Chapter I: Introduction and General Background

The most exciting phrase to hear in science, the one that heralds the most discoveries, is not Eureka! but ‘That’s funny…’

(Isaac Asimov)

Promoter: Prof. dr. ir. Guy Smagghe Laboratory of Agrozoology Department of Crop protection Faculty of Bioscience Engineering Ghent University

Chair of the examination committee:

Prof. Dr. Els J.M. Van Damme Department of Molecular Biotechnology Faculty of Bioscience Engineering Ghent University

Members of the examination committee:

Prof. Dr. ir. Peter Bossier Department of Animal Production Faculty of Bioscience Engineering Ghent University

Prof. Dr. Dirk de Graaf Department of Physiology Faculty of Sciences Ghent University

Prof. Dr. Jozef Vanden Broeck Department of Biology Faculty of Science Catholic University of Leuven

Prof. Dr. Luc Swevers Institute of Biology National Centre for Scientific Research “Demokritos”, Athens, Greece

Dean: Prof. dr. ir. Guido Van Huylenbroeck

Rector: Prof. dr. Paul Van Cauwenberge

Key role and diversity of EcR/USP and other nuclear receptors in selected Arthropoda species

ir. Olivier Christiaens

Thesis submitted in fulfillment of the requirements for the degree of Doctor (PhD) in Applied Biological Sciences

Dutch translation of the title: Sleutelrol en diversiteit van EcR/USP en andere nucleaire receptoren in geselecteerde geleedpotigen

Front cover: (Above) Compilation of: a schematic picture of chromosomes in a cell unwinding into a DNA double helix strand (right-under), the canonical 1-dimensional structure of nuclear receptors (right-above), the holo- and apo-structure of a steroid receptor ligand binding domain (left-under) and the structure of 20-hydroxy-ecdysone (left-above). (Below) A banner of pictures (from left to right) from Acyrthosiphon pisum (picture by Alex Wild), Tetranychus urticae (picture by Eric Erbe, USDA) and Bombus terrestris (www.wikipedia.com).

Please refer to this work as follows: Christiaens, O. (2013). Key role and diversity of EcR/USP and other nuclear receptors in selected Arthropoda species. Ghent University, Ghent, Belgium.

ISBN: 978-90-5989-618-5

This study was financially supported by a PhD grant from the Bijzonder Onderzoeksfonds (BOF)

The author and the promoters give the permission to use this thesis for consulation and to copy parts of it for personal use. Every other use is subject to the copyright laws, more specifically the source must be extensively specified when using results from this thesis.

Table of contents

List of abbreviations

Objectives and outline of this study

Chapter I: Introduction and general background ...... - 1 - 1. General introduction on arthropod phylogeny ...... - 2 - 1.1. Hexapoda ...... - 5 - 1.2. Chelicerata ...... - 7 - 2. Nuclear receptors ...... - 7 - 2.1. Definition ...... - 7 - 2.2. Structure and working mechanism of nuclear receptors ...... - 8 - 2.3. Diversity and evolution of nuclear receptors ...... - 14 - 2.4. Overview and function of the different NRs in arthropods ...... - 22 - 3. Ecdysone signaling pathway ...... - 32 - 3.1. Insect growth, moulting and metamorphosis ...... - 32 - 3.2. Ecdysone biosynthesis ...... - 34 - 3.3. Ecdysone signaling cascade ...... - 34 - 3.4. The ecdysone receptor as a pest control target ...... - 37 - 4. Arthropods used in this thesis ...... - 38 - 4.1. Acyrthosiphon pisum ...... - 38 - 4.2. Bombus terrestris ...... - 41 - 4.3. Tetranychus urticae ...... - 44 - 5. RNAi ...... - 48 - 5.1. Molecular mechanism ...... - 48 - 5.2. RNAi efficiency and efficacy ...... - 49 - 5.3. RNAi as a possible pest control tool ...... - 50 - 5.4. DsRNA delivery ...... - 52 - 5.5. RNAi in Hemiptera and aphids ...... - 55 -

Chapter II: Diversity of nuclear receptors in the animal kingdom . - 57 - 1. Introduction...... - 59 - 2. Material and methods ...... - 60 - 2.1. Acyrthosiphon pisum ...... - 60 -

2.2. Bombus terrestris and Bombus impatiens ...... - 62 - 2.3. Tetranychus urticae ...... - 62 - 2.4. Phylogenetic analysis ...... - 64 - 3. Results and Discussion ...... - 64 - 3.1. Nuclear receptors in Acyrthosiphon pisum ...... - 64 - 3.2. Nuclear receptors in Bombus ...... - 72 - 3.3. Nuclear receptors in Tetranychus urticae ...... - 78 - 4. Conclusion ...... - 86 -

Chapter III: The ecdysone receptor heterodimer ...... - 91 - 1. Introduction...... - 93 - 2. Material and Methods ...... - 93 - 2.1. Sequence analysis ...... - 93 - 2.2. Phylogenetic analysis ...... - 93 - 2.3. 3D-modeling and ligand docking of the ligand binding pockets of ApEcR and TuEcR ...... - 94 - 3. Results and Discussion ...... - 95 - 3.1. EcR ...... - 95 - 3.2. RXR/Ultraspiracle ...... - 104 - 4. Conclusions ...... - 112 -

Chapter IV: Ecdysone biosynthesis in Acyrthosiphon pisum and

Tetranychus urticae ...... - 113 - 1. Introduction...... - 115 - 2. Material and methods ...... - 116 - 2.1. Annotation of Halloween genes ...... - 116 - 2.2. Confirmation of expression ...... - 117 - 2.3. Identification of PonA as a major ecdysteroid in T. urticae ...... - 117 - 3. Results and Discussion ...... - 118 - 3.1. Halloween genes in Acyrthosiphon pisum...... - 118 - 3.2. Ecdysteroid biosynthesis in T. urticae ...... - 121 - 4. Conclusion ...... - 124 -

Chapter V: RNAi ...... - 125 - 1. Introduction...... - 126 - 2. Development of a microinjection protocol ...... - 126 -

2.1. Microinjection setup ...... - 126 - 2.2. Sedation and immobilization ...... - 127 - 2.3. Microcapillaries ...... - 128 - 2.4. Needle insertion ...... - 129 - 2.5. Volume ...... - 129 - 2.6. Toxicity of injected water, buffer and dye ...... - 129 - 3. RNAi experiments ...... - 130 - 3.1. Material and Methods ...... - 130 - 3.2. Results and discussion ...... - 135 - 4. RNAi efficiency in the pea aphid ...... - 140 - 4.1. Material and Methods ...... - 140 - 4.2. Results and Discussion ...... - 142 - 5. Conclusion ...... - 151 -

Chapter VI: General Conclusions and future perspectives ...... - 155 - 1. Nuclear receptors: diversity and conservation in Arthropoda ...... - 156 - 2. Ponasterone A is the main ecdysteroid in T. urticae ...... - 157 - 3. The pea aphid exhibits a low efficiency for RNAi ...... - 158 - 4. RNAi as a possible pest control strategy ...... - 161 -

References………...... - 163 -

Summary………...... - 187 -

Samenvatting…… ...... - 191 -

Curriculum vitae ...... - 195 -

Dankwoord……...... - 201 -

III

List of abbreviations

2,22,25dE 2,22,25-trideoxyecdysone 2,22dE 2,22-dideoxyecdysone 20E 20-hydroxy-ecdysone 2dE 2-deoxyecdysone 7dC 7-dehydrocholesterol AA Amino acid Ala Alanine Arg Arginine Asn Asparagine Asp Aspartic acid Bt Bacillus thuringiensis cDNA Complementary DNA CYP Cytochrome P450 Cys Cysteine DBD DNA-binding domain DNA Deoxyribonucleic acid dsRNA Double stranded RNA E Ecdysone EcR Ecdysone receptor EDTA Ethylenediaminetetraacetic acid EST Expressed sequence tags GFP Green fluorescent protein Gln Glutamine Glu Glutamic acid His Histidine HPLC High pressure liquid chromatography HSP Heat shock protein Ile Isoleucine LBD ligand-binding domain LBP Ligand-binding pocket Leu Leucine Met Methionine miRNA Micro RNA NLS Nuclear localization signal NR Nuclear receptor PCR Polymerase chain reaction Phe Phenylalanine PNR Photoreceptor cell-specific nuclear receptor PonA Ponasterone A PTU Phenylthiourea qPCR Quantitative polymerase chain reaction RISC RNA-induced silencing complex RNA Ribonucleic acid

RNAi RNA interference RT-PCR Reverse transcription polymerase chain reaction RXR Retinoid X receptor Ser Serine siRNA Small interfering RNA Thr Threonine Trp Tryptophan Tyr Tyrosine USP Ultraspiracle Val Valine VIGS Virus-induced gene silencing viRNA Viral RNA

Preface

Objectives and outline of this study

VII

Our understanding of arthropods in general and insects in particular is deep, but still narrow. Some model organisms such as the fruitfly Drosophila melanogaster, the red flour beetle Tribolium castaneum or the silkmoth Bombyx mori are thoroughly described and their genome is sequenced and well-annotated. However, all these model insects that had their genome sequenced thus far were holometabolic. On the side of the hemimetabolic insects and also other non-insect arthropods, our knowledge is still limited. This also counts for the nuclear receptors (NRs). In fact, when we started this research project four years ago, the available knowledge on nuclear receptors outside the Holometabola was very fragmentary and no full non-holometabolic set of NRs was described. The main goal of this thesis was to identify and characterize the nuclear receptors outside of holometabolic insects and subsequently see if any remarkable differences between holometabolic insects and other arthropods are to be discovered.

Three organisms have been chosen to investigate the diversity of nuclear receptors throughout the Arthropoda phylum, namely the hemimetabolous pea aphid (Acyrthosiphon pisum), the holometabolic bumblebee (Bombus terrestris) and the chelicerate spider mite (Tetranychus urticae). These three species represent a wide scope in the Arthropoda phylum and allow a comparison between the different sets of nuclear receptors in a chelicerate organism, a hemimetabolous insect and a holometabolous insect.

In Chapter 1, a general introduction on nuclear receptors will be presented; their structure, working mechanism, characteristics, evolution and their link to the signaling cascade of the ecdysteroids, known as the moulting hormones, will be elaborated on. Additionally, an introduction on RNA interference (RNAi), a functional genomics tool used further on in this thesis, will be presented as well.

Chapter 2 highlights the paradox that nuclear receptors in arthropods exhibit great conservation, as well as some remarkable diversity. The full sets of nuclear receptors from the aforementioned species will be presented and differences between the different species and taxonomical groups which were discovered, will be discussed.

Special interest in this thesis will go towards the ecdysteroid signaling pathway. Centrally in this pathway is the binding of the major insect moulting hormone 20-hydroxyecdysone (20E) with the functional ecdysone receptor complex, formed by the nuclear receptors EcR and VIII

Usp/RXR. This heterodimer and its components will be discussed in Chapter 3. The EcR and RXR/USP of the pea aphid and the two-spotted spider mite will be examined more closely. Also, modeling and docking studies, where the 3D-structure of the protein as well as the possible interaction with different ligands are predicted in silico, will be presented. Structural differences between the insect and chelicerate EcR may then reflect the presence of the different moulting hormones.

In Chapter 4, we will investigate what happens upstream, namely the 20E biosynthesis. A group of enzymes, called the Halloween genes, which are responsible for the conversion of cholesterol to the functional 20E will be investigated.

In a final part, the objective was to investigate the functionality of nuclear receptors, especially EcR and USP. RNA interference (RNAi) was chosen as an ideal technique to elucidate functional information regarding the nuclear receptor genes that were found in the pea aphid. Our first goal was to set up a microinjection and feeding protocol for dsRNA- delivery to aphids. The second goal was to attempt silencing of EcR and USP and examine the phenotypic effects. Despite intensive attempts, RNAi experiments did not provide the expected results and RNAi proved to be inefficient in aphids. Therefore, we investigated possible reasons behind these observations that the RNAi mechanism did not seem to be as robust as for example in T. castaneum and Apis mellifera. The results of these experiments towards a functional RNAi experimental setup are presented in Chapter 5.

