverse primer

 pDNR-LIB 3 – 25 REV GCCAAACGAATGGTCTAGAAAG 63.6 -

Table S4: Forward and reverse primers used to obtain the 3’ UTR sequence. The locations of the primers along the gene are also indicated and are based upon the nucleotide sequence numbering presented in Figure 2.

Primer Sequence Tm Location

(°C)

Gene-specific forward primers

 FOR3UTR GATTCCATTTACTCGAAGC 56.2 395-413

 FOR7DH31 TCATTTTGCATATACATAACTCCCTTAGC 65.4 957-985

 FOR8DH31 CAAACAGTTTCTACGAATAACAATAGCG 65.2 1054-

1081

Gene-specific reverse primer

 DH31-3UTR-6 GGTATAAATTCCATAGAAAGGG 57.1 1422-

1443

 DH31-3UTR-10 CAAAAGATAGACCCAAAGATAACATGC 65.2 1384-

1410

Chapter 3: Isolation and functional characterization of

calcitonin-like diuretic hormone receptors in Rhodnius

prolixus

The proceeding chapter was published in PLoS One. No permission to reproduce the paper as a part of this dissertation was required as per the PLoS One author rights and permissions policy.

Full citation details:

Isolation and functional characterization of calcitonin-like diuretic hormone receptors in

Rhodnius prolixus.

Zandawala M, Li S, Hauser F, Grimmelikhuijzen CJ, Orchard I.

PLoS One. 2013 Nov 29;8(11):e82466. doi: 10.1371/journal.pone.0082466.

Abstract

Several families of diuretic hormones exist in insects, one of which is the calcitonin-like diuretic hormone (CT/DH) family. CT/DH mediates its effects by binding to family B G-protein coupled receptors (GPCRs). Here we isolate and functionally characterize two R. prolixus CT/DH receptor paralogs (Rhopr-CT/DH-R1 and Rhopr-CT/DH-R2) using a novel heterologous assay utilizing a modified human embryonic kidney 293 (HEK293) cell line. Rhopr-CT/DH-R1 is orthologous to the previously characterized D. melanogaster CT/DH receptor (CG17415) while

Rhopr-CT/DH-R2 is orthologous to the D. melanogaster receptor (CG4395), an orphan receptor whose ligand was unknown until now. We determine the cDNA sequences of three splice variants encoding Rhopr-CT/DH-R1 (Rhopr-CT/DH-R1-A, Rhopr-CT/DH-R1-B and Rhopr-

CT/DH-R1-C) and two splice variants encoding Rhopr-CT/DH-R2 (Rhopr-CT/DH-R2-A and

Rhopr-CT/DH-R2-B). Rhopr-CT/DH-R1-A and Rhopr-CT/DH-R2-A encode truncated receptors that lack six and seven of the characteristic seven transmembrane domains, respectively. Rhopr-

CT/DH-R1-B and Rhopr-CT/DH-R1-C, which only differ by 2 amino acids in their C-terminal domain, can both be activated by Rhopr-CT/DH at equal sensitivities (EC50 = 200-300nM).

Interestingly, Rhopr-CT/DH-R2-B is much more sensitive to Rhopr-CT/DH (EC50 = 15nM) compared to Rhopr-CT/DH-R1-B/C and also yields a much greater response (amplitude) in our heterologous assay. This is the first study to reveal that insects possess at least two CT/DH receptors, which may be functionally different. Quantitative PCR demonstrates that Rhopr-

CT/DH-R1 and Rhopr-CT/DH-R2 have distinct expression patterns, with both receptors expressed centrally and peripherally. Moreover, the expression analysis also identified novel target tissues for this neuropeptide, including testes, ovaries and prothoracic glands, suggesting a possible role for Rhopr-CT/DH in reproductive physiology and development.

Introduction

Various neurohormone families have been implicated in regulating diuresis in insects. One such family is the calcitonin-like diuretic hormone (CT/DH) family which is related to the mammalian calcitonin and calcitonin gene-related peptide hormonal system. The first member of this peptide family in insects was isolated and functionally characterized in Diploptera punctata (Furuya et al., 2000). This peptide was originally termed diuretic hormone 31 (DH31) due to its ability to stimulate Malpighian tubule (MT) secretion in certain insects and due to the fact that it is comprised of 31 amino acids (Coast et al., 2005; Coast et al., 2001; Furuya et al., 2000). As is the case with many peptides that are named because of a particular bioassay involved in their isolation, regulating diuresis may not be their function in other insects. Thus, CT/DHs do not stimulate MT secretion in Acrosternum hilare and Podisus maculiventris (Coast et al., 2011).

Moreover, the role of Rhodnius prolixus CT/DH (Rhopr-CT/DH) in diuresis is also questionable as it does not stimulate water reabsorption across the midgut, and only stimulates MT secretion at a rate which is 1.5% of maximum (Te Brugge et al., 2009; Te Brugge et al., 2005); however, it may play a broad role in feeding-related physiological events in various insects. For example,

Rhopr-CT/DH has been shown to have myostimulatory effects on hindgut, dorsal vessel and salivary glands whereas the Drosophila melanogaster CT/DH is required for peristalsis in the larval midgut (Brugge et al., 2008; LaJeunesse et al., 2010; Orchard, 2009). Furthermore, D. punctata CT/DH analogs have anorexigenic effects in Locusta migratoria nymphs (Kaskani et al., 2012). It is thus evident that CT/DHs, like several other neuropeptides, are pleiotropic in nature. Hence, in order to elucidate additional physiological roles for these hormones, it is important to identify and characterize their receptors and determine their expression patterns.

Insect CT/DH receptors (CT/DH-Rs) belong to the family of secretin-like (family B 1) G-protein coupled receptors (GPCRs) (Hewes and Taghert, 2001). Johnson et al. characterized the first insect CT/DH-R from D. melanogaster in 2005 (Johnson et al., 2005). Signaling through this receptor was shown to be dependent on accessory proteins (receptor activity modifying proteins

(RAMPs) and receptor component protein (RCP)), in a manner analogous to mammals (Evans et al., 2000; McLatchie et al., 1998). Recently, a receptor orthologous to this was functionally characterized in Aedes aegypti (AedaeGPCRCAL1) via RNAi-based knockdown (Kwon et al.,

2012; Kwon and Pietrantonio, 2013). RNAi treated females showed a 30% reduction in fluid excretion (relative to control groups) following a blood meal and the hindguts exhibited a 50% reduction in contraction frequency in response to A. aegypti CT/DH (Aedae-CT/DH) compared to controls. Moreover, a 57% decrease in fluid secretion in response to Aedae-CT/DH was also observed in MTs in which AedaeGPCRCAL1 was knocked-down (Kwon et al., 2012).

In the present study, we have isolated and characterized a CT/DH-R from R. prolixus that is orthologous to the previously characterized CT/DH-Rs in D. melanogaster and A. aegypti

(Johnson et al., 2005; Kwon et al., 2012). We propose to rename these receptors as CT/DH-R1.

Moreover we have also isolated and characterized another family B1 GPCR from R. prolixus that is orthologous to the D. melanogaster receptor (CG4395), hector. This orphan receptor is also related to insect CT/DH-Rs and is activated by Rhopr-CT/DH. Hence, we propose to name these receptors as CT/DH-R2. Rhopr-CT/DH-R2 is much more sensitive to Rhopr-CT/DH compared to Rhopr-CT/DH-R1 in our heterologous assay utilizing human embryonic kidney (HEK)-293 cells stably expressing a modified cyclic nucleotide-gated (CNG) channel (HEK293/CNG). We obtained robust and sensitive responses in these cells without having to co-express any accessory proteins, making it ideal to study CT/DH-Rs and, perhaps, deorphanize other family B1 GPCRs.

To our knowledge, this is the first study to reveal that insects possess at least two CT/DH receptors, which may be functionally different. Quantitative PCR demonstrates that Rhopr-

CT/DH-R1 and Rhopr-CT/DH-R2 have distinct expression patterns, with both receptors expressed centrally and peripherally.

Material and methods

Animals

Fifth-instar and adult R. prolixus (4-5 weeks post-feeding) were raised in a long standing colony that was maintained in incubators at 60% humidity and 25°C. The insects were routinely fed artificially once in each instar on defibrinated rabbit blood (Hemostat Laboratories, Dixon, CA,

USA; supplied by Cedarlane Laboratories Inc., Burlington, ON, Canada).

Isolation of cDNA sequences encoding R. prolixus CT/DH receptors

Supercontigs in FASTA format, representing the R. prolixus preliminary genome assembly (June

2009 release), were downloaded from the genome server at The Genome Institute at Washington

University (http://genome.wustl.edu/pub/organism/Invertebrates/Rhodnius_prolixus/). These supercontigs were then imported into Geneious Pro 4.7.6 and used to perform local tBLASTn search, with the D. melanogaster CT/DH receptor (CG17415, accession no: NP_725278.1) protein sequence acting as the query. Hits along two different supercontigs were obtained; these represent two putative CT/DH receptors. Primers specific to the hit regions were designed (see

Appendices: Table S1) and used to amplify the partial cDNA sequence encoding Rhopr-CT/DH-

R1 and Rhopr-CT/DH-R2. Template for the PCR was cDNA synthesized using total RNA extracted from individually-dissected tissues (see below). PCR was performed using s1000 thermal cycler (Bio-Rad Laboratories, Mississauga, ON, Canada) with a temperature-cycling

72 profile that consisted of an initial denaturation (94°C for 3 min) and 35 cycles of denaturation

(94°C for 30 sec), annealing (59°C for 30 sec) and extension (72°C for 1 min); a final 10 min extension at 72°C was also included. Gel electrophoresis was used to visualize the PCR product which was then extracted using the EZ-10 Spin Column DNA Gel Extraction Kit (Bio Basic Inc.,

Markham, ON, Canada). The gel extracted product was cloned and sequenced using the methods described earlier (Zandawala et al., 2012).

Complete cDNA sequences encoding the two receptors were obtained using a modified 5´ and 3´ rapid amplification of cDNA ends (RACE) PCR technique, as described earlier (Zandawala et al., 2012). Primers used for 5´ and 3´ RACE PCRs have been listed in Table S2 and Table S3, respectively (see Appendices). Lastly, the largest cDNA fragments encoding the receptors were amplified using the primers listed in Table S4 (see Appendices) and a proof-reading Taq polymerase. The PCR products were cloned and sequenced as explained earlier (Zandawala et al., 2012).

Sequence analysis

The intron-exon boundaries were predicted using a combination of a BLAST search of the R. prolixus genome and Genie, an online software for splice site prediction (Reese et al., 1997).

Membrane topology of the receptors was predicted using the Transmembrane Prediction Tool plugin for Geneious. The potential phosphorylation sites were predicted using the NetPhos 2.0

Server (Blom et al., 1999) and the potential N-linked glycosylation sites predicted using the

NetNGlyc 1.0 Server. Clustal Omega (http://www.ebi.ac.uk/Tools/msa/clustalo/ - last accessed on August 1, 2013) was used to align Rhopr-CT/DH-R1 (KC660148, KC660149 and KC660150) and R2 isoforms (KF446640 and KF494337) with homologs from D. melanogaster

(NP_725278.1 and NP_572843.2) and Aedes aegypti (AEU12191.1). The alignment figure was obtained using the BOXSHADE 3.21 server

(http://www.ch.embnet.org/software/BOX_form.html - last accessed on August 1, 2013).

Additional family B1 GPCR amino acid sequences were included for phylogenetic analysis.

These included corticotropin releasing-factor (CRF)-related diuretic hormone (CRF/DH) receptors (CRF/DH-Rs), pigment dispersing factor (PDF) receptors (PDF-Rs) and CT/DH-Rs from a variety of insects (see Figure 5 for the accession numbers and species names). Moreover,

CRF receptors (NP_001138618.1 and NP_001189404.1), calcitonin receptor (CTR)

(NP_001158209.1) and calcitonin receptor-like receptor (CRLR) (NP_001258680.1) from Homo sapiens were also included in the analysis and D. melanogaster metabotropic glutamate receptor

(NP_524639.2) was utilized as an outgroup. ClustalX2 was used to align these sequences and the alignment exported to MEGA5 (Larkin et al., 2007; Tamura et al., 2011). A maximum parsimonious tree was constructed using Close-Neighbor-Interchange (CNI) analysis and the bootstrap values obtained were based on 1000 replicates.

Preparation of expression vectors

The largest cDNA fragments encoding Rhopr-CT/DH-R1 transcript variants and Rhopr-CT/DH-

R2-B were amplified as described earlier (see above). The PCR products from these reactions were used as a template to amplify the ORF and introduce a Kozak translation initiation sequence at the 5´ end using the primers listed in Table S5 (see Appendices). The resulting products were cloned into pGEM-T Easy vector (Promega, Madison, WI, USA). These were then subcloned into either pIRES2-ZsGreen1 (Clontech, Mountain View, CA, USA) or pcDNA

3.1+ (Life Technologies Corporation, Carlsbad, CA, USA) for expression in mammalian cells.

Cell culture and transfections

Human embryonic kidney (HEK)-293 cells stably expressing a modified cyclic nucleotide-gated

(CNG) channel (HEK293/CNG) (previously available through BD Biosciences, Mississauga,

ON, Canada) were used to functionally characterize the receptors (Visegrady et al., 2007).

HEK293/CNG cells were grown in Dulbecco’s Modified Eagle Medium Nutrient Mixture F12-

Ham (DMEM/F12) and supplemented with 10% heat-inactivated fetal bovine serum (FBS), 1% penicillin and streptomycin and 250µg/mL G418 (Life Technologies Corporation, Carlsbad, CA,

USA). The cells were incubated at 37°C in 5% CO2. X-tremeGENE HP DNA transfection reagent (Roche Applied Science, Indianapolis, IN, USA) was used to transiently co-transfect the cells with the expression vectors containing receptor transcript variant and cytoplasmic luminescent reporter aequorin at ratio of 2:1 (transfection reagent to expression vectors) using the manufacturer recommended protocol. For negative control, empty expression vector without any receptor transcript was also used to transfect the cells. Cells were incubated for 48 hours and then used to perform the bioluminescence assay.

Alternatively spliced transcript variants of mammalian calcitonin receptors have been known to dimerize with normal functional receptors and inhibit either their surface expression or ligand- induced intracellular cAMP production (Nag et al., 2007; Seck et al., 2003). To determine if

Rhopr-CT/DH-R1-A interacted with Rhopr-CT/DH-R1-B/C, we utilized Chinese hamster ovary

(CHO) cells stably expressing the human G-protein G16 (CHO/G16) (Stables et al., 1997).

CHO/G16 cells stably expressing either Rhopr-CT/DH-R1-B or Rhopr-CT/DH-R1-C were grown and transfected as described earlier (Collin et al., 2013). These cells were then transiently co-transfected with Rhopr-CT/DH-R1-A and aequorin (Collin et al., 2013).

Bioluminescence assay

Bioluminescence assay using CHO/G16 cells was performed as described previously (Collin et al., 2013; Stables et al., 1997; Staubli et al., 2002). To perform the assay using HEK293/CNG cells, they were first harvested by incubating in a PBS-EDTA solution and resuspended in bovine serum albumin (BSA) media (DMEM/F12 containing 1% BSA and 1% penicillin and streptomycin). Coelenterazine h (Promega, Madison, WI, USA) was then added to the cells at a

5µM final concentration and incubated for 3 hours with stirring in the dark. The cells were then diluted 10-fold using BSA media and used to perform the assay. Various doses of peptides were prepared in BSA media and plated in triplicates across a 96-well plate. Cells were loaded in each well using an automated injector and the luminescence recorded over 20 seconds using a Wallac

Victor2 plate reader (Perkin Elmer, San Diego, CA, USA). Rhopr-CT/DH

(GLDLGLSRGFSGSQAAKHLMGLAAANYAGGPamide) and Rhopr-CRF/DH

(MQRPQGPSLSVANPIEVLRSRLLLEIARRRMKEQDASRVSKNRQYLQQIamide) used for the assay were custom synthesised by GenScript (Piscataway, NJ, USA) at > 95% purity. D. melanogaster PDF (NSELINSLLSLPKNMNDAamide) was custom synthesized by GeneMed

Synthesis (San Antonio, TX, USA) at > 95% purity. Dose-response curves were obtained and the

EC50 values determined using Prism5 software.

Quantitative PCR tissue profiling

The following tissues were individually-dissected from fifth-instar R. prolixus of both sexes and used for spatial expression analysis: (1) CNS, (2) dorsal vessel, (3) fat body, abdominal nerves, diaphragms and trachea, (4) foregut, (5) salivary glands, (6) anterior midgut, (7) posterior midgut, (8) MTs, (9) hindgut, (10) immature testes, (11) immature ovaries and (12) prothoracic glands. The following tissues were dissected from adult R. prolixus to determine the expression

76 pattern in reproductive tissues: (1) testes (2) rest of the male reproductive tissues (3) ovaries and

(4) rest of the female reproductive tissues. Total RNA was isolated from these tissues using

PureLink® RNA Mini Kit (Life Technologies Corporation, Carlsbad, CA, USA) which was then used to synthesize cDNA using iScript™ Reverse Transcription Supermix for RT-qPCR (Bio-

Rad Laboratories Ltd., Mississauga, ON, Canada). This cDNA was diluted 10-fold and subsequently used as a template for the qPCR reaction. Primers specific for each receptor variant were designed over exon-exon boundaries to determine expression levels for each transcript (see

Appendices: Table S6). Since the difference between Rhopr-CT/DH-R1-B and Rhopr-CT/DH-

R1-C cDNA sequences was minor (see below), primers differentiating these two transcripts could not be designed. The primer efficiencies for each target were calculated and delta-delta Ct method was used to determine the relative expression of each transcript. Geometric averaging of the transcript levels of three housekeeping genes (alpha-tubulin, beta-actin and ribosomal protein

49) was used to normalize the expression levels of the receptor transcripts. Experiments were performed using MX4000 Quantitative PCR System (Stratagene, Mississauga, ON, Canada) with a temperature-cycling profile that consisted of an initial denaturation (95°C for 30 sec) and 40 cycles of denaturation (95°C for 5 sec) and annealing/extension (60°C for 24 sec); this was followed by a melt curve analysis (60°C - 95°C). SsoFastTM EvaGreen® Supermix with Low

ROX (Bio-Rad Laboratories Ltd., Mississauga, ON, Canada) was used to perform all experiments, which included a no template control and 2 technical replicates per reaction.

Reactions for each target were run on a gel to confirm the amplicon size. The products were also verified by sequencing.

Results

Rhopr-CT/DH receptors

We have isolated cDNA sequences for three splice variants encoding Rhopr-CT/DH-R1 (Rhopr-

CT/DH-R1-A, Rhopr-CT/DH-R1-B and Rhopr-CT/DH-R1-C) and two splice variants encoding

Rhopr-CT/DH-R2 (Rhopr-CT/DH-R2-A and Rhopr-CT/DH-R2-B). Rhopr-CT/DH-R1-A, Rhopr-

CT/DH-R1-B and Rhopr-CT/DH-R1-C are 1746, 1664 and 1301 nucleotides long and encode receptors comprised of 143, 411 and 409 amino acids, respectively (Figure S1 and Figure 1A).

The untranslated regions for Rhopr-CT/DH-R1-C could not be cloned via RACE PCRs as primers differentiating between Rhopr-CT/DH-R1-B and Rhopr-CT/DH-R1-C could not be designed. All three variants contain a polyadenylation signal sequence in their 3´ UTR. Rhopr-

CT/DH-R1-B, Rhopr-CT/DH-R1-C and Rhopr-CT/DH-R2-B all contain seven transmembrane domains, an extracellular N-terminus and an intracellular C-terminus, typical of all GPCRs

(Figure 1A, B). They also contain 6 highly-conserved cysteine residues (typical of family B1

GPCRs) and 3 predicted N-linked glycosylation sites in their N-terminus, and various predicted phosphorylation sites in their intracellular domains. Rhopr-CT/DH-R1-A is a truncated version of the receptor and only contains the extracellular N-terminus and a single transmembrane domain (Figure S1). The gene encoding this truncated receptor comprises 14 exons that are separated by 13 introns (Figure 2A). Rhopr-CT/DH-R1-A contains exon 8 that is absent in the other two variants and results in a premature stop codon. The ORF for this variant spans across exons 4 to 8. Rhopr-CT/DH-R1-B and Rhopr-CT/DH-R1-C differ by only 6 nucleotides within their ORF, which results in a 2 amino acid difference between these variants within their intracellular C-terminal domain. Rhopr-CT/DH-R1-C utilizes an alternate splice site in exon 13 which results in that exon being 6 nucleotides shorter at the 3´ end. The ORF for these two variants spans across exons 4 to 14.

Rhopr-CT/DH-R2-A and Rhopr-CT/DH-R2-B are 1146 and 1639 nucleotides long and yield proteins comprised of 122 and 410 amino acids, respectively (Figure S2 and Figure 1B). Rhopr-

CT/DH-R2-A comprises a partial N-terminus (contains only 5 cysteine residues) and lacks the seven transmembrane domains (Figure S2). Rhopr-CT/DH-R2-B, on the other hand, contains all the characteristics of a family B1 GPCR (Figure 1B). Moreover, it contains 7 predicted N-linked glycosylation sites and one predicted phosphorylation site. The gene encoding these 2 receptor variants is made up of 9 exons (Figure 2B). Exon 4 is absent in Rhopr-CT/DH-R2-A which results in a frame shift and truncated ORF. The ORF for this variant spans across exons 1 to 6 whereas the one for Rhopr-CT/DH-R2-B spans across all 9 exons. The untranslated regions for

Rhopr-CT/DH-R2-A could not be cloned due to its low expression.

Figure 1: cDNA sequences and the deduced amino acid sequences of Rhopr-CT/DH-Rs. Rhopr-

CT/DH-R1-B (A) and Rhopr-CT/DH-R2-B (B). The numbering for each sequence is shown at right. Within the nucleotide sequence, the exon-exon boundaries are shaded in gray and the potential polyadenylation signal is double-underlined. Within the amino acid sequence, the initial methionine start codon has been capitalized, the six conserved cysteine residues are shaded in red, the potential phosphorylation sites are shaded in black, the potential N-linked glycosylation sites are boxed and the seven predicted transmembrane domains are underlined. The two amino acid residues (valine and serine) that are absent in Rhopr-CT/DH-R1-C are dash underlined in

Rhopr-CT/DH-R1-B.

Figure 2: Rhopr-CT/DH-R1 and Rhopr-CT/DH-R2 splicing. Molecular organization of Rhopr-

CT/DH-R1 (A) and Rhopr-CT/DH-R2 (B) splice variants based on BLAST analysis and splice site prediction. The boxes represent exons (drawn to scale). The regions shaded in gray represent the open reading frame while the unshaded regions represent the untranslated regions. The dashed boxes represent predicted regions that were not cloned.

Functional receptor assay

To confirm that Rhopr-CT/DH is the ligand for the isolated putative Rhopr-CT/DH-R1 and

Rhopr-CT/DH-R2, we expressed these receptors in HEK293/CNG and monitored ligand- receptor interaction using a calcium mobilization assay. Only the receptor isoforms which contained all the 7 transmembrane domains were used in this assay. Rhopr-CT/DH-R1-B and

Rhopr-CT/DH-R1-C, which only differ by 2 amino acids in the C-terminus, were both activated by Rhopr-CT/DH with EC50 values ranging from 150-300nM (Figure 3A). The maximum luminescence response obtained for both these receptors following the addition of Rhopr-CT/DH was at least 42-fold higher compared to the addition of medium alone. Interestingly, Rhopr-

CT/DH-R2-B is much more sensitive to Rhopr-CT/DH (EC50 = 15nM) compared to Rhopr-

CT/DH-R1-B/C and results in a greater response (191-fold higher than basal response) in our heterologous assay (Figure 3B). None of these receptors were activated by Rhopr-CRF or

Drome-PDF (data not shown). Moreover, no response was observed following the addition of

Rhopr-CT/DH to the cells that were transfected with empty vector.

Transfection of Rhopr-CT/DH-R1-A in CHO/G16 cells stably-expressing either Rhopr-CT/DH-

R1-B or Rhopr-CT/DH-R1-C did not influence their sensitivity or kinetics of the response following the addition of Rhopr-CT/DH (data not shown). Since we had stably expressed Rhopr-

CT/DH-R1-B in CHO/G16 cells we also compared the kinetics of the response in these cells with that of the HEK293/CNG cells. Rhopr-CT/DH produced a rapid response, with the peak response for HEK/CNG cells and CHO/G16 cells between 5-10 seconds and 0-5 seconds, respectively (Figure S3).

Figure 3: Functional assay of R. prolixus CT/DH receptor isoforms (Rhopr-CT/DH-R1-B,

Rhopr-CT/DH-R1-C and Rhopr-CT/DH-R2-B) transiently expressed in HEK293/CNG cell lines.

Dose-dependent effect on the bioluminescence response after addition of Rhopr-CT/DH to

HEK293/CNG cells expressing Rhopr-CT/DH-R1-B and Rhopr-CT/DH-R1-C (A) and Rhopr-

CT/DH-R2 (B). Vertical bars represent SEM (n=3). Rhopr-CT/DH-R2-B is much more sensitive to Rhopr-CT/DH (EC50 = 15nM) compared to Rhopr-CT/DH-R1-B or C (EC50 = 200-300nM) and results in a greater response.

Sequence and phylogenetic analysis

Rhopr-CT/DH-R amino acid sequences were aligned along with those of Drome-CT/DH-R1,

Aedae-CT/DH-R1 (previously referred to as AedaeGPCRCAL1) and Drome-CT/DH-R2

(previously an orphan and also referred to as hector). The multiple sequence alignment illustrates high conservation across both the receptors (Figure 4). This conservation is localized not only over the seven transmembrane domains but also over the N´-terminal extracellular domain.

Positions of the two predicted N-linked glycosylation sites (positions 85 and 100 in Rhopr-

CT/DH-R1-A) and the 6 cysteine residues are conserved across most sequences. Since the N´- terminus forms part of the ligand-binding domain, it is not surprising that CT/DH activates both the receptors (Conner et al., 2004; Venkatakrishnan et al., 2013).

Phylogenetic analysis of family B1 GPCRs reveals three main monophyletic groups (Figure 5).

These groups represent the three main receptor types – CRF/DH-Rs, PDF-Rs and CT/DH-Rs. All insect CT/DH-Rs are sister to human CTR and CRLR. This further supports the suggestion that these hormonal systems are evolutionarily related. Within the clade comprising CT/DH-Rs,

CT/DH-R1s form a monophyletic group and CT/DH-R2s form another monophyletic group, which suggests that these two receptor subtypes arose from a recent duplication in insects, independent from the one in deuterostomes.

Figure 4: Multiple sequence alignment of select insect CT/DH receptors. Identical and similar amino acids across 50% of the sequences have been highlighted in black and gray, respectively.

The six highly-conserved cysteine residues in N-terminal domain have been marked with an asterisk. Two N-linked glycosylation sites which are conserved across all sequences have been highlighted in red. The predicted locations of the seven transmembrane domains of Rhopr-

CT/DH-R1 and Rhopr-CT/DH-R2 have been indicated using green lines and blue dashed lines, respectively.

Figure 5: A cladogram of family B1 GPCRs obtained following a maximum parsimonious analysis (1000 bootstrap replicates). The taxa are labelled using GenBank accession numbers and the species names. Drosophila melanogaster metabotropic glutamate receptor is utilized as an outgroup. Black symbols are used to denote sequences from Homo sapiens and colored symbols denote insect sequences. Note that the three receptor subtypes (CRF, PDF and CT/DH) form monophyletic clades.

Spatial expression profile of Rhopr-CT/DH-R transcript variants

In order to discover physiological targets of Rhopr-CT/DH, qPCR analysis was performed to determine the spatial expression pattern of Rhopr-CT/DH-R transcript variants (Figure 6). Rhopr-

CT/DH-R1-A, despite lacking the characteristic seven transmembrane domains, is highly expressed in the testes (Figure 6A). Rhopr-CT/DH-R1-B/C, on the other hand, is highly enriched in the CNS and dorsal vessel and expressed at lower levels in the foregut, salivary glands, hindgut, testes, ovaries and prothoracic glands. Rhopr-CT/DH-R2-A is only expressed in the

CNS and in low amounts (Figure 6B). Rhopr-CT/DH-R2-B has the highest abundance in the

CNS and this is over 600 fold higher compared to any other Rhopr-CT/DH-R transcript levels in any tissue. Interestingly, Rhopr-CT/DH-R2-B and not Rhopr-CT/DH-R1-B/C, is expressed in

MTs. Rhopr-CT/DH-R2-B is also expressed at much lower levels in the salivary glands, testes, ovaries and prothoracic glands. With regards to the adult reproductive tissues, both Rhopr-

CT/DH-R1-B/C and Rhopr-CT/DH-R2-B have the highest expression in ovaries and are expressed at lower levels in the testes and female reproductive tissues minus the ovaries (Figure

7). Similar to the fifth-instar, Rhopr-CT/DH-R1-A is highly expressed in the adult testes (Figure

7A).

Figure 6: Spatial expression analysis of Rhopr-CT/DH-Rs in fifth instar R. prolixus determined using quantitative PCR. Rhopr-CT/DH-R1 (A) and Rhopr-CT/DH-R2 (B) expression profile.

Expression was analyzed in the following tissues: CNS (central nervous system), DV (dorsal vessel), Pool (fat bodies, abdominal nerves, diaphragms and trachea), FG (foregut), SG (salivary glands), AMG (anterior midgut), PMG (posterior midgut), MTs (Malpighian tubules), HG

(hindgut), TST (testes), OV (ovaries) and PG (prothoracic glands). Expression of each variant for both the receptors is shown relative to Rhopr-CT/DH-R1-B transcript levels in salivary glands cDNA.

Figure 7: Spatial expression analysis of Rhopr-CT/DH-Rs in R. prolixus adult reproductive tissues determined using quantitative PCR. Rhopr-CT/DH-R1 (A) and Rhopr-CT/DH-R2 (B) expression profile. Expression was analyzed in the following tissues: TST (testes), M.R. (rest of the male reproductive tissues), OV (ovaries) and F.R. (rest of the female reproductive tissues).

Expression of each variant for both the receptors is shown relative to Rhopr-CT/DH-R1-B transcript levels in testes cDNA.

Discussion

In the present study, we have found that R. prolixus, and perhaps other insects, contain two

CT/DH-Rs. Rhopr-CT/DH-R1 encodes three splice variants while Rhopr-CT/DH-R2 encodes two splice variants. We functionally characterized full-length receptor isoforms encoded by these transcript variants in HEK293/CNG cells. The assay used to characterize these receptors monitored calcium mobilization into the cells from the extracellular medium. Since Rhopr-

CT/DH is thought to mediate its effects via cAMP as the secondary messenger, its binding to the receptor would result in receptor activation and a subsequent increase in intracellular cAMP levels (Brugge et al., 2008). This cAMP would then bind to the CNG channel, resulting in its opening and an influx of calcium from the extracellular medium. The amount of calcium mobilized was detected using the reporter, aequorin. Using this assay, we confirmed that Rhopr-

CT/DH-R1-B, Rhopr-CT/DH-R1-C and Rhopr-CT/DH-R2-B were all activated by Rhopr-

CT/DH. This work therefore essentially de-orphans the D. melanogaster hector and related receptors in other insects. Responses that were robust (ranging from 42 to 191-fold higher than basal response) and sensitive (EC50 values in the low nanomolar range) were obtained in this cell line. This is surprising because according to a previous report, HEK293 cells expressing Drome-

CT/DH-R1 were not responsive to Drome-CT/DH until RCP, an accessory protein, was co- expressed with the receptor (Johnson et al., 2005). Even after the expression of D. melanogaster

RCP along with the receptor, the maximum response obtained was under 2-fold that of the basal response. An improved sensitivity (EC50 = 82nM) and a greater response (7-fold) to Drome-

CT/DH was only observed when the human RCP and RAMPs were expressed along with the receptor. This led to the conclusion that perhaps RCP and RAMPs are required for signalling through insect CT/DH-Rs just like mammalian calcitonin receptors; however, our results with a similar cell line, coupled with the fact that proteins with sequence homology to human RCP and

RAMPs are not found in the R. prolixus genome, suggests that Rhopr-CT/DH-Rs may not require accessory proteins for effective signalling.

Comparing the responses of the two Rhopr-CT/DH-Rs, Rhopr-CT/DH-R2-B is at least 10-fold more sensitive to Rhopr-CT/DH than are Rhopr-CT/DH-R1-B or Rhopr-CT/DH-R1-C; the EC50 values of Rhopr-CT/DH-R1-B and Rhopr-CT/DH-R1-C range from 150-300nM. These values are still relatively high and hence we don’t rule out the possibility that other endogenous ligands may also activate this receptor. Another possibility is that Rhopr-CT/DH-R1 and Rhopr-CT/DH-

R2 may interact as heterodimers as is so common of family B1 GPCRs (Mikhailova et al., 2007).

Either way, Rhopr-CT/DH-R1-B and Rhopr-CT/DH-R1-C do not seem to be functionally different. Rhopr-CT/DH-R1-A is highly expressed in the testes and such high expression may suggest a role for this truncated receptor isoform rather than it being just an error in splicing.