Chapter I: Introduction and general background

Chapter I

1. General introduction on arthropod phylogeny

Arthropods are invertebrates that are characterized by a bilateral symmetry, a segmented body, jointed appendages and an exoskeleton, providing both structure and protection. This phylum is the largest and most diverse taxonomical group in the animal kingdom. It is also a very successful group of species. They have been around for over 500 billions of years, are still evolving, show a great diversity, are found in almost all possible environments and are inhabiting our planet in huge numbers. Both the ecological and economic importance of arthropods cannot be underestimated. They play a huge role in agriculture, be it as pest or as a beneficial insect, and are also known carriers of some devastating diseases.

Taxonomically, the Arthropoda are part of the larger Superphylum Ecdysozoa, which is the group of Protostomia that includes all animals which undergo ecdysis during their development (Fig. 1.1). The arthropods are generally divided into five subphyla;

 The extinct Trilobitomorpha  the Chelicerata including all spiders, scorpions and mites  the Myriapoda containing millipedes and centipedes  the Crustacea containing all shrimp, crab and lobster species  the Hexapoda which consists mainly out of insects

Despite the fact that a lot of genomic and evolutionary data are available, there is still some debate about the relationships between these different arthropod groups. During the past few decades, a lot of attention has gone towards these phylogenetic relationships. A clear consensus however has not been met. One of the first successes in this field was the observation that Hexapoda are more closely related to the aquatic Crustacea than to the terrestrial Myriapoda (Friedrich & Tautz, 1995; Boore et al., 1998), confirming the hypothesis of the Pancrustacea group (Hexapoda + Crustacea).

- 2 -

Introduction and general background

Figure 1.1 General phylogeny of Animales highlighting the Ecdysozoa clade to which the Arthropoda belong (after Sanetra et al., 2005). CNS: central nervous system; PDA: protostome-deuterostome ancestor.

The position of the Myriapoda however was still not so clear. Are they more closely related to the Pancrustacea, forming a Mandibulata clade, as was long thought based on morphological observations? Or do they have a common ancestor with the Chelicerata and form the Paradoxopoda clade, as is suggested by some molecular research (Friedrich & Tautz, 1995; Hwang et al., 2001; Mallatt et al., 2004)? Figure 1.2 gives a representation of both theories. For a long time, these observations and conclusions about phylogenetic relationships were based on morphological criteria and later also sequence data obtained from limited taxa and genes. In recent years however, high throughput analyses have enabled us to get a more complete and correct view on these relationships. Regier et al. (2010) have used sequence

- 3 -

Chapter I information from 62 single-copy nuclear protein-coding genes from 75 arthropod species in order to get a clearer view on the molecular phylogeny of Arthropda. Their results strongly supported both the Pancrustacea and the Mandibulata (Myriapoda plus Pancrustacea) hypotheses.

Figure 1.2 Phylogenetic trees visualizing the Paradoxopoda (left) and the Mandibulata (right) hypotheses (Regier et al., 2010).

Another point of debate in arthropod phylogeny was the relationship within the Pancrustacea group, between the Hexapoda and Crustacea. Both were long considered to be monophyletic sister groups, but recent evidence also supported the idea that the insects could belong to a subset of some Crustacea (Carapelli et al., 2007; Cook et al., 2005; Regier et al., 2005, 2010). Based on mitochondrial genome studies, Carapelli et al. (2007) also places the Collembola outside of the clade formed by the other hexapods and considered them as another Crustacea clade. Thus, at this time the arthropod phylogeny is far from clear and a consensus has not been met. However, the volume of data used in the research by Regier et al. (2010) makes that their conclusions have a solid base of evidence and that they could provide a clear framework for the phylogeny in this group (Fig. 1.3).

- 4 -

Introduction and general background

Figure 1.3 Phylogenetic tree showing the full Arthropoda phylogeny, based on Regier et al. (2010). The major taxonomical groups are indicated on the right. Arthropoa, Mandibulata and Pancrustacea nodes are indicated on the tree.

1.1. Hexapoda

This group of six-legged arthropods is generally considered to contain four taxonomical groups, namely the Insecta, Diplura, Protura, Collembola. These last three groups of ametabolous hexapods were all once considered insects, but are now considered separate orders within the class of Entognatha (Fig. 1.3). Even though this class was traditionally - 5 -

Chapter I created based on the morphological feature of internal mouth parts, recent molecular evidence seems to confirm this hypothesis (Regier et al., 2010).

Besides the very small group of wingless and ametabolous insects belonging to the Entognatha, the Hexapoda can generally be divided into two main groups based on their growth and development. The holometabolous insects or Endopterygota undergo a large metamorphosis during the pupal stage resulting in adults that have very large morphological differences compared to the larval stages. However, in hemimetabolous insects, known as Exopterygota, the adult stage organisms have a very similar morphology as their juvenile stages, called nymphs. The only main difference is the formation of wings in the adult stage. Figure 1.4 gives an overview of some of the most important orders belonging to both superorders. In this thesis, the hemipteran aphid Acyrthosiphon pisum (Hemimetabola) and the hymenopteran bumblebee Bombus terrestris (Holometabola) will feature prominently as examples of species from each of these two superorders and the phylogeny of the Hemiptera and Hymenoptera orders is shown in more detail.

Figure 1.4 Simplified phylogenetic tree showing the relationship between the most important insect orders. The phylogenetic position of the three organisms used in this thesis is shown. The Holometabola and Hemimetabola groups are indicated on the right.

- 6 -

Introduction and general background

1.2. Chelicerata

Another important arthropod group is the Chelicerata. They are a large and old group generally divided into three classes. The Pycnogonida contain the sea spiders, the Xiphosura are the horseshoe crabs but the most important class is that of the Arachnida, a large and diverse group of which the spiders (order Araneae), scorpions (order Scorpiones) and the subclass of Acari, containing ticks and mites, are the most well-known taxonomic groups. The Acari are subdivided into two main superorders, the Acariformes (containing most herbivorous mites) and the Parasitiformes (containing ticks). Tetranychus urticae was chosen as a model organism to characterize the nuclear receptors in chelicerates and thus the clade of Acari is indicated in the phylogenetic tree (Fig. 1.5).

Figure 1.5 Simplified phylogenetic tree of the Chelicerata clade, showing the most important and largest taxonomical groups belonging to this subphylum. The Acari clade, containing Acariformes and Parasitiformes, is indicated on the tree itself.

2. Nuclear receptors

2.1. Definition

The nuclear receptor superfamily is a group of transcription factors which are found in all metazoans or Animalia. Until now, no members of this protein superfamily have been identified in plants, fungi, or unicellular eukaryotes, despite the many genomes that have been sequenced. They are involved in a vast array of biological processes such as molting and

- 7 -

Chapter I metamorphosis (Ashburner, 1973), embryonic development (King-Jones and Thummel, 2005), cell differentiation (Siaussat et al., 2007), reproduction (Raikhel et al., 1999), and are therefore also considered as important novel targets in pest insect control (Palli et al., 2005; Billas et al., 2009). Transcriptional regulation of many of these nuclear receptors is dependent on binding of receptor ligands, which are usually small, lipophilic compounds such as steroids, retinoids, thyroid hormones and fatty acids. These nuclear receptors therefore form a direct link between signaling molecules that control these processes and the transcriptional responses that need to follow. In addition, there is a large group of NRs, the so called orphan receptors for which no ligand has been identified. In some cases, the NR is simply incapable of binding with a ligand, due to the absence of a ligand binding pocket (LBP). Examples are the Drosophila HR38 and Tribolium USP. These are called true orphans. Some other nuclear receptors have been speculated as being ligand independent mediators of gene transcription, but time will tell if this hypothesis remains standing (Bunce and Campbell, 2010).

2.2. Structure and working mechanism of nuclear receptors

2.2.1. The modular structure of NRs

The NRs have a typical structure consisting of five or six regions which have a modular character, that is, they are composed of distinct domains, each responsible for one or more specific molecular functions (Fig. 1.6).

Figure 1.6 Schematic overview of the NR modular structure including the different functional domains.

The A/B-domain The N-terminal A/B-domain harbors a transcriptional activation function (AF-1) and can interact with other transcriptional factors. Distinction between the A- and B-domain is often not evident and this is the region where most of the difference between isoforms of NRs is

- 8 -

Introduction and general background found. This region is poorly conserved, is extremely variable in length and its structure is largely unkown.

The C-domain or DNA-binding domain (DBD)

The highly conserved C-domain, also known as the DNA-binding domain, contains two typical non-equivalent C4 zinc finger structures. These structures are responsible for the sequence-specific DNA recognition and binding to the hormone receptor response element (HRE) on the DNA.

The D-domain or hinge region The variable D-domain acts as a hinge between the DBD and the E-domain, which is the ligand-binding domain (LBD). This region contains the nuclear localization signal (NLS) which may overlap partly with the C-domain.

The E-domain or ligand-binding domain (LBD) In sequence, this LBD is less conserved compared to the DBD but still more than the other domains. The tertiary structure of the LBD however is more strongly conserved. This LBD harbors the second transcriptional activation function (AF-2), a second NLS and is important for dimerization and recruitment of co-regulators.

The F-domain In some cases, for example the ecdysone receptor (EcR) in Drosophila melanogaster, these NRs also contain an F-domain at the carboxy-terminal end but its function is unknown. Just like the A/B and D regions, this domain is poorly conserved and its structure has not been elucidated yet.

Exceptions Some unusual NRs that are missing either a DBD or an LBD have been characterized as well. In vertebrates, there are two NRs that are missing a DBD (DAX1 and SHP; NR0B1, 2) and in insects, three NRs have been found that do not possess a LBD (Knirps, Knirps-like and EGON/Eagle; NR0A1, 2, 3). In the case of the former NRs, It is hypothesized that they perform their function by dimerization with another NR, that is then responsible for DNA binding. In the case of the NRs lacking a LBD, activation will most likely be due to

- 9 -

Chapter I dimerization as well, but these NRs do have the ability to recognize and bind to genomic DNA and execute their function as a transcription factor.

2.2.2. Working mechanism

Classical signal transduction can be seen as a multistep procedure where a signaling molecule binds to a receptor, usually in the cell membrane, which is followed by a cascade of reactions in the cytoplasm eventually leading to the physiological response. This response can happen in the form of gene regulation or a change in metabolism. However, NRs bypass this “second messenger” signaling system and act more directly, due to their unique structure. Indeed, NRs contain both a ligand binding domain and a DNA-binding domain, which means they can effectively act as a receptor and also cause the immediate physiological response due to their ability to regulate gene expression. The general working mechanisms behind both sides of the NR will briefly be discussed. Note that most of our understanding in this field comes from research on vertebrate NRs which are very important in drug research. Research on the working mechanism in insect NRs is thus far limited to the ecdysone receptor (EcR), the Retinoid X Receptor/Ultraspiracle (RXR/USP) and a few other NRs which are involved in the ecdysteroid-cascade.

Ligand binding NRs are intracellular proteins and their ligands are small lipophilic molecules which have the ability to enter the cell through passive diffusion. Binding of these signaling molecules to the receptor happens in the LBD region of the protein. The crystal structure has been determined for many vertebrate NRs, both liganded (holo-form) and unliganded (apo-form), helping us to understand the mechanisms involved in ligand binding. The first X-ray crystallography results for NRs were published in the mid-1990s (Bourguet et al., 1995; Renaud et al., 1995; Wagner et al., 1995; Brzozowski et al., 1997) and since then the number of NRs to have their crystal structure elucidated, both in apo- and holo-form rapidly grew. Most of these were vertebrate receptors. However, during the last decade the crystal structure of some insect NRs has also been published, including the Heliothis, Drosophila and Tribolium USP (Billas et al., 2001; Clayton et al., 2001, Iwema et al., 2007) and the Heliothis EcR (Billas et al., 2003).

- 10 -

Introduction and general background

Figure 1.7 Schematic drawing representing the three-dimensional structure of the LBD (a) The unliganded (apo) retinoid X receptor (RXR) LBD with the open ligand binding pocket. (b) The liganded (holo) retinoic acid receptor (RAR) LBD. α-Helices are represented by the cilinders while β-sheets are represented by broad arrows (Bourguet et al., 2000).