Since Rhopr-CT/DH-R1-A and Rhopr-CT/DH-R1-B/C are expressed in the CNS, dorsal vessel, testes and ovaries, we questioned whether Rhopr-CT/DH-R1-A could interact with Rhopr-

CT/DH-R1-B and Rhopr-CT/DH-R1-C and affect their signalling. If Rhopr-CT/DH-R1-A influenced (either inhibited or stimulated) the surface expression of Rhopr-CT/DH-R1-B/C, one would expect their EC50 values to be altered. However, Rhopr-CT/DH-R1-A did not influence the signalling through Rhopr-CT/DH-R1-B and Rhopr-CT/DH-R1-C in our heterologous assay utilizing CHO/G16 cells. Hence, the role of Rhopr-CT/DH-R1-A, if any, is still unclear.

Spatial expression analysis of the two receptors using qPCR demonstrates that they have distinct expression patterns, with both receptors expressed centrally and peripherally. We utilized Rhopr-

CT/DH-R1-B transcript levels in salivary glands and testes cDNA to show the relative expression of each variant for both the receptors in fifth-instar tissues and adult reproductive tissues,

98 respectively. We acknowledge that this method is not as effective as absolute quantification to compare the expression of two different genes and hence this comparison between the two receptors is only an approximation. Nonetheless, at least one of Rhopr-CT/DH-R1-B, Rhopr-

CT/DH-R1-C or Rhopr-CT/DH-R2-B is expressed in the dorsal vessel, salivary glands, hindgut and MTs which have previously been shown to be targets of Rhopr-CT/DH (Brugge et al., 2008;

Orchard, 2009; Te Brugge et al., 2005). Rhopr-CT/DH causes, at most, a 17-fold increase in the rate of fluid secretion by MTs compared to saline controls (Te Brugge et al., 2005); however, this rate is only 1.5% of the maximum rate stimulated by serotonin and Rhopr-CRF/DH. Perhaps

Rhopr-CT/DH is used by the insect at times when diuresis needs to be maintained at a low level, such as the period after the rapid post-feeding diuresis and during digestion. Based on the expression patterns, Rhopr-CT/DH-R2 appears to be the one responsible for MT secretion in R. prolixus, whereas CT/DH-R1 is responsible for the diuretic function in D. melanogaster and A. aegypti (Johnson et al., 2005; Kwon et al., 2012). FlyAtlas (www.flyatlas.org) data indicates that

Drome-CT/DH-R2 is not expressed in D. melanogaster MTs. This observation suggests that the specialized functions, diuresis mediated through CT/DH-R2 in R.prolixus and through CT/DH-

R1 in dipterans, have evolved independently. The phylogeny of these two receptor clusters has been discussed previously (Park, 2012). It remains to be seen whether CT/DH-R2 mediates diuresis in other non-dipteran species. Rhopr-CT/DH also increases muscle contractions of dorsal vessel, salivary glands and hindgut, all of which indicate a role for Rhopr-CT/DH in feeding-related physiological events (Brugge et al., 2008; Orchard, 2009). For instance, the contraction of salivary glands would aid in the release of saliva at the time of feeding. An increase in dorsal vessel contractions would result in increased circulation of the haemolymph, as well as the diuretic hormones that are present in the haemolymph following feeding (Orchard,

1989). Hindgut contractions aid in expulsion of waste, reduce unstirred layers around MTs and

99 also increase haemolymph circulation. Hence increased contractility of dorsal vessel and hindgut could indirectly aid in post-feeding diuresis.

All the receptor transcripts are expressed within the CNS but it is unclear which neural circuits

Rhopr-CT/DH-Rs may be involved with. The qPCR analysis also identified novel target tissues of Rhopr-CT/DH; these include testes, ovaries and prothoracic glands. Rhopr-CT/DH, thus, may regulate reproductive physiology and ecydsteroidogenesis. CT/DH-like immunoreactivity is not associated with male or female reproductive tissues of R. prolixus (Zandawala and Orchard, unpublished). Hence the effect, if any, of Rhopr-CT/DH on reproductive tissues will most-likely be mediated via a hormonal route. Drome-CT/DH-R2 (hector) is expressed in a subset of fruitless neurons and has been shown to be critical for male courtship (Li et al., 2011). Consistent with this role, its transcript is enriched in D. melanogaster brain and male accessory glands

(Chintapalli et al., 2007). It remains to be examined if Rhopr-CT/DH-R2 is also involved in courtship behaviour considering its expression in the CNS and reproductive tissues.

This is the first study to reveal that insects possess at least two CT/DH receptors, which may be functionally different. Our expression analysis suggests that Rhopr-CT/DH-Rs may mediate feeding-related physiological events, some of which must await further investigation. Moreover, we also identified novel target tissues for this neuropeptide, including testes, ovaries and prothoracic glands, suggesting a possible role for Rhopr-CT/DH in reproductive physiology and development.

100

References

Blom, N., Gammeltoft, S. and Brunak, S. (1999). Sequence and structure-based prediction of eukaryotic protein phosphorylation sites. J Mol Biol. 294, 1351-62.

Brugge, V.A., Schooley, D.A. and Orchard, I. (2008). Amino acid sequence and biological activity of a calcitonin-like diuretic hormone (DH31) from Rhodnius prolixus. J Exp Biol. 211, 382-90.

Chintapalli, V.R., Wang, J. and Dow, J.A. (2007). Using FlyAtlas to identify better Drosophila melanogaster models of human disease. Nat Genet. 39, 715-20.

Coast, G.M., Nachman, R.J. and Lopez, J. (2011). The control of Malpighian tubule secretion in a predacious hemipteran insect, the spined soldier bug Podisus maculiventris (Heteroptera, Pentatomidae). Peptides. 32, 493-9.

Collin, C., Hauser, F., de Valdivia, E.G., Li, S., Reisenberger, J., Carlsen, E.M., Khan, Z., Hansen, N.O., Puhm, F., Sondergaard, L., Niemiec, J., Heninger, M., Ren, G.R. and Grimmelikhuijzen, C.J. (2013). Two types of muscarinic acetylcholine receptors in Drosophila and other arthropods. Cell Mol Life Sci. 70, 3231-42.

Conner, A.C., Simms, J., Hay, D.L., Mahmoud, K., Howitt, S.G., Wheatley, M. and Poyner, D.R. (2004). Heterodimers and family-B GPCRs: RAMPs, CGRP and adrenomedullin. Biochem Soc Trans. 32, 843-6.

101

Evans, B.N., Rosenblatt, M.I., Mnayer, L.O., Oliver, K.R. and Dickerson, I.M. (2000). CGRP-RCP, a novel protein required for signal transduction at calcitonin gene-related peptide and adrenomedullin receptors. J Biol Chem. 275, 31438-43.

Hewes, R.S. and Taghert, P.H. (2001). Neuropeptides and neuropeptide receptors in the Drosophila melanogaster genome. Genome Res. 11, 1126-42.

Johnson, E.C., Shafer, O.T., Trigg, J.S., Park, J., Schooley, D.A., Dow, J.A. and Taghert, P.H. (2005). A novel diuretic hormone receptor in Drosophila: evidence for conservation of CGRP signaling. J Exp Biol. 208, 1239-46.

Kaskani, C., Poulos, C.P. and Goldsworthy, G.J. (2012). The effects of linear and cyclic analogs of Locmi-DH, Dippu-DH(46) and Dippu-DH(31) on appetitive behavior in Locusta migratoria. Peptides. 34, 258-61.

Kwon, H., Lu, H.L., Longnecker, M.T. and Pietrantonio, P.V. (2012). Role in diuresis of a calcitonin receptor (GPRCAL1) expressed in a distal-proximal gradient in renal organs of the mosquito Aedes aegypti (L.). PLoS ONE. 7, e50374.

Kwon, H. and Pietrantonio, P.V. (2013). Calcitonin receptor 1 (AedaeGPCRCAL1) hindgut expression and direct role in myotropic action in females of the mosquito Aedes aegypti (L.). Insect biochemistry and molecular biology. 43, 588-93.

LaJeunesse, D.R., Johnson, B., Presnell, J.S., Catignas, K.K. and Zapotoczny, G. (2010). Peristalsis in the junction region of the Drosophila larval midgut is modulated by DH31 expressing enteroendocrine cells. BMC Physiol. 10, 14.

Larkin, M.A., Blackshields, G., Brown, N.P., Chenna, R., McGettigan, P.A., McWilliam, H., Valentin, F., Wallace, I.M., Wilm, A., Lopez, R., Thompson, J.D., Gibson, T.J. and Higgins, D.G. (2007). Clustal W and Clustal X version 2.0. Bioinformatics. 23, 2947-8.

102

Li, Y., Hoxha, V., Lama, C., Dinh, B.H., Vo, C.N. and Dauwalder, B. (2011). The hector G- protein coupled receptor is required in a subset of fruitless neurons for male courtship behavior. PLoS ONE. 6, e28269.

McLatchie, L.M., Fraser, N.J., Main, M.J., Wise, A., Brown, J., Thompson, N., Solari, R., Lee, M.G. and Foord, S.M. (1998). RAMPs regulate the transport and ligand specificity of the calcitonin-receptor-like receptor. Nature. 393, 333-9.

Mikhailova, M.V., Mayeux, P.R., Jurkevich, A., Kuenzel, W.J., Madison, F., Periasamy, A., Chen, Y. and Cornett, L.E. (2007). Heterooligomerization between vasotocin and corticotropin-releasing hormone (CRH) receptors augments CRH-stimulated 3',5'-cyclic adenosine monophosphate production. Mol Endocrinol. 21, 2178-88.

Nag, K., Sultana, N., Kato, A. and Hirose, S. (2007). Headless splice variant acting as dominant negative calcitonin receptor. Biochem Biophys Res Commun. 362, 1037-43.

Orchard, I. (1989). Serotonergic neurohaemal tissue in Rhodnius prolixus: Synthesis, release and uptake of serotonin. J Insect Physiol. 35, 943-947.

Orchard, I. (2009). Peptides and serotonin control feeding-related events in Rhodnius prolixus. Front Biosci (Elite Ed). 1, 250-62.

Park, Y. (2012). Endocrine regulation of insect diuresis in the early postgenomic era. Can J Zool. 90, 507-520.

Reese, M.G., Eeckman, F.H., Kulp, D. and Haussler, D. (1997). Improved splice site detection in Genie. J Comput Biol. 4, 311-23.

Seck, T., Baron, R. and Horne, W.C. (2003). The alternatively spliced deltae13 transcript of the rabbit calcitonin receptor dimerizes with the C1a isoform and inhibits its surface expression. J Biol Chem. 278, 23085-93.

Stables, J., Green, A., Marshall, F., Fraser, N., Knight, E., Sautel, M., Milligan, G., Lee, M. and Rees, S. (1997). A bioluminescent assay for agonist activity at potentially any G-protein- coupled receptor. Anal Biochem. 252, 115-26.

103

Staubli, F., Jorgensen, T.J., Cazzamali, G., Williamson, M., Lenz, C., Sondergaard, L., Roepstorff, P. and Grimmelikhuijzen, C.J. (2002). Molecular identification of the insect adipokinetic hormone receptors. Proc Natl Acad Sci U S A. 99, 3446-51.

Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M. and Kumar, S. (2011). MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol. 28, 2731-9.

Te Brugge, V., Ianowski, J.P. and Orchard, I. (2009). Biological activity of diuretic factors on the anterior midgut of the blood-feeding bug, Rhodnius prolixus. Gen Comp Endocrinol. 162, 105-12.

Te Brugge, V.A., Lombardi, V.C., Schooley, D.A. and Orchard, I. (2005). Presence and activity of a Dippu-DH31-like peptide in the blood-feeding bug, Rhodnius prolixus. Peptides. 26, 29-42.

Venkatakrishnan, A.J., Deupi, X., Lebon, G., Tate, C.G., Schertler, G.F. and Babu, M.M. (2013). Molecular signatures of G-protein-coupled receptors. Nature. 494, 185-94.

Visegrady, A., Boros, A., Nemethy, Z., Kiss, B. and Keseru, G.M. (2007). Application of the BD ACTOne technology for the high-throughput screening of Gs-coupled receptor antagonists. J Biomol Screen. 12, 1068-73.

Zandawala, M., Lytvyn, Y., Taiakina, D. and Orchard, I. (2012). Cloning of the cDNA, localization, and physiological effects of FGLamide-related allatostatins in the blood-gorging bug, Rhodnius prolixus. Insect Biochem Mol Biol. 42, 10-21.

104

Acknowledgements

The authors wish to thank Nikki Sarkar for maintaining the colony and Laura Sedra for dissecting adult reproductive tissues for qPCR analysis.

105

Appendices

Table S1: Primers used to amplify the partial cDNA sequence for Rhopr-CT/DH-R1 and Rhopr-

CT/DH-R2.

Primer Tm Sequence (5’-3’) Rhopr-CT/DH-R1 DH31R1FOR1 62.1 TCTGTGGCTTCTCTGGTATAGG DH31R1REV1 62.1 AAGTTACTTGGAAGGACGCTG Rhopr-CT/DH-R2 DH31R2FOR1 62.9 GTTTGCCGCCAATAATCTG DH31R2REV1 62.8 CCTGTAAAGAAACAAGAAGTGCTG

106

Table S2: Primers used to perform 5’ RACE PCR reactions.

Primer Tm Sequence (5’-3’) Plasmid-specific primers DNR-LIB FOR1 63.9 GTGGATAACCGTATTACCGCC DNR-LIB FOR2 64.5 ACGGTACCGGACATATGCC Rhopr-CT/DH-R1 DH31R1-5RACE-REV1 64.0 CATACACCGTTTTCCAGAATTACC DH31R1-5RACE-REV2 64.2 CAGAATTACCTCCGGAAATGG DH31R1-5RACE-REV3 64.4 ACCAGCCTATACCAGAGAAGCC DH31R1-5RACE-REV4 63.8 ACCAAATTTGTCCAGATTCAGG DH31R1-5RACE-REV5 64.3 GATTCAGGATGTCTAAACCATGTACC DH31R1-5RACE-REV6 64.5 GTTGTTAGACATTGAAGGTAGCGTG DH31R1-5RACE-REV7 65.2 AGTTCAGCATGTGGGTCCAG Rhopr-CT/DH-R2 DH31R2-5RACE-REV1 64.4 CAAGATGTAGATGCAATCCTTCG DH31R2-5RACE-REV2 64.5 CAATCCTTCGCAGAACATCC DH31R2-5RACE-REV3 64.7 ATCAGATTATTGGCGGCAAAC DH31R2-5RACE-REV4 63.8 ATGTCGCTGGGCAGTACAC DH31R2-5RACE-REV6 64.2 TGTTGACTGCTTTTCGAAACTTAAG DH31R2-5RACE-REV7 64.4 TGTCAATGCAGGTCGTATAATTTG DH31R2-5RACE-REV8 63.2 GTAGTATTCCAACATGACCAGCC

107

Table S3: Primers used to perform 3’ RACE PCR reactions.

Primer Tm Sequence (5’-3’) Plasmid-specific primers pDNR-LIB 3 -25 REV 63.6 GCCAAACGAATGGTCTAGAAAG pDNR-LIB 3 -88 REV 63.3 AGTCATACCAGGATCTCCTAGGG Rhopr-CT/DH-R1 DH31R1-3RACE-FOR1 63.6 GGTTGTTGGTAAGTTAAGAGCTGG DH31R1-3RACE-FOR2 64.2 GACCTTCAAGAGCCTTACTACAAGC DH31R1-3RACE-FOR3 64.9 GCTTGGGCTTAACTATCTTTTGACTC Rhopr-CT/DH-R2 DH31R2-3RACE-FOR1 64.1 TTGGTCTTAGGAAGGCTGTACG DH31R2-3RACE-FOR2 63.9 GAAAGCACCAGGAGAACGAG DH31R2-3RACE-FOR3 64.4 TCAGCACTTCTTGTTTCTTTACAGG

108

Table S4: Primers used to amplify the largest cDNA fragments.

Primer Tm Sequence (5’-3’) Rhopr-CT/DH-R1 DH31-R1-FOR2 62.1 GAAGTTGTGCAAAGTTTGTGG DH31-R1-REV3 61.5 TGTGAAACATCTAATGGACAAAAC Rhopr-CT/DH-R2 DH31R2FOR5 55.5 ACCACCTCCGAGTACAC DH31R2REV3 54.3 CCTTACAACAACATATAATTCATA

109

Table S5: Primers used to amplify full ORF and introduce Kozak sequence.

Primer Tm Sequence (5’-3’) Rhopr-CT/DH-R1 DH31R1-KOZAK-FOR 62.4 GCCACCATGTCGGATG DH31R1-ORF-REV 60.8 CTACATTATTTTATATTGAGACCCATTTC DH31-R1-A-REV1 60.4 AAATAACTTCCTGGCCATTG Rhopr-CT/DH-R2-B DH31R2-KOZAK-FOR 61.0 GCCACCATGAGAAATGTAGAC DH31R2-ORF-REV 60.6 CGAAATGATCTAGTCTCGCA

110

Table S6: Primers used for qPCR reactions.

Primer Tm Sequence (5’-3’) Rhopr-CT/DH-R1-A qPCR-DH31R-A-FOR1 56.1 GATGATTTAAATTTAAGGCAAC qPCR-DH31R-A-REV1 56.5 GGACATCTTAAAGATTTTTATTCTAC Rhopr-CT/DH-R1-B/C qPCR-DH31R-B-FOR2 58.1 CTTTGATCCAAATCGATTG qPCR-DH31R-B-REV1 56.0 CATCTTAAAGATTTAAAATAAGATAGG Rhopr-CT/DH-R2-A qPCR-DH31R2A-FOR 61.7 CAGGATTTGATATAAATATTTCGAAAAG qPCR-DH31R2A-REV 61.7 TGGCAAATTAACTGATTTTCTTG Rhopr-CT/DH-R2-B qPCR-DH31R2B-FOR2 61.4 TCGCAAATGTTTAGAAAATGG qPCR-DH31R2A-REV 61.7 TGGCAAATTAACTGATTTTCTTG Rhopr-alpha-tubulin alphaTUB-qPCR-F 64.3 GTGTTTGTTGATTTGGAACCTACAG alphaTUB-qPCR-R 64.4 CCGTAATCAACAGACAATCTTTCC Rhopr-beta-actin Actin5c-qPCR-F 62.9 AGAGAAAAGATGACGCAGATAATGT Actin5c-qPCR-R 63.6 ATATCCCTAACAATTTCACGTTCG Rhopr-ribosomal protein 49 rp49-qPCR-F 63.7 GTGAAACTCAGGAGAAATTGGC rp49-qPCR-R 65.0 AGGACACACCATGCGCTATC

111

Figure S1: Rhopr-CT/DH-R1-A cDNA sequence and the deduced amino acid sequence. The numbering for each sequence is shown at right. Within the nucleotide sequence, the exon-exon boundaries are shaded in gray and the potential polyadenylation signal is double-underlined.

Within the amino acid sequence, the initial methionine start codon has been capitalized, the six conserved cysteine residues are shaded in red, the potential N-linked glycosylation sites are boxed and the predicted transmembrane domain is underlined.

112

113

Figure S2: Rhopr-CT/DH-R2-A cDNA sequence and the deduced amino acid sequence. The numbering for each sequence is shown at right. Within the nucleotide sequence, the exon-exon boundaries are shaded in gray. Within the amino acid sequence, the initial methionine start codon has been capitalized, the conserved cysteine residues are shaded in red and the potential N-linked glycosylation sites are boxed.

114

115

Figure S3: Kinetics of the bioluminescence responses of HEK/CNG (A) and CHO/G16 (B) cells expressing Rhopr-CT/DH-R1-B. Bioluminescence was recorded for every 5 seconds for 15 seconds following the addition of phosphate-buffered saline (PBS) or 10-6M peptide. Vertical bars represent SEM (n=3). Rhopr-CT/DH produced a rapid response, with the peak response for

HEK/CNG cells and CHO/G16 cells between 5-10 seconds and 0-5 seconds, respectively. The assay was performed using the methods described earlier (Stables et al., 1997; Staubli et al.,

2002).

116

117

Chapter 4: Structure-activity relationships of two Rhodnius

prolixus calcitonin-like diuretic hormone analogs

The proceeding chapter was reprinted/reproduced with permission from Elsevier.

Full citation details:

Structure-activity relationships of two Rhodnius prolixus calcitonin-like diuretic hormone analogs.

Zandawala M, Poulos C, Orchard I.

Peptides. 2014 Apr 2. pii: S0196-9781(14)00104-1. doi: 10.1016/j.peptides.2014.03.019.

118

Abstract

The calcitonin-like diuretic hormone (CT/DH) in Rhodnius prolixus influences various tissues associated with feeding-related physiological events. The receptors for this peptide have also been identified and shown to be expressed in these tissues. In the present study, we have investigated the effects of two R. prolixus CT/DH analogs (full-length form and N-terminal truncated form) on hindgut contractions and in a heterologous receptor expression system. The analogs contained the amino acid methyl-homoserine in place of methionine in order to prevent them from being oxidized and thus increase their stability. The full-length form of the analog retained all of its activity in our assays when compared to the endogenous peptide. Truncated analog displayed no activity in our assays.

119

Introduction

Rhodnius prolixus undergoes a period of rapid post-feeding diuresis (urine production) after ingesting large blood meals. It releases Trypanosoma cruzi, a parasite which causes Chagas disease, in its urine near the site of the bite after a blood meal. Infection by the parasite occurs when it enters the bloodstream of the host through the site of the bite. This post-feeding diuresis and enhanced haemolymph circulation involves a variety of tissues including the anterior midgut, Malpighian tubules (MTs), hindgut and dorsal vessel (see Orchard, 2009). Various diuretic and anti-diuretic hormones (DHs and ADHs respectively) modulate these tissues in R. prolixus, including serotonin (5-hydroxytryptamine), corticotropin releasing-factor (CRF)-related diuretic hormone (CRF/DH), calcitonin-like diuretic hormone (CT/DH), kinin and CAPA (see

Orchard, 2009). Since these hormones control diuresis, they also influence the transmission of

Chagas disease. The study of receptor-ligand interactions is important as it has great potential for screening many analogs. This will help in the development of pathway-specific peptidomimetics which could be used to control diuresis and prevent the spread of Chagas disease.

Rhodnius prolixus CT/DH (Rhopr-CT/DH) has been shown to stimulate low levels of MT secretion and to increase contractions of the anterior midgut, hindgut and dorsal vessel, in vitro

(see Zandawala, 2012). Recently, we isolated and characterized two CT/DH receptors from R. prolixus (Rhopr-CT/DH-R1 and Rhopr-CT/DH-R2) (Zandawala et al., 2013). Spatial expression analysis showed that both receptors are expressed in feeding-related tissues and hence provides further evidence for the involvement of this signaling system in post-feeding diuresis. In the present research, Rhopr-CT/DH analogs were synthesized and their activities compared with the native peptide using an in vitro biological assay and a cell-based functional receptor assay. We focused on the importance of peptide length and biostability, since peptide mimetics need to be 120 stable and small in size (lower production costs) for their use in pest-control to be economically feasible. The importance of the peptide length was assessed by using an N’-terminal truncated analog, which lacked the first 16 amino acid residues ([Hse(Me)]20 -Rhopr-CT/DH(17-31)).

Moreover, the methionine residue was replaced by a methyl-homoserine residue to reduce its susceptibility to oxidation, since oxidized peptides show reduced biological activity, as has been shown for the vertebrate calcitonin (see Torosantucci et al., 2014).

Material and methods

Animals

Fifth-instar R. prolixus (8-10 weeks post-feeding as fourth instars) were obtained from a long standing colony at the University of Toronto Mississauga. Insects were reared in incubators in high relative humidity at 25°C and routinely fed on defibrinated rabbit blood (Cedarlane

Laboratories Inc., Burlington, ON, Canada).

Peptide synthesis

Rhopr-CT/DH (GLDLGLSRGFSGSQAAKHLMGLAAANYAGGPamide), which is identical in amino acid sequence to Diploptera puntata CT/DH (Dippu-CT/DH), was custom synthesised by

GenScript (Piscataway, NJ, USA) at > 95% purity. [Hse(Me)]20 -Rhopr-CT/DH and [Hse(Me)]20

-Rhopr-CT/DH(17-31) were synthesized as described earlier (Kaskani et al., 2009). The purity of these peptide analogs was > 99%, as estimated by HPLC. Stock solutions (1 mM) of all the peptides were made by dissolving them in water.

Hindgut contraction assay

121

Contraction assays were performed on isolated hindguts from unfed fifth-instar R. prolixus maintained under physiological saline (Te Brugge et al., 2008; Zandawala and Orchard, 2013).

Hindguts were incubated in 200 µL saline and secured by pinning a small piece of the ventral cuticle on a Sylgard (Dow Corning, Midland, Michigan, USA) coated dish. One end of a fine silk thread was tied between the anterior portion of the hindgut and a small portion of the posterior midgut while the other end of the thread was hooked to a miniature force transducer

(AksjeselskapetMikro-elektronikk, Horten, Norway). Tissues were allowed to equilibrate in saline for a minimum of 10 minutes. The frequency of longitudinal contractions was monitored and recorded for 200 seconds using PicoLog recorder (Pico Technology, St. Neots,

Cambridgeshire, UK) after the application of either saline or peptide. One-way ANOVA and

Tukey post-test were used for the statistical analysis (P<0.05).

Expression vector construction

The ORFs encoding Rhopr-CT/DH-R1-B and Rhopr-CT/DH-R2-B were amplified as described previously (Zandawala et al., 2013). Kozak translation initiation sequence was also introduced at the 5’ end of these sequences; the resulting sequences were cloned into pGEM-T Easy vector

(Promega, Madison, WI, USA). These were then subcloned into pIRES2-ZsGreen1 (Clontech,

Mountain View, CA, USA) for mammalian cell expression.

Cell culture, transfections and bioluminescence assay

Chinese hamster ovary (CHO) cells stably expressing the human G-protein G16 (CHO/G16) used for the bioluminescence assay were cultured according to the conditions mentioned earlier

(Stables et al., 1997). X-tremeGENE HP DNA transfection reagent (Roche Applied Science,

Indianapolis, IN, USA) was used to transiently transfect CHO/G16 cells with the receptor

122 expression vector (Rhopr-CT/DH-R2-B) and an expression vector containing cytoplasmic luminescent reporter aequorin at ratio of 2:1 (transfection reagent to expression vectors) according to the protocol supplied by the manufacturer. Since no difference in response was detected between CHO/G16 cells expressing Rhopr-CT/DH-R1-B either stably or transiently, previously established CHO/G16 cells stably expressing Rhopr-CT/DH-R1-B were used here

(Zandawala et al., 2013). These cells were then transiently transfected with aequorin as described above. The cells were incubated for 48 hours post-transfection and used to perform the bioluminescence assay as mentioned earlier (Zandawala et al., 2013).

Results

Hindgut contraction assay

Previous work has shown that Rhopr-CT/DH increases the contraction frequency of R. prolixus hindgut in a dose-dependent manner (Te Brugge et al., 2008). Hence, we utilized this assay to compare the activity of the analogs against the native peptide. Rhopr-CT/DH and [Hse(Me)]20 -

Rhopr-CT/DH at 10-6 M both increased the frequency of hindgut contractions, with no significant difference in the response to the two peptides (Figure 1). There was no definite increase in the amplitude of contractions or the basal tonus. [Hse(Me)]20 -Rhopr-CT/DH(17-31) on the other hand had no effect on hindgut contractions compared to the saline controls (Figure

1).

Bioluminescence assay

We expressed Rhopr-CT/DH-R1 and Rhopr-CT/DH-R2 in CHO/G16 cells and monitored ligand-receptor interactions using a calcium mobilization assay. Rhopr-CT/DH and [Hse(Me)]20 -

Rhopr-CT/DH were both able to activate Rhopr-CT/DH-R1-B (EC50 = 260-340 nM) and Rhopr-

123

CT/DH-R2-B (EC50 = 1.7-3.3 nm) at equal sensitivities (Figure 2). No response, however, was observed following the addition of [Hse(Me)]20 -Rhopr-CT/DH(17-31) to the cells that were transfected with either receptor (Figure 2).

124

Figure 1: Effect of Rhopr-CT/DH and its analogs on frequency of R. prolixus hindgut contractions. (A-D) Sample traces. Control contractions observed in saline prior to the application of 10-6 M Rhopr-CT/DH (A), [Hse(Me)]20 -Rhopr-CT/DH (B) or [Hse(Me)]20 -Rhopr-

CT/DH(17-31) (C). Each preparation was thoroughly washed between different peptide applications. Saline was applied at the filled arrow and peptides were applied at the open arrow.

(D) Bar graph showing the effects of the three peptides on the frequency of R. prolixus hindgut contractions. Bars represent mean ± s.e.m. of 6 preparations. Scale bars represent tension and time.

125

126

Figure 2: Functional assay of R. prolixus CT/DH receptor isoforms (Rhopr-CT/DH-R1-B and

Rhopr-CT/DH-R2-B) expressed in CHO/G16 cells. Dose-dependent effect on the bioluminescence response after addition of Rhopr-CT/DH, [Hse(Me)]20 -Rhopr-CT/DH or

[Hse(Me)]20 -Rhopr-CT/DH(17-31) to CHO/G16 cells expressing Rhopr-CT/DH-R1-B (A) and

Rhopr-CT/DH-R2-B (B). Vertical bars represent SEM (n=3). Both receptors are activated by

Rhopr-CT/DH and [Hse(Me)]20 -Rhopr-CT/DH but not by [Hse(Me)]20 -Rhopr-CT/DH(17-31).

127

128

Discussion

In the present work, biostable analogs of Rhopr-CT/DH were synthesized and their activities compared with the native peptide. The methionine residue at position 20 was replaced by a methyl-homoserine residue to reduce its susceptibility to oxidation and thus improve peptide- stability. The full length form of the analog was as effective as the endogenous peptide when their activities were compared in both the cell-based receptor functional assay and the hindgut contraction assay; however, the N’-terminal truncated analog was inactive. Our results are in agreement with previous work where effects of similar modifications to Dippu-CT/DH were examined on secretion by D. punctata MTs (Kaskani et al., 2009). Replacement of methionine with methyl-homoserine in Dippu-CT/DH did not alter the secretion rate (Kaskani et al., 2009), whereas the N’-terminal truncated analog had no effect on MT secretion. This is not surprising as there must be a selective pressure on maintaining the length of the peptide which has been conserved across protostomes and deuterostomes (see Zandawala, 2012); however, it is interesting that injection of the N-terminal truncated analog into locust nymphs was anorexigenic

(Kaskani et al., 2012), although it is not known which other receptors might be activated.

Development of novel insecticides is important due to various problems associated with current insect-control strategies. Neonicotinoids represent a class of neuro-active chemical insecticides that are currently the most widely used insecticide around the globe; however, several insect pests have already developed resistance to them (see Nauen and Denholm, 2005) and non-target beneficial species are also susceptible, including important pollinators (Krupke et al., 2012;

Whitehorn et al., 2012). Hence there is an urgent demand for novel environmentally-friendly and highly-selective insecticides. Peptide-based insecticides represent excellent candidates especially since some occur naturally as part of spider venoms (Hardy et al., 2013). Most peptides mediate

129 their action via G protein-coupled receptors (GPCRs) to alter the physiology and behavior of insects. Targeting these receptors using biostable, species-specific peptide mimetics that alter the normal physiology of insects is a promising avenue to control insect populations (Jiang et al.,

2013; Nachman et al., 2012). In order for this approach of pest-control to be economically feasible, it is important that the peptide mimetics be small in size in order to reduce the production costs. Hence, future work should focus on CT/DH-R antagonists as opposed to agonists as the peptide length appears to be critical for activation of these receptors.

130

References

Hardy, M.C., Daly, N.L., Mobli, M., Morales, R.A. and King, G.F. (2013). Isolation of an orally active insecticidal toxin from the venom of an Australian tarantula. PLoS ONE. 8, e73136.

Jiang, H., Wei, Z., Nachman, R.J. and Park, Y. (2013). Molecular cloning and functional characterization of the diapause hormone receptor in the corn earworm Helicoverpa zea. Peptides.

Kaskani, C., Poulos, C.P., Zhang, J. and Tobe, S.S. (2009). The synthesis and biological activity of linear and cyclic analogs of the two diuretic peptides of Diploptera punctata. Peptides. 30, 603-7.

Krupke, C.H., Hunt, G.J., Eitzer, B.D., Andino, G. and Given, K. (2012). Multiple routes of pesticide exposure for honey bees living near agricultural fields. PLoS ONE. 7, e29268.

Nachman, R.J., Hamshou, M., Kaczmarek, K., Zabrocki, J. and Smagghe, G. (2012). Biostable and PEG polymer-conjugated insect pyrokinin analogs demonstrate antifeedant activity and induce high mortality in the pea aphid Acyrthosiphon pisum (Hemiptera: Aphidae). Peptides. 34, 266-73.

Nauen, R. and Denholm, I. (2005). Resistance of insect pests to neonicotinoid insecticides: current status and future prospects. Arch Insect Biochem Physiol. 58, 200-15.

Orchard, I. (2009). Peptides and serotonin control feeding-related events in Rhodnius prolixus. Front Biosci (Elite Ed). 1, 250-62.

131

Te Brugge, V.A., Schooley, D.A. and Orchard, I. (2008). Amino acid sequence and biological activity of a calcitonin-like diuretic hormone (DH31) from Rhodnius prolixus. J Exp Biol. 211, 382-90.

Torosantucci, R., Schoneich, C. and Jiskoot, W. (2014). Oxidation of therapeutic proteins and peptides: structural and biological consequences. Pharm Res. 31, 541-53.

Whitehorn, P.R., O'Connor, S., Wackers, F.L. and Goulson, D. (2012). Neonicotinoid pesticide reduces bumble bee colony growth and queen production. Science. 336, 351-2.