These crystal structures show a LBD, generally made up of 12 α-helices (numbered H1-H12) associated with a short hairpin of two antiparallel β-strands, creating a ligand binding pocket (LBP). There can be some variation though. For example, the human RARγ has no H2 (Renaud et al., 1995) and of the 21 known D. melanogaster NRs, two do not possess the 12 α- helices (King-Jones and Thummel, 2005). Based on the structure of the apo- and holo- conformations, the mechanism of ligand binding itself has often been described as a mousetrap-like mechanism. Upon ligand binding, some helices rearrange themselves and H12 will fold over the ligand-binding cavity, creating a sort of lid on the LBP. This H12 mobility explains the lack of entry site into the LBP that is to be seen in crystal structures of liganded receptors (Moras and Gronemeyer, 1998). The 3D-structure of the LBP both in holo- and apo- form is represented in Figure 1.7. This schematic drawing shows the positional changes of helices H11 and H12 after binding of the ligand.

Dimerization and DNA binding Most NRs function as dimers, either as homodimers or as heterodimers, often partnered with RXR (USP in insects). Both the DBD and the LBD contribute to the dimerization interface of the receptor. In the former, one of the zinc fingers stabilizes the dimer and in the latter, helices - 11 -

Chapter I

9 and 10 are the most important regions for dimerization but some residues on H7 and H11 are also involved, as well as parts of loop L8-9 and L9-10 (Bourguet et al., 2000; Gampe et al., 2000).

Figure 1.8 Schematic drawing of the DBD of the Estrogen receptor (ER) showing the two zinc fingers and the location of the P- and D-box upon binding of the dimer to the HRE. Amino acids in blue can vary depending on the NR, as shown in the table (Vanden Heuvel, 2009).

Two important elements in the DBD are the two zinc-fingers (Fig. 1.8). These are small protein domains forming folds that are created and stabilized due to the binding of specific amino acids to a zinc atom. Each zinc atom coordinates with four cysteine residues. The first zinc finger contains the proximal- or P-box which is an α-helix responsible for the recognition of the highly conserved ‘core half-site’ of the response element. The second contains the distal- or D-box and is mainly responsible for the stability of the NR complex and dimerization. Both the P- and D-box are highly conserved.

The HREs to which the receptors bind have a common hexanucleotide consensus sequence originally described as AGGTCA, but variations do occur. Specificity of DNA binding between different NRs is achieved in part through variation in the nucleotide sequence 5’ to the HRE as well as in the number of nucleotides between half-sites. Most steroid receptors for example recognize a slightly different consensus AGAACA sequence (Bain et al., 2007). The response elements for these dimeric NRs are organized as direct repeats

- 12 -

Introduction and general background

(AGGTCA…AGGTCA), inverted repeats (AGGTCA…ACTGGA) or everted repeats (ACTGGA…AGGTCA) of these hexanucleotide ‘half-sites’, each separated by a variable number of non-consensus nucleotides, somewhere between 0 and 10. In the case of monomers, such as FTZ-F1, HR3 or E75, the NR can bind to DNA through a single core sequence. In this case these flanking nucleotides are also critical.

Co-repressors, co-activators and gene transcription activation The NRs can generally be divided into three different groups based on their working mechanism and ligand type: the steroid receptor family (type I), the thyroid/retinoid receptor family (type II) and the orphan receptor family (type III). These three groups distinguish themselves in the way they bind to DNA, their localization in the cell and their dimerization as well (Fig. 1.9).

Figure 1.9 Overview of the working mechanisms of Type I and Type II NRs. (a). Steroid receptors (Type I) bind their ligands in the cytosol and upon binding, migrate to the cell nucleus where they will bind the DNA as homodimers and eventually activate transcription. (b). Thyroid/retinoid receptors (Type II) are always present in the nucleus and are always bound to the DNA. In unliganded form, they act as transcription inhibitors due to the action of the corepressors. Upon binding with the ligand, the corepressors are replaced by coactivators and transcription is activated (Vanden Heuvel, 2009).

- 13 -

Chapter I

Type I NRs are typically located in the cytosol in a protein complex which includes several proteins (hsp70, hsp90, FKBP52/51 and others). Upon binding with a ligand, the NR homodimer becomes free from the complex and migrates to the nucleus. The receptors belonging to the type II group on the other hand are always located in the nucleus and typically act as heterodimers, often partnered with the RXR. The remaining NRs, all orphan receptors, are grouped as Type III NRs. These receptors can either bind the HRE as a heterodimer, usually partnering RXR, or as monomers. The NRs contain two activation factors (AF). AF-1 is situated in the unconserved A/B domain, just before the DBD and the other one, AF-2 is located in the LBD. Both elements are capable of recruiting coactivators and activating transcription of the target gene. Not much is known about the weakly conserved AF-1, its working mechanism and function. In contrast, the ligand binding- dependent AF-2 is well described, being situated in the LBD.

Upon binding of a ligand, the conformational changes in the LBD give rise to a hydrophobic groove which is formed by several helices of the LBD, including helix 12 which is indispensable for the formation of this coactivator binding surface. This hydrophobic groove is then capable of recruiting proteins called coactivators. Most of these coactivators that contain helical LXXLL motifs (where L is leucine and X is any amino acid), such as those from the p160 family, can then bind with the groove through hydrophobic interactions. In the absence of a ligand, H12 is positioned away from the LBD core structure and the hydrophobic groove is thus unable to take form. These coregulators often appear to function in large protein complexes and it is suggested that the transcriptional activity of the NRs involves both the sequential and the combinatorial cooperation of multiple of these coregulators (Glass and Rosenfeld, 2000).

2.3. Diversity and evolution of nuclear receptors

2.3.1. Introduction

Until the 1990s, when first the human genome and the fruit fly’s genome were sequenced, information on NRs remained scarce, especially for insects. Research towards NRs in humans had started as early as the late 1950s, when Edward Jensen managed to isolate one of the estrogen receptors (Jensen & Jacobson, 1960), but our knowledge on this protein family remained fragmentary until the 80s and 90s, when the importance of NRs in for example drug

- 14 -

Introduction and general background research became apparent. Recent genome projects of both vertebrate, nematode, arthropod and some more basal species contributed a lot in identifying the different NRs and providing us with a number of complete sets of nuclear receptors, allowing us to have an overview of the diversity of this protein superfamily throughout the animal kingdom. In total, 48 NRs are known in humans (Robinson-Rechavi, 2001), 49 in mouse, 47 in rat (Zhang et al., 2004), 68 in the pufferfish (Maglich et al., 2003) and over 284 are present in Caenorhabditis elegans (Gissendanner, 2004). In insects on the other hand, the number of NRs found is significantly lower compared to vertebrates and nematodes. In Drosophila, only 21 NR genes have been identified, while 20 were found in Anopheles, 22 in Apis, 19 in Bombyx and 21 in Tribolium (Adams et al., 2000; Holt et al. 2002; Velarde et al., 2006; Bonneton et al., 2008; Cheng et al., 2008).

2.3.2. Nomenclature and classification of nuclear Receptors

After about two decades of NR research, it became apparent that there was a need for a clear nomenclature and classification system. Until then, NRs were given names by the researchers describing them and in many cases different names for the same NRs were prevalent. In order to organize this growing diversity of NRs, the nuclear receptors Nomenclature Committee was brought to life and, together with the International Union of Pharmacology Committee on Receptor Nomenclature and Drug Classification (NC-IUPHAR), proposed a clear system in 1999, similar to the one that is used for the cytochrome P450 superfamily as well. Studies of NRs in both mammals and arthropods revealed seven distinct subfamilies (NR0-NR6) in which they could be classified, mainly based on phylogenetic data (Fig. 1.10) (Laudet, 1997; Nuclear Receptors Nomenclature Committee, 1999; Aranda & Pascual, 2001). Each of these subfamilies is a monophyletic group and all receptors belonging to one subfamily have a single common ancestor. The proposed names start with the capital letters ‘NR’ followed by the subfamily numeral. NRs that group together within a subfamily are then further identified using a capital letter followed by Arabic numerals for the individual receptors.

What follows is a short overview of the different subfamilies, their characteristics and a brief summary of some important and well described NRs. A full overview of all subfamilies and their vertebrate NRs can be found on the website of the nuclear receptor Research (http://nrresource.org/pathways-2.html), together with a plethora of information about their

- 15 -

Chapter I function and the pathways in which they are involved. The Arthropoda NRs will be discussed into more detail later.

Figure 1.10 Summarized phylogenetic tree of the NR superfamily, containing all human, Drosophila and a number of nematode NRs. The NR0 subfamily has been left out of this tree. On the right, gene groups with official gene nomenclature (Nuclear Receptors Nomenclature Committee 1999) are listed. Open circles indicate genes that are inferred to have existed in the common Bilaterian ancestor. Hatched circles represent alternative ancestral genes. The broken lines leading to EcR, UNC-55, RevErb, NHR67 and the coral TLL/DSF indicate the lack of significant resolution of phylogenetic methods to position these genes (adapted from Bertrand et al., 2004). - 16 -

Introduction and general background

Subfamily NR1 This subfamily is the largest and contains many of the Type I NRs, such as the thyroid receptor (TR) and retinoic acid receptor (RAR), peroxisome proliferator-activated receptor (PPAR), Rev-Erb as well as the Vitamin D receptor (VDR). Most members of this subfamily form heterodimers with RXR and bind to direct repeats of the core motif.

Subfamily NR2 All members of subfamily 2 are able to form homodimers on direct repeat sequences. Members of this group include Hepatocyte Nuclear Factor 4 (HNF4), the tailless homologue TLX and RXR, which is an important heterodimerisation partner for many NRs.

Subfamily NR3 The Type I progesterone receptor (PR), androgen receptor (AR) and estrogen receptor (ER) are part of subfamily 3. This group clusters members of the steroid receptor group that are able to dimerize on palindromic elements as homodimers.

Subfamily NR4, NR5 and NR6 These small subfamilies contain nerve growth factor IB (NGFIB), nuclear receptor related protein 1 (NURR1), neuron-derived orphan receptor 1 (NOR1) (all NR4), liver receptor homolog 1 (LRH1), steroidogenic factor 1 (SF1) (both NR5) and the germ cell nuclear factor (GCNF1) from subfamily 6.

Subfamily NR0 This subfamily contains the NRs that have an exceptional structure. SHP and DAX1 which do not have a DBD are the only vertebrate members of this subfamily.

2.3.3. Evolution of nuclear receptors

2.3.3.1. Diversity through gene losses and duplications

The growing amount of sequenced genomes from species occupying key positions in the metazoan clade makes it possible to find out when and where gene families originated and how they differentiated. Bertrand et al. (2004) found by inference based on the available sequence data at the time that the Urbilateria, common ancestor of chordates, nematodes and

- 17 -

Chapter I arthropods, must have possessed around 25 nuclear receptors. Recently, sequence data from species belonging to ancient metazoan branches has become available that allows us to look a bit further back (Fig. 1.11).

Metazoa Eumetazoa

(/) Q) Bi lateria ...... (/) ro (/) Q) Q) Q) (/) 0> 0> (/) Q) E ro c c (/) ...... 0 4= 0 ro c E (/) 0 0. 0 ro 0 0 (/) · ~ ...... c N (/) Q) ro 0 0 ro 0 ..... 0 E ü :-2 ...... :::! ..c Q) ro c ...... 0 Q) ü 0 0.... ü 0.... 0

SF-1/ Steroid HR39 GCNF receptors/ ERR TR2/TR4 FXR/ LXR/EcR VDRJPXRJCAR/HR96 ThR NR1/NR4/ RAR INR* Rev-Erb/E75 E78 ROR/HR3 PPAR NR4s

Stercid receptors/ERR** SF-1/GCNF RXR INR2c-f

HNF4 Origin of ancestral NR

Figure 1.11 Overview of the origin of NRs and their evolution within the Metazoa clade. Green bars: gene duplication; Red bars: gene losses. The names next to the coloured boxes represent the NRs or NR groups they generated in those lineages. The NR1/NR4/INR ortholog lost in the placozoans (marked *) was generated in the duplication marked ** (Bridgham et al., 2010).

- 18 -

Introduction and general background

The genome from the cnidarian Nematostella vectensis for example contains 17 NRs and that of the placozoan Trichoplax adhaerens contains no more than 4 NRs (Bridgham et al., 2010; Reitzel and Tarrant, 2009). However, the demosponge Amphimedon queenslandica, belonging to the Porifera, is part of an even more ancient branch of the metazoan group. This sponge seems to possess only 2 NRs. Bridgham et al. (2010) included these early metazoan NRs in their phylogenetic analysis, together with no less than 275 NR sequences of species strategically located in the metazoan tree of life. Rooting the phylogeny of the NR superfamily between these two NRs, they found that AqNR2 is orthologous to the fatty acid binding HNF4 family, which is present in all metazoan clades. Their phylogeny indicates that the last common ancestor of all Metazoa had two NRs, one orthologous to HNF4 and one NR that eventually gave rise to all other known NRs through a series of gene duplications.