Zandawala, M. (2012). Calcitonin-like diuretic hormones in insects. Insect Biochem Mol Biol. 42, 816-25.

Zandawala, M., Li, S., Hauser, F., Grimmelikhuijzen, C.J. and Orchard, I. (2013). Isolation and Functional Characterization of Calcitonin-Like Diuretic Hormone Receptors in Rhodnius prolixus. PLoS ONE. 8, e82466.

Zandawala, M. and Orchard, I. (2013). Post-feeding physiology in Rhodnius prolixus: The possible role of FGLamide-related allatostatins. Gen Comp Endocrinol. 194, 311-7.

132

Acknowledgements

The authors wish to thank Nikki Sarkar for maintaining the colony. This work was supported by a Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery Grant to

I.O. and a NSERC Canadian Graduate Scholarship to M.Z.

133

Chapter 5: Cloning of the cDNA, localization, and physiological effects of FGLamide-related allatostatins in the

blood-gorging bug, Rhodnius prolixus

The proceeding chapter was reprinted/reproduced with permission from Elsevier.

Full citation details:

Cloning of the cDNA, localization, and physiological effects of FGLamide-related allatostatins in the blood-gorging bug, Rhodnius prolixus.

Zandawala M, Lytvyn Y, Taiakina D, Orchard I.

Insect Biochem Mol Biol. 2012 Jan;42(1):10-21. doi: 10.1016/j.ibmb.2011.10.002.

134

Abstract

Allatostatins (ASTs) are insect neuropeptides that were first identified as inhibitors of juvenile hormone biosynthesis by the corpora allata. There are three families of ASTs in insects, defined by their C-terminus conserved regions, one of which is FGLamide. Here we determine, for the first time in a hemipteran, the complete 1013bp cDNA sequence encoding the Rhodnius prolixus

FGLa/ASTs (Rhopr-FGLa/ASTs), and confirm the transcript size using northern blot.

Phylogenetic analysis suggests that the Rhopr-FGLa/AST prepropeptide is most similar to the

FGLa/AST precursors identified in Hymenoptera. Reverse-transcriptase PCR demonstrates that the Rhopr-FGLa/AST transcript is highly expressed in the central nervous system (CNS) in unfed fifth-instar R. prolixus, and is reduced in expression in CNS dissected from one day old blood- fed insects. Fluorescent in situ hybridization shows transcript expression in neurons in each ganglion of the CNS, but also in cells located on peripheral nerves. Rhopr-FGLa/ASTs dose- dependently inhibit contractions of the anterior midgut and hindgut, suggesting a role in feeding- related physiological events.

135

Introduction

Allatostatins (ASTs) were first identified by their ability to inhibit the biosynthesis of juvenile hormone (JH) by the corpora allata (CA) in insects (see Bendena et al., 1999; Bendena et al.,

1997; Tobe and Bendena, 2006). There are three types of ASTs that differ in their C-terminus conserved regions: the cockroach type (FGLa/ASTs), cricket type (MIP/ASTs), and moth type

(PISCF/ASTs). The FGLa/ASTs were the first to be identified when they were isolated from

Diploptera punctata (Pratt et al., 1991; Woodhead et al., 1989). This neuropeptide family inhibits JH biosynthesis in only cockroaches, crickets, termites and locusts (see Elliott et al.,

2010). FGLa/ASTs are now known to be present in numerous other insect orders (although currently not known in Coleoptera), where they possess other physiological roles, especially as inhibitors of visceral muscle contraction (see Tobe and Bendena, 2006). Indeed, it is now considered that their role as inhibitors of JH biosynthesis is a secondarily evolved function, and that inhibition of visceral muscle contraction is likely the ancestral function (see Tobe and

Bendena, 2006). Myoinhibitory activity of FGLa/ASTs on the gut has been found in a variety of species. For example, FGLa/ASTs inhibit spontaneous contractions of the foregut in Leucophaea maderae (Duve et al., 1995), proctolin-induced muscle contractions of the D. punctata midgut

(Fuse et al., 1999) and spontaneous and proctolin-induced hindgut muscle contractions of D. punctata (Lange et al., 1995; Lange et al., 1993). FGLa/AST-like immunoreactivity is found in neurons within the central nervous system (CNS), but is also associated with the foregut, midgut, and hindgut, as well as open-type midgut endocrine cells in a variety of insects (see Robertson and Lange, 2010), which suggest that ASTs play an important role in feeding-related physiological events.

136

FGLa/AST encoding genes (FGLa/ASTs) have also been identified in a variety of insect orders, including Orthoptera (Belles et al., 1999; Meyering-Vos et al., 2001; Vanden Broeck et al.,

1996), Diptera (East et al., 1996; Lenz et al., 2000; Veenstra et al., 1997), Lepidoptera (Abdel-

Latief et al., 2004; Davey et al., 1999; Duve et al., 1997; Secher et al., 2001), and Blattaria

(Belles et al., 1999; Ding et al., 1995; Donly et al., 1993). The FGLa/AST peptides are all derived from a single polypeptide precursor, a prepropeptide, except for the blowfly, Calliphora vomitoria, which has two FGLa/ASTs (Bendena et al., 1997; Davey et al., 1999; see Tobe and

Bendena, 2006).

Recently, a putative FGLa/AST (Rhopr-FGLa/AST) prepropeptide, as well as some Rhopr-

FGLa/AST peptides, were identified in the medically-important bug, Rhodnius prolixus (Ons et al., 2009; Ons et al., 2011). These are the first to be reported for any hemipteran, but are particularly important in this case because of the involvement of FGL/ASTs in insect feeding activities coupled to the blood-gorging feeding habits of R. prolixus, and the fact that the transmission of Chagas’ disease follows feeding on the human host.

In the present study we have extended the work of Ons et al., (2009, 2011) and used a CNS cDNA library of R. prolixus to clone the complete Rhopr-FGLa/AST cDNA sequence; the first time in a hemipteran. Reverse transcriptase PCR (RT-PCR) was performed to determine the spatial and temporal expression pattern of Rhopr-FGLa/AST in fifth-instar R. prolixus, and fluorescent in situ hybridization (FISH) was also performed to localize cell-specific expression of Rhopr-FGLa/AST within the CNS. Two of the Rhopr-FGL/ASTs were found to be potent inhibitors of both anterior midgut and hindgut contractions in R. prolixus, confirming the involvement of this family of peptides in feeding-related activities.

137

Material and methods

Animals

Male and female fifth-instar R. prolixus were obtained from a long standing colony at the

University of Toronto Mississauga. Insects were reared at high relative humidity at 25°C in incubators and routinely fed on defibrinated rabbit blood (Cedarlane Laboratories Inc.,

Burlington, ON, Canada). All tissues were dissected from 4 - 6 weeks post-fed fifth-instar insects, in nuclease-free phosphate-buffered saline (PBS) (Sigma-Aldrich, Oakville, ON,

Canada), or in saline for the physiological assays. For temporal expression analysis, CNSs were dissected from the following fifth-instar insects: unfed fifth-instars 3 weeks prior to and immediately prior to feeding the remaining insects; 1 day post-feeding, 2 days post-feeding and

10 days post-feeding. Tissues were stored at -20°C in RNA laterTM RNA stabilization reagent

(Qiagen Inc., Mississauga, ON, Canada) prior to use.

Screening of fifth-instar CNS cDNA library

Rhopr-FGLa/AST peptide sequences (Ons et al., 2009) were used to search the R. prolixus preliminary genome assembly using BLAST (Altschul et al., 1990). Gene-specific primers,

RhoprAST-FOR1 and RhoprAST-REV1 (see Appendices: Table S1), were designed and used to amplify the Rhopr-FGLa/AST open reading frame (ORF). A fifth-instar R. prolixus CNS cDNA library (Paluzzi et al., 2008) was used as the template. All PCR reactions were performed using s1000 thermal cycler (Bio-Rad Laboratories, Mississauga, ON, Canada) with the following temperature-cycling profile: initial denaturation at 95°C for 3 min, followed by 35 cycles of

94°C for 30 sec, 60°C for 30 sec and 72°C for 1 min, followed by 10 min final extension at

72°C. The PCR product was gel extracted using the EZ-10 Spin Column DNA Gel Extraction

138

Kit (Bio Basic Inc., Markham, ON, Canada). The gel extracted product was cloned using the pGEM-T Vector System (Promega, Madison, WI, USA). Plasmid DNA, isolated from the overnight culture using EZ-10 Spin Column Plasmid DNA Miniprep Kit (BioBasic Inc.,

Markham, ON, Ontario), was sequenced at the Centre of Applied Genomics at the Hospital for

Sick Children (MaRS Centre, Toronto, Ontario, Canada).

Next, a modified 3’ and 5’ rapid amplification of cDNA ends (RACE) PCR approach using fifth- instar R. prolixus CNS cDNA library was employed to obtain the complete Rhopr-FGLa/AST cDNA sequence. Series of forward and reverse gene-specific primers for 3’ and 5’ RACE, respectively, were designed using the ORF and used in combination with plasmid-specific primers (see Appendices: Table S2 and S3). For 3’ RACE, nested PCR was performed using one plasmid-specific reverse primer (pDNR-LIB3 -88REV) and two gene-specific forward primers

(AST5 FOR1, AST5 FOR2). One plasmid-specific forward primer (pDNR-LIB FOR1) and three gene-specific reverse primers (AST3 REV1, AST3 REV2, and AST3 REV3) were used for 5’

RACE. The primers were used in succession to amplify a specific product using a nested PCR approach. PCR product of each reaction was column purified using EZ-10 Spin Column PCR

Product Purification Kit (Bio Basic Inc., Markham, ON, Canada) and used for the subsequent

PCR reaction. The products from the final RACE reactions were gel extracted, cloned, and sequenced as described earlier.

Lastly, a forward primer, RhoprAST-FOR5 and a reverse primer, RhoprAST-REV5 (see

Appendices: Table S4) were used to amplify the largest possible Rhopr-FGLa/AST cDNA fragment and thus confirm the full cDNA sequence obtained through RACE. The R. prolixus

CNS cDNA library was used as a template in a PCR reaction with 55°C annealing temperature.

139

The resulting PCR product was cloned and sequenced as described earlier. Sequencing results were confirmed from several independent clones to ensure base accuracy.

cDNA sequence analysis

The deduced Rhopr-FGLa/AST prepropeptide sequence was examined for a potential signal peptide for processing in a secretory pathway using SignalP 3.0 (Bendtsen et al., 2004). The intron-exon boundaries were predicted using a BLAST search of the R. prolixus genome assembly and confirmed using Genie, online software for splice site prediction (Reese et al.,

1997). Dibasic arginine and lysine cleavage sites were predicted using ProP 1.0 online software

(Duckert et al., 2004). ClustalW2 (http://www.ebi.ac.uk/Tools/msa/clustalw2/ - last accessed on

June 26, 2011) was used to align the Rhopr-FGLa/AST prepropeptide (JN559385), with its homologs from: Drosophila melanogaster (AAF97792.1), Glossina morsitans morsitans

(ADD20145.1), Aedes aegypti (AAB08870.1), Anopheles gambiae (XP_313511.3), D. punctata

(AAA18260.1), Blaberus craniifer (AAC72893.1), Blattella germanica (AAC72892.1), Supella longipalpa (AAC72894.1), Periplaneta americana (CAA62500.1), Blatta orientalis

(AAC72895.1), Reticulitermes flavipes (ACN42938.1), Helicoverpa armigera (AAB94674.1),

Spodoptera frugiperda (Q5ZQK7), Bombyx mori (NP_001037036.1), Schistocerca gregaria

(CAA91232.1), Apis mellifera (ADE45320.1), Gryllus bimaculatus (CAC83078.1),

Harpegnathos saltator (EFN80332.1), Camponotus floridanus (EFN68211.1), and Acromyrmex echinatior (EGI57352.1). An alignment figure was obtained using the BOXSHADE 3.21 server

(http://www.ch.embnet.org/software/BOX_form.html - last accessed on June 26, 2011). Multiple sequence alignment was performed and exported to produce a phylogenetic tree using MEGA5

(Tamura et al., 2011). Neighbour-joining (NJ) analysis was performed to obtain an unrooted phylogram. Bootstrap values were obtained based on 1000 replicates.

140

Northern blot hybridization

Northern blot hybridization was performed with total RNA isolated from the CNS using RNeasy

Mini Kit (Qiagen Inc., Mississauga, ON, Canada). 1µg of CNS RNA was used to perform northern hybridization using the protocol described earlier (Zandawala et al., 2011), with the following modifications. To prepare the RNA probe, a 679bp partial Rhopr-FGLa/AST cDNA fragment was amplified using RhoprAST-FOR1 and RhoprAST-REV1 primers (see Appendices:

Table S1) and cloned into pGEM-T Vector System. Plasmid DNA was then isolated, linearized using NcoI and purified using EZ-10 Spin Column PCR Product Purification Kit (Bio Basic Inc.,

Markham, ON, Canada). DIG-labeled anti-sense RNA probe was synthesized from this linearized recombinant plasmid DNA via in vitro transcription using SP6 RNA polymerase supplied with the DIG RNA labeling kit SP6/T7 (Roche Applied Science, Mannheim, Germany).

Hybridization was performed using hybridization solution containing denatured DIG-labeled

RNA probe (~1µg/mL). The blots were exposed at room temperature for various times ranging from 1 minute to 1 hour using Bioflex Scientific Imaging Films (Clonex Corporation, Markham,

ON, Canada).

RT-PCR tissue profiling

The following tissues were used for spatial expression analysis: (1) dorsal vessel, fat body, abdominal nerves, diaphragms and trachea, (2) salivary glands, (3) immature ovaries, (4) immature testes, (5) foregut, (6) anterior midgut, (7) posterior midgut, (8) Malpighian tubules

(MT), (9) hindgut, and (10) CNS. Moreover, different parts of the CNS were also dissected separately: (1) brain and suboesophageal ganglion (SOG), (2) prothoracic ganglion (PRO), and

(3) mesothoracic ganglionic mass (MTGM). For temporal expression analysis, CNSs were

141 dissected at different times pre- and post-feeding, as described earlier (see above). Manufacturer supplied protocol was followed to isolate total RNA using RNeasy Mini Kit (Qiagen Inc.,

Mississauga, ON, Canada) and at least 20ng of total RNA was used as a template for cDNA synthesis using iScript™ Reverse Transcription Supermix for RT-qPCR (Bio-Rad Laboratories

Ltd., Mississauga, ON, Canada). An aliquot of this single-stranded cDNA was used as a template in a subsequent PCR reaction. A 679bp segment of Rhopr-FGLa/AST was amplified using the forward primer, RhoprAST-FOR1 and the reverse primer, RhoprAST-REV1 (see Appendices:

Table S1). A 318bp fragment of Rhopr-β-actin (housekeeping gene) was also amplified using the forward primer, ACTIN FOR1 and the reverse primer, ACTIN REV1 (see Appendices: Table

S5). The temperature cycling profile used was similar to that described earlier, but amplification was performed for 30 cycles instead of 35 (see above).

Expression localization using fluorescent in situ hybridization (FISH)

Cell-specific expression in CNS of unfed fifth-instar R. prolixus was assessed using FISH protocols described earlier (Paluzzi and Orchard, 2010; Paluzzi et al., 2008; Zandawala et al.,

2011), with the following adaptations. DIG-labeled anti-sense RNA probe was synthesized as described earlier for northern hybridization; however, a different set of primers (RhoprAST-

FOR3 and RhoprAST-REV2) were used to amplify a 155bp Rhopr-FGLa/AST cDNA fragment

(see Appendices: Table S6). Similarly, DIG-labeled sense RNA probe was synthesized for a negative control using the same recombinant plasmid that was linearized with NdeI instead of

NcoI. CNSs were dissected in PBS and fixed in a freshly prepared 4% paraformaldehyde (in

PBS) solution for no longer than 2 hours at room temperature; female and male tissues were kept separate throughout the entire protocol. Following the quenching of the endogenous peroxidase activity, tissues were incubated in 4% Triton X-100 (Sigma Aldrich, Oakville, Ontario, Canada)

142 in PBS for 1.5 hours at room temperature. Prehybridization was performed for 2 hours at 56°C and hybridization was performed in hybridization solution containing probe at a final concentration of approximately 1.75ng/µL. Biotin-SP-conjugated IgG fraction monoclonal mouse antidigoxin (Jackson ImmunoResearch Laboratories Inc., West Grove, PA, USA) was used for in situ hybridization signal detection. Samples were incubated with Alexa Fluor 568 tyramide solution diluted 1:200 with amplification buffer. ZEN 2009 LE software (Zeiss, Jena,

Germany) was used to obtain images which were then analyzed using ImageJ (Collins, 2007).

Radioimmunoassay (RIA) quantification of FGLa/ASTs

Peptides were extracted from CNS and an RIA was performed as previously described (see Yagi et al., 2005) using an anti-Dippu-AST-7 antiserum. Data was expressed as Dippu-AST-7 equivalents.

Muscle contraction assays

The anterior midgut, hindgut and dorsal vessel from unfed fifth-instar R. prolixus were set up for monitoring muscle contractions as previously described (Sarkar et al., 2003; Te Brugge et al.,

2009; Te Brugge and Orchard, 2002; Te Brugge et al., 2002). The amplitude and frequency of hindgut contractions were monitored before and after application of synthetic Rhopr-FGLa/ASTs by way of a force transducer (Aksjeselskapet Mikro-elektronikk, Horten, Norway). The contractions were recorded using BIOPAC MP100 system hardware and AcqKnowledge MP100

Manager software (BIOPAC Systems, Inc, Santa Barbara, CA, USA). The frequency of anterior midgut and dorsal vessel contractions was monitored before and after application of Rhopr-

FGLa/ASTs by way of an impedance monitor (UFI, Morro Bay, CA, USA). Rhopr-FGLa/AST-2

143

(LPVYNFGLamide) and Rhopr-FGLa/AST-3 (AHNEGRLYSFGLamide) were custom

synthesized by GenScript (Piscataway, NJ, USA) at > 95% purity.

Results

Rhopr-FGLa/AST transcript

The complete cDNA encoding Rhopr-FGLa/AST was cloned and sequenced (Figure 1A). The

(A) sequence is at least 1013bp long, which includes a 612bp ORF. The size of the RNA transcript (B) was confirmed using northern hybridization, which estimated a transcript size, including the

poly(A) tail, of approximately 1300bp (Figure 1C). The 5’ and 3’ UTRs are 44bp and 357bp

long, respectively. Within the 3’ UTR is a poly(A) tail that is at least 29bp long. There are 5

polyadenylation consensus sequences (AATAAA) in the 3’ UTR of Rhopr-FGLa/AST (Figure

1A). Proteins that cleave the 3’ end of RNA recognize and bind this consensus sequence,

following which the poly(A) tail is added (Murthy and Manley, 1995; Zhao et al., 1999).

Molecular organization of Rhopr-FGLa/AST was determined using a combination of a BLAST

search and splice site prediction. Rhopr-FGLa/AST comprises 4 exons which are 33bp, 375bp,

193bp and 383bp long (Figure 1B). These exons are separated by 3 introns which are ~30670bp,

~4984bp, and 1818bp long. The complete gene is predicted to be approximately 38.5kb long.

The large size of Rhopr-FGLa/AST is a result of its large introns.

Sequence and phylogenetic analysis of Rhopr-FGLa/AST prepropeptide

The Rhopr-FGLa/AST ORF encodes a 203 amino acid long prepropeptide. The prepropeptide

contains a highly predicted signal peptide, which is most likely cleaved between alanine at

position 20 and isoleucine at position 21. The prepropeptide contains 7 predicted Rhopr-

FGLa/AST peptides that could result following post-translational cleavage at dibasic and

144 tribasic, lysine and arginine residues (Table 1). The first Rhopr-FGLa/AST peptide (Rhopr-

FGLa/AST-1) has an unusual LGL-NH2 C-terminus. Rhopr-FGLa/AST-1 is flanked by the signal peptide at the N-terminus and lysine and arginine dibasic amino acids at the C-terminus.

All the subsequent Rhopr-FGLa/AST peptides are flanked by lysine and arginine dibasic amino acids at the N-terminus and glycine followed by lysine and arginine dibasic amino acids at the C- terminus. The glycine residue at the C-terminus is predicted to undergo amidation. In addition to the common lysine and arginine cleavage sites, there are two unconventional cleavage sites present in the prepropeptide; there is a rare lysine-lysine dibasic cleavage site and a lysine- arginine-lysine tribasic cleavage site (Figure 1A). The lysine-lysine site is predicted to be non- functional in this case as it lacks a cysteine residue at -3 position (Veenstra, 2000). As a result, this dibasic residue pair is part of Rhopr-FGLa/AST-1 (Table 1). The lysine-arginine-lysine tribasic cleavage site is also unique in the sense that it is predicted to undergo cleavage at either the lysine-arginine pair or at all 3 residues. Depending on the position of the cleavage, the length of Rhopr-FGLa/AST-7 would be affected (Table 1); the partial cleavage could result in some peptides with lysine at the N-terminus and some without. Peaks corresponding to the predicted masses for both these peptides are obtained with mass spectrometry analysis, suggesting partial cleavage at this site, although there was not sufficient material for sequencing to confirm this.

The 7 predicted Rhopr-FGLa/AST peptides consist of 8 (Rhopr-FGLa/AST-2) to 17 (Rhopr-

FGLa/AST-1) amino acids.

Predicted or cloned FGLa/AST prepropeptide sequences from other insect species were aligned to study conservation of the FGLa/AST prepropeptide. Identical and similar amino acids in the aligned sequences were highlighted with a 50% conservation cut-off (Figure 2). The alignment shows that FGLa/AST prepropeptide is not well conserved across insects. The FGLa/AST

145 prepropeptide sequences in insects are well conserved only at the C-terminus of each encoded

FGLa/AST peptide. However, Rhopr-FGLa/AST-2 (peptide #2 on the alignment) is most highly conserved across all the species observed. Interestingly, the prepropeptide region between

Rhopr-FGLa/AST-1 and Rhopr-FGLa/AST-2 is well conserved in all species, except the dipterans. There is also a great variation in the length of FGLa/AST prepropeptide sequences across various insects. The sequence lengths vary from 151 amino acids in D. melanogaster, to

380 amino acids in R. flavipes, with R. prolixus sequence (203 amino acids) lying in the middle of the spectrum (see Appendices: Table S7). There is also variation in the number of FGLa/AST peptides that could result from the prepropeptide; four FGLa/AST peptides are predicted in several dipteran and hymenopteran species, including D. melanogaster, and 14 peptides in several cockroach species including B. craniifer and B. germanica. Rhodnius prolixus is in the middle of the range, where 7 Rhopr-FGLa/AST peptides are predicted.

For phylogenetic analysis, 21 FGLa/AST prepropeptide sequences from various insects across several orders were used. Apart from Rhopr-FGLa/AST prepropeptide sequence, no other hemipteran FGLa/AST sequences are fully available. Our analysis indicates that Rhopr-

FGLa/AST prepropeptide sequence is sister to a monophyletic group comprising hymenopteran sequences (Figure 3). Moreover, the clade comprising the R. prolixus and hymenopteran sequences is sister to all other insect sequences. Dipteran sequences form a monophyletic group and so do the lepidopteran and orthopteran sequences. Interestingly, sequences representing the cockroach species (Blattaria) do not form a monophyletic group. The FGLa/AST precursor sequence of B. germanica (Blattaria) is most similar to the R. flavipes (Isoptera) sequence.

146

Table 1: Rhopr-FGLa/AST predicted peptides.

No. Peptide sequence

1 1 INDREDDFNKKLTELGL-NH2

2 LPVYNFGL-NH2

3 AHNEGRLYSFGL-NH2

4 AAKMYSFGL-NH2

5 LPSIKYPEGKMYSFGL-NH2

6 SNPNGHRFSFGL-NH2

2 7 KGERSMQYSFGL-NH2

1 non-functional KK dibasic cleavage site

2 K residue that is partially cleaved

147

Figure 1: Rhopr-FGLa/AST sequence and structure. (A) Rhopr-FGLa/AST cDNA and its deduced amino acid sequence. The numbering for each sequence is on the right. Nucleotides highlighted in black denote exon-exon boundaries. Polyadenylation consensus sequences are underlined, while the partial poly(A) tail is dashed underlined. Capitalized methionine codon indicates the translation start site. Amino acid sequences of the 7 predicted Rhopr-FGLa/AST peptides are bolded. The dibasic and tribasic, lysine and arginine residues, denoting post- translation cleavage sites, are shaded in grey. The lysine-lysine cleavage site (underlined) is not predicted to get cleaved while the lysine-arginine-lysine cleavage site (dashed underlined) could get cleaved at either the first two or all three residues. Glycine residues needed for amidation are boxed. The predicted site for signal sequence cleavage is shown with an arrow. (B) Rhopr-

FGLa/AST structure determined using BLAST and intron prediction. Boxes represent exons and lines or dashed lines (estimated length) represent introns. (C) Northern blot analysis of Rhopr-

FGLa/AST. Hybridization was performed with an anti-sense probe complimentary to the Rhopr-

FGLa/AST ORF. The size of RNA molecular weight markers is indicated.

148

149

Figure 2: Multiple sequence alignment of the Rhopr-FGLa/AST prepropeptide from different insects. ClustalW2 was used to align FGLa/AST prepropeptide sequences from the fruit fly, D. melanogaster (AAF97792.1), the tsetse fly, G. morsitans morsitans (ADD20145.1), the yellow fewer mosquito, A. aegypti (AAB08870.1), the African malaria mosquito, A. gambiae

(XP_313511.3), the Pacific beetle cockroach, D. punctata (AAA18260.1), the death’s head cockroach, B. craniifer (AAC72893.1), the German cockroach, B. germanica (AAC72892.1), the brown-banded cockroach, S. longipalpa (AAC72894.1), the American cockroach, P. americana (CAA62500.1), the oriental cockroach, B. orientalis (AAC72895.1), the Eastern subterranian termite, R. flavipes (ACN42938.1), the kissing bug, R. prolixus (JN559385), the cotton bollworm, H. armigera (AAB94674.1), the fall army worm, S. frugiperda (Q5ZQK7), the silkworm, B. mori (NP_001037036.1), the desert locust, S. gregaria (CAA91232.1), the Western honey bee, A. mellifera (ADE45320.1), the field cricket, G. bimaculatus (CAC83078.1), the

Jerdon's jumping ant, H. saltator (EFN80332.1), the Florida carpenter ant, C. floridanus

(EFN68211.1), and the leaf-cutter ant, A. echinatior (EGI57352.1). Identical and similar amino acids in the aligned sequences are highlighted with a 50% majority rule. Location of the 7

Rhopr-FGLa/AST peptides has been indicated.

150

151

Figure 3: An unrooted phylogram obtained using neighbour-joining analysis, showing evolutionary relationships of FGLa/AST prepropeptide from 21 species. Boot strap values

(above 50) based on 1000 replicates are indicated at the split of the branches. The tree is depicted to scale, with branch lengths proportional to the evolutionary distances used to infer the phylogenetic tree.

152

153

Rhopr-FGLa/AST expression profiling

Rhopr-FGLa/AST expression profiling was performed using RT-PCR. Rhopr-FGLa/AST is highly expressed in the CNS and expressed at a lower level in a pool of tissues comprising dorsal vessel, fat bodies, trachea, diaphragms and abdominal nerves (Figure 4A). Within the CNS,

Rhopr-FGLa/AST is expressed at similar levels in brain and SOG, PRO and MTGM (Figure 4B).

Weak expression was also associated with peripheral tissues such as the ovaries, anterior midgut, posterior midgut, Malpighian tubules (MTs) and hindgut. Lastly, Rhopr-FGLa/AST expression in the CNS was determined at different time points pre- and post-feeding. Interestingly, low Rhopr-

FGLa/AST expression was observed in the CNS one day post-feeding, while similar levels of expression were observed for all the other time points, pre- and post-feeding (Figure 4C).

Rhopr-FGLa/AST expression in fifth-instar CNS

FISH was performed to localize Rhopr-FGLa/AST expression within the CNS (Figure 5A, 5B and 5C). The diagrammatic representation of Rhopr-FGLa/AST expression shows that it is expressed in approximately 105 neurons throughout all ganglia of the CNS, that are organized in a bilaterally symmetrical manner (Figure 6), except for 5 dorsal unpaired median (DUM) neurons in the MTGM (Figure 5C, 6). None of these neurons stain in control CNSs hybridized with DIG-labeled sense RNA probe (not shown). Moreover, no obvious staining differences were observed between males and females. Approximately 58 cells in the brain show Rhopr-

FGLa/AST expression. The SOG contains 16 Rhopr-FGLa/AST-expressing cells, including a pair along the midline on the ventral side which are strongly stained. The PRO contains 8 bilaterally- paired cells, of which 2 are located on the dorsal side and 6 on the ventral side. The MTGM contains Rhopr-FGLa/AST expression in 23 cells, of which 16 bilaterally-paired cells are located ventrally. Strong staining was observed in 5 DUM neurons in the distal region of the posterior

154

MTGM. Peripheral Rhopr-FGLa/AST-expressing cells were also observed on abdominal nerves

(Figure 5D).

RIA quantification

FGLa/AST content of the CNS was measured by RIA using an antiserum against the cockroach

Dippu-AST-7. The whole CNS contains about 800 + 50 fmol of Dippu-AST-7 equivalents. This amount is distributed amongst the various ganglia, with the brain containing about 500 + 120 fmol, the SOG 130 + 8 fmol, the PRO 150 + 10 fmol and the MTGM 200 + 10 fmol when assayed separately. Dippu-AST-7 equivalents were also measured in the abdominal nerves, with about 50 fmol present.

155

Figure 4: Expression analyses of Rhopr-FGLa/AST in fifth-instar R. prolixus. (A) Spatial expression pattern. The following tissues were dissected from 10 insects: Pool (dorsal vessel, fat body, abdominal nerves, diaphragms and trachea), SG (salivary glands), OV (ovaries), TS

(testes), FG (foregut), AM (anterior midgut), PM (posterior midgut), MT (Malpighian tubules),

HG (hindgut) and CNS (central nervous system). NT represents an additional negative control with no template used for the cDNA synthesis reaction (n = 4 replicates). (B) Expression pattern within the CNS. The following tissues were dissected from 10 insects: B and SOG (brain and suboesophageal ganglion), PRO (prothoracic ganglion), and MTGM (mesothoracic ganglionic mass). (n=3). (C) Temporal expression pattern within the CNS. CNS was dissected from insects at -3w (3 weeks before feeding), 0 (right before feeding), +1d (1 day post-feeding) +2d (2 days post-feeding) and +10d (10 days post-feeding). (n=2). For all experiments, -ve represents a negative control, where no template was used for the PCR reaction, and Rhopr-β-actin was amplified as a positive control to test the quality of the cDNA.

156

157

Figure 5: Fluorescent in situ hybridization portraying Rhopr-FGLa/AST expression in fifth-instar

R. prolixus dorsal CNS: (A) brain, (B) suboesophageal ganglion (C) mesothoracic ganglionic mass and (D) abdominal nerve. Arrow indicates the cell-specific staining in the abdominal nerve.

Scale bars: A,B: 50µm and C,D: 25µm.

158

159

Figure 6: Diagrammatic representation of Rhopr-FGLa/AST expression on (A) the ventral and

(B) dorsal aspect of the CNS: brain, suboesophageal ganglion (SOG), prothoracic ganglion

(PRO), and mesothoracic ganglionic mass (MTGM). The map is based on at least 20 preparations. Scale bar: 200µm.

160

161

Effects of Rhopr-FGLa/ASTs on muscle contraction

Anterior midguts from fifth instars that still contained blood content from the previous feed as fourth instars consistently produced phasic peristaltic contractions. Both Rhopr-FGLa/AST-2 and

3 resulted in a dose-dependent inhibition in the frequency of these spontaneous contractions

(Figures 7, 8). The effects were reversible upon washing in physiological saline. As can be seen from the dose-response curves, Rhopr-FGLa/AST-3 is more potent than Rhopr-FGLa/AST-2,

-13 -12 -11 with the threshold of the former being between 10 and 10 M, and IC50 at approximately 10

M. Total inhibition is achieved at doses of approximately 10-9 M.

Hindguts from R.prolixus are not spontaneously myoactive; however, consistent longitudinal muscle contractions can be produced by the addition of insect kinins, such as leucokinin 1 (see

Te Brugge and Orchard, 2002). Leucokinin 1 (LK1) produces a change in basal tonus and an increase in frequency and amplitude of superimposed phasic contractions (Figure 9). Rhopr-

FGLa/AST-2 and 3 are both capable of inhibiting 10-8 M LK1- induced hindgut contractions in a dose-dependent manner (Figures 9, 10). Again, Rhopr-FGLa/AST-3 is more potent with

-11 -10 threshold for inhibition of amplitude of phasic contractions between 10 and 10 M, and IC50 at approximately 10-8 M. Maximum inhibition is achieved at doses of approximately 10-5M.

Rhopr-FGLa/AST-3 is also more potent than Rhopr-FGLa/AST-2 when measuring the frequency of contractions (Figure 11). Rhopr-FGLa/AST-2 was also capable of inhibiting the frequency of

-10 -9 -8 heart-beat, with threshold between 10 M and 10 M and EC50 at 1.5 X 10 M (not shown).

162

Figure 7: Sample traces showing the effects of Rhopr-FGLa/AST-2 on spontaneous contractions of R. prolixus anterior midgut. (A,B) Before and after application of 10-10 M Rhopr-FGLa/AST-

2. (C,D) Before and after application of 10-7 M Rhopr-FGLa/AST-2. Note the inhibitory effect of the peptide.