The last common ancestor of all bilateria probably contained about 22-25 receptors (Bertrand et al., 2004). This is close to the number of NRs found in arthropods. However, arthropods should not be seen as a kind of primitive state within the Bilateria when it concerns NRs. Between this common ancestor and the Protostomia/Deuterostomia branch, a whole history of gene loss and gene duplication took place and many changes have occurred after this separation as well. Figure 1.12 gives a summary of these diversification events inside the Bilateria clade. Branches where a high degree of gene duplication is seen are the vertebrate branch, the fish branch and the time before the diversification into the major bilaterian clades (Bertrand et al., 2004). We know that most gene duplications in the vertebrate and fish branches is due to the genome duplication events (Escriva et al., 2003, Bertrand et al., 2007). Incidences of intense gene loss can be situated in the sea squirt and nematode lineages, as well as a parallel gene loss event in the chordate and ecdysozoan lineages. A special case is that of the apparent gene duplication in nematodes. Of the 280+ known NRs in C. elegans, about 250 belong to the NR2A group (Robinson-Rechavi et al., 2005). However, at the same time of this lineage-specific duplication event, many different NRs were lost as well in the nematode branch.

- 19 -

Chapter I

Figure 1.12 Overview of gene loss and gene duplication/diversification in Bilateria based on the Ecdysozoa phylogeny (Bertrand et al., 2004). Branch lengths are arbitrary. Above each species, the total number of NRs is noted. SRs: Steroid receptors. Gene losses are indicated in blue, gene duplication events are indicated in green. Note the intense gene duplication in the vertebrate and fish branches, coinciding with the events of genome duplication.

- 20 -

Introduction and general background

2.3.3.2. Evolution of ligand binding

Evolution of NRs seems to be connected to DNA-binding and dimerization characteristics. For example, the receptors that can interact/dimerize with RXR in vertebrates seem to cluster together in subfamilies 1 and 4. This could imply that this function is not shared by the whole family (Brelivet et al., 2004). Another example is subfamily 3, containing the NRs that are able to bind on palindromic elements. In contrast, no link with ligand binding has been detected so far. In the phylogenetic tree of the NR superfamily, receptors with the same ligands do not group together but are interspersed with receptors that bind completely different type of molecules. Steroid receptors for example are separated in multiple subfamilies (1 and 3). Moreover, receptors that are closely related can have completely different ligands as is the case in the NR1 subfamily where NR1A binds thyroid hormones, NR1B binds retinoids and NR1C has prostaglandids as ligand. This led to the hypothesis that NRs have evolved from an ancestral orphan receptor and that the ligand binding capacity of many receptors has been acquired several times during their evolution independently (Laudet, 1997; Escriva et al., 2000).

Bridgham et al. (2010) however found that NRs probably evolved from a ligand-activated receptor near the base of the Metazoa, with fatty acids as a possible ancestral ligand. Their robust phylogeny, thanks to functional data on NRs in basal metazoan lineages, allowed them to infer a model for the ancestral NR (AncNR). They proposed a NR that has the ability to activate transcription and be activated by binding a ligand. The fact that no ligand- independent activators are found so far in these basal lineages such as the sponges and cnidarians seems to support this idea. This model would of course implicate that the ligand- independent activity that is seen in some NRs has been derived repeatedly over different clades and for multiple receptors.

In the same publication, an attempt was also made to try and reconstruct the AncNR’s LBD structure (Bridgham et al., 2010). They found strong support for the fact that the ancestor’s LBD possessed a classical NR fold consisting of three layers of helices in highly conserved positions, that it possessed an open ligand binding pocket and a surface for binding coactivator proteins. The speculation about the possible ligand is a bit more difficult, but several of the most basal NRs bind fatty acids and the necessary hydrogen bond between the

- 21 -

Chapter I

FA’s carboxyl-group oxygen and the Arg side chain on H5 is conserved in several basal lineages and receptors, including HNF4s, RXRs and AqNR1 (Bridgham et al., 2010).

2.4. Overview and function of the different NRs in arthropods

2.4.1. The ecdysone receptor

The functional ecdysone receptor is a heterodimer consisting of EcR (NR1H1) and RXR/USP (NR2B4) (Yao et al., 1993). Of all arthropod NRs, these two are by far the best known and have been studied intensively. EcR seems to be present in all arthropods and recently some functional orthologues have been discovered in a nematode as well (Shea et al., 2010). A whole list of known and sequenced EcR or USP proteins can be found in Henrich, 2012.

EcR (NR1H1) The arthropod EcR is distantly related to the vertebrate farnesol X receptor (FXR; Forman et al., 1995) and liver X receptor (LXR; Willy et al., 1995) and first described in Drosophila. The main ligand for this NR in most insects is 20E, but it can also bind numerous other ecdysteroids and even nonsteroidal compounds used in pest management (Yao et al., 1992, 1993; Thomas et al., 1993). In decapod Crustacea for example, the main ecdysteroids regulating the different ecdysteroid-dependent processes are 20-hydroxyecdysone (20E) as well as ponasterone A (Pon A, 25-Deoxy-20-hydroxyecdysone) (Chung, 2010). Crystal structures of EcR bound to PonA have been described for H. virescens (Billas et al., 2003), B. tabaci (Carmichael et al., 2005) and T. castaneum (Iwema et al., 2007) and in the case of HvEcR, the crystal structure bound to 20E (Browning et al., 2007) and to a nonsteroidal agonist (Billas et al., 2003) was published as well. The EcR LBD is very flexible and can adapt the shape of its ligand binding pocket to the ligand that is bound to it, whether this is an ecdysteroid or a nonsteroidal compound (Billas et al., 2003). The gene coding for EcR has two different promoters in most insects, coding for two different isoforms EcR-A and EcR-B. This can vary however. In Drosophila for example, three different isoforms have been found; A, B1 and B2. The two B-isoforms are a result of alternative splicing.

RXR/USP (NR2B4) USP, just like its vertebrate homolog RXR seems to act as a heterodimerization partner for many other NRs. Until now, no ligand for USP has been found. Research into the evolution of

- 22 -

Introduction and general background the USP LBD and EcR/USP dimerization interface showed that the divergence rate of this LBD was high in Mecopterida, the major insect terminal group, containing the Diptera and Lepidoptera (Iwema et al., 2009). An increase in evolutionary rate for some NRs, including USP, was already reported in these lineages (Bonneton et al., 2008). They have proposed a classification of USP into three categories: (1) the mecopteridan USP which underwent considerable evolutionary divergence and seems to possess a large LBP (Carmichael et al., 2005; Iwema et al., 2007); (2) the coleopteran and hymenopteran USP which underwent a slower divergence and has no LBP; (3) the USP of more basal insects, which possesses an LBP and is reported to be able to bind the RXR-ligand 9-cis retinoic acid (Nowickyj et al., 2008). These shape modifications in the Mecopterida USP LBD also proved to significantly alter the heterodimerization surface and also modifiy the structure of the functional EcR/USP complex. These changes could also explain why the non-Mecopterida USP and EcR are both able to form homodimers while the Mecopterida USP and EcR are unable to do so (Lezzi et al., 2002; Graham et al., 2007; Iwema et al., 2007, Minakuchi et al., 2007; Henrich, 2012). Earlier research had also showed that Mecopterida EcR needs USP for stabilization and solubilization (Li et al., 1997; Henrich et al., 2012), contrary to the non-Mecopterida EcR. The dimerization between those receptors has co-evolved and seems to have reinforced the need for heterodimerization between these two NRs (Iwema, 2009).

There might not be a bona fide ligand yet for the Mecopterida USP, but some research suggests methyl farnesoate (MF) could be a candidate. Experiments in Drosophila showed some affinity of the receptor for MF (Jones et al., 2006). Furthermore, Wang and Leblanc (2009) showed that the RXR of the crustacean Daphnia magna when complexed with EcR is responsive to MF. Interestingly, in the USP LBP crystal structures of Drosophila and Heliothis, a phospholipid was discovered (Billas et al., 2001, 2003; Clayton et al., 2001). One hypothesis to explain this is that this phospholipid could be a structural cofactor for USP, in a similar was as fatty acids are for HNF-4 (Benoit et al., 2004). Iwema et al. (2009) also showed that this phospholipid is essential for the special structure of Mecopterida USP LBD.

EcR-RXR/USP heterodimer Insect EcR proteins can also form functional heterodimers with several USP and RXR proteins from other species. The chimaeric dimer constructed with Drosophila EcR and the mammalian RXR becomes responsive to muristerone A (murA), but not 20E. This suggests that USP plays a role in the specificity of EcR ligand binding in this dimer (Christopherson et - 23 -

Chapter I al., 1992; Yao et al., 1993). Even though no direct evidence for this influence is available, it is known that for example heterologous pairings of EcR with USP from different species can significantly change the levels of transcriptional activity (Beatty et al., 2009).

Figure 1.13 A schematic diagram showing the points of interactions between EcR and USP in three species from different orders (Hemiptera: Bemisia tabaci; Coleoptera: Tribolium castaneum; Lepidoptera: Heliothis virescens). The hydrophobic core is presented as dashed lines, conserved electrostatic and polar interactions as black lines. The electrostatic and polar interactions that are specific to one species are in red, whereas Heliothis/Bemisia-specific interactions are in blue. (Iwema et al., 2009).

The natural response elements for the ecdysone receptor that have been discovered so far were asymmetric elements composed of either palindromic inverted or direct repeats. In vitro work has shown that the receptor can also bind on symmetric response elements. One of these natural response elements is a palindromic inverted nucleotide repeat sequence situated in the ecdysteroid-inducible promoter of the 27 kDa heat shock protein (hsp27) in D. melanogaster (Riddihough and Pelham, 1987). However, recent research revealed that EcR and USP co- localize at multiple early puff sites as well as hundreds of other sites (Gauhar et al., 2009).

- 24 -

Introduction and general background

2.4.2. Subfamily 1

This group contains a few core NRs of the ecdysone signaling cascade, including EcR, E75, E78 and HR3. Together with the NR2 member USP/RXR, these are by far the most intensively studied NRs in arthropods. The fifth member of this subfamily is HR96. Three important groups of vertebrate subfamily 1 NRs have been lost in ecdyzoans, namely the PPAR-, RAR- and thyroid hormone receptor group (NR1C, NR1B and NR1A respectively).

E75 (NR1D3) E75 is a transcriptional repressor, present in all insect genomes which have been sequenced up until now and a homologue of the vertebrate REVERB-α and REVERB-β NRs, which are involved in the pacemaker controlling the circadian clock in vertebrates. However, a similar role for E75 in insects has not been reported yet. It is known that in Drosophila and Bombyx, the ecdysone-inducible E75 acts as a repressor of HR3, another member of the ecdysone pathway (White et al., 1997; Swevers et al., 2002a), probably through direct interaction. E75 also seems to act as a link in the ecdysone and juvenile hormone (JH) pathways (Dubrovsky et al., 2004). Other processes E75 is thought to be involved in are embryogenesis, metamorphosis, oogenesis and vitellogenesis (Bonneton & Laudet, 2012) but the complex role of this gene is not fully understood yet. A remarkable recent discovery is the presence of a heme prosthetic group in Drosophila E75 and the fact that E75 is regulated by the binding of gases (NO and CO) (Reinking et al., 2005; de Rosny et al., 2006). Binding of CO by the receptor seems to interfere with the ability of E75 to interact with HR3. Also interesting to mention is that some E75 isoforms do not present the typical domain structure. Drosophila E75B and E75D for example lack one and two zinc-fingers in the DBD respectively, making them incapable of binding to DNA (Jiang et al., 2000).

E78 (NR1E1) The NR1E gene and paralog of E75 seems to be present in all and only in protostomes. Just like its paralog, it is ecdysone-induced in Drosophila (Stone and Thummel, 1993). However, not much is known about E78. Mutant phenotypes in Drosophila seem to be viable and fertile with some minor defects in regulation of some puffs and formation of dorsal chorionic appendages (Russel et al., 1996; Bryant et al., 1999). In Tribolium, knockdown of this gene resulted in a partially blocked embryogenesis, but this did not seem to have an effect on

- 25 -

Chapter I reproduction or metamorphosis which means it may possibly not be involved in the ecdysone cascade in this beetle (Tan and Palli, 2008; Xu et al., 2010).

HR3 (NR1F4) HR3 is a homolog of the vertebrate and ROR NRs. These transcriptional activators bind to the same response elements as REVERB and together they play a role in circadian rhythm (Jetten et al., 2009). Similar to E75, it is still unknown whether or not HR3 could be involved in insect circadian rhythm. HR3 is an ecdysone-inducible early-late gene which acts as a repressor on early genes and inducor of the late gene βFTZ-F1 in Drosophila (White et al., 1997). Its role in moulting and metamorphosis could be conserved throughout the arthropods and even ecdysozoa, since inhibition of the nematode HR3 homologue nhr-23 is necessary for moulting (Kostrouchova et al., 2001). Just like E75, HR3 also seems to act as a link in the interplay between ecdysone and JH pathways (Siaussat et al., 2004). Since this NR is expressed in many different tissues and life stages (embryo, larval, adult), it is expected to have other functions as well, besides its involvement in the ecdysone cascade.

HR96 (NR1J1) The hr96 gene has first been discovered in Drosophila (Fisk and Thummel, 1995) and has since been found in all insect genomes. Also in non-insect ecdysozoans, this gene is present. In the crustacean Daphnia pulex, an extra three divergent duplicates have been found besides the true orthologue DpHR96 (Thomson et al., 2009) and in C. elegans, three homologues of HR96 are known, namely DAF-12, NHR-8 and NHR-48 (Sluder and Maina, 2001). In vertebrates, all of these ecdysozoan genes are related to the NR1I group containing PXR, CAR and the vitamin D receptor VDR. These receptors are known to bind a wide variety of xenobiotics (Moore et al., 2006). In Drosophila, results with loss of function mutants revealed that HR96 plays a role in the response to certain xenobiotics, such as the insecticide dichlorodiphenyl-trichloroethane (DDT) (King-Jones et al., 2006). This role has recently been confirmed in Daphnia as well (Karimullina et al., 2012) and a similar function has also been reported for one of the HR96 homologs in C. elegans, namely NHR-8 (Lindblom et al., 2001). Another parallel is that these receptors all seem to interact with some lipid metabolism elements (Horner et al., 2009; Motola et al., 2006). Regulation of the lipid metabolism and the response to xenobiotics could be an ancestral function of the NR1I/J group.

- 26 -

Introduction and general background

2.4.3. Subfamily 2

This group contains most of the ecdysone-independent or ecdysone pathway-inhibiting NRs. Many members in this group are involved in neural development during embryogenesis. The three most conserved NRs throughout the metazoans (HNF4, SVP and TLL) are also member of this subfamily.

HNF4 (NR2A4) This receptor is very well conserved among all animals. There are three homologs in vertebrates and two in mammals (HNF4α and HNF4γ) which do not only share a high sequence identity with the insect HNF4, but the expression pattern is similar as well. In Drosophila, HNF4 can be found in developing fat body, Malpighian tubules and the midgut during organogenesis (Zhong et al., 1993). In B. mori and Aedes aegypti, this receptor has been found in fat body, gut and ovaries (Swevers & Iatrou, 1998; Kapitskaya et al., 1998). Not much is known about its function in insects, but in Drosophila it seems to play a role in the activation of genes involved in lipid catabolism and β-oxidation required for energy production (Palanker et al., 2009).

HR78 (NR2D1) HR78 is a repressor of the 20E signaling cascade. In Drosophila, this receptor is able to bind to a number EcR/USP binding sites and in Bombyx the receptor heterodimerizes with BmUSP in vitro (Hirai et al., 2002). Functional analysis of Drosophila HR78 has also shown the receptor to act as a repressor of the 20E signaling cascade in Drosophila (Fisk and Thummel, 1998; Astle et al., 2003). Despite a constant expression of this gene throughout embryonic development, the receptor does not seem to be crucial in early development. RNAi experiments showed that this is also the case in Tribolium, where knockdown of the HR78 gene resulted in a decrease of egg production and 40% larval lethality but did not affect the embryogenesis (Tan and Palli, 2008; Xu et al., 2010). HR78 also seems to be involved in growth (Baker et al., 2007) which could be a conserved function of the NR2C/D group, since TR4 mutants in mice show similar defects.

TLL (NR2E2) TLL and especially its involvement in embryogenesis has been studied intensively in Drosophila. The NR is one of the terminal gap genes of the fruitfly and is involved in

- 27 -

Chapter I posterior and anterior patterning (Pignoni et al., 1990). Even though TLL has been shown to function only as transcriptional repressor, it is also able to activate the expression of certain genes, probably by repressing other repressors (Moran and Jimenez, 2006). Just like its vertebrate homolog TLX, TLL is also involved in the neural development in later stages of development, such as the formation of protocerebral neuroblasts and in eye formation (Rudolph et al., 1997; Daniel et al., 1999).

HR51(NR2E3) The gene coding for HR51 is called unfulfilled and was only discovered during the genome annotation of the D. melanogaster genome. Similar to E75, this receptor contains a heme iron and was shown to bind NO and CO (de Rosney et al., 2008). In Drosophila, this gene is expressed in the nervous system during developmental stages but unlike its vertebrate homolog Photoreceptor-specific nuclear receptor (PNR), this NR is not involved in the development of the visual system. Mutations in this gene in Drosophila affect wing expansion and fertility (Sung et al., 2009).

DSF (NR2E4) The dissatisfaction gene in Drosophila is reported to be expressed throughout all stages of development, but its expression is limited to a small number of different neurons (Finley et al., 1998). In Drosophila, DSF seems necessary for normal courtship behaviours and sex-specific neuronal differentiation (Finley et al., 1997). In Tribolium, DSF has proven to be required for egg production and also for embryogenic development (Tan and Palli, 2008; Xu et al., 2010).

HR83 (NR2E5) Similar to HR51, this other PNR-homolog was found thanks to annotation of some insect genomes. HR83 has been found in all insects, except B. mori and the human body louse Pediculus humanus. In the recently sequenced genome of the crustacean Daphnia pulex, this receptor seems to be missing as well. Not much is known about HR83 or its functions. In Drosophila and Tribolium, no LBD has been found in the sequence for this receptor. In contrast, a LBD is present in the HR83 of the hymenopterans Nasonia vitripennis and A. mellifera. In Drosophila, the expression in all developmental stages is very low which could mean the expression is tissue restricted (Sullivan and Thummel, 2003).

- 28 -

Introduction and general background

NR2E6 This is a NR discovered very recently in the honeybee A. mellifera, where it was found to be expressed in the brain and the compound eye of pupa and adult bees (Velarde et al., 2006). It has therefore also been named PNR-like. Even though it seems to be present in most insect genomes, it is absent in Drosophila. Because of this, not much is known about the function of this NR. The only species in which functional studies have been done is Tribolium, but silencing of this gene delivered no phenotype (Tan & Palli, 2008).

SVP (NR2F3) The seven-up gene belongs to the group of COUP-TFs. These orphan receptors are repressors that show a wide variety of function and are widely expressed throughout all stages. During embyogenesis, SVP has a crucial role in neural development. It is involved in the establishment of the embryonic nervous system and the development of four photoreceptor cells (Mlodzik et al., 1990). Besides the nervous system, SVP in Drosophila embryos was also found in the Malpighian tubes and the fat body (Hoshizaki et al., 1994; Broadus and Doe, 1995; Kerber et al., 1998; Sudarsan et al., 2002). In Callosobruchus maculates, SVP is shown to be involved in the response of this beetle against certain plant defenses. This response seems to depend on interaction of SVP with HNF-4, which can act as an activator (Ahn et al., 2007; 2010). Just like the vertebrate COUP-TF receptors can inhibit the RXR signaling pathways, the same can be said about SVP and USP/RXR (Zelhof et al., 1995b). In Drosophila, no direct interaction between SVP and USP was detected but in vitro tests with the mosquito Aedes showed that the repression of the ecdysone pathway is probably due to direct interaction between both NRs (Miura et al., 2002; Zhu et al., 2003). In Tribolium, SVP is proven to be involved in the ecdysone pathway as well, since the receptor was required to complete metamorphosis (Tan and Palli, 2008).

2.4.4. Subfamily 3

ERR (NR3B4) The estrogen related receptor (ERR) is the only member of the NR3 group found in insects. In vertebrates, the group is relatively large, containing the NR3C group of steroid receptors (AR, PR, GR and MR), the NR3A group of estrogen receptors (ERs) and the group of ERRs. The receptors in this last group share many similarities in structure and function with ERs and are

- 29 -

Chapter I considered orphan receptors since their ligands are not known yet. Not much is known about this NR, other than the fact that this gene is expressed during mid-embryogenesis and the third larval stage of D. melanogaster (Sullivan and Thummel, 2003). This last upregulation seems to coincide with the start of preparation for metamorphosis, but functional data is lacking in order to make any hypothesis about its possible role herein. Since the gene is also shown to be expressed in the adult brain of honeybees (Velarde et al., 2006), a role in the nervous system is likely.

2.4.5. Subfamily 4

HR38 (NR4A4) Subfamily 4 is a small group of NRs containing HR38 and the vertebrate Nurr1 and NGFI-B NRs. Just like their vertebrate relatives, HR38 can bind the DNA either as a monomer, or as a heterodimer with RXR/USP. Another parallel between vertebrate and insect NR4 receptors is that these lack a functional LBP, which means these receptors are true orphans. Just like HR78, HR38 is a repressor of the 20E cascade. It can indirectly inhibit the ecdysone response by its ability to bind to the same response elements as the functional ecdysone receptor and also because of its ability to compete with EcR in binding with USP. This heterodimer seems to respond not only to 20E, but also to a number of other ecdysteroids such as α-ecdysone, 3- epi-20E and 3-dehydromakisterone A. We know that this alternative ecdysone pathway does not depend on binding of the ligand with HR38 though since the crystal structure of the LBD showed that the receptor is unable to bind a ligand due to the lack of a LBP. It is hypothesized that the working mechanism would involve interaction with another protein that is able to bind these ecdysteroids (Baker et al., 2003). In Drosophila, HR38 is widely expressed throughout all stages and has a role in metamorphosis, but also in cuticle-forming as shown by null mutant experiments as well as the fact that HR38 is reported to activate some epidermal genes involved in cuticle formation (Bruey-Sedano et al., 2005; Davis et al., 2007). In Tribolium, HR38 is important in embryogenesis and for both larva to pupa and pupa to adult transitions (Tan and Palli, 2008; Xu et al., 2010).

- 30 -

Introduction and general background

2.4.6. Subfamily 5

FTZ-F1 (NR5A3) The Drosophila FTZ-F1 gene has two promotors, causing the expression of two different isoforms, αFTZ-F1 and βFTZ-F1. The former is only expressed during early embryogenesis, where it has a role in the segmentation of the early embryo. βFTZ-F1 on the other hand is expressed during late embryogenesis but also during the larval and pupal development stages. It has multiple reported functions. It is crucial for reproduction and embryogenesis but it is best known as being the late gene in the ecdysone signaling cascade. βFTZ-F1 is expressed just before each developmental transition in all insects thus far investigated (Martin, 2010). It is repressed by 20E and expression always occurs when 20E titers are low (Yamada et al., 2000; Sullivan and Thummel, 2003; Palanker et al., 2006). In holometabolic insects, βFTZ-F1 also plays a role as a competence factor for stage-specific transcriptional responses to 20E (Yamada et al., 2000). This is not the case in more primitive insects however. In Drosophila and Blattella, βFTZ-F1 has also been proven to play a role in the onset of ecdysteroid production during the intermolt periods (Parvy et al., 2005; Cruz et al., 2008). Interestingly, in C. elegans, the homologs of HR3 and FTZ-F1, NHR-23 and NHR-25 respectively, seem to have a similar cooperation which could suggest this function is conserved throughout Ecdysozoa.