163

164

Figure 8: Dose-response curves for the effects of Rhopr-FGLa/AST-2 (■) and Rhopr-

FGLa/AST-3(▲) on the frequency of spontaneous contractions of R. prolixus anterior midgut.

Results are expressed as the percentage relative to the saline controls.

165

166

Figure 9: Sample traces showing the effects of Rhopr-FGLa/AST-3 on leucokinin 1 (LK1)- induced contractions of R. prolixus hindgut. Control contractions were induced by 10-8 M LK1

(A,C,E) prior to the application of increasing doses of Rhopr-FGLa/AST-3 (B,D,F). Each preparation was thoroughly washed between the control and experimental peptide application.

Peptides applied at the arrow heads. Note the inhibitory effect of Rhopr-FGLa/AST-3 on the frequency of contractions and amplitude of induced basal contraction.

167

168

Figure 10: Dose-response curves for the effects of Rhopr-FGLa/AST-2 (■) and Rhopr-

FGLa/AST-3 (▲) on the amplitude of 10-8 M LK1-induced contractions of R. prolixus hindgut.

Results are expressed as a percentage relative to 10-8 M LK1 alone.

169

170

Figure 11: Dose-response curves for the effects of Rhopr-FGLa/AST-2 (■) and Rhopr-

FGLa/AST-3(▲) on the frequency of 10-8 M LK1-induced phasic contractions of R. prolixus hindgut. Results are expressed as the percentage relative to 10-8 M LK1 alone.

171

172

Discussion

We have isolated and cloned the complete Rhopr-FGLa/AST cDNA sequence from R. prolixus - the first time in a hemipteran. Previously, Ons et al., (2011) submitted the AST A precursor mRNA sequence (GQ856315.1) to GenBank, that is referred to as Rhopr-FGLa/AST; however our extended sequence contains the 3’ UTR. Our sequencing results differ slightly from that of

Ons et al., (2011) by 1 nucleotide at position 185, which is T rather than C confirmed in multiple sequenced clones and from the R. prolixus genome. The 1013bp cDNA sequence contains 5 polyadenylation signal consensus sequences (AATAAA). Since only 1 band was observed in the northern blot analysis, it is most likely that only 1 signal sequence is used for cleavage and addition of a poly(A) tail.

The 203 amino acid long Rhopr-FGLa/AST prepropeptide contains 7 predicted FGLa-like peptides, one of which is an unusual LGLa. Such a peptide has been found in only one other species; that of C. floridanus. Five of these 7 peptides have been de novo sequenced, namely

Rhopr-FGLa/AST- 2,3,4,5 and 6 (Ons et al., 2011). Rhopr-FGLa/AST-2 is highly conserved in all species examined. Rhopr-FGLa/AST-7 has an ambiguous cleavage site (KRK) that might result in the presence of an N-terminus K residue.

Expression of Rhopr-FGLa/AST was predominantly found in the CNS of both female and male insects. Weak expression was observed in the ovaries, anterior midgut, posterior midgut, hindgut, MTs and the pool of dorsal vessel, fat body, diaphragms, abdominal nerves and trachea.

The midgut data is predictable, since previously we have shown FGLa/AST-like immunoreactive endocrine cells in this tissue (Sarkar et al., 2003), as is found in a variety of other insects (see

Tobe and Bendena, 2006). The other tissues are more intriguing, although interestingly

173 expression of Dippu-FGLa/AST has been shown in the oviducts of D. punctata, and has also been shown to change over the reproductive cycle (Garside et al., 2002). Caution must be exercised, however, in attributing expression to the actual tissue. For example, we have found peripheral cells expressing the transcript on peripheral nerves, and we do not know the extent of their distribution. Peripheral neurons expressing a variety of neuropeptides (including the

FGLa/ASTs) are well described in insects (Cantera and Nassel, 1992; Fifield and Finlayson,

1978; O'Brien and Taghert, 1998; Te Brugge et al., 2001; Yoon and Stay, 1995; Zitnan et al.,

1996). Thus, peripheral nerves are likely present in our dissected tissues and any peripheral cells might contaminate the tissue. In addition, FGLa/AST in situ distribution has been detected in a small population of haemocytes in D. punctata (Skinner et al., 1997), although the function of these AST-producing haemocytes is unknown. It is possible that weak Rhopr-FGLa/AST expression could be due to haemocyte contamination of tissues prior to RNA extraction, assuming that R. prolixus haemocytes might also produce Rhopr-FGLa/ASTs. Clearly additional investigations are needed to clarify these data. Lastly, the relative expression levels need to be confirmed using quantitative RT-PCR as the classical RT-PCR method used here is only qualitative.

FISH was performed to localize Rhopr-FGLa/AST expression within the CNS. Expression was consistent with RT-PCR analysis and RIA data, which showed expression in all parts of the

CNS: brain, SOG, PRO, and MTGM. Such a wide distribution is typical in insects (see Stay et al., 1994; Tobe and Bendena, 2006). In addition, expression is consistent with the clusters of cells that are also FGLamide-like immunoreactive (Sarkar et al., 2003). All of the cells detected by FISH were also observed in immunohistochemistry. The immunohistochemical analysis

(Sarkar et al., 2003) reveals a few more cells than are localized by FISH. This might indicate low

174 transcript expression in these cells. Among the brightest neurons stained by FISH were the lateral neurosecretory cells that project over the corpus cardiacum/CA to the dorsal vessel, and 5

DUM neurons in the MTGM. These DUM neurons produce neurohaemal sites over the abdominal nerves, and are distinct from 5 DUM neurons in the MTGM that are serotonergic (see

Orchard et al., 1989; Sarkar et al., 2003) and responsible for the regulation of diuresis (see

Orchard, 2006; Orchard, 2009). These latter serotonergic DUM neurons release serotonin at the time of feeding that stimulates absorption of salts and water across the anterior midgut, secretion of salts and water by the upper MTs, and reabsorption of KCl by the lower MTs. Serotonin also stimulates contraction of anterior midgut and hindgut (see Orchard, 2006; Orchard, 2009). The location of the 5 Rhopr-FGLa/AST expressing DUM neurons alongside the 5 serotonergic DUM neurons in the MTGM presents the interesting possibility of neural integration of the two groups and therefore the possible involvement of Rhopr-FGLa/ASTs in feeding-related activities (see later). There is about 800 fmol of Dippu-AST-7 equivalents in the CNS, and also about 50 fmol associated with the abdominal nerves and presumably their neurohaemal sites. These values are not dissimilar to those reported for other insects (e.g. Yagi et al., 2005; Yu et al., 1993). The absolute values cannot be determined because the antiserum is generated against a peptide that is an FGLamide, but does not have the same amino acid sequence as the native Rhopr-FGLa/ASTs, and therefore relies upon cross-reactivity. Clearly, though, Rhopr-FGLa/ASTs are present throughout the CNS, and are likely to act as neurohormones, in addition to their involvement as brain-gut peptides.

There is currently no evidence for a functional AST controlling JH production in R. prolixus (see

Sarkar et al., 2003). Indeed, severing the nervi corpori cardiaci II in fifth instars does not inhibit progression to the adult stage (Chiang, 2000), and there are no AST-like immunoreactive

175 processes associated with the CA, other than those running over the surface to the dorsal vessel

(Sarkar et al., 2003). Previously it has been suggested that species in which FGLa/ASTs are allatostatic have CAs that are directly innervated by FGLa/AST neurons (see Nassel, 2002).

Interestingly though, the Rhopr-FGLa/AST temporal expression profile shows reduced expression within one day post-feeding, with the levels recovering on the second day post- feeding. This indicates that the Rhopr-FGLa/AST expression might be down-regulated and then up-regulated after feeding, thereby altering the amount of peptide available for release. Blood feeding is the stimulus for growth and development in R. prolixus, but also initiates short term physiological changes associated with digestion, and salt and water balance (see Orchard, 2009).

Interestingly, with this in mind, we have demonstrated that Rhopr-FGLa/ASTs are biologically active on tissues associated with feeding, namely the anterior midgut and hindgut, in addition to the dorsal vessel. This confirms some previous data on the myoinhibitory nature of non-native

FGLs/ASTs in R. prolixus (Sarkar et al., 2003). Gut contractions during feeding produce mixing of the ingested blood meal as well as the surrounding haemolymph. Rhodnius prolixus MTs have been shown to have large unstirred layers surrounding them in vitro, while in vivo these unstirred layers are significantly reduced (see Orchard, 2009). It has been suggested that contractions of the anterior midgut and hindgut reduce the unstirred layers, as well as increase the circulation of hormones within the haemolymph (see Orchard, 2009). These tissues may be considered accessory hearts. Presumably the contractions of anterior midgut might also aid in the absorption of salts and water. Release of Rhopr-FGLa/ASTs into the haemolymph might therefore occur at the cessation of diuresis in R. prolixus, when diuretic hormone release has stopped and urine production is being halted. In this physiological state, the insect has already concentrated the blood meal in the anterior midgut and now needs to begin the process of digestion (which takes several days) and to also conserve water. The rapid mixing of haemolymph and gut-contents

176 during diuresis is no longer required and a decrease in gut contractions might return the insect to a more stable physiological state. It is not known if Rhopr-FGLa/ASTs actually participate in physiological activities involved in the cessation of diuresis in R. prolixus; however, our results suggest this may be a role for Rhopr-FGLa/ASTs in R. prolixus, and this possibility deserves further investigation. There are also other potential physiological states that may require a decrease in gut activity, such as during and after ecdysis or starvation stress. Clearly though, these Rhopr-FGLa/ASTs would appear to be well suited to act as instruments for fine-tuning the contractile activity of the insect’s gut. FGLa/ASTs have previously been shown to be inhibitors of visceral muscle contraction in insects (see Bendena et al., 1999; Duve and Thorpe, 1994;

Lange et al., 1995) where they directly innervate the tissue, and it has been suggested that this may be their ancestral function (see Tobe and Bendena, 2006). In addition they are inhibitors of heart-beat frequency, and so may fine-tune circulation induced by the heart.

The C-terminal FGLamide sequence is highly conserved and therefore implies an important physiological relevance, whereas the N-terminus is not particularly well conserved in insects.

This is consistent with previous studies showing that N-terminally truncated peptides retain their activity provided that the minimal pentapeptide is preserved (Stay et al., 1994; see Tobe and

Bendena, 2006). Considerable variation of the N-terminus has previously been shown to account for difference in relative potencies of FGLa/ASTs in their effect on JH biosynthesis and on muscle activity (Lange et al., 1995; Stay et al., 1994). A similar phenomenon occurs in R. prolixus whereby Rhopr-FGLa/AST-2 and 3 vary in their effectiveness in inhibiting contractions of the anterior midgut and hindgut in R. prolixus.

177

Clearly R. prolixus possesses the FGLa/AST family of peptides which can be considered brain- gut peptides, as well as hormones as suggested by their presence in neurosecretory and endocrine cells, as also seen in other insects. At present, there is no direct evidence that Rhopr-FGLa/ASTs possess allatostatic activity in R. prolixus, and studies to investigate this will have to await the chemical identification of the R. prolixus JH which is currently unknown.

178

References

Abdel-Latief, M., Meyering-Vos, M. and Hoffmann, K.H. (2004). Type-A allatostatins from the fall armyworm, Spodoptera frugiperda: molecular cloning, expression and tissue-specific localization. Arch Insect Biochem Physiol. 56, 120-32.

Altschul, S.F., Gish, W., Miller, W., Myers, E.W. and Lipman, D.J. (1990). Basic local alignment search tool. J Mol Biol. 215, 403-10.

Belles, X., Graham, L.A., Bendena, W.G., Ding, Q.I., Edwards, J.P., Weaver, R.J. and Tobe, S.S. (1999). The molecular evolution of the allatostatin precursor in cockroaches. Peptides. 20, 11-22.

Bendena, W.G., Donly, B.C. and Tobe, S.S. (1999). Allatostatins: a growing family of neuropeptides with structural and functional diversity. Ann N Y Acad Sci. 897, 311-29.

Bendena, W.G., Garside, C.S., Yu, C.G. and Tobe, S.S. (1997). Allatostatins: diversity in structure and function of an insect neuropeptide family. Ann N Y Acad Sci. 814, 53-66.

Bendtsen, J.D., Nielsen, H., von Heijne, G. and Brunak, S. (2004). Improved prediction of signal peptides: SignalP 3.0. J Mol Biol. 340, 783-95.

Cantera, R. and Nassel, D.R. (1992). Segmental peptidergic innervation of abdominal targets in larval and adult dipteran insects revealed with an antiserum against leucokinin I. Cell Tissue Res. 269, 459-71.

Chiang, R.G. (2000). Neural inhibition of the corpus allatum in the last larval instar differs from that in adults in the blood-feeding insect, Rhodnius prolixus. Arch Insect Biochem Physiol. 44, 1- 6.

Collins, T.J. (2007). ImageJ for microscopy. Biotechniques. 43, 25-30.

Davey, M., Duve, H., Thorpe, A. and East, P. (1999). Characterisation of the helicostatin peptide precursor gene from Helicoverpa armigera (Lepidoptera: Noctuidae). Insect Biochem Mol Biol. 29, 1119-27.

179

Ding, Q., Donly, B.C., Tobe, S.S. and Bendena, W.G. (1995). Comparison of the allatostatin neuropeptide precursors in the distantly related cockroaches Periplaneta americana and Diploptera punctata. Eur J Biochem. 234, 737-46.

Donly, B.C., Ding, Q., Tobe, S.S. and Bendena, W.G. (1993). Molecular cloning of the gene for the allatostatin family of neuropeptides from the cockroach Diploptera punctata. Proc Natl Acad Sci U S A. 90, 8807-11.

Duckert, P., Brunak, S. and Blom, N. (2004). Prediction of proprotein convertase cleavage sites. Protein Eng Des Sel. 17, 107-12.

Duve, H., Johnsen, A.H., Maestro, J.L., Scott, A.G., Winstanley, D., Davey, M., East, P.D. and Thorpe, A. (1997). Lepidopteran peptides of the allatostatin superfamily. Peptides. 18, 1301-9.

Duve, H., Johnsen, A.H., Scott, A.G. and Thorpe, A. (1995). Isolation, identification and functional significance of [Hyp2]Met-callatostatin and des Gly-Pro Met-callatostatin, two further post-translational modifications of the blowfly neuropeptide Met-callatostatin. Regul Pept. 57, 237-45.

Duve, H. and Thorpe, A. (1994). Distribution and functional significance of Leu-callatostatins in the blowfly Calliphora vomitoria. Cell Tissue Res. 276, 367-79.

East, P., Tregenza, K., Duve, H. and Thorpe, A. (1996). Identification of the dipteran Leu- callatostatin peptide family: characterisation of the prohormone gene from Calliphora vomitoria and Lucilia cuprina. Regul Pept. 67, 1-9.

Elliott, K.L., Chan, K.K. and Stay, B. (2010). Evidence for a Phe-Gly-Leu-amide-like allatostatin in the beetle Tenebrio molitor. Peptides. 31, 402-7.

Fifield, S.M. and Finlayson, L.H. (1978). Peripheral Neurons and Peripheral Neurosecretion in the Stick Insect, Carausius morosus. Proc R Soc Lond (Biol). 200, 63-85.

Fuse, M., Zhang, J.R., Partridge, E., Nachman, R.J., Orchard, I., Bendena, W.G. and Tobe, S.S. (1999). Effects of an allatostatin and a myosuppressin on midgut carbohydrate enzyme activity in the cockroach Diploptera punctata. Peptides. 20, 1285-93.

180

Garside, C.S., Koladich, P.M., Bendena, W.G. and Tobe, S.S. (2002). Expression of allatostatin in the oviducts of the cockroach Diploptera punctata. Insect Biochem Mol Biol. 32, 1089-99.

Lange, A.B., Bendena, W.G. and Tobe, S.S. (1995). The effect of the thirteen Dip-allatostatins on myogenic and induced contractions of the cockroach (Diploptera punctata) hindgut. J Insect Physiol. 41, 581-588.

Lange, A.B., Chan, K.K. and Stay, B. (1993). Effect of allatostatin and proctolin on antennal pulsatile organ and hindgut muscle in the cockroach, Diploptera punctata. Arch Insect Biochem Physiol. 24, 79-92.

Lenz, C., Williamson, M. and Grimmelikhuijzen, C.J. (2000). Molecular cloning and genomic organization of an allatostatin preprohormone from Drosophila melanogaster. Biochem Biophys Res Commun. 273, 1126-31.

Meyering-Vos, M., Wu, X., Huang, J., Jindra, M., Hoffmann, K.H. and Sehnal, F. (2001). The allatostatin gene of the cricket Gryllus bimaculatus (Ensifera, Gryllidae). Mol Cell Endocrinol. 184, 103-14.

Murthy, K.G. and Manley, J.L. (1995). The 160-kD subunit of human cleavage- polyadenylation specificity factor coordinates pre-mRNA 3'-end formation. Genes Dev. 9, 2672- 83.

Nassel, D.R. (2002). Neuropeptides in the nervous system of Drosophila and other insects: multiple roles as neuromodulators and neurohormones. Prog Neurobiol. 68, 1-84.

O'Brien, M.A. and Taghert, P.H. (1998). A peritracheal neuropeptide system in insects: release of myomodulin-like peptides at ecdysis. J Exp Biol. 201, 193-209.

Ons, S., Richter, F., Urlaub, H. and Pomar, R.R. (2009). The neuropeptidome of Rhodnius prolixus brain. Proteomics. 9, 788-92.

Ons, S., Sterkel, M., Diambra, L., Urlaub, H. and Rivera-Pomar, R. (2011). Neuropeptide precursor gene discovery in the Chagas disease vector Rhodnius prolixus. Insect Mol Biol. 20, 29-44.

181

Orchard, I. (2006). Serotonin: a coordinator of feeding-related physiological events in the blood-gorging bug, Rhodnius prolixus. Comp Biochem Physiol A Mol Integr Physiol. 144, 316- 24.

Orchard, I. (2009). Peptides and serotonin control feeding-related events in Rhodnius prolixus. Front Biosci (Elite Ed). 1, 250-62.

Paluzzi, J.P. and Orchard, I. (2010). A second gene encodes the anti-diuretic hormone in the insect, Rhodnius prolixus. Mol Cell Endocrinol. 317, 53-63.

Pratt, G.E., Farnsworth, D.E., Fok, K.F., Siegel, N.R., McCormack, A.L., Shabanowitz, J., Hunt, D.F. and Feyereisen, R. (1991). Identity of a second type of allatostatin from cockroach brains: an octadecapeptide amide with a tyrosine-rich address sequence. Proc Natl Acad Sci U S A. 88, 2412-6.

Reese, M.G., Eeckman, F.H., Kulp, D. and Haussler, D. (1997). Improved splice site detection in Genie. J Comput Biol. 4, 311-23.

Robertson, L. and Lange, A.B. (2010). Neural substrate and allatostatin-like innervation of the gut of Locusta migratoria. J Insect Physiol. 56, 893-901.

Sarkar, N.R., Tobe, S.S. and Orchard, I. (2003). The distribution and effects of Dippu- allatostatin-like peptides in the blood-feeding bug, Rhodnius prolixus. Peptides. 24, 1553-62.

Secher, T., Lenz, C., Cazzamali, G., Sorensen, G., Williamson, M., Hansen, G.N., Svane, P. and Grimmelikhuijzen, C.J. (2001). Molecular cloning of a functional allatostatin gut/brain receptor and an allatostatin preprohormone from the silkworm Bombyx mori. J Biol Chem. 276, 47052-60.

182

Skinner, J.R., Fairbairn, S.E., Woodhead, A.P., Bendena, W.G. and Stay, B. (1997). Allatostatin in hemocytes of the cockroach Diploptera punctata. Cell Tissue Res. 290, 119-28.

Stay, B., Tobe, S.S. and Bendena, W.G. (1994). Allatostatins: identification, primary structures, functions and distribution. Adv Insect Physiol. 25, 267-337.

Te Brugge, V., Ianowski, J.P. and Orchard, I. (2009). Biological activity of diuretic factors on the anterior midgut of the blood-feeding bug, Rhodnius prolixus. Gen Comp Endocrinol. 162, 105-12.

Te Brugge, V.A., Nassel, D.R., Coast, G.M., Schooley, D.A. and Orchard, I. (2001). The distribution of a kinin-like peptide and its co-localization with a CRF-like peptide in the blood- feeding bug, Rhodnius prolixus. Peptides. 22, 161-73.

Te Brugge, V.A. and Orchard, I. (2002). Evidence for CRF-like and kinin-like peptides as neurohormones in the blood-feeding bug, Rhodnius prolixus. Peptides. 23, 1967-79.

Te Brugge, V.A., Schooley, D.A. and Orchard, I. (2002). The biological activity of diuretic factors in Rhodnius prolixus. Peptides. 23, 671-81.

Tobe, S.S. and Bendena, W.G., 2006. Chapter 31 - Allatostatins in the Insects, in: Abba, J.K. (Ed.), Handbook of Biologically Active Peptides. Academic Press, Burlington, pp. 201-206.

Vanden Broeck, J., Veelaert, D., Bendena, W.G., Tobe, S.S. and De Loof, A. (1996). Molecular cloning of the precursor cDNA for schistostatins, locust allatostatin-like peptides with myoinhibiting properties. Mol Cell Endocrinol. 122, 191-8.

Veenstra, J.A. (2000). Mono- and dibasic proteolytic cleavage sites in insect neuroendocrine peptide precursors. Arch Insect Biochem Physiol. 43, 49-63.

Veenstra, J.A., Noriega, F.G., Graf, R. and Feyereisen, R. (1997). Identification of three allatostatins and their cDNA from the mosquito Aedes aegypti. Peptides. 18, 937-42.

183

Yagi, K.J., Kwok, R., Chan, K.K., Setter, R.R., Myles, T.G., Tobe, S.S. and Stay, B. (2005). Phe-Gly-Leu-amide allatostatin in the termite Reticulitermes flavipes: content in brain and corpus allatum and effect on juvenile hormone synthesis. J Insect Physiol. 51, 357-65.

Yoon, J.G. and Stay, B. (1995). Immunocytochemical localization of Diploptera punctata allatostatin-like peptide in Drosophila melanogaster. J Comp Neurol. 363, 475-88.

Yu, C.G., Stay, B., Joshi, S. and Tobe, S.S. (1993). Allatostatin content of brain, corpora allata and haemolymph at different developmental stages of the cockroach, Diploptera punctata: Quantitation by ELISA and bioassay. J Insect Physiol. 39, 111-122.

Zandawala, M., Paluzzi, J.P. and Orchard, I. (2011). Isolation and characterization of the cDNA encoding DH(31) in the kissing bug, Rhodnius prolixus. Mol Cell Endocrinol. 331, 79-88.

Zitnan, D., Kingan, T.G., Hermesman, J.L. and Adams, M.E. (1996). Identification of ecdysis-triggering hormone from an epitracheal endocrine system. Science. 271, 88-91.

184

Acknowledgements

This research was funded through an NSERC Discovery Grant to I.O. We would like to thank

Rodney Kwok for performing the RIAs, Angela Wang for performing the dorsal vessel assays, and Nikki Sarkar for maintaining the colony and providing assistance with the feeding experiments.

185

Appendices

Table S1: Primers used to sequence Rhopr-FGLa/AST ORF.

Primer Tm %GC Sequence (5’-3’) RhoprAST-FOR1 64.3 47.62 GAATCCGCTACAATGATGCTG RhoprAST-REV1 64.8 42.86 TGATTTTGCTGGATGTTGAGG

186

Table S2: Primers used to perform 3’ RACE reactions.

Primer Tm %GC Sequence (5’-3’) AST5 FOR1 63.6 43.48 AGATGAACGAGAAGAGAAAAGGG AST5 FOR2 63.3 37.50 TTGGATTAGGAAAAAGAACTCAGC pDNR-LIB3 -88REV 63.3 52.17 AGTCATACCAGGATCTCCTAGGG

187

Table S3: Primers used to perform 5’ RACE reactions.

Primer Tm %GC Sequence (5’-3’) pDNR-LIB FOR1 63.9 52.38 GTGGATAACCGTATTACCGCC AST3 REV1 63.3 47.62 GTTTTCCTAATCCCAGTTCGG AST3 REV2 64.0 52.63 CGTTAATGGCTTGTGCACC AST3 REV3 65.6 50.00 CACTAGCAGGACAATGAATGGC

188

Table S4: Primers used to amplify the complete Rhopr-FGLa/AST cDNA sequence.

Primer Tm %GC Sequence (5’-3’) RhoprAST-FOR5 58.0 45.45 CTCAGTTAACTTGGTGACTACG RhoprAST-REV5 58.5 6.90 TTTGATCAAAAAATTTATTTATTTTTAAT

189

Table S5: Primers used to amplify Rhopr-β-actin (housekeeping gene) in RT-PCR reactions.

Primer Tm %GC Sequence (5’-3’) ACTIN FOR1 63.7 50.0 ACACCCAGTTTTGCTTACGG ACTIN REV1 63.6 55.56 GTTCGGCTGTGGTGATGA

190

Table S6: Primers used to synthesize RNA probe for fluorescent in situ hybridization.

Primer Tm %GC Sequence (5’-3’) RhoprAST-FOR3 64.4 39.1 TGATTCATCAGAAGAAAGCAAGC RhoprAST-REV2 64.2 39.3 GATCTAGCTGAGTTCTTTTTCCTAATCC

191

Table S7: Summary of FGLa/AST prepropeptides used for phylogenetic analysis.

Species Common name Length FGLa/AST Accession (amino peptides number acids) predicted R. flavipes Eastern subterranian termite 380 14 ACN42938.1 P. americana American cockroach 379 14 CAA62500.1 B. germanica German cockroach 375 14 AAC72892.1 B. craniifer Death’s head cockroach 371 14 AAC72893.1 S. longipalpa Brown-banded cockroach 350 14 AAC72894.1 G. bimaculatus Field cricket >345 14 CAC83078.1 B. orientalis Oriental cockroach 316 14 AAC72895.1 D. punctata Pacific beetle cockroach 370 13 AAA18260.1 S. gregaria Desert locust 283 10 CAA91232.1 H. armigera Cotton bollworm 225 8 AAB94674.1 B. mori Silkworm 204 8 NP_001037036.1 R. prolixus Kissing bug 203 7 JN559385 S. frugiperda Fall army worm 186 6 Q5ZQK7 A. gambiae African malaria mosquito 201 5 XP_313511.3 A. mellifera Western honey bee 197 5 ADE45320.1 A. aegypti Yellow fewer mosquito 197 5 AAB08870.1 C. floridanus Florida carpenter ant 193 5 EFN68211.1 A. echinatior Leaf-cutter ant 190 4 EGI57352.1 H. saltator Jerdon's jumping ant 179 4 EFN80332.1 G. morsitans morsitans Tsetse fly 163 4 ADD20145.1 D. melanogaster Fruit fly 151 4 AAF97792.1

192

Chapter 6: Post-feeding physiology in Rhodnius prolixus: the

possible role of FGLamide-related allatostatins

The proceeding chapter was reprinted/reproduced with permission from Elsevier.

Full citation details:

Post-feeding physiology in Rhodnius prolixus: the possible role of FGLamide-related allatostatins.

Zandawala M, Orchard I.

Gen Comp Endocrinol. 2013 Dec 1;194:311-7. doi: 10.1016/j.ygcen.2013.10.005.

193

Abstract

Allatostatins (ASTs) are neuropeptides that were first identified as inhibitors of juvenile hormone biosynthesis by the corpora allata of some insect species. The FGLamide-related ASTs

(FGLa/ASTs) belong to one of three families of insect ASTs. Previously, we showed that

Rhodnius prolixus FGLa/ASTs (Rhopr-FGLa/ASTs) are present throughout the R. prolixus central nervous system and are associated with 5 dorsal unpaired median (DUM) neurons in the mesothoracic ganglionic mass. A similar set of neurons contain serotonin which is a diuretic hormone in R. prolixus. Rhopr-FGLa/ASTs inhibit both spontaneous contractions of the anterior midgut and leucokinin-1-induced hindgut contractions. Since these tissues are involved with post-feeding diuresis, these data suggest a possible role for FGLa/ASTs in events associated with feeding, and a possible interaction with serotonin. To investigate this possibility, we have examined the DUM neurons in more detail with regard to their peptide content, examined the potential release of Rhopr-FGLa/ASTs into the haemolymph following feeding, and further investigated the effects of Rhopr-FGLa/ASTs on feeding-related tissues. There are 10 DUM neurons in the abdominal neuromeres, 5 of which express serotonin-like immunoreactivity and the other 5 express FGLa/AST-like immunoreactivity. FGLa/AST-like immunoreactivity is reduced in the 5 DUM neuron cell bodies and their neurohaemal sites on abdominal nerves at 3-5 hours post feeding. Rhopr-FGLa/ASTs do not inhibit serotonin-stimulated anterior midgut absorption or Malpighian tubule secretion but do inhibit hindgut contractions induced by an endogenous kinin, suggesting that they may only indirectly affect post-feeding diuresis in R. prolixus.

194

Introduction

The ability of certain neuropeptides (referred to as allatostatins, ASTs) to inhibit the biosynthesis of juvenile hormone (JH) by the corpora allata (CA) was first shown in Diploptera punctata

(Pratt et al., 1991; Woodhead et al., 1989). Since that time a number of AST neuropeptides have been isolated that arise from three peptide families. These families are encoded by three separate genes and processed into neuropeptides with distinct sequences and pleiotropic activities across arthropods. The families are referred to as the cockroach type (FGLa/ASTs), cricket type

(MIP/ASTs), and moth type (PISCF/ASTs), each based upon their C-terminal characteristic sequences (see Bendena et al., 1999; Tobe and Bendena, 2006). The FGLa/ASTs are present in numerous insect orders where they have physiological roles distinct from controlling JH biosynthesis, such as inhibition of visceral muscle contraction (see Tobe and Bendena, 2006).

These latter authors suggest that the FGLa/AST role as inhibitors of JH biosynthesis is a secondarily-evolved function, and that inhibition of visceral muscle contraction might be their ancestral function (see Tobe and Bendena, 2006). It is also interesting to observe that the cricket type MIP/ASTs were first described based upon their ability to inhibit visceral muscle contraction. FGLa/AST-like immunoreactivity is found in neurons within the central nervous system (CNS), but is also associated with processes over the foregut, midgut, and hindgut, as well as open-type midgut endocrine cells (see Robertson and Lange, 2010; Sarkar et al., 2003), which suggest that ASTs might play an important role in feeding-related physiological events.

Myoinhibitory activity of FGLa/ASTs on the gut has been found in a variety of species. For example, FGLa/ASTs inhibit spontaneous contractions of the foregut in Leucophaea maderae

(Duve et al., 1995), proctolin-induced muscle contractions of the D. punctata midgut (Fuse et al.,

1999) and spontaneous and proctolin-induced hindgut muscle contractions of D. punctata (Lange et al., 1993).

195

Recently, the FGLa/AST gene transcript as well as FGLa/AST peptides were identified in the medically-important bug, Rhodnius prolixus (Ons et al., 2009; Ons et al., 2011; Zandawala et al.,

2012). This is of some importance because of the possible involvement of FGL/ASTs in insect feeding activities and the fact that this blood-gorging bug transmits the parasite Trypanosoma cruzi to humans in its urine, causing Chagas disease. Zandawala et al. (2012) used fluorescent in situ hybridization (FISH) to localize cell-specific expression of the Rhopr-FGLa/AST transcript within the CNS, and in particular identified 5 dorsal unpaired median (DUM) neurons in abdominal neuromeres of the mesothoracic ganglionic mass (MTGM). These same cells are positive for FGLa/AST-like immunoreactivity (Sarkar et al., 2003). This is of some interest since a similar set of 5 DUM neurons in the abdominal neuromeres of the MTGM contain serotonin

(5-hydroxytryptamine, 5-HT) which is a diuretic hormone in R. prolixus (Maddrell et al., 1991;

Orchard et al., 1989), colocalized with a neuropeptide – a calcitonin-like diuretic hormone

(CT/DH) (Te Brugge et al., 2005). In light of their similar location, we were interested in further examining the DUM neurons of the MTGM, and examining their properties and possible interactions in post-feeding diuresis and other feeding-related physiological events.

Material and methods

Animals

Fifth-instar R. prolixus were obtained from a long standing colony at the University of Toronto

Mississauga. Insects were reared in incubators in high relative humidity at 25°C and routinely fed on defibrinated rabbit blood (Cedarlane Laboratories Inc., Burlington, ON, Canada).

Immunohistochemistry and time-course analysis

196

The insects were fed on defibrinated rabbit blood or saline containing 1mM ATP and 150mM

NaCl at pH 7 (Friend and Smith, 1982). CNS, anterior midgut, posterior midgut and hindgut were used for the immunohistochemical analysis from the following insects: 3-4 week unfed fifth-instars, 1 hour post-feeding on blood or saline, and 3-5 hours post-feeding on blood or saline. The fixation and staining of the tissues for FGLa/AST-like immunoreactivity was done as described previously (Sarkar et al., 2003) with some minor modifications. Tissues were incubated for approximately 60 hours at 4°C in a 1:1000 polyclonal rabbit antiserum generated against Dippu-AST-7 (formerly 1) (Vitzthum et al., 1996) diluted in 0.4% Triton X-100 with 2%

BSA and 10% NGS. The preparations were then washed with phosphate-buffered saline (PBS) for 4-6 hours at room temperature. Next, the tissues were incubated overnight at 4°C in affinity purified goat anti-rabbit antibody conjugated to Cy3 at a dilution of 1:600 in PBS containing 2%

NGS. The tissues were then washed in PBS for 4-18 hours and cleared with a glycerol series before mounting in 100% glycerol. Images were taken using a confocal microscope equipped with ZEN 2009 LE software (Zeiss, Jena, Germany). Staining intensity of the DUM neurons and the background was determined using the ImageJ Software. Ratios of cell intensity to background intensity were calculated for all 5 DUM neurons and their means plotted as a bar graph. An unpaired t-test was used for the statistical analyses (P<0.05). Double-label immunohistochemistry was performed on CNS as described earlier (Sarkar et al., 2003), using anti-dippu-AST-7 antiserum and anti-serotonin antiserum (DiaSorin, Stillwater, MN, USA).

Neurophysiology

Intracellular recordings and Lucifer yellow injection of DUM neurons was performed as previously described, as was immunohistochemistry following such treatments (Orchard et al.,

1989). Briefly, for intracellular recordings, the dorsal thoracic and abdominal cuticle as well as

197 the gut were removed to expose the MTGM. The MTGM was stabilized with a metal spoon and kept moist with physiological saline. Electrodes were filled with 5% (w/v) Lucifer yellow CH in distilled water. The electrode resistances varied between 60-200 MΩ. The electrodes were connected to a World Precision Instruments S-7071A electrometer (Sarasota, Florida, USA) and oscilloscope, which allowed for recording or current injection via a bridge circuit. Lucifer yellow was injected into the cells by passing hyperpolarising direct current (4-7 nA) for up to 20 minutes. The ganglia were then fixed in 2% paraformaldehyde and processed for immunohistochemistry.

Peptides

Rhopr-FGLa/AST-2 (LPVYNFGLamide), Rhopr-FGLa/AST-3 (AHNEGRLYSFGLamide) and

Rhopr-kinin-2 (AKFSSWGamide) were custom synthesised by GenScript (Piscataway, NJ,

USA) at > 95% purity.

Anterior midgut absorption assay

Fluid transport across the anterior midgut was measured using the absorption assay described earlier (Te Brugge et al., 2009). Briefly, the anterior midgut was exposed by removing the dorsal cuticle and diaphragm. The anterior end of the anterior midgut (foregut and anterior midgut juncture) was ligated using a silk thread. Next, the anterior midgut was nicked at the juncture with posterior midgut and 30-50 µl physiological saline (150 mM NaCl, 8.6 mM KCl, 2 mM

CaCl2, 4 mM NaHCO3, 34 mM glucose, 8.5 mM MgCl2, 5 mM HEPES [pH 7.2]) containing

0.01% methylene blue was injected into the anterior midgut before ligating its posterior end.

Methylene blue was used to check for leakage. The anterior midgut was gently blotted and weighed on a Mettler AE 240 balance. Once weighed, the tissues were placed in a micro-

198 centrifuge tube with 1 mL saline, saline containing only serotonin (10-7 M) or both serotonin (10-

7 M) and peptide solution (10-6 and 10-7 M). Following 30 min incubation, the tissues were blotted and weighed again. The difference in weight (initial – final) was used to calculate the absorption rate (nL/min), assuming a specific gravity of 1. The results are expressed as mean ±

SE. One-way ANOVA and Tukey post-test were used for the statistical analyses (P<0.05).

Upper Malpighian tubule (MT) secretion assay

The upper MT fluid secretion assay was performed using the methods described earlier, with few modifications (Donini et al., 2008; Paluzzi and Orchard, 2006). Briefly, upper MTs were dissected using a fine glass probe under saline that contained (mmol l–1): 129 NaCl, 8.6 KCl, 4.0

NaHCO3, 4.3 NaH2PO4, 8.5 MgCl2, 2 CaCl2, 8.6 HEPES and 20 glucose at pH 7. Upper tubule segments were transferred to a Sylgard-lined dish containing 90 μl drops of saline covered with water-saturated paraffin oil. The open end of the tubules were pulled out of the saline and wrapped around a nearby minuten pin. Next, 10 μl of saline (control) or saline containing 10-6 M peptide (Rhopr-FGLa/AST-2 or Rhopr-FGLa/AST-3) were added to the bathing saline droplet

(final concentration of the peptide was 10-7 M). The tubules were then incubated for 15 minutes.

Following this incubation, 10 μl of saline was removed from the bathing droplet and replaced with 10 μl of saline containing 10-6 M serotonin alone (control) or combined with 10-7 M peptide

(Rhopr-FGLa/AST-2 or Rhopr-FGLa/AST-3). Droplets of secreted fluid that formed at the pin after 30 min were collected using fine glass probes. The diameter (d) of the droplet was measured using an ocular micrometer and used to calculate the volume of secreted fluid using the equation V=(π/6)d3. The volume was divided by the time over which the droplets formed to obtain the rate of secretion. One-way ANOVA and Tukey post-test were used for the statistical analyses (P<0.05).

199

Hindgut contraction assay

The contraction assay using hindgut from unfed fifth-instar R. prolixus was performed as described previously (Te Brugge et al., 2002; Zandawala et al., 2012), except that contractions were monitored and recorded using PicoLog recorder (Pico Technology, St Neots,

Cambridgeshire, UK). Two-tailed unpaired t test was used for the statistical analysis (P<0.05).

Briefly, hindgut assays were conducted on isolated fifth-instar hindguts maintained under physiological saline. The preparation consisted of a small piece of ventral cuticle surrounding the anus to secure the hindgut to a Sylgard (Dow Corning, Midland Michigan, USA) coated dish, while the anterior end of the hindgut and a small portion of the posterior midgut was tied by a fine thread to a miniature force transducer (AksjeselskapetMikro-elektronikk, Horten, Norway).

Longitudinal contractions were recorded on the PicoLog. Hindguts from unfed R. prolixus rarely contract spontaneously in vitro, and so Rhopr-kinin-2 (Bhatt et al., 2013) was used to stimulate contraction before testing the inhibitory effects of Rhopr-FGLa/AST-2 or -3).

Results

Serotonin-like and FGLa/AST-like immunoreactivity in DUM neurons

Double-label immunohistochemistry indicates that the 5 serotonin-like DUM neurons in the

MTGM are distinct from the 5 FGLa/AST-like DUM neurons (Figure 1A). Each of these DUM neurons produces extensive neurohaemal sites on their corresponding abdominal nerve (see

Orchard, 2009), and there is no co-localization of FGLa/AST-like and serotonin-like immunoreactivity in these neurohaemal sites (Figure 1B). There are, however, other cells in the

CNS which are immunoreactive for both serotonin and FGLa/ASTs (Figure 1A).

200

Figure 1: FGLa/AST-like (red) and serotonin-like (green) double-label immunohistochemistry.

(A) Dorsal view of MTGM. The 5 FGLa/AST-like immunoreactive DUM neurons are distinct from the 5 serotonin-like DUM neurons. The DUM neurons lie dorsal/ventral to each other and so overlap in the image. Each neuron has been circled to add clarity. Note the larger diameter of the serotonergic DUM cell bodies. Other neurons which are double-labelled (in yellow) for both serotonin-like and FGLa/AST-like immunoreactivity are indicated with an arrow head. Scale bar:

50 μm. (B) Abdominal nerves: note the absence of FGLa/AST-like and serotonin-like immunoreactivity co-localization in the neurohemal sites on the abdominal nerves. Scale bar: 10

μm.

201

202

Neurophysiology

Intracellular recordings from the FGLa/AST-like immunoreactive DUM neurons (confirmed by

Lucifer yellow injection and immunohistochemistry) revealed that they have smaller cell bodies than the serotoninergic DUM neurons (30 µm diameter versus 40 µm diameter). They are capable of supporting overshooting action potentials of between 40-60 mV amplitude, with large after hyperpolarisations (Figure 2). There are likely multiple spike initiation sites resulting in various spike types recorded from the cell body (Figure 2). Lucifer yellow did not pass between the DUM neurons and so it is unlikely that they are electrically-coupled.

FGLa/AST-like immunoreactivity pre- and post-feeding

The FGLa/AST-like staining intensity of the DUM neuron cell bodies is significantly reduced at

3-5 hours post-feeding (Figure 3A-C). The bilateral axons were more visible in neurons 1 hour after feeding. In addition, the number and intensity of neurohemal sites on the abdominal nerves decreased at 3-5 hours post-feeding (Figure 4A-C). FGLa/AST-like immunoreactive open-type endocrine cells are present in the anterior and posterior midguts (Figure 5A, C) whereas the hindgut only contains immunoreactive nerve processes (Figure 5E). The FGLa/AST-like immunoreactive endocrine cells remained brightly stained in anterior and posterior midgut from

1 hour post-fed (not shown) to 3-5 hours post-fed (Figure 5B, D). Similarly, there was no difference in the staining-intensity of immunoreactive processes on the hindgut from unfed and fed insects at the times tested (Figure 5F). Similar results were obtained from blood fed or saline fed insects.

203

Figure 2: Immunohistochemistry, Lucifer yellow injection and physiology of a DUM neuron in the abdominal neuromeres of the mesothoracic ganglionic mass of R. prolixus. (A) Lucifer yellow injection of a DUM neuron that does not exhibit serotonin-like immunoreactivity (arrow), and therefore is an FGLa/AST-like immunoreactive DUM neuron. Serotonin DUMs 1-5 are labeled 1-5. (B) Intracellular recording from the cell body of this neuron showing spontaneous overshooting action potentials and (C) Multiple-component action potentials. Scale bars: A, 50

µm; B, C, 10 mV, 20 msec

204

205

Figure 3: FGLa/AST-like immunoreactivity in the DUM neurons in the posterior MTGM of unfed (A) and 3-5 hours post-fed with saline (B) fifth-instar R. prolixus. The 5 DUM neurons are numbered in (A). The arrow in each figure shows cell-specific staining which indicates that the decrease in DUM neuron staining was not due to a procedure artifact. Scale bars: 50 μm. (C) Bar graph showing the average staining-intensity in DUM neurons relative to the background. Bars represent mean ± s.e.m. of 5-6 preparations. Note the decrease in staining-intensity in DUM neurons from 3-5 hours post-fed insects (P<0.05, unpaired t-test). Different letters denote significantly different.

206

207

Figure 4: FGLa/AST-like immunoreactivity in the abdominal nerves of unfed (A), 1 hour post- fed with saline (B) and 3-5 hours post-fed with saline (C) fifth-instar R. prolixus. Note the decrease in the number and intensity of neurohemal sites that stained after 3-5 hours post- feeding. Scale bars: 20 μm.

208

209

Figure 5: FGLa/AST-like immunoreactivity in the gut of unfed (A, C and E) and 3-5 hours post- fed with saline (B, D and F) fifth-instar R. prolixus; anterior midgut (A and B), posterior midgut

(C and D) and hindgut (E and F). PM = posterior midgut, HG = hindgut. Scale bars – (A, B and

D): 50 μm and (C, E and F): 100 μm.

210

211

Anterior midgut absorption assay

The anterior midgut absorption assay is a simple yet sensitive assay used to determine the rate of water transport across the midgut epithelium. The fluid transport rate across the anterior midgut incubated in saline is negligible but 10-7 M serotonin increased the absorption rate to 53.9 ± 4.1 nL/min (Figure 6). This rate was not significantly altered when midguts were incubated in either

10-7 M or 10-6 M Rhopr-FGLa/AST-2 or 3 (Figure 6).

MT secretion assay

We also tested the effects of Rhopr-FGLa/ASTs on serotonin-stimulated MT secretion. Tubules incubated in 10-7 M Rhopr-FGLa/AST-2 or 3 along with 10-7 M serotonin, showed no significant decrease compared to those incubated in 10-7 M serotonin alone (Figure 7).

Hindgut contraction bioassay

Rhopr-kinin-2 stimulated a sustained basal contraction of hindgut with phasic contractions superimposed (Figure 8). Rhopr-FGLa/AST-2 and 3 at 10-6 M both reduced the amplitude of the basal contraction and inhibited the frequency and amplitude of the phasic contractions stimulated using 2.5 × 10-9 M Rhopr-kinin-2 (Figure 8). Rhopr-FGLa/ASTs also increased the delay between time of peptide application and the first contraction (Figure 8).

212

Figure 6: Effects of Rhopr-FGLa/ASTs on serotonin-stimulated anterior midgut absorption in fifth-instar R. prolixus. Anterior midguts were incubated with saline, 10-7 M serotonin, or serotonin plus Rhopr-FGLa/ASTs (10-6 M and 10-7 M). Bars represent mean ± s.e.m. of 6 – 10 preparations. Different letters denote significantly different.

213

214

Figure 7: Effects of Rhopr-FGLa/ASTs on serotonin-stimulated Malpighian tubule (MT) secretion in fifth-instar R. prolixus. Secretion rates of MTs incubated in 10-7 M serotonin or serotonin plus Rhopr-FGLa/ASTs (10-7 M). Bars represent mean ± s.e.m. of 7 or more.

215

216

Figure 8: Effects of Rhopr-FGLa/ASTs on Rhopr-kinin-2 (RK2)-induced contractions of R. prolixus hindgut. (A-D) Sample traces. Control contractions were induced by 2.5 × 10-9 M RK2

(A and C) prior to the application of 10-6 M Rhopr-FGLa/AST-2 (B) or Rhopr-FGLa/AST-3 (D).

Each preparation was thoroughly washed (open arrow) between the control and experimental peptide application. Peptides were applied at the filled arrow. Note the inhibitory effects of

Rhopr-FGLa/ASTs on the frequency and amplitude of RK2-induced contractions. (E) Bar graph showing the effects of Rhopr-FGLa/AST-2 (black bars) and Rhopr-FGLa/AST-3 (grey bars) on

Rhopr-kinin-2-induced contractions of R. prolixus hindgut. The following aspects were analyzed: frequency, amplitude and delay – the time between peptide application and the first contraction.

Bars represent mean ± s.e.m. of 5 preparations. Scale bars represent tension and time.

217

218

Discussion

There are 10 DUM neurons in the abdominal neuromeres of the MTGM of R. prolixus; 5 of which express FGLa/AST-like immunoreactivity, and the other 5, larger DUMs, express serotonin-like immunoreactivity. Both sets of DUMs produce neurohaemal sites on the abdominal nerves, indicating potential release of their contents into the haemolymph.

Interestingly, even though serotonin and FGLa/ASTs are not co-localized in these DUMs, other cells in the MTGM do show co-localization. This would indicate that the two neuroactive chemicals can be released from the same neurons as co-transmitters, but not in the case of the

DUMs. The FGLa/AST-containing DUM neurons share electrophysiological properties with all other DUM neurons examined (electrically-excitable cell bodies, overshooting action potentials of long duration, pronounced after-hyperpolarization, multiple spike initiating zones), and so the electrophysiological properties are due to the morphology and membrane properties of DUM neurons and not to their amine/peptide content. The two sets of 5 DUMs are not electrically- coupled and so are not neurally-integrated as a common unit.

Serotonin is a regulator of feeding-related physiological events in R. prolixus (see Orchard,

2006) and is responsible, along with peptidergic diuretic hormones, for diuresis (see Orchard,

2006; Orchard, 2009). These 5 serotonergic DUM neurons release serotonin into the haemolymph at the time of feeding. Serotonin then stimulates absorption of salts and water across the anterior midgut into the haemolymph, secretion of salts and water by the upper MTs, and reabsorption of KCl by the lower MTs. Serotonin controls other feeding-related tissues, and stimulates contractions of salivary glands, anterior midgut, hindgut and heart (see Orchard, 2006;

Orchard, 2009). The location of the 5 FGLa/AST DUM neurons alongside the 5 serotonergic

DUM neurons in the MTGM presented the interesting possibility of neural integration of the two

219 groups and therefore the possible involvement of Rhopr-FGLa/ASTs in feeding-related activities. Initially we examined the possible release of FGLa/ASTs from the DUM neuron cell bodies and their neurohaemal sites using immunohistochemistry. Immunohistochemistry can give an indication of release from neurohaemal sites, but also indications of synthesis and restocking. Feeding on either rabbit blood or saline (containing ATP) resulted in a reduction in staining of neurohaemal sites at 3-5 hours post-feeding (towards the end of the time of diuresis), indicative of the release of FGLa/AST into the haemolymph in response to feeding. Interestingly the cell bodies were less intensely stained with axons becoming more visible, indicative of transport of FGLa/AST towards the release sites. The similar results with rabbit blood and saline containing ATP indicate that the stimulus for these post-feeding events does not lie in the nutrient component of the meal.

Previously, we have examined the serotonergic DUM neurons and shown that their neurohaemal sites are reduced in intensity 15 mins after the start of a blood meal – a timing that matches the peak of serotonin titer in the haemolymph (Lange et al., 1989). Serotonin initiates anterior midgut absorption and MT secretion, which is maintained at a high rate by the presence of a peptidergic DH, corticotropin releasing-factor-related diuretic hormone (Rhopr-CRF/DH), acting with serotonin to continue post-prandial diuresis for about 3 hours. The timing of apparent release of the FGLa/ASTs from abdominal nerve neurohaemal sites at 3-5 hours post-feeding implies a role for FGLa/AST distinct from diuresis itself. In addition, FGLa/AST-like immunoreactive endocrine cells remained brightly stained in anterior and posterior midgut at 3-5 hours post-feeding, suggesting that release from these cells must occur later in the feeding cycle.

Digestion takes place in the posterior midgut in R. prolixus some days after feeding, and so possibly FGLa/ASTs from these cells might be released at this later stage. Similarly, there was

220 no difference in the staining-intensity of immunoreactive processes on the hindgut from unfed and fed insects at the times tested.

In light of the apparent release of FGLa/ASTs from abdominal nerve neurohaemal sites associated with feeding, we examined for physiological effects of Rhopr-FGLa/ASTs on feeding-related tissues. Initially we questioned whether Rhopr-FGLa/ASTs might be antidiuretic hormones, since release appears to occur at a time similar to the antidiuretic hormone, Rhopr-

CAPA-2, which is released into the haemolymph at a time which signals the end of diuresis in

R. prolixus (Paluzzi and Orchard, 2006); however Rhopr-FGLa/ASTs did not mimic Rhopr-

CAPA-2 by inhibiting serotonin-stimulated absorption from anterior midgut, or serotonin- stimulated secretion from MTs. On the other hand, Rhopr-FGLa/ASTs were potent inhibitors of hindgut contractions induced by Rhopr-kinin-2. Hindgut contractions result in expulsion of urine during diuresis, and we have previously suggested that Rhopr-kinins are involved in the control of these contractions. Kinins are released from abdominal nerve neurohaemal sites following feeding and are also in the nerve supply to the hindgut (see Orchard, 2009). We did not observe changes in the staining intensity of the FGLa/AST-like immunoreactive process on the hindgut following feeding. This observation might imply that any inhibition of hindgut contractions 3-5 hours after feeding may be a result of FGLa/ASTs acting as neurohormones, released from the 5

DUM neurons. The FGLa/AST-like immunoreactive processes on the hindgut might therefore result in inhibition at other stages in the physiology of the insect. Anterior midgut contractions and heart-beat frequency are also inhibited by Rhopr-FGLa/ASTs (Zandawala et al., 2012).

Thus, although Rhopr-FGLa/ASTs appear to play no direct role in diuresis, they can play an indirect role by inhibiting contractions of muscles that are actively involved in feeding-related events; dorsal vessel contractions are vital for circulation of the haemolymph and the diuretic

221 hormones released following feeding, the anterior midgut contracts to remove unstirred layers, mix the blood meal, and improve circulation of haemolymph, and hindgut contraction is used to expel the urine (see Orchard, 2006; Orchard, 2009). Serotonin (released from 5 DUM neurons in the MTGM) is stimulatory on each of the above tissues, whereas Rhopr-FGLa/ASTs (released from a different set of 5 DUM neurons of the MTGM) are inhibitory. Rhopr-CAPA-2 is a potent inhibitor of serotonin-stimulated anterior midgut absorption and MT secretion (Ianowski et al.,

2010; Paluzzi et al., 2008); however, unlike Rhopr-FGLa/ASTs, Rhopr-CAPA-2 does not inhibit contraction of muscles associated with feeding-related events. Hence, Rhopr-FGLa/ASTs and

Rhopr-CAPA-2 seem to complement each other during the cessation of diuresis in R. prolixus, when most of the excess water and salts from the blood meal have been excreted, the release of diuretic hormones into the haemolymph has stopped, and urine production needs to be terminated in order to conserve water and essential ions.

The act of blood feeding in R. prolixus initiates short-term physiological changes associated with digestion, and salt and water balance, but is also the stimulus for growth and development (see

Orchard, 2009). Rhopr-FGLa/ASTs are biologically active inhibitors of tissues associated with feeding. These include anterior midgut, hindgut, and dorsal vessel. It is likely that there are other physiological states that are associated with inhibition of gut activity, including ecdysis or stress associated with starvation. Rhopr-FGLa/ASTs might be important messengers at these times, acting to fine-tune the contractile activity of visceral and cardiac muscle.

222

References

Bendena, W.G., Donly, B.C. and Tobe, S.S. (1999). Allatostatins: a growing family of neuropeptides with structural and functional diversity. Ann N Y Acad Sci. 897, 311-29.

Bhatt, G., da Silva, R., Nachman, R.J. and Orchard, I. (2013). The molecular characterization of the kinin transcript and the physiological effects of kinins in the blood-gorging insect, Rhodnius prolixus. Peptides.

Donini, A., O'Donnell, M.J. and Orchard, I. (2008). Differential actions of diuretic factors on the Malpighian tubules of Rhodnius prolixus. J Exp Biol. 211, 42-8.

Duve, H., Wren, P. and Thorpe, A. (1995). Innervation of the foregut of the cockroach Leucophaea maderae and inhibition of spontaneous contractile activity by callatostatin neuropeptides. Physiol Entomol. 20, 33-44.

Friend, W.G. and Smith, J.J.B. (1982). ATP analogues and other phosphate compounds as gorging stimulants for Rhodnius prolixus. J Insect Physiol. 28, 371-376.

Ianowski, J.P., Paluzzi, J.P., Te Brugge, V.A. and Orchard, I. (2010). The antidiuretic neurohormone RhoprCAPA-2 downregulates fluid transport across the anterior midgut in the blood-feeding insect Rhodnius prolixus. Am J Physiol Regul Integr Comp Physiol. 298, R548-57.

Lange, A.B., Orchard, I. and Michael Barrett, F. (1989). Changes in haemolymph serotonin levels associated with feeding in the blood-sucking bug, Rhodnius prolixus. J Insect Physiol. 35, 393-399.

223

Maddrell, S.H., Herman, W.S., Mooney, R.L. and Overton, J.A. (1991). 5- Hydroxytryptamine: a second diuretic hormone in Rhodnius prolixus. J Exp Biol. 156, 557-66.

Ons, S., Richter, F., Urlaub, H. and Pomar, R.R. (2009). The neuropeptidome of Rhodnius prolixus brain. Proteomics. 9, 788-92.

Ons, S., Sterkel, M., Diambra, L., Urlaub, H. and Rivera-Pomar, R. (2011). Neuropeptide precursor gene discovery in the Chagas disease vector Rhodnius prolixus. Insect Mol Biol. 20, 29-44.

Orchard, I. (2006). Serotonin: A coordinator of feeding-related physiological events in the blood-gorging bug, Rhodnius prolixus. Comp Biochem Physiol Part A Mol Integr Physiol. 144, 316-324.

Orchard, I. (2009). Peptides and serotonin control feeding-related events in Rhodnius prolixus. Front Biosci. 1, 250-62.

Paluzzi, J.P. and Orchard, I. (2006). Distribution, activity and evidence for the release of an anti-diuretic peptide in the kissing bug Rhodnius prolixus. J Exp Biol. 209, 907-915.

Robertson, L. and Lange, A.B. (2010). Neural substrate and allatostatin-like innervation of the gut of Locusta migratoria. J Insect Physiol. 56, 893-901.

224

Sarkar, N.R., Tobe, S.S. and Orchard, I. (2003). The distribution and effects of Dippu- allatostatin-like peptides in the blood-feeding bug, Rhodnius prolixus. Peptides. 24, 1553-62.

Te Brugge, V.A., Ianowski, J.P. and Orchard, I. (2009). Biological activity of diuretic factors on the anterior midgut of the blood-feeding bug, Rhodnius prolixus. Gen Comp Endocrinol. 162, 105-112.

Te Brugge, V.A., Lombardi, V.C., Schooley, D.A. and Orchard, I. (2005). Presence and activity of a Dippu-DH31-like peptide in the blood-feeding bug, Rhodnius prolixus. Peptides. 26, 29-42.

Te Brugge, V.A., Schooley, D.A. and Orchard, I. (2002). The biological activity of diuretic factors in Rhodnius prolixus. Peptides. 23, 671-681.

Tobe, S.S. and Bendena, W.G., 2006. Chapter 31 - Allatostatins in the Insects, in: Abba, J.K. (Ed.), Handbook of Biologically Active Peptides. Academic Press, Burlington, pp. 201-206.

Vitzthum, H., Homberg, U. and Agricola, H. (1996). Distribution of Dip-allatostatin I-like immunoreactivity in the brain of the locust Schistocerca gregaria with detailed analysis of immunostaining in the central complex. J Comp Neurol. 369, 419-37.

225

Acknowledgements

The authors wish to thank Nikki Sarkar for maintaining the colony and providing assistance with the feeding experiments, Dr. Jean-Paul Paluzzi for his guidance on the anterior midgut absorption assay and Dr. Andrew Donini for his guidance on the MT secretion assay. We kindly acknowledge the use of the Dippu-AST-7 antiserum from Professor Agricola.

226

Chapter 7: Isolation and functional characterization of

FGLamide-related allatostatin receptor in Rhodnius prolixus

The proceeding chapter will be submitted for publication in Molecular and Cellular

Endocrinology.

227

Abstract

FGLamide-related ASTs (FGLa/ASTs) are a family of brain/gut peptides with numerous physiological roles, including inhibition of juvenile hormone (JH) biosynthesis by the corpora allata and inhibition of visceral muscle contraction. FGLa/ASTs mediate their effects by binding to a rhodopsin-like G-protein coupled receptor that is evolutionarily related to the vertebrate galanin receptor. Here we determine the cDNA sequence encoding FGLa/AST receptor

(FGLa/AST-R) from the Chagas disease vector, Rhodnius prolixus (Rhopr-FGLa/AST-R), and determine its spatial expression pattern using quantitative PCR. Our analysis indicates that

Rhopr-FGLa/AST-R is highly expressed in the central nervous system. The receptor is also expressed in various peripheral tissues including the dorsal vessel, midgut, hindgut and reproductive tissues of both males and females, suggesting a role in processes associated with feeding and reproduction. The possible involvement of Rhopr-FGLa/ASTs in the inhibition of

JH biosynthesis is also implicated due to presence of the receptor transcript in the R. prolixus corpora cardiaca/corpora allata complex.

228

Introduction

Allatostatins (ASTs) are insect neuropeptides that were first identified for their ability to inhibit juvenile hormone (JH) biosynthesis by the corpora allata (CA) (see Bendena and Tobe, 2012).

Three families of ASTs exist and these are characterized by their conserved C-terminus sequences. These include the FGLamide-related ASTs (FGLa/ASTs), myoinhibitory peptide

(MIP)/ASTs and PISCF-related ASTs (PISCF/ASTs); these are also commonly referred to as

AST-A, AST-B and AST-C, respectively (Coast and Schooley, 2011). Although most insects have been shown to contain all three AST families, only one type of AST possesses allatostatic activity, if any, in any given insect. Hence, only the FGLa/ASTs have been shown to possess allatostatic activity in cockroaches, crickets, termites and locusts (see Bendena and Tobe, 2012).

It is now evident that FGLa/ASTs have numerous other physiological roles and that their role as inhibitors of JH biosynthesis is only a secondary evolved function. One such major role is the inhibition of visceral muscle contraction. Consequently, myoinhibitory activity of FGLa/ASTs has been shown in various insects including Diploptera punctata (Fuse et al., 1999; Lange et al.,

1993), Locusta migratoria (Robertson et al., 2012), Drosophila melanogaster (Vanderveken and

O'Donnell, 2014) and Rhodnius prolixus (Sarkar et al., 2003; Zandawala et al., 2012) amongst others. Moreover, FGLa/ASTs influence processes associated with feeding (Aguilar et al., 2003;

Hergarden et al., 2012; Wang et al., 2012; Zandawala and Orchard, 2013) and reproduction

(Veelaert et al., 1996; Woodhead et al., 2003).

Regardless of which biological process they influence, FGLa/ASTs mediate their effects by binding to a G-protein coupled receptor (GPCR). FGLa/AST receptors (FGLa/AST-Rs) belong to the family of rhodopsin-like GPCRs. The first FGLa/AST-R was isolated and functionally characterized from D. melanogaster and was originally referred to as DAR-1 (Birgul et al.,

229

1999). Soon after that, another FGLa/AST-R (DAR-2) was also discovered in D. melanogaster

(Larsen et al., 2001; Lenz et al., 2000). Both these receptors were shown to be evolutionarily related to the vertebrate galanin receptors. Since then only a single FGLa/AST-R has been isolated and functionally characterized from each of Periplaneta americana (Auerswald et al.,

2001), D. punctata (Lungchukiet et al., 2008), Bombyx mori (Secher et al., 2001) and

Caenorhabditis elegans (Bendena et al., 2008). Knockdown of C. elegans FGLa/AST-R and D. melanogaster DAR-1 via RNAi has been shown to affect foraging behavior in these species

(Bendena et al., 2008; Wang et al., 2012).

In the present study, we have isolated and characterized a cDNA sequence encoding FGLa/AST-

R from R. prolixus (Rhopr-FGLa/AST-R). Phylogenetic analysis indicates that this receptor is more closely related to the previously characterized receptor from P. americana than to either of the two receptors from D. melanogaster. Most insects, with the exception of dipterans, appear to possess only one FGLa/AST-R. Quantitative PCR (qPCR) was also used to determine the spatial expression pattern of Rhopr-FGLa/AST-R and consequently unravel novel target tissues for this peptide.

Material and methods

Animals

Fifth-instar and adult R. prolixus of both sexes were taken from a long standing colony at the

University of Toronto Mississauga. Insects were raised in incubators maintained at 60% humidity and 25°C, and were artificially fed on defibrinated rabbit blood (Hemostat

Laboratories, Dixon, CA, USA; supplied by Cedarlane Laboratories Inc., Burlington, ON,

Canada).

230

Isolation of a cDNA sequence encoding R. prolixus FGLa/AST receptor

Supercontig sequences representing the R. prolixus genome assembly (January 2012 release) were downloaded from VectorBase (https://www.vectorbase.org/). The sequences were imported to Geneious 4.7.6 and used to perform local BLAST searches. Drosophila melanogaster

FGLa/AST-R-1 (DAR-1) amino acid sequence was used as a query to search the R. prolixus genome. A forward gene-specific primer (Rpr-ASTR-FOR1: 5’

AGAACTTACTGAAAAAATAGTGGCG 3’) and a reverse gene-specific primer (Rpr-ASTR-

REV1: 5’ GTTCTGGTGTTTTTGGTTAAGGC 3’) were designed based on the resultant hits and used to amplify the partial cDNA sequence encoding Rhopr-FGLa/AST-R. A fifth-instar R. prolixus CNS cDNA library (Paluzzi et al., 2008) was used as the template for this PCR. The resultant product was purified, cloned and sequenced, as described earlier (Zandawala et al.,

2013). Rhodnius prolixus genome assembly was used to predict the splice sites upstream and downstream of the regions encoding this partial Rhopr-FGLa/AST-R so that additional sequence for the first and last exons could be obtained (see below for splice site prediction). Primers (Rpr-

ASTR-FOR8: 5’ ATCGAGATGAACGGATCAC 3’ and Rpr-ASTR-REV4: 5’

TAACACGAACCTAGGCAGTACAG 3’) were designed with these predicted splice sites in mind so as to avoid them spanning an intron. These were then used to amplify the complete cDNA sequence encoding Rhopr-FGLa/AST-R with a proof-reading Taq polymerase using the methods described earlier (Zandawala et al., 2013; Zandawala et al., 2011).

Sequence and phylogenetic analysis

The intron-exon boundaries of Rhopr-FGLa/AST-R were predicted using a combination of

BLAST and Genie, a splice site prediction software (Reese et al., 1997). The following software

231

programs were used for predicting various features of Rhopr-FGLa/AST-R: Geneious 4.7.6 for the membrane topology of the receptor, NetPhos 2.0 Server (Blom et al., 1999) for the potential phosphorylation sites and NetNGlyc 1.0 Server (http://www.cbs.dtu.dk/services/NetNGlyc/ - last accessed on June 8, 2014) for the potential N-linked glycosylation sites.

Clustal Omega (http://www.ebi.ac.uk/Tools/msa/clustalo/ - last accessed on June 8, 2014) was used to align Rhopr-FGLa/AST-R with its homologs from D. melanogaster (NP_524700.1 and

NP_524544.1), P. americana (AAK52473.1) and B. mori (AAG44631.1). The alignment figure was obtained using the BOXSHADE 3.21 server

(http://www.ch.embnet.org/software/BOX_form.html - last accessed on June 8, 2014).

Additional sequences used for the phylogenetic analysis were obtained by performing a protein

BLAST search using Rhopr-FGLa/AST-R sequence as the query. Additionally, galanin receptors from Homo sapiens (NP_001471.2, NP_003848.1 and NP_003605.1) were also included in the analysis and D. melanogaster AST-C receptor (NP_649039.4) was utilized as an outgroup.