HR39 (NR5B1) HR39 is the result of a duplication of the ftz-f1 gene very early during bilaterian evolution (Bertrand et al., 2004). They are somewhat similar in sequence and structure. Both bind to DNA as monomers and recognize the same response elements (Ohno and Petkovich, 1993; Crispi et al., 1998) which means they are effectively in competition with each other. They also seem to function completely differently. While FTZ-F1 is a transcriptional activator, HR39 is probably a repressor (Palanker et al., 2006). Their expression patterns are also rather different. While βFTZ-F1 is a ‘late gene’ in the ecdysteroid signaling cascade, HR39 seems to act like an ‘early gene’ during the onset of metamorphosis. In Drosophila, HR39 null mutants are viable. In Tribolium however, RNAi experiments showed that HR39 is necessary for embryonic and post-embryonic development (Tan and Palli, 2008; Xu et al., 2010). Furthermore, HR39 is also involved in sexual development in insects, just like βFTZ-F1 (Allen and Spradling, 2008; Xu et al., 2010).

- 31 -

Chapter I

2.4.7. Subfamily 6

HR4 (NR6A6) HR4 is the only member of this subfamily in insects. This transcriptional repressor is homologous to Germ Cell Nuclear Factor (GCNF; NR6A1), an orphan receptor in vertebrates. It is found in all insect genomes sequenced and could have a similar function as HR3. Loss of function mutants in Drosophila and Blattella point out that HR4 is involved in the repression of a few early or early/late genes such as E75 and HR3 and that HR4 is required for the transcriptional activation of the late gene FTZ-F1 (King-Jones et al., 2005; Martin et al., 2010). RNAi experiments in Tribolium also proved an involvement in vitellogenesis and oogenesis (Xu et al., 2010).

2.4.8. Subfamily 0

The three members of this subfamily are Knirps (Kni; NR0A1), knirps-like (Knrl; NR0A2) and eagle/Egon (NR0A3). These are special NRs which do not have the normal canonical NR structure. All three NRs are missing their LBD. Kni and Knrl are both involved in early embryonic development where they are found to be expressed in posterior and anterior domains at the blastoderm stage. Interestingly, while mutations in both genes resulted in defects in head morphogenesis, these defects were not observed in the single mutants, meaning one can take over from the other at least what the role in head patterning is concerned. Drosophila has so far been the only insect in which all three NRs have been found. All other insect genomes sequenced so far, including some other Diptera, do not contain Kni which suggests that the gene coding for this NR is probably a very recent duplication of Knrl.

3. Ecdysone signaling pathway

3.1. Insect growth, moulting and metamorphosis

Most insects are oviparous or egg laying. After hatching, insects go through a series of juvenile stages before turning adult. To overcome the rigid constraints of their exoskeleton, they undergo one or more moulting events during their development in which the old exoskeleton is shed and a new and bigger one is created. This moulting event is a necessary though potentially dangerous moment in the arthropod’s life. Failure to execute this process in a correct way can mean the death of the animal and intervening in this process hence draws - 32 -

Introduction and general background the attention from researches working in the field of pest control. In the Hexapoda subphylum, three different types can be observed. The ametabolous (Collembola,…), the hemimetabolous (Hemiptera, Orthoptera, Dermaptera,…) and the holometabolous development (Diptera, Coleoptera, Lepidoptera, Hymenoptera,…) (Fig. 1.14).

The ametabolous hexapods do not go through any metamorphosis. Their juvenile stages look exactly the same as the adult stage, albeit smaller. Similar to the ametabolous, the instar stages of the hemimetabolous species (called nymphs) look just like their adult stages as well. The main difference with the aforementioned ametabolous group is the formation of wings in the adult stages. Holometabolous species undergo a complete metamorphosis and have a pupal stage between the juvenile stages (also called larval stages) and the adult stage. In some cases, the insects have a constant number of instar stages (somewhere between 3 and 15) but in other cases this can vary depending on temperature, availability of food or other environmental factors, as is the case in the red flour beetle T. castaneum.

Fig 1.14 Schematic representation of the three types of development in arthropods. Ametabolic insects are characterized by juvenile stages that look exactly like the adult stage, only smaller. In hemimetabolous insects, the juveniles also look like the adults, but lack wings while the adults develop wings during the last moult. Holometabolic insects are characterized by a complete metamorphosis between the last larval stage and the adult stage, involving a pupal stage.

- 33 -

Chapter I

3.2. Ecdysone biosynthesis

Growth and post-embryonic development of insects, including moulting, are regulated through a number of pathways activated by juvenile hormones, ecdysteroids or neuropeptides, such as the insulin-like peptides. The critical endocrinological signal that switches the animal from its feeding stage or ‘intermoult’ to the pharate stage or moulting stage is the elevation of ecdysteroid levels. During the feeding stage, the insect grows until internal mechanisms cause the brain to send a signal to go to the next stage. This signal starts with the secretion of the prothoracicotropic hormone (PTTH) in the brain. This PTTH, probably in combination with a number of other neuropeptides such as insulin-like peptides (ILPs), triggers the prothoracic glands to start the production of the ecdysteroid. The biosynthesis of ecdysone and other ecdysteroids starts with cholesterol and the conversions are mediated by a group of cytochrome P-450 enzymes which are the product of the so-called Halloween genes (Petryk et al., 2003; Rewitz et al., 2006a; Iga and Smagghe, 2009). In most insects, these prothoracic glands produce ecdysone, but some larval Lepidoptera are also known to secrete 3- dehydroecdysone which is then later converted to ecdysone by enzymes present in the hemolymph (Fescemeyer et al., 1995). Ecdysone itself is a prohormone, and is converted to the active ecdysteroid in most insects, which is 20E, by Shade, one of the P-450 enzymes encoded for by the Halloween genes.

3.3. Ecdysone signaling cascade

Which ecdysteroid exactly is responsible for the moult-trigger depends on the species. In most insects this is 20-hydroxy-ecdysone (20E), but there are some arthropods (e.g. decapod Crustacea) where ponasterone A (Pon A, 25-Deoxy-20-hydroxyecdysone) is a main active ecdysteroid as well (Fig. 1.15). Epidermal cells respond to these rising levels of ecdysteroid by changing shape and expressing a set of genes involved in the moulting process. These changes will eventually cause a new cuticle to be formed underneath the old one after which the latter will eventually be shed.

- 34 -

Introduction and general background and signal transduction proteins. Many of these are genes encoding nuclear receptors, including DHR3, DHR39, E75, E78, FTZ-F1 (Fig. 1.16). These proteins will be responsible for the activation of certain processes which are necessary in the upcoming ecdysis sequence, such as air swallowing or certain movements which will ultimately cause the cuticle shedding.

Figure 1.15 Structure of ecdysone (E), 20E and Pon A

On a molecular level, the whole moulting process is triggered by the binding of the active ecdysteroid hormone to its functional receptor, a heterodimer of EcR RXR/USP which sets off the expression of a number of genes in the so called ‘early’, ‘early-late’ and ‘late’ phases, resulting in the production of a series of peptides and receptors, peptide processing enzymes and signal transduction proteins. Many of these are genes encoding nuclear receptors, including DHR3, DHR39, E75, E78, FTZ-F1 (Fig. 1.16). These proteins will be responsible for the activation of certain processes which are necessary in the upcoming ecdysis sequence, such as air swallowing or certain movements which will ultimately cause the cuticle shedding.

In salivary glands of Drosophila, many of these genes have been observed to be located in the polytene chromosome puffs. Polytene chromosomes result from several rounds of DNA amplification without cell division following. The so-called puffs are regions on the chromosome where the DNA is unraveled, allowing RNA transcription. Many insights in this pathway have been gained due to observations of the location, timing and size of these puffs during insect development since these changes are linked to the transcriptional activity.

- 35 -

Chapter I

Ashburner et al. (1973) proposed a first model on how ecdysone controls the activation of the early and late puffs in Drosophila melanogaster. Binding of ecdysone to its receptor would induce the expression of a few early genes while at the same time it represses the expression of the late genes. Then, when the proteins encoded by the early genes are sufficiently abundant, they repress their own promoter while at the same time activating those of the late genes. Since Ashburner presented his initial ideas, many aspects of the model have been confirmed and extended by molecular analyses. The active receptor that binds ecdysone in the first place is since identified as a heterodimer of the EcR and Usp/RXR, which has been observed to co-localize to early puff sites (Yao et al., 1993) but also to late puff sites (Talbot et al., 1993). Huet et al. (1995) have integrated the inter-moult and early-late puffs in this original Ashburner model.

Figure 1.16 Summary of the ecdysteroid regulatory cascade. Binding of 20E to the EcR-Usp complex starts the ecdysteroid cascade with the expression of the so called ‘early’ genes (EcR, E75, BR, E74 and E93) that will then be responsible for the upregulation of a set of ‘early-late’ genes (including HR3, HR4, HR38 and E78). Via FTZ-F1, the signal will eventually be passed on to the ‘late’ genes. The nuclear receptors are boxed. Redrafted after Bonneton et al. (2008).

- 36 -

Introduction and general background

While the ecdysteroid pathway is mainly known for its role in moulting and metamorphosis and its action in these processes have been studied well, it has been implicated in the regulation of many other biological processes as well. One of these processes in which 20E is thought to be involved is reproduction and more specifically in oogenesis, vitellogenesis and spermatogenesis (Raikhel et al., 2005). Additionally, this pathway plays a critical role in the control of cellular proliferation and differentiation (Siaussat et al., 2009), embryogenesis (Kozlova and Thummel, 2003), diapause (Denlinger et al., 2005) and polyphenism (Hartfelder and Emlen, 2012) as well.

3.4. The ecdysone receptor as a pest control target

Over the last three decades a number of chemically novel insecticides called diacylhydrazines (DAHs) have been discovered and developed. They were first discovered in 1983 by Rohm and Haas Company and appeared to be very specific in their toxicity towards certain insect orders such as Lepidoptera, Coleoptera and Diptera. Other non-target organisms were generally unaffected. Wing et al. (1988) showed that this class of compounds acted as ecdysteroid agonists, mimicking the action of ecdysteroids. A range of different DAH compounds have been developed and commercialized over the years, such as tebufenozide (RH-5992), halofenozide (RH-0345) and methoxyfenozide (RH-2485). Even though there are some differences in spectrum and potency between these different compounds, their mode of action and effect were generally the same. The effects on larvae were comparable with those that are expected from a state of ecdysteroid excess, called hyperecdysonism. They are therefore also called molting-accelerating compounds (MACs). Their activity is based on binding to EcR and inducing premature lethal molting in larval stages and aborting reproduction in adults (Dhadialla et al. 1998, 2005; Pineda et al. 2007; Smagghe et al., 2013). The DAH-based ecdysone agonist insecticides are active against several economically important lepidopteran and coleopteran pests, and considered as environmentally friendly compounds with an excellent margin of safety to non-target organisms (Dhadialla et al. 1998; 2005; Smagghe et al., 2013). Consequently, these pesticides are usually included in IPM programs targeting different pests and crops all over the world.

An important process in the selectivity of these compounds is the specific binding of the MACs to the target EcR that is governed by a lock-and-key principle (Dhadialla et al. 2005). For instance, according to Carlson et al. (2001) the binding affinity is high in targeted - 37 -

Chapter I

Lepidoptera pests, whereas binding is low or not detectable in non-targeted insects. In agreement with these results, previous studies have found tebufenozide to have an extremely low affinity for hemipteran EcRs belonging to the Sternorrhyncha, namely B. tabaci and the aphid Myzus persicae (Carmichael et al. 2005; Graham et al. 2007). In this respect, it appears that differences in the architecture of LBP may provide the differential binding affinities of the DAH-based compounds among different taxonomic orders (Carmichael et al. 2005).

4. Arthropods used in this thesis

4.1. Acyrthosiphon pisum

4.1.1. Introduction

Aphids comprise about 4,000 species, most of them living in the temperate regions of the northern hemisphere. Compared to 60,000 species of weevils, 10,000 species of grasshoppers or 12,000 species of geometrid moths, this is a relatively small group. Of those 4,000, only about 100 are really the cause of significant agricultural damage. Despite being a relatively small group in terms of species number, they still pose a big concern for agriculture. They are versatile, sap-sucking hemipterans that possess a high capacity to adapt to their environment and most of them have the ability to reproduce asexually as well. They are also characterized by their associations with the mutualistic bacterium Buchnera aphidicola, which provides the aphid with essential amino acids which are not present in the phloem sap (Akman Gündüz and Douglas, 2009). Their life cycle, characterized by a high fecundity and short generation time, combined with their adaptation abilities make it one of the most destructive pests on cultivated plants.