ClustalX2 was used to align these sequences and the alignment exported to MEGA5 (Tamura et al., 2011). A maximum likelihood tree was constructed using Close-Neighbor-Interchange (CNI) analysis and the bootstrap values obtained were based on 1000 replicates.

Quantitative PCR tissue profiling

Quantitative PCR (qPCR) was used to determine the spatial expression profile of Rhopr-

FGLa/AST-R using the method described earlier (Zandawala et al., 2013). Briefly, tissues were individually-dissected in phosphate-buffered saline (PBS) and stored in RNAlater® Stabilization

Solution (Life Technologies Corporation, Carlsbad, CA, USA) until RNA extraction. Total RNA was extracted from these tissues using PureLink® RNA Mini Kit (Life Technologies

232

Corporation, Carlsbad, CA, USA) which was then used to synthesize cDNA with iScript™

Reverse Transcription Supermix for RT-qPCR (Bio-Rad Laboratories Ltd., Mississauga, ON,

Canada). The resulting cDNA was diluted 10-fold and used as template for qPCR.

A 237bp Rhopr-FGLa/AST-R fragment was amplified using the forward primer, ASTR-qPCR-

FOR (5’ AGTGGCTTTTCAGGTTTCATTC 3’) and the reverse primer, ASTR-qPCR-REV (5’

CCGATACTTTTCATTACCAGAATCAC 3’). Delta-delta Ct method was used to determine the relative expression of Rhopr-FGLa/AST-R, which was normalized via the geometric averaging of the transcript levels of alpha-tubulin, beta-actin and ribosomal protein 49 (Zandawala et al.,

2013).

Results

Rhopr-FGLa/AST-R

We have isolated a 1550bp cDNA sequence (Rhopr-FGLa/AST-R) encoding Rhopr-FGLa/AST-

R (Figure 1). The open reading frame (ORF) is 1215bp long which encodes a receptor comprised of 404 amino acids. The ORF spans three exons which are separated by two large introns (Figure

2). Exons 1 and 3 are at least 753bp and 599bp, respectively, while exon 2 is 198bp long. A polyadenylation signal sequence is absent in the 3’ UTR suggesting that it is incomplete. The 5’

UTR is also partial and despite the absence of an in-frame stop codon upstream of the start codon, we have sufficient evidence to believe that the ORF is complete (see below under discussion). As is typical of all functional GPCRs, Rhopr-FGLa/AST-R is predicted to have an extracellular N-terminus, an intracellular C-terminus and seven transmembrane domains. The receptor is also predicted to undergo phosphorylation at five residues in its intracellular loops and N-linked glycosylation at four sites in its N-terminus and extracellular loop.

233

Figure 1: Rhopr-FGLa/AST-R cDNA sequence and the deduced amino acid sequence. The numbering for each sequence is shown at right. Within the nucleotide sequence, the exon-exon boundaries are shaded in gray. Within the amino acid sequence, the initial methionine start codon has been capitalized, the potential phosphorylation sites are shaded in black, the potential N- linked glycosylation sites are boxed and the seven putative transmembrane domains are underlined.

234

235

Figure 2: Rhopr-FGLa/AST-R gene structure based on BLAST analysis and splice site prediction. The gene comprises three exons (represented by boxes) that are separated by two introns.

236

237

Sequence and phylogenetic analysis

Rhopr-FGLa/AST-R amino acid sequence was aligned along with its orthologs from D. melanogaster, B. mori and P. americana that have been previously cloned (Figure 3). The region between the seven transmembrane domains is highly conserved across all sequences, with very little conservation observed in the N- and C-termini. The five phosphorylation sites predicted in

Rhopr-FGLa/AST-R are also conserved across all the other receptors which suggests that they may very well be functional. The predicted N-linked glycosylation sites, on the other hand, are less conserved across the receptors examined here.

A comprehensive phylogenetic analysis of FGLa/AST-Rs and galanin receptors confirms the evolutionary relatedness between invertebrate FGLa/AST-Rs and vertebrate galanin receptors

(Figure 4); the clade comprising the three human galanin receptor subtypes is sister to the clade comprising all the invertebrate FGLa/AST-Rs. The putative Rhopr-FGLa/AST-R is sister to a clade which includes receptors from Nilaparvata lugens and P. americana. Since the receptor from P. americana has been functionally characterized as a receptor for FGLa/ASTs, it indicates that the putative Rhopr-FGLa/AST-R would most-likely be activated by Rhopr-FGLa/ASTs. It is also interesting to note that most insects, with the exception of dipterans, possess one

FGLa/AST-R. Most dipteran species, including Anopheles gambiae, D. melanogaster and several other Drosophila species, have two FGLa/AST-Rs that form two distinct monophyletic clades. This suggests that these two receptors arose from a recent duplication in dipterans.

Lastly, the four FGLa/AST-Rs found in Ixodes scapularis represents an independent duplication event in this species.

238

Figure 3: Multiple sequence alignment of select insect FGLa/AST receptors. Identical and similar amino acids across 80% of the sequences have been highlighted in black and gray, respectively. The predicted locations of the seven transmembrane domains of Rhopr-FGLa/AST-

R have been indicated using green lines. The five predicted phosphorylation sites that are conserved across all sequences are highlighted in red and the four predicted N-linked glycosylation sites are indicated using an asterisk.

239

240

Figure 4: A phylogram of FGLa/AST and galanin GPCRs obtained following a maximum likelihood analysis (1000 bootstrap replicates). The taxa are labelled using GenBank accession numbers and the species names. The number following the species name indicates the receptor subtype (only those that are experimentally verified). Drosophila melanogaster allatostatin C receptor (ASTC-R) was utilized as outgroup.

241

242

Spatial expression profile of Rhopr-FGLa/AST-R

To identify the possible target tissues of Rhopr-FGLa/ASTs, spatial expression profile of Rhopr-

FGLa/AST-R was determined via qPCR. Within the fifth-instar, Rhopr-FGLa/AST-R was highly expressed in the CNS and expressed at moderate levels in the dorsal vessel and posterior midgut

(Figure 5). Low levels of the transcript were also detected in the foregut, salivary glands, anterior midgut and hindgut (Figure 5). Looking at the expression in adult reproductive tissues, Rhopr-

FGLa/AST-R was highly expressed in female reproductive tissues, comprised of bursa, oviducts, spermatheca and cement gland (Figure 6). The receptor transcript was moderately expressed in the male reproductive tissues, comprised of vas deferens, seminal vesicle, accessory glands and ejaculatory duct, while low levels of the transcript were detected in both testes and ovaries

(Figure 6).

In order to determine the possible involvement of Rhopr-FGLa/ASTs in JH biosynthesis in R. prolixus, we examined the presence of the receptor transcript in the CC/CA complex using qPCR. Rhopr-FGLa/AST-R is expressed in the CC/CA complex of both fifth-instars and adults although at lower levels than in fifth instar and adult CNS (Figure 7).

243

Figure 5: Spatial expression analysis of Rhopr-FGLa/AST-R in fifth instar R. prolixus determined using quantitative PCR. Expression was analyzed in the following tissues: CNS

(central nervous system), DV (dorsal vessel), Pool (fat bodies, abdominal nerves, diaphragms and trachea), FG (foregut), SG (salivary glands), AMG (anterior midgut), PM (posterior midgut),

MTs (Malpighian tubules), HG (hindgut), TST (testes), OV (ovaries) and PG (prothoracic glands). Expression is shown relative to transcript levels in PM cDNA.

244

245

Figure 6: Spatial expression analysis of Rhopr-FGLa/AST-R in R. prolixus adult reproductive tissues determined using quantitative PCR. Expression was analyzed in the following tissues:

TST (testes), M.R. (rest of the male reproductive tissues), OV (ovaries) and F.R. (rest of the female reproductive tissues). Expression is shown relative to transcript levels in TST cDNA.

246

247

Figure 7: Spatial expression analysis of Rhopr-FGLa/AST-R in R. prolixus corpora cardiaca/corpora allata (CC/CA) complex determined using quantitative PCR. Expression was analyzed in the CNS (central nervous system) and CC/CA from fifth-instar and adult R. prolixus.

Expression is shown relative to transcript levels in fifth-instar CC/CA cDNA.

248

249

Discussion

We have isolated and characterized a cDNA sequence encoding FGLa/AST-R from the Chagas disease vector, R. prolixus (Rhopr-FGLa/AST-R). Even though an in-frame stop codon upstream of the start codon is absent in our sequence, we believe that the ORF is complete. This is based on the fact that the receptor topology prediction yields seven transmembrane domains, an extracellular N-terminus and an intracellular C-terminus. Moreover, the multiple sequence alignment of FGLa/AST-Rs shows that the predicted N-linked glycosylation sites in the N- terminus of Rhopr-FGLa/AST-R are also conserved across other receptors and Rhopr-

FGLa/AST-R doesn’t appear to be any shorter compared to other sequences. However, experiments to functionally characterize this putative Rhopr-FGLa/AST-R are needed, and thus in progress, which will confirm that the receptor is activated by Rhopr-FGLa/ASTs and is thus complete and functional. Phylogenetic analysis indicates that the putative Rhopr-FGLa/AST-R is more closely related to the previously characterized receptor from P. americana than to either of the two previously characterized receptors from D. melanogaster. Moreover, the analysis also shows that most dipteran species examined here have two FGLa/AST-Rs that form two distinct monophyletic clades, suggesting that these two receptor subtypes may have originated from a recent duplication in dipterans. These two receptor types are functionally different as evident from their responses in Chinese Hamster Ovary (CHO) cells (Larsen et al., 2001). The two receptors, DAR-1 and DAR-2, when activated by FGLa/ASTs, lead to multiple signaling pathways, including the Gi/o alpha subunit mediated pathway that involves the inhibition of adenylate cyclase. However, these receptors show different preferences for coupling to specific

G alpha subunits. Functional characterization of Rhopr-FGLa/AST-R in mammalian cells, as done previously (Larsen et al., 2001; Zandawala et al., 2013), will enable us to see which of the

250

two D. melanogaster receptors is more functionally similar to Rhopr-FGLa/AST-R and thus provide a better insight of their evolution.

Spatial expression analysis of Rhopr-FGLa/AST-R via qPCR identified several target tissues of

FGLa/ASTs in R. prolixus. The receptor is expressed in the dorsal vessel, anterior midgut and hindgut, all of which whose contractility is inhibited by Rhopr-FGLa/ASTs (Zandawala et al.,

2012; Zandawala and Orchard, 2013). The receptor is also expressed in the foregut, salivary glands and posterior midgut. FGLa/ASTs have been shown to inhibit spontaneous contractions of the foregut in Leucophaea maderae (Duve et al., 1995), proctolin-induced foregut contractions in L. migratoria (Robertson et al., 2012), proctolin-induced midgut contractions in

D. punctata (Fuse et al., 1999) and gut contractions in D. melanogaster (Vanderveken and

O'Donnell, 2014). Hence Rhopr-FGLa/ASTs may also inhibit contractions of the foregut, posterior midgut and salivary glands, all of which are surrounded by muscles in R. prolixus.

Moreover, FGLa/ASTs inhibit K+ absorption across D. melanogaster midgut (Vanderveken and

O'Donnell, 2014), K+ absorption across L. migratoria hindgut (Robertson, personal communication) and may also affect ion transport in Aedes aegypti midgut (Onken et al., 2004).

Therefore, Rhopr-FGLa/ASTs may also influence digestion of the blood meal and/or the subsequent absorption of K+ across the R. prolixus posterior midgut. Not surprisingly, the receptor is not expressed in Malpighian tubules (MTs), which is in agreement with the previous findings where Rhopr-FGLa/ASTs failed to inhibit serotonin-stimulated MT secretion

(Zandawala and Orchard, 2013). This expression pattern of their receptor further supports the theory that Rhopr-FGLa/ASTs play a role in feeding-related physiological events.

251

Rhopr-FGLa/AST-R is also expressed in male reproductive tissues without the testes and female reproductive tissues without the ovaries. The male reproductive tissue sample comprises the vas deferens, seminal vesicle, accessory glands and ejaculatory duct while the female reproductive tissue sample comprises the bursa, oviducts, spermatheca and cement gland. This suggests that

Rhopr-FGLa/ASTs could influence reproductive physiology. Indeed, they have been shown to inhibit spontaneous contractions of R. prolixus oviduct and bursa, in vitro (Sedra, personal communication), and thus may eventually contribute to the inhibition of egg-laying. The effects of Rhopr-FGLa/ASTs on male reproductive tissues have not yet been examined. One prediction is that they may affect the transfer of sperm and seminal fluids to females during copulation.

Moreover, FGLa/AST-like immunoreactivity is not associated with male or female reproductive tissues of R. prolixus (Sedra and Lange, unpublished). Hence the effect, if any, of Rhopr-

FGLa/ASTs on reproductive tissues will most-likely be mediated via a hormonal route.

Previous studies have not been able to provide any conclusive evidence for Rhopr-FGLa/ASTs involvement, or lack thereof, in JH biosynthesis in R. prolixus. FGLa/AST-like immunoreactive processes are not associated with the CA of fifth-instar and adult R. prolixus, as is the case in species in which FGLa/ASTs are allatostatic, but are associated with the CC and aorta close by

(see Nassel, 2002). Moreover, it is usually the brain neurons associated with nervi corpori cardiaci II (NCCII) that are responsible for this neural inhibition of CA; however, transecting the

NCCII in fifth-instar R. prolixus does not affect the insect’s normal progression into an adult stage (Chiang, 2000), although transecting them in adults does influence egg production (Chiang,

1998). Lastly, it has not been possible to obtain any direct evidence confirming that Rhopr-

FGLa/ASTs possess allatostatic activity as the chemical identification of R. prolixus JH is still unknown. Hence to address this question, we investigated the presence of Rhopr-FGLa/AST-R in

252

the CC/CA complex via qPCR. Our analysis indicates that the receptor is indeed expressed in the

CC/CA of both fifth-instars and adults. Thus Rhopr-FGLa/ASTs may be acting on the CC/CA of

R. prolixus to inhibit JH production; however, other possibilities must also be taken into consideration. For instance, Rhopr-FGLa/ASTs and their receptor may also influence the release of other hormones found in the CC, such as the adipokinetic hormone, as has been shown in L. migratoria (Clark et al., 2008). Additional investigations, possibly involving the knockout of

Rhopr-FGLa/AST-R via RNAi, are needed to further address this question and thus determine the factor(s) controlling JH production and development in R. prolixus.

253

References

Auerswald, L., Birgul, N., Gade, G., Kreienkamp, H.J. and Richter, D. (2001). Structural, functional, and evolutionary characterization of novel members of the allatostatin receptor family from insects. Biochem Biophys Res Commun. 282, 904-9.

Bendena, W.G., Boudreau, J.R., Papanicolaou, T., Maltby, M., Tobe, S.S. and Chin-Sang, I.D. (2008). A Caenorhabditis elegans allatostatin/galanin-like receptor NPR-9 inhibits local search behavior in response to feeding cues. Proc Natl Acad Sci U S A. 105, 1339-42.

Bendena, W.G. and Tobe, S.S. (2012). Families of allatoregulator sequences: a 2011 perspective. Can J Zoolog. 90, 521-544.

Birgul, N., Weise, C., Kreienkamp, H.J. and Richter, D. (1999). Reverse physiology in Drosophila: identification of a novel allatostatin-like neuropeptide and its cognate receptor structurally related to the mammalian somatostatin/galanin/opioid receptor family. EMBO J. 18, 5892-900.

Blom, N., Gammeltoft, S. and Brunak, S. (1999). Sequence and structure-based prediction of eukaryotic protein phosphorylation sites. J Mol Biol. 294, 1351-62.

Chiang, R.G. (1998). Partial localization of a brain factor inhibiting egg production in the blood‐ feeding insect, Rhodnius prolixus. Arch Insect Biochem Physiol. 39, 126-131.

Clark, L., Lange, A.B., Zhang, J.R. and Tobe, S.S. (2008). The roles of Dippu-allatostatin in the modulation of hormone release in Locusta migratoria. J Insect Physiol. 54, 949-58.

254

Coast, G.M. and Schooley, D.A. (2011). Toward a consensus nomenclature for insect neuropeptides and peptide hormones. Peptides. 32, 620-31.

Hergarden, A.C., Tayler, T.D. and Anderson, D.J. (2012). Allatostatin-A neurons inhibit feeding behavior in adult Drosophila. Proc Natl Acad Sci U S A. 109, 3967-72.

Larsen, M.J., Burton, K.J., Zantello, M.R., Smith, V.G., Lowery, D.L. and Kubiak, T.M. (2001). Type A allatostatins from Drosophila melanogaster and Diplotera puncata activate two Drosophila allatostatin receptors, DAR-1 and DAR-2, expressed in CHO cells. Biochem Biophys Res Commun. 286, 895-901.

Lenz, C., Williamson, M. and Grimmelikhuijzen, C.J. (2000). Molecular cloning and genomic organization of a second probable allatostatin receptor from Drosophila melanogaster. Biochem Biophys Res Commun. 273, 571-7.

Lungchukiet, P., Donly, B.C., Zhang, J., Tobe, S.S. and Bendena, W.G. (2008). Molecular cloning and characterization of an allatostatin-like receptor in the cockroach Diploptera punctata. Peptides. 29, 276-85.

Nassel, D.R. (2002). Neuropeptides in the nervous system of Drosophila and other insects: multiple roles as neuromodulators and neurohormones. Prog Neurobiol. 68, 1-84.

255

Onken, H., Moffett, S.B. and Moffett, D.F. (2004). The anterior stomach of larval mosquitoes (Aedes aegypti): effects of neuropeptides on transepithelial ion transport and muscular motility. J Exp Biol. 207, 3731-9.

Reese, M.G., Eeckman, F.H., Kulp, D. and Haussler, D. (1997). Improved splice site detection in Genie. J Comput Biol. 4, 311-23.

Robertson, L., Rodriguez, E.P. and Lange, A.B. (2012). The neural and peptidergic control of gut contraction in Locusta migratoria: the effect of an FGLa/AST. J Exp Biol. 215, 3394-402.

Sarkar, N.R., Tobe, S.S. and Orchard, I. (2003). The distribution and effects of Dippu- allatostatin-like peptides in the blood-feeding bug, Rhodnius prolixus. Peptides. 24, 1553-62.

Vanderveken, M. and O'Donnell, M.J. (2014). Effects of diuretic hormone 31, drosokinin, and allatostatin A on transepithelial K(+) transport and contraction frequency in the midgut and hindgut of larval Drosophila melanogaster. Arch Insect Biochem Physiol. 85, 76-93.

Veelaert, D., Devreese, B., Schoofs, L., Van Beeumen, J., Vanden Broeck, J., Tobe, S.S. and De Loof, A. (1996). Isolation and characterization of eight myoinhibiting peptides from the desert locust, Schistocerca gregaria: new members of the cockroach allatostatin family. Mol Cell Endocrinol. 122, 183-90.

256

Wang, C., Chin-Sang, I. and Bendena, W.G. (2012). The FGLamide-allatostatins influence foraging behavior in Drosophila melanogaster. PLoS One. 7, e36059.

Woodhead, A.P., Thompson, M.E., Chan, K.K. and Stay, B. (2003). Allatostatin in ovaries, oviducts, and young embryos in the cockroach Diploptera punctata. J Insect Physiol. 49, 1103- 14.

Zandawala, M., Li, S., Hauser, F., Grimmelikhuijzen, C.J. and Orchard, I. (2013). Isolation and functional characterization of calcitonin-like diuretic hormone receptors in Rhodnius prolixus. PLoS One. 8, e82466.

Zandawala, M. and Orchard, I. (2013). Post-feeding physiology in Rhodnius prolixus: the possible role of FGLamide-related allatostatins. Gen Comp Endocrinol. 194, 311-7.

Zandawala, M., Paluzzi, J.P. and Orchard, I. (2011). Isolation and characterization of the cDNA encoding DH(31) in the kissing bug, Rhodnius prolixus. Mol Cell Endocrinol. 331, 79-88.

257

Acknowledgements

The authors wish to thank Nikki Sarkar for maintaining the colony and Amir Haddad for technical support. This work was supported by a Natural Sciences and Engineering Research

Council of Canada (NSERC) Discovery Grant to I.O. and a NSERC Canadian Graduate

Scholarship to M.Z.

258

Chapter 8: General Discussion

259

Major conclusions

This section summarizes the major findings of all the research chapters and provides some preliminary results addressing the questions raised in this thesis.

Rhopr-CT/DH and its receptors

Three splice variants encoding Rhopr-CT/DH are found in the CNS

In chapter 2, I determined the cDNA sequences of three Rhopr-CT/DH splice variants (A, B and

C) and characterized their expression using various molecular techniques. Various analyses confirmed that the three splice variants, all of which encode an identical mature peptide, are expressed in the CNS of fifth-instar Rhodnius prolixus; however, it is unknown if the three variants exhibit a differential expression pattern, either spatially or temporally. Additional spatial and temporal expression profiling of Rhopr-CT/DH has now been performed using semi- quantitative RT-PCRs. Rhopr-CT/DH-A, -B and -C are all expressed within an individual CNS of both male and female R. prolixus. Hence, these variants do not appear to be expressed differentially between the two sexes. To determine if the three variants are differentially expressed within the CNS, spatial expression analysis was performed using different parts of the

CNS. Within the CNS, Rhopr-CT/DH-A, -B and -C are all expressed within the brain and suboesophageal ganglion, the prothoracic ganglion and the mesothoracic ganglionic mass.

Lastly, Rhopr-CT/DH-A, -B and -C are all expressed in the CNS at different time points pre- and post-feeding (ranging from unfed to 10 days post-feeding). Thus, it is still unclear what the biological significance is, if any, of having three splice variants all of which encode the same mature peptide.

CT/DH mediates its actions via two functionally different receptors

260

In chapter 3, I described the isolation, spatial expression profile and functional characterization of two CT/DH receptors in R. prolixus (Rhopr-CT/DH-R1 and Rhopr-CT/DH-R2). I determined the cDNA sequences of three splice variants encoding Rhopr-CT/DH-R1 (A, B and C) and two splice variants encoding Rhopr-CT/DH-R2 (A and B). This work clarified some previous reports and provided novel insights into CT/DH signaling in insects. For the sake of simplicity, when discussing Rhopr-CT/DH-R1 and R2 below, I refer to only the variants that encode a typical and functional GPCR, namely variants Rhopr-CT/DH-R1-B/C and Rhopr-CT/DH-R2-B.

Rhopr-CT/DH-R1 is orthologous to the previously characterized Drosophila melanogaster

CT/DH receptor (now referred to as Drome-CT/DH-R1) (Johnson et al., 2005). Johnson et al.

(2005) proposed that insect CT/DH-Rs require accessory proteins for proper functioning, analogous to the mammalian calcitonin (CT) receptor (CTR) signaling. In humans, there are two calcitonin-like receptors, referred to as the CTR and calcitonin receptor-like receptor (CRLR).

These two receptors interact with two types of accessory proteins: receptor component protein

(RCP) and receptor-activity-modifying protein (RAMP). RCP is required for efficient coupling to the signal transduction pathway whereas the three RAMPs (1, 2 and 3) are required for trafficking the receptor to the cell surface and for ligand specificity (Prado et al., 2002). There are several peptides that form the calcitonin-family of peptides (calcitonin, calcitonin gene- related peptide (CGRP), adrenomedullin 1 and 2, and amylin). The pharmacological profile of the two receptors changes depending on which type of RAMP subunit they interact with. Hence interaction of CTR with RAMP1 results in an amylin receptor but when no RAMPs are present,

CTR behaves as a receptor for calcitonin (Couvineau and Laburthe, 2012). Johnson et al. (2005) expressed Drome-CT/DH-R1 in HEK-293 cells and measured cAMP levels following the addition of Drome-CT/DH. However, addition of 10-6M Drome-CT/DH failed to elicit any

261 response. A minute response (under two-fold) was only observed when the receptor was co- transfected with D. melanogaster RCP. The response was much greater (about ten-fold) when the receptor was co-transfected with human RCP. In either case, the EC50 value for the response was in the high nanomolar range, which is uncharacteristic of a hormone. This study led me to initially question whether Drome-CT/DH-R1 was indeed a CT/DH receptor and whether accessory proteins are required for proper signaling. Although the recent functional characterization of Aedes aegypti CT/DH-R1 (orthologous to Rhopr-CT/DH-R1) inferred from

RNAi experiments was able to eliminate the doubts about the identity of the receptor (Kwon et al., 2012), doubts still remained about its signaling properties. Kwon et al. (2012) showed that

MTs isolated from female mosquitoes in which Aedae-CT/DH-R1 was knocked-down exhibited up to 57% decrease in fluid secretion in response to Aedae-CT/DH. Hence Aedae-CT/DH-R1 and its orthologs from other insects (including Drome-CT/DH-R1 and Rhopr-CT/DH-R1) are indeed CT/DH receptors. Since Kwon et al. (2012) did not functionally characterize Aedae-

CT/DH-R1 in a cell-based assay, the role of accessory proteins, if any, was unclear. Hence I functionally characterized Rhopr-CT/DH-R1 in two different cells lines in order to see if accessory proteins are required for efficient signaling. If accessory proteins are indeed required, then the response in two cell lines of different backgrounds (and hence different endogenous accessory proteins) would likely be different (Johnson et al., 2005). I found a similar response with Rhopr-CT/DH when Rhopr-CT/DH-R1 was expressed in cell lines of HEK-293 and CHO backgrounds (Zandawala et al., 2013; Zandawala et al., 2014). This, coupled with the fact that orthologs of RCP and other peptides of the calcitonin-family are not found in insect genomes, including that of R. prolixus, certainly suggests that insect CT/DH-Rs do not require accessory proteins for efficient coupling and to alter their pharmacological profile.

262

Rhopr-CT/DH-R2 is orthologous to the D. melanogaster receptor (CG4395), an orphan receptor whose ligand was unknown until now. The only study which examined its role in insects comes from D. melanogaster, where the receptor was shown to be expressed in neurons that are critical for male courtship behavior (Li et al., 2011). When Rhopr-CT/DH-R2 was tested in a heterologous expression system, it was found to be more sensitive to Rhopr-CT/DH compared to

Rhopr-CT/DH-R1 in both the cell lines (Zandawala et al., 2013; Zandawala et al., 2014).

Assuming that the differences in response are not attributable to different transfection efficiencies of the two receptors and/or any shortcomings of the heterologous expression system, these receptors appear to be functionally different. Perhaps the high EC50 value of the response for Rhopr-CT/DH-R1 may be physiologically relevant if that receptor is expressed in tissues that experience a high dose of Rhopr-CT/DH locally (such as when it would be released more locally or at a synapse). This would also mean that Rhopr-CT/DH-R2 might be expressed in tissues that are not innervated (e.g. MTs) and hence the lower dose of Rhopr-CT/DH that it would encounter

(via a hormonal route) would be sufficient for its activation. Thus, one would predict the two receptors to have distinct expression patterns, with each receptor expressed exclusively in various tissues. I examined the spatial expression pattern of both the receptors using qPCR.

Rhopr-CT/DH-R1 is expressed in the CNS, dorsal vessel, salivary glands and hindgut; these are the same tissues where Rhopr-CT/DH-like immunoreactivity is present, and hence could experience a high dose of Rhopr-CT/DH, locally (Te Brugge et al., 2005; Zandawala et al.,

2013). Moreover, Rhopr-CT/DH-R2 and not Rhopr-CT/DH-R1 is expressed in MTs, consistent with this theory.

Interestingly, there are several tissues, including the CNS, prothoracic glands, salivary glands, testes, ovaries and female reproductive tissues in which both the receptors are expressed. This

263 situation is unlike that in D. melanogaster, where the two receptors have distinct expression patterns; Drome-CT/DH-R1 is highly expressed in the CNS, midgut, hindgut, heart and MTs, whereas Drome-CT/DH-R2 is only found in the CNS and male accessory glands (Chintapalli et al., 2007). Why would both the receptors be present in the same tissue in R. prolixus, if one is more sensitive than the other? Could the response to Rhopr-CT/DH be altered if both the receptors are present in a given tissue and perhaps the same cell? To address these questions,

Rhopr-CT/DH-R2 was transiently transfected into CHO/G16 cells stably-expressing Rhopr-

CT/DH-R1. Preliminary results indicate that presence of both receptors does indeed result in a much more sensitive response; the EC50 value for the response when both the receptors are present is lower than when only one of the receptor types is present (Figure 1). Moreover, having more of the same receptor type by transiently-transfecting Rhopr-CT/DH-R1 in cells stably- expressing Rhopr-CT/DH-R1 also lowers its EC50 value. These results indicate that having more receptors results in an increased sensitivity to Rhopr-CT/DH. However, it does not rule out a possible interaction between the receptors of the same type as well as the two subtypes. There is a possibility that the receptors could form dimers (homodimers and heterodimers), as is common for class-B GPCRs (Ng et al., 2012). More controls are needed to determine if the receptors interact in this manner and result in a synergistic response.

264

Figure 1: Functional assay of R. prolixus CT/DH receptors (Rhopr-CT/DH-R1 and Rhopr-

CT/DH-R2) to investigate any possible interaction between the receptors. Influx of calcium from the endoplasmic reticulum following receptor activation was detected using the reporter, aequorin. Four different transfections were performed: (1) empty vector was transiently transfected in CHO/G16 cells stably-expressing Rhopr-CT/DH-R1 (control 1), (2) Rhopr-

CT/DH-R1 was transiently transfected in CHO/G16 cells stably-expressing Rhopr-CT/DH-R1

(control 2), (3) Rhopr-CT/DH-R2 was transiently transfected in CHO/G16 cells (control 3) and

(4) Rhopr-CT/DH-R2 was transiently transfected in CHO/G16 cells stably-expressing Rhopr-

CT/DH-R1. The graphs show a dose-dependent effect on the bioluminescence response

(normalized) after addition of Rhopr-CT/DH (see Zandawala et al. (2013) for a detailed method).

Co-expression moves the dose-response curve to the left.

265

266

Another functional significance of having two receptors is that different intracellular signaling cascades could be activated depending on which type of G-protein alpha subunit the receptor couples with. So aside from the scenarios discussed above, Rhopr-CT/DH-R1 and Rhopr-

CT/DH-R2 may couple with different G-protein alpha subunits. Previously, Rhopr-CT/DH was shown to increase the cAMP content of anterior midgut, hindgut and dorsal vessel, in vitro; however, no change was observed in the cAMP content of R. prolixus MTs (see Orchard, 2009).

Thus, Rhopr-CT/DH-R2 probably does not couple with the Gs alpha subunit (which activates the cAMP pathway) since it is the only one expressed in MTs, whereas Rhopr-CT/DH-R1 probably couples with the Gs alpha subunit. To test this hypothesis, the heterologous receptor assay utilizing HEK293/CNG cells was performed in the presence of a Gs alpha subunit inhibitor,

NF449. If the receptor indeed couples with the Gs alpha subunit, then no response would be elicited in the presence of NF449; however, the results of this assay were inconclusive. Another option to decipher the signaling properties of the two receptors is to perform the same assay in calcium-free saline, supplemented with a calcium chelator (to chelate any calcium released by the cells). If Rhopr-CT/DH-R1 works via cAMP, then its activation would result in an increase in intracellular cAMP, which would cause an opening of the CNG channel and allow extracellular calcium to enter the cell. This calcium would then be detected using the reporter, aequorin. Therefore, no response should be observed if there is no calcium in the extracellular media but an increased response would be seen in normal media.

Lastly, the spatial expression analysis of the two receptors sheds some light on potential novel target tissues of Rhopr-CT/DH. The receptor transcripts were detected in prothoracic glands

(which also contain fat body) and reproductive tissues of both males and females, suggesting novel roles for this neuropeptide. Since R. prolixus oviducts are muscular and exhibit

267 spontaneous contractions (Sedra and Lange, 2014), it was worth examining if Rhopr-CT/DH was able to stimulate their contraction frequency. Preliminary work indicates that Rhopr-CT/DH does not influence the oviduct contraction frequency (Zandawala, Sedra and Orchard, unpublished).

Alternatively, Rhopr-CT/DH could influence the release of ecdysteroids by ovaries and prothoracic glands and thus play a role in egg development and metamorphosis (Cardinal-Aucoin et al., 2013; Vafopoulou and Steel, 1989). Experiments to validate this scenario are on-going.

Rhopr-CT/DH peptide length is critical for receptor activation

In chapter 4, I examined the structure-activity relationships of two Rhopr-CT/DH analogs (full- length form and N-terminal truncated form lacking the first 16 residues) using a hindgut contraction assay and the heterologous receptor expression system. The truncated form was tested to determine the importance of peptide length. A comparison of arthropod CT/DH sequences and vertebrate calcitonin shows that only four amino acid residues are conserved across all the sequences (see Zandawala, 2012). Furthermore, Rhopr-CT/DH is only 19% identical to the human CT. When looking at the alignment of all the sequences, two features stand out. Firstly, the C-terminal proline and glycine (the latter undergoes amidation) are conserved across all sequences. Secondly, the length of the sequences is also conserved, ranging from 31 to 34 amino acid residues. Compromising the peptide length using the N-terminal truncated analog resulted in a peptide that failed to activate the receptor, indicating that peptide length is critical. It will be interesting to see the importance of proline-amide C-terminus in receptor activation using alanine substitutions.

FGLa/ASTs and their putative receptor

Rhopr-FGLa/ASTs are inhibitors of visceral muscle contractions

268

In chapter 5, I cloned and characterized the cDNA encoding Rhopr-FGLa/ASTs and also examined their physiological effects using various biological assays. Molecular analyses indicated that Rhopr-FGLa/AST is expressed in the CNS and peripheral tissues such as the midgut. The results of the physiological assays demonstrated that Rhopr-FGLa/ASTs are biologically active on tissues associated with post-feeding physiology, namely the anterior midgut, hindgut and dorsal vessel. Contractions of these tissues aid in circulation of the haemolymph (and the diuretic hormones in it), mixing of the blood meal, and reducing unstirred layers surrounding MTs, all of which may contribute to the rapid post-feeding diuresis. Hence inhibition of these contractions by Rhopr-FGLa/ASTs would perhaps contribute to an anti- diuretic strategy when the insect has eliminated the excess water and salts.

Rhopr-FGLa/ASTs indirectly aid the anti-diuretic hormone, Rhopr-CAPA-α2, during the cessation of diuresis

In chapter 6, I determined the possible roles of Rhopr-FGLa/ASTs during the rapid post-feeding diuresis in R. prolixus. Immunohistochemical analyses show that Rhopr-FGLa/ASTs are released into the haemolymph from the CNS following feeding, possibly at a time when the antidiuretic hormone, Rhopr-CAPA-α2, is also released into the haemolymph (see Paluzzi, 2012). Rhopr-

CAPA-α2 is a potent inhibitor of serotonin-stimulated anterior midgut absorption and MT secretion but Rhopr-FGLa/ASTs have no such effect. Thus, although Rhopr-FGLa/ASTs appear to play no direct role in anti-diuresis, they can indirectly aid Rhopr-CAPA-α2 by inhibiting contractions of muscles that are actively involved in feeding-related events; dorsal vessel contractions are vital for circulation of the haemolymph and the diuretic hormones released following feeding, the anterior midgut contracts to remove unstirred layers, mix the blood meal, and improve circulation of haemolymph, and hindgut contraction is used to expel the urine.

269

Release of Rhopr-FGLa/ASTs into the haemolymph might therefore occur at the cessation of diuresis in R. prolixus, when diuretic hormone release has stopped and urine production is being halted.

Since Rhopr-FGLa/AST-R is not expressed in MTs (see Chapter 7) it is clear that Rhopr-

FGLa/ASTs do not influence MT secretion; however, Rhopr-FGLa/AST-R is expressed in the anterior midgut. As mentioned earlier, Rhopr-CAPA-α2 inhibits serotonin-stimulated absorption across the anterior midgut and serotonin-stimulated MT secretion but is unable to inhibit Rhopr-

CRF/DH stimulated MT secretion. So one could wonder whether Rhopr-FGLa/ASTs are able to inhibit Rhopr-CRF/DH (but not serotonin) stimulated anterior midgut absorption or if inhibition of the anterior midgut contractions is their sole action. Similarly, Rhopr-FGLa/ASTs inhibit

Rhopr-kinin-2 (which does not work via cAMP) stimulated hindgut contractions. Whether they can inhibit hindgut contractions stimulated by Rhopr-CT/DH, Rhopr-CRF/DH or serotonin (all of which predominantly work via cAMP) remains to be seen. Additional investigations are needed to clarify these uncertainties.

Rhopr-FGLa/ASTs could influence processes associated with feeding, reproduction and development via a single GPCR

In chapter 7, I described the isolation and spatial expression profile of the transcript encoding a putative Rhopr-FGLa/AST-R. Expression profiling via qPCR indicated that the receptor is expressed in various tissues associated with post-feeding physiology (dorsal vessel, hindgut, midgut and foregut), reproduction (male and female reproductive tissues) and development

(CC/CA complex).

270

Integrating the whole

A model for the rapid post-feeding diuresis in R. prolixus, summarizing the data presented in this thesis and previous reports (see Orchard, 2009; Paluzzi, 2012), is shown in Figure 2. Rhopr-

CT/DH, serotonin, Rhopr-CRF/DH and Rhopr-kinin contribute towards the diuretic strategy whereas Rhopr-FGLa/ASTs and Rhopr-CAPA-α2 contribute towards the anti-diuretic strategy.

Rhopr-CT/DH and Rhopr-FGLa/ASTs don’t seem to play any direct role during the rapid post- feeding diuresis. Perhaps they may be involved with the long term physiological changes post- feeding.

The potential physiological roles of FGLa/ASTs and CT/DH in R. prolixus are summarized in

Figure 3. Some of these roles are predicted based on the spatial expression analyses of the receptor transcripts (see previous chapters) and previous reports examining these roles in other species; however, they need to be experimentally-verified in R. prolixus. Hence, Rhopr-CT/DH stimulates contractions of various tissues, whereas Rhopr-FGLa/ASTs inhibit these contractions as part of the short-term physiological changes associated with feeding. In the long term, these two neuropeptides may influence digestion of the blood meal and/or the subsequent absorption of K+ across the R. prolixus posterior midgut. They may also influence metamorphosis through their effects on JH (Rhopr-FGLa/ASTs) and ecdysteroid synthesis (Rhopr-CT/DH).

271

Figure 2: Schematic of the R. prolixus alimentary canal summarizing the physiological roles played by various diuretic and anti-diuretic hormones during the rapid post-feeding diuresis.

Data pertaining to serotonin, CRF/DH, kinin and Rhopr-CAPA-α2 has been reviewed previously

(see Orchard, 2009; Paluzzi, 2012). Data pertaining to the neuropeptides characterized in this thesis, FGLa/ASTs and CT/DH, are shown in red and green, respectively. Modified figure from

Mykles et al. (2010) (alimentary canal originally drawn by Zach McLaughlin).

272

273

Figure 3: A model summarizing the potential physiological roles of Rhopr-FGLa/ASTs and

Rhopr-CT/DH. The model is based on previous reports (see Orchard, 2009) and the data presented in this thesis. Rhopr-CT/DH is stimulatory whereas Rhopr-FGLa/ASTs are inhibitory.

There are tick marks next to the roles that have been experimentally verified and question marks next to predicted roles.

274

275

Future directions

Investigating the in vivo effects on post-feeding diuresis

Most of the current knowledge on water and ion homeostasis, and especially the post-feeding diuresis in insects has been based on in vitro bioassays. The recent advances in gene knockdown have facilitated the in vivo studies (Benoit et al., 2014; Kersch and Pietrantonio, 2011; Kwon et al., 2012), but there is still a huge void that needs to be addressed. Consequently, we have started to utilize RNAi to knockdown the receptors for various hormones and examine their possible involvement in post-feeding diuresis in R. prolixus, in vivo. Preliminary experiments involving the knockdown of a serotonin receptor that is highly expressed in MTs indicates that, as expected from previous data, serotonin is not the only hormone responsible for the rapid post-feeding diuresis in R. prolixus, in vivo. The surprising result was that the insect appears to compensate for the reduction in serotonin receptors. Thus, the models based on in vitro assays cannot always be assumed to reflect the in vivo physiology. So in order to determine the roles of Rhopr-CT/DH and Rhopr-FGLa/ASTs, it would be worthwhile to knockdown their receptors via RNAi and examine the effect on post-feeding diuresis and metamorphosis.

Identifying other hormones regulating diuresis and anti-diuresis in R. prolixus

Only the roles of the four classical diuretic hormone families (CRF/DH, CT/DH, kinin, CAPA) and serotonin have been examined in much detail in R. prolixus. But there is a high possibility that other hormones may also be involved in post-feeding diuresis in R. prolixus. Based on the in vitro assays, only serotonin and Rhopr-CRF/DH act as “true” diuretic hormones, whereas Rhopr-

CT/DH and Rhopr-kinins may only indirectly stimulate diuresis in R. prolixus. They may do so by stimulating contractions of various tissues which helps reduce unstirred layers around the

MTs and increases haemolymph (and diuretic hormone) circulation. Perhaps there are other

276 hormones that may be compensating for the lack of any direct effect of Rhopr-CT/DH and

Rhopr-kinins on diuresis. Hence, tachykinin and tyramine, both of which have been shown to stimulate MT secretion in other insects (Blumenthal, 2003; Cabrero et al., 2013), could well play a part in R. prolixus diuresis. Aside from these hormones, there is also another candidate which has not yet been discovered in Protostomia. A class-B GPCR orthologous to the human parathyroid hormone receptor (PTHR) has been identified in genomes of various insects (see

Tanaka et al., 2014). Hardly anything is known about this receptor (or its possible ligand) in insects because the gene encoding this receptor has been lost in several model insects including

D. melanogaster, Anopheles gambiae, Bombyx mori and a hemipteran, Acyrthosiphon pisum (Li et al., 2013). Although this receptor was also thought to have been lost in R. prolixus (Li et al.,

2013), I have been able to identify a single putative PTHR in the R. prolixus genome, in silico.

The human parathyroid hormone is a polypeptide comprising 84 amino acids and regulates blood calcium levels amongst other things. Moreover, phylogenetic analysis shows that insect PTHRs are located between CT/DH-Rs and CRF/DH-Rs, both of which have been shown to be involved in regulating water balance. Hence PTHR, and its yet unidentified ligand, may contribute to diuresis in R. prolixus.

With regards to the antidiuretic mechanism in insects, a universal anti-diuretic hormone family is yet to be discovered in insects. Hence, although, Rhopr-CAPA-α2 is anti-diuretic in R. prolixus, this family of peptides is usually associated with stimulating diuresis in other insects (see

Schooley et al., 2012). Moreover, Rhopr-CAPA-α2 only negates the actions of serotonin but not

Rhopr-CRF/DH. Since Rhopr-FGLa/ASTs do not play any direct part in the inhibition of post- feeding diuresis in R. prolixus, I believe that there must be another factor involved in the anti- diuretic mechanism, which may possibly counteract the actions of Rhopr-CRF/DH. One such

277 candidate is the heterodimeric glycoprotein hormone, GPA2/GPB5, which has recently been shown to regulate ion transport in A. aegypti hindgut (Paluzzi et al., 2014).

Clearly, with the advances brought about by the R. prolixus genome project, the future is exciting and more detailed information will undoubtedly be forthcoming.

278

References

Benoit, J.B., Hansen, I.A., Attardo, G.M., Michalkova, V., Mireji, P.O., Bargul, J.L., Drake, L.L., Masiga, D.K. and Aksoy, S. (2014). Aquaporins are critical for provision of water during lactation and intrauterine progeny hydration to maintain tsetse fly reproductive success. PLoS Negl Trop Dis. 8, e2517.

Blumenthal, E.M. (2003). Regulation of chloride permeability by endogenously produced tyramine in the Drosophila Malpighian tubule. Am J Physiol Cell Physiol. 284, C718-28.

Cabrero, P., Richmond, L., Nitabach, M., Davies, S.A. and Dow, J.A. (2013). A biogenic amine and a neuropeptide act identically: tyramine signals through calcium in Drosophila tubule stellate cells. Proc Biol Sci. 280, 20122943.

Cardinal-Aucoin, M., Rapp, N. and Steel, C.G. (2013). Circadian regulation of hemolymph and ovarian ecdysteroids during egg development in the insect Rhodnius prolixus (Hemiptera). Comp Biochem Physiol A Mol Integr Physiol. 166, 503-9.

Chintapalli, V.R., Wang, J. and Dow, J.A. (2007). Using FlyAtlas to identify better Drosophila melanogaster models of human disease. Nat Genet. 39, 715-20.

Couvineau, A. and Laburthe, M. (2012). The family B1 GPCR: structural aspects and interaction with accessory proteins. Curr Drug Targets. 13, 103-15.

Kersch, C.N. and Pietrantonio, P.V. (2011). Mosquito Aedes aegypti (L.) leucokinin receptor is critical for in vivo fluid excretion post blood feeding. FEBS Lett. 585, 3507-12.

279

Li, C., Chen, M., Sang, M., Liu, X., Wu, W. and Li, B. (2013). Comparative genomic analysis and evolution of family-B G protein-coupled receptors from six model insect species. Gene. 519, 1-12.

Mykles, D.L., Adams, M.E., Gade, G., Lange, A.B., Marco, H.G. and Orchard, I. (2010). Neuropeptide action in insects and crustaceans. Physiol Biochem Zool. 83, 836-46.

Ng, S.Y., Lee, L.T. and Chow, B.K. (2012). Receptor oligomerization: from early evidence to current understanding in class B GPCRs. Front Endocrinol (Lausanne). 3, 175.

Orchard, I. (2009). Peptides and serotonin control feeding-related events in Rhodnius prolixus. Front Biosci. 1, 250-62.

Paluzzi, J.P. (2012). Anti-diuretic factors in insects: the role of CAPA peptides. Gen Comp Endocrinol. 176, 300-8.

Paluzzi, J.P., Vanderveken, M. and O'Donnell, M.J. (2014). The heterodimeric glycoprotein hormone, GPA2/GPB5, regulates ion transport across the hindgut of the adult mosquito, Aedes aegypti. PLoS One. 9, e86386.

Prado, M.A., Evans-Bain, B. and Dickerson, I.M. (2002). Receptor component protein (RCP): a member of a multi-protein complex required for G-protein-coupled signal transduction. Biochem Soc Trans. 30, 460-4.

Schooley, D.A., Horodyski, F.M. and Coast, G.M., 2012. 9 - Hormones Controlling Homeostasis in Insects, in: Gilbert, L.I. (Ed.), Insect Endocrinology. Academic Press, San Diego, pp. 366-429.

Sedra, L. and Lange, A.B. (2014). The female reproductive system of the kissing bug, Rhodnius prolixus: arrangements of muscles, distribution and myoactivity of two endogenous FMRFamide-like peptides. Peptides. 53, 140-7.

280

Tanaka, Y., Suetsugu, Y., Yamamoto, K., Noda, H. and Shinoda, T. (2014). Transcriptome analysis of neuropeptides and G-protein coupled receptors (GPCRs) for neuropeptides in the brown planthopper Nilaparvata lugens. Peptides. 53, 125-33.

Te Brugge, V.A., Lombardi, V.C., Schooley, D.A. and Orchard, I. (2005). Presence and activity of a Dippu-DH31-like peptide in the blood-feeding bug, Rhodnius prolixus. Peptides. 26, 29-42.

Vafopoulou, X. and Steel, C.G. (1989). Developmental and diurnal changes in ecdysteroid biosynthesis by prothoracic glands of Rhodnius prolixus (Hemiptera) in vitro during the last larval instar. Gen Comp Endocrinol. 74, 484-93.

Zandawala, M. (2012). Calcitonin-like diuretic hormones in insects. Insect Biochem Mol Biol. 42, 816-25.

Zandawala, M., Poulos, C. and Orchard, I. (2014). Structure-activity relationships of two Rhodnius prolixus calcitonin-like diuretic hormone analogs. Peptides.

281

Appendix A: Calcitonin-like diuretic hormone in insects

The proceeding chapter was reprinted/reproduced with permission from Elsevier.

Full citation details:

Calcitonin-like diuretic hormones in insects.

Zandawala, M.

Insect Biochem Mol Biol. 2012 Oct;42(10):816-25. doi: 10.1016/j.ibmb.2012.06.006

282

Abstract

Insect neuropeptides control various biological processes including growth, development, homeostasis and reproduction. The calcitonin-like diuretic hormone (CT/DH) is one such neuropeptide that has been shown to affect salt and water transport by Malpighian tubules of several insects. With an increase in the number of sequenced insect genomes, CT/DHs have been predicted in several insect species, making it easier to characterize the gene encoding this hormone and determine its function in the species in question. This mini review summarizes the current knowledge on insect CT/DHs, focusing on mRNA and peptide structures, distribution patterns, physiological roles, and receptors in insects.

283

Introduction

Neuropeptides represent the largest class of signaling compounds amongst insects that control several critical biological processes including growth, development, homeostasis and reproduction. Based on the annotation of complete insect genomes, it is estimated that about 30-

40 neuropeptide precursor-encoding genes are present in a given insect (see Roller et al., 2008).

Calcitonin-like diuretic hormone (CT/DH) is one such neuropeptide that has been shown to affect salt and water transport by Malpighian tubules (MTs) of several insects. As the number of complete sequenced genomes has increased greatly in recent times, opportunities for genome mining are plentiful. Consequently, CT/DHs and their associated receptors have been predicted in several insect species, as well as crustaceans, chelicerates and other members of Arthropoda

(Christie, 2008; Christie et al., 2010a; Christie et al., 2011; Gard et al., 2009; Hauser et al., 2006;

Hauser et al., 2008). This greatly facilitates the cloning and characterization of genes encoding this hormone and its receptor. For example, primers and probes can be generated using the predicted sequences to perform quantitative-PCR and in situ hybridization, respectively, to determine the expression pattern of the transcripts encoding these peptides and their receptors.

RNAi can also be used to knockdown these transcripts and study their effect on whole animal physiology. More importantly, knowing the tissues in which the receptor is expressed allows one to perform bioassays using synthetic peptides (as opposed to crude tissue extracts or biochemically purified peptides) on actual target tissues. Hence knowing their gene sequence will aid in elucidating the function(s) of CT/DHs in the species in question. Since little is known about the biological activities of CT/DHs, it is worthwhile to summarize the molecular and physiological knowledge of these peptides in order to create avenues for future research to determine their function(s). In this mini review, the current knowledge on insect CT/DHs is

284 summarized, particularly focusing on mRNA and peptide structures, as well as their distribution patterns, physiological roles, and associated receptors.

Discovery

Leptinotarsa decemlineata (El-Salhy et al., 1983; Veenstra et al., 1985). However, it wasn’t until

2000; Te Brugge et al., 2005). Although it has low sequence identity to the vertebrate CT, it has the conserved C-terminal Glycine-X-Proline-NH2 and they are similar in length (see Table S1)

(Furuya et al., 2000). These are both important features in terms of their bioactivity as has been shown using vertebrate CTs (see Andreotti et al., 2006). Residues 9-19 of salmon calcitonin are involved in forming a stable α-helix which interacts with the C-terminus (Amodeo et al., 1999).

Furthermore, the α-helical region has been shown to directly interact with the N-terminus of the receptor and varying the length of the helix causes a reduction in the peptide’s bioactivity

(Andreotti et al., 2006; Stroop et al., 1996). Since their isolation from D. punctata, CT/DHs have been identified and cloned in several insect species (Table S1) and their role as diuretic hormones examined in few insects (see section: Biological activity). Due to minimal sequence

285 similarity between vertebrate CTs and insect CT/DHs, as well as their inability to act as “true” diuretic hormones (they are mostly only able to stimulate MT secretion) in most of the species examined so far (see section: Biological activity), one might question the appropriateness of the name attributed to this peptide family. Due to the functional similarity between the vertebrate calcitonin receptors and insect CT/DH receptor(s) (see section: Receptors), the CT-like nature of this peptide family remains assured. However, their role as diuretic hormones in insects is still questionable.

286

Table S1: Structures of mature CT/DHs (deduced or sequenced).

Species Peptide structure Reference Insects 1 Rhodnius prolixus GLDLGLSRGFSGSQAAKHLMGLAAANYAGGP-NH2 (Te Brugge et al., 2008) 1 Diploptera punctata GLDLGLSRGFSGSQAAKHLMGLAAANYAGGP-NH2 (Furuya et al., 2000) 1 Apis mellifera GLDLGLSRGFSGSQAAKHLMGLAAANYAGGP-NH2 (Schooley et al., 2005) 1 Solenopsis invicta GLDLGLSRGFSGSQAAKHLMGLAAANYAGGP-NH2 Predicted 1 Camponotus floridanus GLDLGLSRGFSGSQAAKHLMGLAAANYAGGP-NH2 (Gruber and Muttenthaler, 2012) 1 Apis florea GLDLGLSRGFSGSQAAKHLMGLAAANYAGGP-NH2 Predicted 1 Megachile rotundata GLDLGLSRGFSGSQAAKHLMGLAAANYAGGP-NH2 Predicted 1 Bombus terrestris GLDLGLSRGFSGSQAAKHLMGLAAANYAGGP-NH2 Predicted 1 Bombus impatiens GLDLGLSRGFSGSQAAKHLMGLAAANYAGGP-NH2 Predicted 2 Tribolium castaneum GLDLGLGRGFSGSQAAKHLMGLAAANFAGGP-NH2 (Li et al., 2008) 3 Anopheles gambiae TVDFGLSRGYSGAQEAKHRMAMAVANFAGGP-NH2 (Coast et al., 2005) 3 Aedes aegypti TVDFGLSRGYSGAQEAKHRMAMAVANFAGGP-NH2 (Schooley et al., 2005) 3 Culex quinquefasciatus TVDFGLSRGYSGAQEAKHRMAMAVANFAGGP-NH2 (see Zandawala et al., 2011) 4 Drosophila melanogaster TVDFGLARGYSGTQEAKHRMGLAAANFAGGP-NH2 (Coast et al., 2001) 5 Drosophila virilis TVDFGLARGYSGTQEAKHRMGLAAANFPGGP-NH2 (see Schooley et al., 2012) 6 Nasonia vitripennis GLDLGLNRGFSGSQAAKHLMGLAAANYAGGP-NH2 (Hauser et al., 2010) 6 Nasonia longicornis GLDLGLNRGFSGSQAAKHLMGLAAANYAGGP-NH2 Predicted 6 Nasonia giraulti GLDLGLNRGFSGSQAAKHLMGLAAANYAGGP-NH2 Predicted 6 Pogonomyrmex barbatus GLDLGLNRGFSGSQAAKHLMGLAAANYAGGP-NH2 Predicted 7 Acromyrmex echinatior GLDLGLNRGYSGSQAAKHMMGLAAANYAGGP-NH2 Predicted 7 Atta cephalotes GLDLGLNRGYSGSQAAKHMMGLAAANYAGGP-NH2 (Gruber and Muttenthaler, 2012) Acyrthosiphon pisum GLDLGLSRGYSGTQAAKHLMGMAAANFAGGP-NH2 (see Zandawala et al., 2011)

Nilaparvata lugens GLDLGLSRGFSGSQAAKHLMGLAAANYAAGP-NH2 (see Zandawala et al., 2011) Bombyx mori AFDLGLGRGYSGALQAKHLMGLAAANFAGGP-NH2 (Schooley et al., 2005) Harpegnathos saltator GLDLGLSRGFSGSQSAKHMMGLAAANYAGGP-NH2 (Gruber and Muttenthaler, 2012) Heliothis virescens ALDLGLSRGYSGALQAKHLMGLAAAHYAGGP-NH2 (see Schooley et al., 2012) Manduca sexta ALDLGLSRGYSGALQAKHLIGLAAANYAGGP-NH2 (see Schooley et al., 2012) Pediculus humanus GLDLGLSRGFSGSQAAKHLMGLAAANFAGGP-NH2 (see Schooley et al., 2012) corporis Dendroctonus ponderosae GIDLGLGRGFSGSQAAKHLMGLAAANFAGGP-NH2 (see Schooley et al., 2012) Linepithema humile GLDLGINRGFSGSEAAKHLMGLAAANYAGGP-NH2 Predicted Crustaceans Daphnia pulex GVDFGLGRGYSGSQAAKHLMGLAAANYAIGP-NH2 (Gard et al., 2009) 2 Homarus americanus GLDLGLGRGFSGSQAAKHLMGLAAANFAGGP-NH2 (Christie et al., 2010b) Lepeophtheirus salmonis GLDFGLGRGFSGTQAAKHFMGLAAAKYAGGP-NH2 (see Schooley et al., 2012) Chelicerates Ixodes scapularis AGGLLDFGLSRGASGAEAAKARLGLKLANDPYGP-NH2 (Christie, 2008) Varroa destructor SNGMLDFGLARGMSGVDAAKARLGLKYANDPYGP-NH2 (see Zandawala et al., 2011) Arachnids Tetranychus urticae GLDLGLRRGLSGQRAAKHLVGLANAEFAGGP-NH2 (see Schooley et al., 2012) Vertebrates Gallus gallus domesticus CASLSTCVLGKLSQELHKLQTYPRTDVGAGTP-NH2 (see Furuya et al., 2000) Homo sapiens CGNLSTCMLGTYTQDFNKFHTFPQTAIGVGAP-NH2 (see Furuya et al., 2000)

Amino acids that are shared by all sequences are highlighted in black. Amino acids that are shared by all arthropod sequences are highlighted in gray. Amino acids that are shared by all insect sequences are highlighted in yellow. 1,2,3,6,7 Identical sequences. 4 Drosophila melanogaster sequence is identical to Drosophila pseudoobscura pseudoobscura, Drosophila yakuba, Drosophila rhopaloa, Drosophila persimilis, Drosophila elegans, Drosophila simulans, Drosophila sechellia, Drosophila ficusphila, Drosophila takahashii, Drosophila erecta, Drosophila eugracilis and Drosophila biarmipes sequences. 5 Drosophila virilis sequence is identical to Drosophila grimshawi, Drosophila willistoni, Drosophila mojavensis, Drosophila ananassae, Drosophila kikkawai and Drosophila bipectinata sequences.

287 mRNA and prepropeptide structures

CT/DH mRNA and prepropeptide structures are well conserved across insects. CT/DH transcripts are composed of 4 to 6 exons (Table S2 and Figure S1). In the sequences examined here, exon 1 represents most of the 5’ untranslated region (UTR) and the CT/DH mature peptide is encoded by the last exon, with the exception of Anopheles gambiae where the last two exons

(exons 5 and 6) encode CT/DH (Figure S1). More specifically, the region encoding the CT/DH mature peptide is at the 5’ end of the last exon (3’ end of exon 5 and 5’ end of exon 6 in A. gambiae). The majority of the last exon is the 3’ UTR which can be as large as 937 bp in

Rhodnius prolixus CT/DH-C (Rhopr-CT/DH-C) (Figure S1) (Zandawala et al., 2011). Intron lengths have not been examined as complete genomes are not available for all the species examined here. CT/DH prepropeptides range in size from 103 amino acid residues (aar) in

Bombus terrestris CT/DH-A to 146 aar in Rhopr-CT/DH-C (Figure S2A). CT/DH prepropeptides are predicted to produce four peptides following post-translational proteolytic processing: signal peptide, precursor peptide 1 (PP1), CT/DH and precursor peptide 2 (PP2)

(Figure S2A and S2B). Interestingly, PP1 in Rhopr-CT/DH-B and Rhopr-CT/DH-C are considerably longer than others due to the presence of exon 4 and exons 4 and 5, respectively. It has been previously suggested that Rhopr-CT/DH-B and Rhopr-CT/DH-C are more derived than

Rhopr-CT/DH-A and could have resulted from a possible DNA insertion event in R. prolixus

(Zandawala et al., 2011). Although PP1 is not similar to any other known peptides, its conservation at the C-terminus may suggest a biological role for this peptide. However, attempts to identify this peptide in R. prolixus central nervous system (CNS) extracts via mass spectrometry analysis have been unsuccesful (Zandawala and Orchard, unpublished). The multiple sequence alignment shows that CT/DH and its flanking cleavage sites are very well

288 conserved across all insects (Figure S2A and S2B). Lastly, PP2 is a small non-conserved peptide, ranging from 1 to 10 aar, which most-likely lacks any biological activity.

289

Table S2: Summary of CT/DH transcripts in insects.

Species Variant Accession No. of Exons Exons Reference number exons encoding encoding prepropeptide mature peptide Drosophila A NM_078790 5 3,4 & 5 5 (see Coast, 2006) melanogaster Drosophila C NM_164825 5 3,4 & 5 5 melanogaster Rhodnius A HM030716 4 2,3 & 6 6 (Zandawala et al., prolixus 2011) Rhodnius B HM030715 5 2, 3, 4 & 6 6 (Zandawala et al., prolixus 2011) Rhodnius C HM030714 6 2, 3, 4, 5 & 6 6 (Zandawala et al., prolixus 2011) Anopheles - XM_321755 6 2, 3, 4, 5 & 6 5 & 6 gambiae Bombyx - NM_001130907 5 2, 3, 4 & 5 5 (Roller et al., 2008) mori Acyrthosiphon - XM_001945866 4 2, 3 & 4 4 pisum Nasonia - XM_001599898 4 2, 3 & 4 4 vitripennis Bombus A XM_003395571 4 2, 3 & 4 4 terrestris

290

Figure S1: Molecular organization of insect CT/DHs based on BLAST analysis and gene structure prediction using Webscipio 2.0 (Hatje et al., 2011). CT/DH mRNA sequences from the following insects were used for the analysis: Drosophila melanogaster (variant A -

NM_078790.3), Rhodnius prolixus (variant A - HM030716.1; variant B - HM030715.1; variant

C - HM030714.1), Bombyx mori (NM_001130907.1), Acyrthosiphon pisum (XM_001945866.2),

Nasonia vitripennis (XM_001599898.2), Bombus terrestris (variant A - XM_003395571.1) and

Anopheles gambiae (XM_321755.5). The boxes represent exons which are numbered sequentially. Introns are not included.

291

292

Figure S2: CT/DH prepropeptides in insects. A) Multiple sequence alignment of insect CT/DH prepropeptides: Rhodnius prolixus (variant A - AEA51302.1; variant B - AEA51301.1; variant C

- AEA51300.1), Drosophila melanogaster (variant A - NP_523514.1; variant C - NP_723401.1),

Bombyx mori (NP_001124379.1), Anopheles gambiae (XP_321755.3), Acyrthosiphon pisum (XP

001945901.1), Aedes aegypti (EAT40182.1), Nasonia vitripennis (XP 001599948.1),

Nilaparvata lugens (DB826761), Bombus terrestris (variant A - XP_003395619.1) and

Solenopsis invicta (EFZ22024.1). PP – precursor peptide. The positions of the various peptides are based on their prediction in Rhodnius prolixus (redrawn from Zandawala et al., 2011). B)

Schematic representations (drawn to scale) of R. prolixus and Drosophila melanogaster CT/DH prepropeptides. Arginine and lysine propeptide cleavage sites are indicated by arrows.

293

294

Distribution mRNA distribution

CT/DH transcript distribution in insects has not been studied in great detail. FlyAtlas database

(www.flyatlas.org) reports Drome-CT/DH expression in various tissues of both the larval and adult flies. As evident from the data, Drome-CT/DH is abundantly expressed in the CNS and midgut, which corroborates the presence of Drome-CT/DH-like immunoreactive cells in the

CNS and midgut. In R. prolixus, at least three CT/DH-encoding transcripts are expressed within the CNS (Zandawala et al., 2011). In particular, all three variants are expressed in the brain and various ganglia (Zandawala and Orchard, unpublished). Moreover, all three variants are expressed in an individual R. prolixus with no apparent difference in expression between males and females (Zandawala and Orchard, unpublished). This begs the question as to why three

CT/DH-encoding transcripts are present in R. prolixus CNS considering they all produce the mature CT/DH peptide. Perhaps the answer may lie in any potential roles for PP1 which is variable across the three variants. The lack of mutually-exclusive exons within the three transcripts makes it difficult to determine the cellular localization of the individual transcripts

(see section: mRNA and prepropeptide structures); however, cell-specific localization of Rhopr-

CT/DH (all three transcripts) using fluorescent in situ hybridization (FISH) (Figure S3) reveals a pattern which has good overlap with the immunohistochemical localization of the peptide (Te

Brugge et al., 2005; Zandawala et al., 2011). Immunohistochemical analysis reveals a few more cells than are localized by FISH which might indicate low transcript expression in these cells.

Interestingly, a fourth splice variant (referred to as Rhopr-CT/DH-D from here on) for this gene was recently discovered in ESTs derived from R. prolixus testes (Ons et al., 2011). Rhopr-

CT/DH-D does not encode the CT/DH mature peptide but it may still encode a peptide similar to

Rhopr-CT/DH-C PP1 (see section: mRNA and prepropeptide structures). Additional

295 investigations are needed to determine the presence of Rhopr-CT/DH-D peptide in R. prolixus testes and its physiological role(s), if any.

296

Figure S3: Rhopr-CT/DH expression pattern determined using fluorescent in situ hybridization in fifth instar R. prolixus CNS (redrawn from Zandawala et al., 2011). Putative medial neurosecretory cells (MNSC) and posterior ventral cells (PVC) have been labelled. Scale bar:

200μm.

297

298

Peptide localization

Immunohistochemical analyses to localize CT/DH-like peptides have only been conducted in a handful of species following the isolation of the first peptide from D. punctata brain/corpus cardiacum extracts. As expected, CT/DH-like immunoreactivity is present in the CNS of all the insects in which it has been examined, including Drosophila melanogaster, R. prolixus and

Oncopeltus fasciatus (Park et al., 2008; Te Brugge et al., 2005; Te Brugge and Orchard, 2008).

In D. melanogaster larvae, Drome-CT/DH-like immunoreactivity is present in cells throughout the CNS, except in thoracic segment 1 (Park et al., 2008). Surprisingly, Drome-CT/DH distribution in the adult CNS has not been mapped yet. With regards to the gut, Drome-CT/DH- like immunoreactivity is observed in endocrine cells throughout the larval midgut, whereas

CT/DH-expressing endocrine cells are only found in the last half of the posterior midgut of adult

D. melanogaster (Veenstra, 2009). Within larvae, Drome-CT/DH endocrine cells are more abundant in the anterior midgut and the anterior-middle midgut junction compared to the middle and posterior midgut. Interestingly, some of the tachykinin-expressing endocrine cells in the posterior midgut also express Drome-CT/DH. This Drome-CT/DH distribution pattern in the midgut is similar in both feeding and wandering third instar larvae, but the staining is much reduced in the wandering larvae. It is believed that these CT/DH-expressing endocrine cells are responsible for stimulating fluid secretion by the MTs and the midgut in addition to their role in modulating peristalsis (LaJeunesse et al., 2010; Veenstra, 2009). However, stress caused by starvation and desiccation has been shown to induce the hormonal release of tachykinin, possibly from the endocrine cells in the midgut, which in turn acts on MTs to regulate Drosophila insulin- like peptide 5 signaling (Soderberg et al., 2011). Thus Drome-CT/DH present in these endocrine cells may also be released along with tachykinin during starvation and osmotic stress. This observation is in stark contrast to the role of Drome-CT/DH in fluid secretion.