The pea aphid or Acyrthosiphon pisum is a green and rather large aphid which feeds mainly on legume plants of the Fabaceae family such as beans, clover and alfalfa. Compared to other aphid species, its host plant range is rather small. It is considered a model species for the Hemiptera.

- 38 -

Introduction and general background

4.1.2. Taxonomy

Phylum: Arthropoda Class: Insecta Order: Hemiptera Suborder: Sternorrhyncha Superfamily: Aphidoidea Family: Aphididae Genus: Acyrthosiphon

The phylogenetic relationship within the hemimetabolous order of the Hemiptera is not clear and is still under investigation and discussion. Historically, these insects, which are mainly characterized by the sucking mouth parts, were divided into two large groups, the hemipteran and the homopteran insects. Today, the Homoptera order has been dissolved and divided into two large suborders, the Auchenorryncha and the Sternorrhyncha. The former contains the cicadas, planthoppers, leafhoppers while the latter contain insects such as the whiteflies and the aphids (Aphididae). These two suborders, together with the Coleorrhyncha and the Heteroptera, now form the order of Hemiptera. The pea aphids is a member of the Aphidoidea superfamily, which comprises all sap-sucking aphids.

4.1.3. Polyphenism and life cycle

Aphids are known for their phenotypic plasticity, and show both a wing and a reproductive polyphenism. These changes are induced by environmental conditions such as the quality of the host plant, aphid density, predation and also some abiotic factors.

The aphid life cycle begins in the spring with a female ‘foundress’ hatching from an overwintering egg. This foundress will find a suitable host plant and start a colony. The reproduction in this phase is parthenogenetic, so asexual, and the aphids at this stage are born viviparous. This means that the population starting from one foundress at this point is female and genetically identical, apart from some spontaneous mutations. During the asexual production, certain environmental conditions such as the quality of the host plant, aphid density or predation can cause the females to produce winged offspring as well. These winged aphids can then cover greater distances and find a new host plant to colonize. In the fall, changing light and temperature conditions induce the females to produce males, winged and - 39 -

Chapter I unwinged, and a change from parthenogenetic to sexual reproduction occurs. Females will now lay eggs, which have to overwinter for a few months (Fig. 1.17).

Figure 1.17 Life cycle of aphids highlighting the switch between sexual and asexual reproduction. (www.Aphidbase.com). A female aphid, hatched from an egg, starts a colony in spring by asexual reproduction Throughout summer, the aphids continue to reproduce asexually through parthenogenesis. Seasonal changes will induce the females to produce males as well and sexual reproduction occurs. The resulting eggs will overwinter before hatching again in spring.

4.1.4. Economic impact

Aphids can damage their host plants in more than one way. There is the direct damage to the plant, caused by the feeding of phloem saps but just as important – or perhaps even more so – is the indirect damage caused by the transmission of some of the most devastating plant viruses in cereal crops, vegetables and potatoes. To a lesser extent, they also facilitate the growth of sooty molds, which grow on the honeydew excreted by the aphids.

Since aphids are mainly present in temperate regions, the damage on tropical crops is limited. In temperate climates however, Hill (1987) reported that of the 45 major pests on the 6 main

- 40 -

Introduction and general background food crops in these areas, 26 are aphids. For the UK, direct losses from aphids were estimated to lie between 8-16% in pea, 10-13% in wheat and around 5% in potato (Tatchell, 1989). The indirect damage, caused by virus transmission can be even higher. In sugar beet for example, the aphid-transmitted viruses such as the Beet yellows virus (BYV) and Beet mild yellows virus (BMYV) can decrease sugar production yield by up to 49% (Smith and Hallsworth, 1990). Also in cereal crops and field-grown vegetable crops, the viruses that are transmitted by aphids can cause significant damages. More details and information on the economic impact of aphids on agricultural crops can be found in the review of Dedryver et al. (2010).

4.2. Bombus terrestris

4.2.1. Introduction

As opposed to Acyrthosiphon, which is a pest insect, the bee species belonging to the Bombus genus are mainly beneficial insects and have a large economic impact on agriculture due to their role as a pollinator. They are quite large compared to most other insect species and are covered in a dense fur which allows them to withstand cooler temperatures. Bumblebees in general are mainly found in the northern hemisphere and predominantly at higher altitudes or higher latitudes. Bombus terrestris is the most numerous European Bombus species and can be found pretty much all over the continent. As it is a social insect, albeit a more primitive form of eusociality than in the well-known honeybee A. mellifera, the bumblebee forms colonies, living in nests. Similar to bees and many wasps, bumblebees are characterized by an ovipositor which is modified into a stinger with which they are able to inject venom. Bumblebees are phytophagous and live on nectar and pollen throughout their entire life. Adults feed on the nectar they gather from flowering plants while they forage to collect pollen to take to the nest and feed their young. B. terrestris is a very generalist species as more than 300 plant species have been listed as a possible food source for this insect (Rasmont, 1988). This foraging behavior is very important for us as well, since it causes pollination of many flowering plants, including many of our cultured crops. Moreover, bumblebee colonies are even commercially available for farmers to use in greenhouses.

- 41 -

Chapter I

4.2.2. Taxonomy

Phylum: Arthropoda Class: Insecta Order: Hymenoptera Suborder: Apocrita Superfamily: Apoidea Family: Apidae Subfamily: Apinae Genus: Bombus

Around 250 species comprise the Bombus genus containing all bumblebees. They are part of the important order of Hymenoptera, characterized by usually two pair of membraned wings and strongly developed chewing mouth parts. In some species, including all bees, these mouth parts have evolved into a large proboscis, with wich the insects can drink from plants as well. This order includes important social insects such as bees, ants and wasps which are grouped in the suborder of Apocrita. The Hymenoptera are holometabolous, meaning they undergo a complete metamorphosis between the larval stages and the adult stage. One of the best known families is the Apidae family, containing all bees, including the bumblebees.

4.2.3. Life cycle

Bumblebees generally know an annual colony cycle (Fig. 1.18). This means that the queen, which is the only one of the colony that survives the winter, must start a new colony every year. In spring, the queen ends her hibernation and will start feeding on nectar to replenish her reserves. In the meantime, she will start looking for a suitable location to start her nest. Usually these nests are built underground in the case of bumblebees. After collecting enough nectar and pollen she will lay her first eggs. The eggs which are laid during most of the summer season will usually give rise to the smaller worker bumblebees. The first group of workers will mainly be fed by the queen during their juvenile stages and after they turn adult, they will take over the foraging and feed the later generation of worker bumblebees. Late in summer, the colony will start producing males and other queens as well, which will mate in autumn. These queens will then leave the colony and start looking for a suitable hibernation location.

- 42 -

Introduction and general background

Figure 1.18 Annual life cycle of a bumblebee colony (Goulson, 2010). After overwintering underground, a female foundress will start searching for a suitable place to form a nest and start a colony. During most of the spring and summer, the colony will only produce female workers. Later in summer, males and other queens will be produced as well. The males will mate with the young queens, die, and the queens will find a place to overwinter.

4.2.4. Economic importance

The importance of bees and bumblebees for ecosystem integrity and food security cannot be underestimated. Because of their foraging behavior, collecting pollen as food for their young, they are natural pollinators. A broad variety of crops are dependent on foraging and pollinating insects for their reproduction. Either because they require cross-pollination or because they require insects to move the pollen from the anthers to the stigma. While the honeybee A. mellifera is by far the best known pollinator, other bee species are very important

- 43 -

Chapter I as well. First of all because the honeybee is sometimes unable to pollinate certain plants, often due to anatomical factors. Another difference between the honeybee and the bumblebee is that the former is less active in colder temperatures and rainy days, while bumblebees remain active in such conditions. For that reason, the bumblebee can sometimes be more important in more northern regions.

Bumblebee-pollination became a commercial succes-story when in the 1980s, the Belgian veterinarian and amateur bumblebee researcher Dr. De Jonghe discovered the value of bumblebees for the pollination of greenhouse tomatoes. Until then, tomatoes were pollinated by manual labour, which is time consuming and hence very costly. He founded the company Biobest for the commercial rearing of bumblebees and some others followed soon. Within a few years, virtually all Dutch greenhouse tomatoes were pollinated by these commercially reared bumblebees. Over the last two decades, the sale of bumblebee colonies for this purpose has spread globally and at this moment, the economic value of these bumblebee-pollinated tomato crops is estimated at €12 billion (Goulson, 2010). An excellent review on the domestication and commercialization of bumblebees is the one written by Velthuis and van Doorn (2006).

4.3. Tetranychus urticae

4.3.1. Introduction

The two-spotted spider mite or Tetranychus urticae is one of the most damaging arthropods in many agricultural crops. This chelicerate damages the plant mainly by puncturing the leaf cells with their needle-like mouth parts in order to feed on the cell content. The spider mite is a very polyphagous pest and has a high fecundity, two characteristics partly explaining the success of this species. Another important factor is its adaptive nature and its ability to rapidly develop resistance against pesticides. The name of the spider mite refers to its ability to spin webs. Plants that are heavily colonized by this mite can be seen covered in the silky webs made by this arthropod. In horticulture, this webbing can cause an additional cosmetic damage as well.

- 44 -

Introduction and general background

4.3.2. Taxonomy

Phylum: Arthropoda Subphylum: Chelicerata Class: Arachnida Subclass: Acari Superorder: Acariformes Order: Trombidiformes Family: Tetranychidae Genus: Tetranychus

The arachnid taxon of Acari, containing all the mites and ticks is an old, large and also a very diverse taxonomic group within the Arachnida. The classification of this group has been subject to a lot of discussion that is still ongoing. The present day consensus however is to place this group as a subclass within the Arachnida and to divide it into two or three superorders. There is the large superorder containing most mites, the Acariformes, to which also the Tetranychidae belong, and there is the Parasitiformes superorder, containing the ticks such as Ixodes scapularis and a variety of other mites. The third group, the Opilioacariformes, is a small group comprised of one family and only 20 species of rather large mites. The Acariformes form by far the largest group, containing about 30,000 of the 42,000 known Acari species. It is divided into the Sarcoptiformes and the Trombidiformes, the latter comprising most plant-parasitic mites such as Tetranychus urticae.

4.3.3. Life cycle

Just like most mites, the life cycle of T. urticae is a little more complex compared to most insect species (Fig. 1.19). Mites are oviparous and an adult female can lay up to 20 eggs per day in optimal conditions. These eggs can either be fertilized, giving birth to female mites, or not fertilized, giving birth to males. After 3 to 5 days the eggs will hatch into larvae. Contrary to the later stages, the larva of the mite only has six legs, instead of eight. After moulting for 3 to 10 days, the larva will develop into the first nymphal stage, called the protonymph, which are now eight-legged. The subsequent moult brings the mite to the deutonymph stage after which it will move on to the adult stage. The time of development of a T. urticae individual is very much subject to the environmental conditions, mainly the temperature and humidity. Spider mites prefer rather dry conditions and high temperatures and thrive optimally around a - 45 -

Chapter I temperature around 30°C. At that temperature, they can turn adult as soon as 8 days after their own eggs were laid by their mother.

Figure 1.19 Life cycle of the two spotted spider mite showing the different stages from egg, over six-legged larvae and eight-legged nymphs to adults. (www.al.gov.bc.ca/cropprot/mites.htm)

4.3.4. Economic importance

Herbivorous mites form a big problem in agriculture and T. urticae is probably the most important one. It is known to be one of the most polyphagous pests and has as many as 1100 possible host plants belonging to 140 different plant families. T. urticae itself thrives in warm conditions and is thus mainly a problem in warmer areas and in the greenhouses in more temperate regions. Some of the most important host plants from an economic standpoint are tomatoes, peppers, cucumber, maize, soy, grapes and apples. The success of this pest can be explained by a number of factors. First of all, it has a very high fecundity and short developmental time causing a very quick colonization of a plant. Mites are also characterized by a very high incidence of resistance. T. urticae is sometimes dubbed ‘the most resistant species’ because of its known resistance against a very large and still increasing number of - 46 -

Introduction and general background pesticides. According to the review on this matter by Van Leeuwen et al. (2010), there is proof of resistance against 92 active ingredients.