299

Within the R. prolixus CNS, Rhopr-CT/DH-like immunoreactivity is consistently observed in two lateral neurosecretory cells in the dorsal brain and six cells on the posterior ventral surface of the mesothoracic ganglionic mass (MTGM) (Te Brugge et al., 2005). Interestingly, 5- hydroxytryptamine (serotonin) is co-localized with Rhopr-CT/DH in the 5 dorsal unpaired median (DUM) neurons of the MTGM (Te Brugge et al., 2005). The presence of serotonin in these 5 DUM neurons is unique as all the insect DUM neurons examined so far contain octopamine (see Orchard et al., 1989). These neurons project axons through their corresponding abdominal nerves, where they produce neurohemal sites. Moreover, the staining in DUM neurons is completely abolished in R. prolixus 1 hr after feeding. The neurohemal sites are also much reduced in fed insects compared to unfed. Although serotonin and Rhopr-CT/DH are co- localized, and possibly co-released, they have different actions. Unlike serotonin, Rhopr-CT/DH does not stimulate absorption across the anterior midgut nor does it stimulate MT secretion to the same extent (see section: Biological activity) (Te Brugge et al., 2009; Te Brugge et al., 2002). In addition to the CNS, Rhopr-CT/DH-like immunoreactivity is observed in processes over the hindgut and on salivary glands, specifically the salivary nerve, accessory gland and salivary duct

(Te Brugge et al., 2005). Unlike D. melanogaster, no CT/DH-containing endocrine cells were detected in R. prolixus midgut. Immunohistochemical analysis using O. fasciatus CNS reveals a pattern similar to that observed in R. prolixus (Te Brugge and Orchard, 2008). In particular,

Rhopr-CT/DH-like immunoreactivity is localized to lateral neurosecretory cells in the brain and posterior ventral cells in the MTGM. However, instead of localizing to the DUM neurons,

CT/DH-like immunoreactivity is localized to four pairs of dorsal medial neurosecretory cells in the MTGM of O. fasciatus. Staining is also observed in the processes originating from these cells that extend to the abdominal nerves. Rhopr-CT/DH-like immunoreactivity in the O.

300 fasciatus gut presents an interesting situation. Similar to that seen in R. prolixus, Rhopr-CT/DH- like immunoreactivity is observed in processes over the hindgut (Te Brugge and Orchard, 2008); however, as seen in D. melanogaster (Veenstra, 2009), staining is observed in endocrine-like cells of midgut, specifically the fourth ventricle (Te Brugge and Orchard, 2008). Hence there is a possibility that endocrine-like cells containing CT/DH could be present in R. prolixus but were not detected at the feeding stages examined. Alternatively, this may reflect an inherent difference in peptide expression between herbivorous insects such as D. melanogaster and O. fasciatus and hematophagous insects such as R. prolixus. Additional analyses in other insects are needed to determine the presence of CT/DH-containing endocrine cells in insect midguts.

Biological activity

Diuresis

As mentioned earlier, CT/DH was originally named based on its ability to stimulate MT secretion in D. punctata (Furuya et al., 2000). Since then its role in diuresis has been demonstrated in various other insects including D. melanogaster, A. gambiae and R. prolixus

(Coast et al., 2005; Coast et al., 2001; Te Brugge et al., 2005) (Table S3). However, there are also examples of species such as Acrosternum hilare and Podisus maculiventris in which

CT/DHs do not stimulate diuresis (Coast et al., 2011). Caution must still be exercised in discarding CT/DHs role in diuresis in these insects as only the non-native CT/DHs were tested; the native CT/DHs from these species may still be diuretic. Although CT/DHs and their role as a diuretic factor are somewhat conserved across insects, there are several differences in their underlying mechanisms (Table S3). Firstly, CT/DH employs different secondary messengers in

MTs of different insects; calcium signaling is involved in Locusta migratoria while cAMP is used across various dipterans (Coast et al., 2005; Coast et al., 2001; Furuya et al., 2000). It is still

301 unclear as to what the secondary messengers are in D. punctata or R. prolixus (Furuya et al.,

2000; Te Brugge et al., 2005; Tobe et al., 2005). Secondly, the resulting ion composition of the secreted fluid in MTs stimulated by CT/DHs is dependent upon the species in question. For example, Rhopr-CT/DH has no effect on ion composition of the secreted fluid but mosquito

CT/DHs stimulate increased Na+ secretion (i.e. natriuresis) (Coast et al., 2005; Donini et al.,

2008). Indeed the Aedes aegypti CT/DH was previously referred to as the Mosquito Natriuretic

Peptide (Petzel et al., 1987; Petzel et al., 1985; Petzel et al., 1986). Lastly, the relative potencies of CT/DHs compared to other diuretic factors are also variable. For instance, CT/DHs stimulate maximum secretion rates in mosquitoes but only up to 50% of maximum rates in D. punctata, L. migratoria and D. melanogaster (Coast et al., 2005; Coast et al., 2001; Furuya et al., 2000).

Rhopr-CT/DH is an extreme exception as the secretion rate is only 1.48% of maximum secretion with serotonin. Although this increase in secretion is small relative to that of serotonin, it is still

17-fold higher than basal secretion rates which are minimal in R. prolixus (Te Brugge et al.,

2005).

CT/DH is not the only family of diuretic hormones (DHs) in insects. Other DHs include biogenic amines such as serotonin and tyramine, as well as various peptide families such as the kinins, corticotropin-releasing factor-like DHs (CRF/DHs) and CAP2b-like DHs (see Schooley et al., 2012). The different diuretic factors could work in concert to control MTs and other feeding- related tissues. This may bring about synergism which results in a steeper response at lower concentrations of DHs. Synergism between various diuretic factors has been examined in a few species (Table S4). Not surprisingly, the results are quite variable. For instance, Dippu-CT/DH and Dippu-CRF/DH are synergistic in D. punctata but only additive in L. migratoria (Furuya et al., 2000); however, when Dippu-CRF/DH is replaced with the native Locmi-CRF/DH,

302 synergism is observed. On the other hand, Rhopr-CT/DH and serotonin are only additive in R. prolixus whereas Bommo-CT/DH and Tenebrio molitor CRF/DH are neither synergistic nor additive in T. molitor. Needless to say, synergism between CT/DHs and other diuretic factors needs to be tested in detail before its role as a diuretic can be discarded in any species.

303

Table S3: Summary of CT/DH effects on Malpighian tubule fluid secretion.

Factor Species Diuretic EC50 % of Secondary Cellular effects Reference (nM) maximum messenger (relative to) Dippu- Diploptera Yes 9.8 41% (Dippu- Not cAMP - (Furuya et al., CT/DH punctata CRF/DH) or cGMP 2000; Tobe et al., 2005) Dippu- Locusta Yes 0.56 50% (Dippu- Ca2+ No effect on Na+/K+ (Furuya et al., CT/DH migratoria CRF/DH) ratio of secreted fluid 2000) Drome- Drosophila Yes 4.3 35% cAMP Stimulates an apical (Coast et al., CT/DH melanogaster (Musdo- membrane V-type H+ 2001) Kinin) ATPase in principal cells; equal effect on K+ and Na+ transport across the basolateral membrane Anoga- Anopheles Yes 50 100% cAMP Natriuretic – (Coast et al., CT/DH gambiae selectively stimulates 2005) transepithelial Na+ transport Aedae- Aedes aegypti Yes - 100% cAMP Natriuretic – (Coast et al., CT/DH selectively stimulates 2005) transepithelial Na+ transport Rhopr- Rhodnius Yes <10 1.48% Not cAMP Small lumen-positive (Donini et al., CT/DH prolixus (Serotonin) shift in transepithelial 2008; Te Brugge potential of the upper et al., 2005) tubule; No effect on ion composition of secreted fluid Rhopr- Oncopeltus Yes - - - - (Te Brugge and CT/DH fasciatus Orchard, 2008) Bommo Tenebrio Yes 0.61 50% (cAMP) - - (Holtzhausen and -CT/DH molitor Nicolson, 2007) Anoga- Tenebrio Yes 14 50% (cAMP) - - (Holtzhausen and CT/DH molitor Nicolson, 2007) Dippu- Tenebrio No - 0% - - (Holtzhausen and CT/DH molitor Nicolson, 2007) Dippu- Acrosternum No - 0% - - (Coast et al., CT/DH hilare 2011) Tenmo- Podisus No - 0% - - (Coast et al., CT/DH maculiventris 2011)

304

Table S4: Studies examining synergistic or additive effects between CT/DHs and other diuretic factors on Malpighian tubule secretion (modified from Holtzhausen and Nicolson, 2007).

Factor 1 Factor 2 Species Combined effect Reference Dippu-CT/DH Dippu-CRF/DH Diploptera punctata Synergistic (Furuya et al., 2000) Dippu-CT/DH Dippu-CRF/DH Locusta migratoria Additive (Furuya et al., 2000) Dippu-CT/DH Locmi-CRF/DH Locusta migratoria Synergistic (Furuya et al., 2000) Dippu-CT/DH Locmi-Kinin Locusta migratoria Synergistic (Furuya et al., 2000) Drome-CT/DH Musdo-Kinin Drosophila melanogaster Additive (Coast et al., 2001) Rhopr-CT/DH Serotonin (5-HT) Rhodnius prolixus Additive (Te Brugge et al., 2005) Bommo-CT/DH Tenmo-CRF/DH Tenebrio molitor Not synergistic or (Holtzhausen and additive Nicolson, 2007)

305

Feeding-related physiological events

CT/DHs could also play an important role in feeding-related physiological events in various insects. In R. prolixus, CT/DH causes an increase in frequency of salivary gland muscle contractions (see Orchard, 2009). This may aid in the mixing of salivary gland contents and to propel the saliva out of the principal gland during feeding. Moreover, Rhopr-CT/DH causes an increase in frequency of anterior midgut, hindgut and dorsal vessel contractions, with cAMP possibly acting as the secondary messenger (Te Brugge et al., 2009; Te Brugge et al., 2008).

Interestingly, Rhopr-CT/DH and Rhopr-CRF/DH both increase anterior midgut contractions, but only Rhopr-CRF/DH increases water absorption from the anterior midgut (Orchard, personal communication; Te Brugge et al., 2009; Te Brugge et al., 2011). Thus, Rhopr-CT/DH may play an indirect role in diuresis by increasing haemolymph circulation and mixing of the gut contents, which aids in reducing unstirred layers and facilitating ion transport. Moreover a peptide from this family was also isolated and partially sequenced from the Belgian forest ant, Formica polyctena, before the identification of Dippu-CT/DH in 2000 (see Schooley et al., 2012).

Interestingly, the peptide was isolated based on its ability to stimulate the spontaneous writhing movements of L. migratoria MTs, which could indirectly contribute to diuresis. In D. melanogaster, CT/DH produced by enteroendocrine cells is required for peristalsis in the junction region between the anterior portion and acidic region of the larval midgut (LaJeunesse et al., 2010). Hence CT/DH may regulate digestion in D. melanogaster. On the other hand,

Dippu-CT/DH linear and cyclic analogs have been shown to modulate appetitive behavior in food-deprived nymphs of L. migratoria (Kaskani et al., 2012). It acts as an anorexigenic by reducing the duration of the first meal and increasing the latency to feed.

Other effects

306

Recent work in M. sexta suggests that CT/DH may also be involved in ecdysis (Kim et al.,

2006). A subset of ecdysis-triggering hormone receptor (ETHR)-expressing cells in the abdominal ganglia produce CT/DH. However, their precise physiological role during ecdysis has not yet been examined. Perhaps CT/DH stimulates ecdysis-associated diuresis in M. sexta as such a diuresis has been demonstrated using crude tissue extracts in another lepidopteran, Pieris brassica (Nicolson, 1976). On a separate note, CT/DH may also be involved in modulating the neuronal clock network in D. melanogaster. Drome-CT/DH activates the D. melanogaster pigment-dispersing factor (PDF) receptor (PDFR) in vitro and also causes a large increase in cAMP levels in all PDF-expressing clock neurons (Mertens et al., 2005; Shafer et al., 2008).

Receptors

CT/DH receptor in D. melanogaster

Insect CT/DH receptors belong to the family of secretin-like (family B) G-protein coupled receptors (GPCRs) (Hewes and Taghert, 2001). The first insect CT/DH receptor was functionally characterized in D. melanogaster (CG17415/CG32843) (Johnson et al., 2005). Mammalian CT receptor-like receptor (CLR) signaling is dependent on two types of accessory proteins: receptor activity modifying proteins (RAMPs) and receptor component protein (RCP) (Evans et al., 2000;

McLatchie et al., 1998). Likewise, the Drome-CT/DH receptor is activated by Drome-CT/DH

(EC50 = 5219 nM) when co-expressed in HEK293 cells with the D. melanogaster RCP.

However, an improved sensitivity is observed when the human RCP (hRCP) was co-expressed instead (EC50 = 116 nM) and this sensitivity is further improved when the receptor is co- expressed with hRCP and human RAMPs (EC50 = 82 nM). Unlike mammals, where RCP expression mirrors CLR expression, FlyAtlas data indicates that the D. melanogaster RCP is expressed in almost all tissues which suggests that it may also interact with other receptors 307

(Chintapalli et al., 2007; Ma et al., 2003). Nonetheless, this is the most compelling evidence for the functional conservation between signaling from mammalian CT receptor-like receptors and insect CT/DH receptor.

Immunohistochemical analysis shows that the Drome-CT/DH receptor is expressed in principal cells of D. melanogaster MTs (Johnson et al., 2005). FlyAtlas data indicates that the Drome-

CT/DH receptor transcript is also highly expressed in the adult CNS, crop, midgut, hindgut, and heart (Chintapalli et al., 2007). The relative expression of Drome-CT/DH receptor in the larval

MTs is only about 3% compared to that in adults. Nonetheless, Drome-CT/DH receptor expression in these tissues is consistent with the biological function of this hormone in controlling water and salt homeostasis in D. melanogaster and other insects. In addition, Drome-

CT/DH and Drome-CRF/DH receptors are expressed in neurons that also express the neuropeptide corazonin (Johnson et al., 2005). Corazonin transcript expression is inhibited during starvation and osmotic stress (Harbison et al., 2005; Zhao et al., 2010) and it has been suggested that this inhibition may be caused by signaling through Drome-CT/DH and/or Drome-

CRF/DH (Zhao et al., 2010). Assuming that Drome-CT/DH levels are elevated following feeding and/or during over hydration, it may then cause increased corazonin expression and its subsequent release from the neurons. Drome-CT/DH and corazonin might then increase heart contractions (transcripts for their respective receptors are highly expressed in the heart) to increase circulation of other hormones, including diuretic hormones, that may be present in the haemolymph (Chintapalli et al., 2007). Hence CT/DHs may also have an indirect effect on water and salt balance.

CT/DH receptor in other insects

308

CT/DH receptors have also been previously predicted in A. gambiae (Hill et al., 2002), Tribolium castaneum (Tc 70) (Hauser et al., 2008) and Apis mellifera (Am 55) (Hauser et al., 2006) through genome-wide annotation of neuropeptide GPCRs. Phylogenetic analysis of family B GPCRs

(Figure S4) reveals that CT/DH receptors form a monophyletic clade. This clade is sister to another monophyletic clade that contains receptors orthologous to the D. melanogaster CG4395

GPCR (see section: Receptors). Interestingly, the Acyrthosiphon pisum genome contains two putative CT/DH receptors (receptor 1 [XP_001943752.2] and receptor 2 [XP_001945278.2]), unlike other insects examined which all have one CT/DH receptor. It is unclear whether both the receptors are true CT/DH receptors. This clarification must await functional characterization of these receptors.

309

Figure S4: Phylogenetic tree of insect GPCRs, belonging to family B. The maximum parsimonious tree was constructed using Close-Neighbor-Interchange (CNI) analysis and the bootstrap values obtained were based on 1000 replicates. The receptors are labelled using the associated GenBank accession number, five letter species code and, where available, gene id.

Accession number has not been provided for Rhodnius prolixus CT/DH-R, however, it has been cloned and sequenced (Zandawala and Orchard, unpublished). The tree is rooted using the D. melanogaster metabotropic glutamate receptor (CG11144).

310

311

Classification of family B receptors

There has been a lack of consensus on the classification of family B GPCRs and their ligands

(Hill et al., 2002; Lovejoy et al., 2006; Price et al., 2004). However, recent studies on D. melanogaster family B GPCRs have helped clarify this. The D. melanogaster genome contains 5 peptide GPCRs (CG4395, CG17415, CG13758, CG8422 and CG12370) that belong to family B

(Hewes and Taghert, 2001). Out of these 5 receptors, CG4395 and CG17415 (also known as

CG32843) are most closely related to CT receptor and CT gene related peptide receptor of vertebrates. CG17415 has been functionally characterized as the CT/DH receptor, as mentioned earlier (Johnson et al., 2005). CG4395, which is expressed in a subset of fruitless neurons, is critical for male courtship behavior (Li et al., 2011). FlyAtlas database also reports CG4395 transcript expression in the brain and male accessory glands, which is consistent with this role

(Chintapalli et al., 2007). Three independent reports have characterized CG13758 as the PDFR

(Hyun et al., 2005; Lear et al., 2005; Mertens et al., 2005). Interestingly, despite the lack of sequence similarity between Drome-CT/DH and Drome-PDF, Drome-CT/DH activates the

Drome-PDFR (EC50 = 218.6 nM) when functional ligand-receptor interactions are analyzed in

HEK293 cells (Mertens et al., 2005). However, the receptor activation by 10-6M Drome-CT/DH is less than 40% compared to the response by an equivalent dose of Drome-PDF. In contrast,

Drome-CT/DH receptor is not sensitive to Drome-PDF . This phenomenon is not unique as partial agonism by diverse ligands is a common feature among family B GPCRs (Hay et al.,

2004). Furthermore, de novo predictions of Drome-PDF and Drome-CT/DH tertiary structures show that both these peptides form a helical region with the Drome-PDF helix being shorter than that of Drome-CT/DH (Maupetit et al., 2009). The relationship between a specific helix length and bioactivity has been examined previously using salmon CT where bioactivity was considerably reduced when the helix length was altered (Andreotti et al., 2006). Assuming

312 similar structural requirements for the Drome-CT/DH receptor activation and assuming that the

Drome-PDFR requires a ligand with a helical domain for activation, then this might explain the activation of Drome-PDFR by Drome-CT/DH and the absence of effect of Drome-PDF on

Drome-CT/DH receptor. Clearly, additional work is needed to validate these speculations.

CG8422, on the other hand, encodes a functional CRF/DH receptor (Drome-CRF/DH-R1)

(Johnson et al., 2004). Hence despite the fact that the DH receptors isolated from the tobacco hornworm M. sexta (Reagan, 1994) and the house cricket Acheta domesticus (Reagan, 1996) were originally classified broadly as members of the CT/secretin/CRF receptor family, these have now been determined to be receptors for CRF/DHs (Figure S4) (Johnson et al., 2004).

Likewise, the DH receptor cloned from the rice brown planthopper Nilaparvata lugens is a

CRF/DH receptor (Price et al., 2004). Lastly, CG12370 (isoform A), which is paralogous to

CG8422, has also been functionally characterized as a CRF/DH receptor (Drome-CRF/DH-R2)

(Hector et al., 2009). Phylogenetic analysis groups this receptor along with other CRF/DH receptors (Figure S4), which suggests possible gene duplication. Interestingly, FlyAtlas reports high expression of CG12370 transcript in the crop, midgut, MTs and hindgut but a lack of

CG8422 transcript in these tissues (Chintapalli et al., 2007) which are associated with diuresis.

The fact that Drome-CRF/DH-R1 is expressed in corazonin-expressing neurons, it is safe to assume that Drome-CRF/DH-R1 regulates processes centrally (e.g. corazonin secretion) whereas

Drome-CRF/DH-R2 regulates peripheral physiological processes (e.g. diuresis).

Conclusion and future directions

Diuresis in insects is a process that is tightly regulated by various factors, one of which is the

CT/DH. These factors work in concert to fine-tune the fluid secretion by MTs as well as the contractility of the insect’s gut, which can also indirectly aid in stimulating fluid secretion (see

313

Zandawala et al., 2012). They do so by reducing the unstirred layers around the MTs and by increasing the circulation of other diuretic hormones that may be present in the haemolymph.

Like several other insect neuropeptides, CT/DHs are pleiotropic in nature. Hence in addition to their effect on peripheral tissues, they may also affect processes centrally including the regulation of corazonin transcript expression. Moreover, the presence of CT/DH in the CNS and gut of D. melanogaster and O. fasciatus suggests that this family of peptides might be a new class of brain-gut peptides. It is safe to conclude that CT/DHs are well conserved across insects, both structurally and functionally.

314

References

Chintapalli, V.R., Wang, J. and Dow, J.A. (2007). Using FlyAtlas to identify better Drosophila melanogaster models of human disease. Nat Genet. 39, 715-20.

Christie, A.E. (2008). Neuropeptide discovery in Ixodoidea: an in silico investigation using publicly accessible expressed sequence tags. Gen Comp Endocr. 157, 174-85.

Christie, A.E., Durkin, C.S., Hartline, N., Ohno, P. and Lenz, P.H. (2010a). Bioinformatic analyses of the publicly accessible crustacean expressed sequence tags (ESTs) reveal numerous novel neuropeptide-encoding precursor proteins, including ones from members of several little studied taxa. Gen Comp Endocr. 167, 164-78.

Christie, A.E., Nolan, D.H., Garcia, Z.A., McCoole, M.D., Harmon, S.M., Congdon-Jones, B., Ohno, P., Hartline, N., Congdon, C.B., Baer, K.N. and Lenz, P.H. (2011). Bioinformatic prediction of arthropod/nematode-like peptides in non-arthropod, non-nematode members of the Ecdysozoa. Gen Comp Endocr. 170, 480-6.

Christie, A.E., Stevens, J.S., Bowers, M.R., Chapline, M.C., Jensen, D.A., Schegg, K.M., Goldwaser, J., Kwiatkowski, M.A., Pleasant, T.K., Jr., Shoenfeld, L., Tempest, L.K., Williams, C.R., Wiwatpanit, T., Smith, C.M., Beale, K.M., Towle, D.W., Schooley, D.A. and Dickinson, P.S. (2010b). Identification of a calcitonin-like diuretic hormone that functions as an intrinsic modulator of the American lobster, Homarus americanus, cardiac neuromuscular system. J Exp Biol. 213, 118-27.

Coast, G.M., 2006. Chapter 24 - Insect Diuretic and Antidiuretic Hormones, in: Abba, J.K. (Ed.), Handbook of Biologically Active Peptides. Academic Press, Burlington, pp. 157-162.

315

Donini, A., O'Donnell, M.J. and Orchard, I. (2008). Differential actions of diuretic factors on the Malpighian tubules of Rhodnius prolixus. J Exp Biol. 211, 42-8.

Gruber, C.W. and Muttenthaler, M. (2012). Discovery of defense- and neuropeptides in social ants by genome-mining. PLoS One. 7, e32559.

316

Harbison, S.T., Chang, S., Kamdar, K.P. and Mackay, T.F. (2005). Quantitative genomics of starvation stress resistance in Drosophila. Genome Biol. 6, R36.

Hatje, K., Keller, O., Hammesfahr, B., Pillmann, H., Waack, S. and Kollmar, M. (2011). Cross-species protein sequence and gene structure prediction with fine-tuned Webscipio 2.0 and Scipio. BMC Res Notes. 4, 265.

Hauser, F., Cazzamali, G., Williamson, M., Blenau, W. and Grimmelikhuijzen, C.J. (2006). A review of neurohormone GPCRs present in the fruitfly Drosophila melanogaster and the honey bee Apis mellifera. Prog Neurobiol. 80, 1-19.

Hauser, F., Cazzamali, G., Williamson, M., Park, Y., Li, B., Tanaka, Y., Predel, R., Neupert, S., Schachtner, J., Verleyen, P. and Grimmelikhuijzen, C.J. (2008). A genome- wide inventory of neurohormone GPCRs in the red flour beetle Tribolium castaneum. Front Neuroendocrin. 29, 142-65.

Hauser, F., Neupert, S., Williamson, M., Predel, R., Tanaka, Y. and Grimmelikhuijzen, C.J. (2010). Genomics and peptidomics of neuropeptides and protein hormones present in the parasitic wasp Nasonia vitripennis. J Proteome Res. 9, 5296-310.

Hay, D.L., Conner, A.C., Howitt, S.G., Smith, D.M. and Poyner, D.R. (2004). The pharmacology of adrenomedullin receptors and their relationship to CGRP receptors. J Mol Neurosci. 22, 105-13.

Hector, C.E., Bretz, C.A., Zhao, Y. and Johnson, E.C. (2009). Functional differences between two CRF-related diuretic hormone receptors in Drosophila. J Exp Biol. 212, 3142-7.

Hewes, R.S. and Taghert, P.H. (2001). Neuropeptides and neuropeptide receptors in the Drosophila melanogaster genome. Genome Res. 11, 1126-42.

317

Hill, C.A., Fox, A.N., Pitts, R.J., Kent, L.B., Tan, P.L., Chrystal, M.A., Cravchik, A., Collins, F.H., Robertson, H.M. and Zwiebel, L.J. (2002). G protein-coupled receptors in Anopheles gambiae. Science. 298, 176-8.

Holtzhausen, W.D. and Nicolson, S.W. (2007). Beetle diuretic peptides: the response of mealworm (Tenebrio molitor) Malpighian tubules to synthetic peptides, and cross-reactivity studies with a dung beetle (Onthophagus gazella). J Insect Physiol. 53, 361-9.

Hyun, S., Lee, Y., Hong, S.T., Bang, S., Paik, D., Kang, J., Shin, J., Lee, J., Jeon, K., Hwang, S., Bae, E. and Kim, J. (2005). Drosophila GPCR Han is a receptor for the circadian clock neuropeptide PDF. Neuron. 48, 267-78.

Johnson, E.C., Bohn, L.M. and Taghert, P.H. (2004). Drosophila CG8422 encodes a functional diuretic hormone receptor. J Exp Biol. 207, 743-8.

Kim, Y.J., Zitnan, D., Cho, K.H., Schooley, D.A., Mizoguchi, A. and Adams, M.E. (2006). Central peptidergic ensembles associated with organization of an innate behavior. Proc Natl Acad Sci USA. 103, 14211-6.

Lear, B.C., Merrill, C.E., Lin, J.M., Schroeder, A., Zhang, L. and Allada, R. (2005). A G protein-coupled receptor, groom-of-PDF, is required for PDF neuron action in circadian behavior. Neuron. 48, 221-7.

318

Lovejoy, D.A., Al Chawaf, A. and Cadinouche, M.Z. (2006). Teneurin C-terminal associated peptides: an enigmatic family of neuropeptides with structural similarity to the corticotropin- releasing factor and calcitonin families of peptides. Gen Comp Endocr. 148, 299-305.

Ma, W., Chabot, J.G., Powell, K.J., Jhamandas, K., Dickerson, I.M. and Quirion, R. (2003). Localization and modulation of calcitonin gene-related peptide-receptor component protein- immunoreactive cells in the rat central and peripheral nervous systems. Neuroscience. 120, 677- 94.

Maupetit, J., Derreumaux, P. and Tuffery, P. (2009). PEP-FOLD: an online resource for de novo peptide structure prediction. Nucleic Acids Res. 37, W498-503.

Mertens, I., Vandingenen, A., Johnson, E.C., Shafer, O.T., Li, W., Trigg, J.S., De Loof, A., Schoofs, L. and Taghert, P.H. (2005). PDF receptor signaling in Drosophila contributes to both circadian and geotactic behaviors. Neuron. 48, 213-9.

Nicolson, S.W. (1976). The hormonal control of diuresis in the cabbage white butterfly Pieris brassicae. J Exp Biol. 65, 565-75.

Ons, S., Sterkel, M., Diambra, L., Urlaub, H. and Rivera-Pomar, R. (2011). Neuropeptide precursor gene discovery in the Chagas disease vector Rhodnius prolixus. Insect Mol Biol. 20, 29-44.

319

Orchard, I. (2009). Peptides and serotonin control feeding-related events in Rhodnius prolixus. Front Biosci. 1, 250-62.

Park, D., Veenstra, J.A., Park, J.H. and Taghert, P.H. (2008). Mapping peptidergic cells in Drosophila: where DIMM fits in. PLoS One. 3, e1896.

Petzel, D.H., Berg, M.M. and Beyenbach, K.W. (1987). Hormone-controlled cAMP-mediated fluid secretion in yellow-fever mosquito. Am J Physiol. 253, R701-11.

Petzel, D.H., Hagedorn, H.H. and Beyenbach, K.W. (1985). Preliminary isolation of mosquito natriuretic factor. Am J Physiol. 249, R379-86.

Petzel, D.H., Hagedorn, H.H. and Beyenbach, K.W. (1986). Peptide nature of two mosquito natriuretic factors. Am J Physiol. 250, R328-32.

Price, D.R., Du, J., Dinsmore, A. and Gatehouse, J.A. (2004). Molecular cloning and immunolocalization of a diuretic hormone receptor in rice brown planthopper (Nilaparvata lugens). Insect Mol Biol. 13, 469-80.

Reagan, J.D. (1994). Expression cloning of an insect diuretic hormone receptor. A member of the calcitonin/secretin receptor family. J Biol Chem. 269, 9-12.

Reagan, J.D. (1996). Molecular cloning and function expression of a diuretic hormone receptor from the house cricket, Acheta domesticus. Insect Biochem Mol Biol. 26, 1-6.

Roller, L., Yamanaka, N., Watanabe, K., Daubnerova, I., Zitnan, D., Kataoka, H. and Tanaka, Y. (2008). The unique evolution of neuropeptide genes in the silkworm Bombyx mori. Insect Biochem Mol Biol. 38, 1147-57.

Schooley, D.A., Horodyski, F.M. and Coast, G.M., 2005. 3.10 - Hormones Controlling Homeostasis in Insects, in: Editors-in-Chief: Lawrence, I.G. et al. (Eds.), Comprehensive Molecular Insect Science. Elsevier, Amsterdam, pp. 493-550.

320

Schooley, D.A., Horodyski, F.M. and Coast, G.M., 2012. 9 - Hormones Controlling Homeostasis in Insects, in: Lawrence, I.G. (Ed.), Insect Endocrinology. Academic Press, San Diego, pp. 366-429.

Shafer, O.T., Kim, D.J., Dunbar-Yaffe, R., Nikolaev, V.O., Lohse, M.J. and Taghert, P.H. (2008). Widespread receptivity to neuropeptide PDF throughout the neuronal circadian clock network of Drosophila revealed by real-time cyclic AMP imaging. Neuron. 58, 223-37.

Soderberg, J.A., Birse, R.T. and Nassel, D.R. (2011). Insulin production and signaling in renal tubules of Drosophila is under control of tachykinin-related peptide and regulates stress resistance. PLoS One. 6, e19866.

Te Brugge, V., Ianowski, J.P. and Orchard, I. (2009). Biological activity of diuretic factors on the anterior midgut of the blood-feeding bug, Rhodnius prolixus. Gen Comp Endocr. 162, 105- 12.

Te Brugge, V., Paluzzi, J.P., Schooley, D.A. and Orchard, I. (2011). Identification of the elusive peptidergic diuretic hormone in the blood-feeding bug Rhodnius prolixus: a CRF-related peptide. J Exp Biol. 214, 371-81.

Te Brugge, V.A., Lombardi, V.C., Schooley, D.A. and Orchard, I. (2005). Presence and activity of a Dippu-DH31-like peptide in the blood-feeding bug, Rhodnius prolixus. Peptides. 26, 29-42.

Te Brugge, V.A. and Orchard, I. (2008). Distribution and activity of a Dippu DH31-like peptide in the large milkweed bug Oncopeltus fasciatus. Peptides. 29, 206-13.

Te Brugge, V.A., Schooley, D.A. and Orchard, I. (2002). The biological activity of diuretic factors in Rhodnius prolixus. Peptides. 23, 671-81.

321

Te Brugge, V.A., Schooley, D.A. and Orchard, I. (2008). Amino acid sequence and biological activity of a calcitonin-like diuretic hormone (DH31) from Rhodnius prolixus. J Exp Biol. 211, 382-90.

Tobe, S.S., Zhang, J.R., Schooley, D.A. and Coast, G.M. (2005). A study of signal transduction for the two diuretic peptides of Diploptera punctata. Peptides. 26, 89-98.

Veenstra, J.A. (2009). Peptidergic paracrine and endocrine cells in the midgut of the fruit fly maggot. Cell Tissue Res. 336, 309-23.

Zandawala, M., Paluzzi, J.P. and Orchard, I. (2011). Isolation and characterization of the cDNA encoding DH(31) in the kissing bug, Rhodnius prolixus. Mol Cell Endocrinol. 331, 79-88.

Zhao, Y., Bretz, C.A., Hawksworth, S.A., Hirsh, J. and Johnson, E.C. (2010). Corazonin neurons function in sexually dimorphic circuitry that shape behavioral responses to stress in Drosophila. PLoS One. 5, e9141.

322

Acknowledgements

I would like to thank my advisor, Dr. Ian Orchard, for providing valuable input and support throughout the writing of this review. I am also thankful to Dr. Jean-Paul Paluzzi for his insightful comments on an earlier version of this manuscript. The unpublished work described here was supported by the Natural Sciences and Engineering Research Council of Canada.

323