The damage created by the two-spotted spider mite is directly related to their feeding on the cell content of leaf cells. With their needle-like mouth parts, they pierce through the cell and suck up the cell content. This causes the mesophyll tissue to collapse and little chlorotic or necrotic spots appear on the leaves and eventually the whole leaf will be affected and die. If the infestation is not controlled, complete defoilage of the plant is possible. In horticulture and especially the ornamental plants business, mites can also cause an additional aesthetic form of damage due to their webbing (Fig 1.20).

Figure 1.20 Damage and webbing caused by T. urticae on sweet pepper plants (www.wikipedia.com). The feeding of the spider mites on plants cause necrotic spots on the leaves and eventually the whole leaf can die.

- 47 -

Chapter I

5. RNAi

RNA interference can be defined as a post-transcriptional gene silencing pathway that is triggered by double stranded RNA (dsRNA). This dsRNA can be of endogenous origin, in the form of microRNA (miRNA) or it can be exogenous dsRNA, in the form of viruses, transposons or introduced into a cell for research purposes. Since its discovery in the nematode C. elegans (Fire et al., 1998), RNAi has rapidly developed as a widely used molecular research tool in a variety of insect orders, including Diptera (Lum et al., 2003; Dietzl et al., 2007), Lepidoptera (Yu et al., 2008; Chen et al., 2008; Tian et al., 2009; Terenius et al., 2011), Coleoptera (Arakane et al., 2004; Suzuki et al., 2008) and Hymenoptera (Schluns & Crozier, 2007; Marco Antonio et al., 2008). In 2006, Andrew Z. Fire and Craig C. Mello have also been awarded the Nobel Prize in Physiology or Medicine for their discovery of RNAi. The first RNAi successes with insects were achieved with the fruit fly D. melanogaster. The first experiment was performed in vivo by injection of embryos (Kennerdell & Carthew, 1998) and in vitro by soaking S2 cells in dsRNA-containing medium (Clemens et al., 2000). In 2000, the first genetically transformed D. melanogaster lines were constructed where a heritable RNAi effect was created by expressing dsRNA as an extended hairpin-loop RNA (Kennerdell & Carthew, 2000). These RNAi libraries have become a widely used asset allowing high throughput loss-of-function research in this dipteran model insect. In contrast to D. melanogaster and also the silkmoth B. mori (Kanginakudru et al., 2007; Dai et al., 2008), most insects are much more difficult to transform genetically. Therefore, heritable RNAi by bringing dsRNA to expression in a stable manner is currently not possible in most species which means that these loss-of-function experiments have to be done by introducing the dsRNA from outside of the insect.

5.1. Molecular mechanism

There are two different RNAi pathways. The first one is the miRNA pathway leading to a translation block, which will not be discussed into further detail here. The second pathway is the siRNA pathway, leading to nucleotide degradation. The core of the RNAi pathways is formed by a few important proteins such as the RNaseIII endoribonuclease Dicer (Dcr) and the RNA-induced Silencing Complex (RISC) in which the RNaseH Argonaute (Ago) and R2D2 are the most important components. The pathway involves two phases. In a first phase,

- 48 -

Introduction and general background the dsRNA that enters the cytosol will bind to the Dicer enzyme and will be cleaved into pieces of small interfering RNA (siRNA), each around 20 nucleotides long. In a second step, these smaller transcripts will be recognized and bound by the RISC complex. Subsequently, the siRNA will be unwound to a single strand and will hybridize with the matching endogenous mRNA. The final step then is the degradation of this mRNA by Ago. This process is visually represented in Figure 1.21.

Fiure 1.21 Schematic representation of the siRNA pathway. dsRNA which enters the cell will bind with Dicer, a RNaseIII endoribonuclease and will be cleaved into smaller pieces of RNA. These smaller pieces, called siRNA will associate with the RISC complex and will bind to the complementary endogenous mRNA. Eventually, the RNaseH enzyme Argonaute, which is part of the RISC complex, will then start the cleavage of the mRNA and hence prevent translation (http://www.welgeninc.com/technologies.html).

5.2. RNAi efficiency and efficacy

Despite the success stories of RNAi, there seems to be a rather large variation in the success rate of these experiments for different insect species. Some species, such as the red flour beetle Tribolium castaneum and the honeybee Apis mellifera have a very high success rate for RNAi and especially the former shows a very robust systemic response after introduction of exogenous dsRNA, meaning that the silencing response is observed in tissues throughout the insect body. In some cases, such as the cricket Gryllus bimaculatus (Orthoptera) and the firebrat Thermobia domestica (Thysanura) even a long term silencing response has been - 49 -

Chapter I observed (Moriyama et al., 2008; Uryu et al., 2013). In the case of the cricket, injection of dsRNA targeting period, a circadian clock gene, resulted in a RNAi effect for over 50 days. On the other hand, in many Lepidoptera, such as S. littoralis and B. mori, as well as many species from other orders, RNAi is rather inefficient or the efficiency is very variable at best. The cause is often unknown and could be a combination of many factors such as degradation of the dsRNA, lack of uptake into the cells or insufficient upregulation of the core RNAi genes. During the last few years, more and more research has gone into the factors possibly limiting the use of RNAi in some species. However, the available information is still very fragmentary. For recent reviews and overviews of RNAi in insects, I can refer to the Bellés (2010), Terenius et al., (2011) and Gu & Knipple (2013).

5.3. RNAi as a possible pest control tool

Besides its role as a functional genomics research tool, RNAi could in theory be a very useful tool in crop protection as well. Knockdown of certain genes could cause all sorts of lethal or sublethal effects and one of its main advantages is its selectivity. Indeed, one of the biggest challenges in the search for environmental-friendly pest control agents has always been the selectivity of their toxicity and the attempt to minimize the damage inflicted on non-target organisms, especially beneficial insects such as pollinators and predators. RNAi provides a tool with which you can specifically target single species or larger taxonomical groups. Very important in that aspect is the target fragment design and the conservation of the selected sequence. Using more conserved regions can cause toxicity towards a range of species and the level of conservation can determine how closely related you want the affected species to be.

A number of studies have already shown that transgenic plants producing specific insect dsRNA can effectively reduce the damage caused by insects on these host plants. In 2007, a study by Baum et al. proved that the production of vATPase dsRNA in transgenic maize plants caused a reduction of corn root damage caused by the western corn rootworm Diabrotica virgifera virgifera (Baum et al., 2007). Another study provided evidence by genetically modifying Arabidopsis and tobacco plants to produce a dsRNA specific for a cytochrome P450 gene (CYP6AE14) which is needed by insects for the detoxification of gossypol present in the leaves. Feeding these transgenic leaves to caterpillars of the cotton bollworm Helicoverpa armigera caused a knockdown of this gene and thus decreased the tolerance of the insect to gossypol-containing food. Transgenic cotton plants expressing the - 50 -

Introduction and general background same CYP6AE14-dsRNA later showed to acquire an increased tolerance against the same cotton bollworm (Mao et al., 2007; 2011).

Recently, more research was done towards the mode of action of the RNAi effect in Diabrotica by a research group from Monsanto (Bolognesi et al., 2012). They tested dsRNA delivery by oral feeding, observed what happened to the dsRNA after ingestion and also investigated the importance of the dsRNA fragment length. They found that DvSnf7 dsRNA was taken up by the midgut cells of Diabrotica after delivery via a biofeeding assay and that the signal was carried on to other tissues after 24h. The dsRNA delivery caused high mortality (up to 95% after 12 days compared to no mortality in the control).

Lepidoptera, which are among the most devastating agricultural pests, have proven to be a challenge when it comes to RNAi. In many lepidopteran species the technique exhibits a low or at most a very variable efficiency. A large metastudy on the success of RNAi in Lepidoptera has been published recently by Terenius et al. (2011). Despite the problems that have been experienced with this insect order, a number of successes have been reported as well. In 2012 for example, Zhu et al. have investigated the potential of RNAi in the control of Helicoverpa armigera. They modified tobacco plants to produce dsRNA specific for EcR and noticed that feeding on these plants by Helicoverpa caused a dramatic decrease in EcR mRNA and resulted in moulting defects and larval lethality. Interestingly, when caterpillars of Spodoptera exigua, another lepidopteran species, were fed on these plants producing the same dsRNA fragment, similar effects were observed (Zhu et al., 2012). This was due to the fact that the dsRNA fragment of EcR was sufficiently conserved over both species to cause an RNAi effect in Spodoptera as well as in Helicoverpa. This highlights the importance of the target gene and the chosen region within the gene and also the possibilities that RNAi provides in targeting single species, larger families and even complete orders of insects based on sequence conservation. In 2009, Whyard et al. (2009) published the results of a comparative study on the RNAi effect in four different species from four major insect orders (D. melanogaster, Manduca sexta, T. castaneum, and A. pisum) using ingestion as the delivery method. Their aim was to investigate whether RNAi could have a future as a pest control tool. The target gene in this study was vATPase. They found that silencing caused mortality in all four tested species using realistic concentrations of dsRNA. They also demonstrated that one of the main advantages of using RNAi as a pest control tool is its selectivity. - 51 -

Chapter I

Despite the body of evidence that this technique has the potential to be a tool for designing a new generation of pest control mechanisms, there are important limitations as well using RNAi-based technology for pest control. Even in species for which RNAi has been proven to be relatively efficient, some major challenges remain. One of them is the choice of the target gene and the region in this gene. Another one is the delivery method of dsRNA into the insect. Especially the latter is still a limitation for many species that should be addressed.

5.4. DsRNA delivery

One of the main challenges for successful RNAi is to find easy and reliable methods of dsRNA delivery into the cell. Three main strategies have been applied in entomological research. Delivery through injection, through ingestion or by soaking. All methods have their advantages and disadvantages and the optimal choice depends on several factors, such as the the purpose and scale of the experiment as well as the species type. DsRNA delivery by feeding for example can be very useful in high throughput experiments using insects that are easy to rear on an artificial diet. Microinjection on the other hand is very labor-intensive but has the advantage that exact volumes of dsRNA can be administered and that specific target tissues can be targeted as well. Finally, soaking does not have that much use when working with living animals apart from nematodes, but is the ideal choice for RNAi experiments with cell cultures or embryos. In this thesis, two of these delivery methods have been used in order to perform RNAi in the pea aphid, namely feeding and microinjection.

5.4.1. Feeding

In 1998, Timmon and Fire reported that C. elegans, fed on Escherichia coli bacteria expressing dsRNA, showed the same phenotypes as the corresponding loss-of-function mutants (Timmon & Fire, 1998; Timmon et al., 2001). This discovery created an extra method to introduce dsRNA into organisms for triggering RNAi. After the initial discovery in this nematode, dsRNA-mediated RNAi research by ingested dsRNA was applied in various insects such as Spodoptera exigua, Diabrotica virgifera virgifera and Epiphyas postvittana (Turner et al., 2006; Baum et al., 2007; Tian et al., 2009; Surakasi et al., 2011). Three main strategies can be used in dsRNA-feeding experiments: dsRNAs can either be expressed in bacteria or they can be synthesized in vitro, and then fed to insects either by mixing with food

- 52 -

Introduction and general background or by supplying as solution droplets. Alternatively, transgenic plants can be engineered to express hairpin dsRNAs targeting genes from insects who feed on the plants.

Oral delivery of dsRNA into insects provides several advantages. It is labor-saving, cost- effective and easy to perform (Tian et al., 2009). Also, this method is applicable for high throughput gene screening, especially genes for pest control (Kamath et al., 2000). Another advantage of oral delivery of dsRNA is that it is less invasive and also a more practical method for small insects such as aphids and 1st- and 2nd-instar larvae or nymphs (Araujo et al., 2006; Tian et al., 2009; Walshe et al., 2009). However, delivery of dsRNA by ingestion has several limitations. As observed in previous research, ingestion of dsRNA is less effective for inducing RNAi in C. elegans (Hunter, 1999) and Rhodnius prolixus (Araujo et al., 2006) than injection. Furthermore, oral delivery of dsRNA may not be suitable for all species. The efficiency of RNAi by ingestion of dsRNA varies between different species possibly due to a different gut environment. Therefore, optimization of the concentration of dsRNA used to trigger RNAi is important (Turner et al., 2006). Another limitation of oral delivery of dsRNA is that it is hard to determine the amount of dsRNA brought inside the insect through ingestion (Surakasi et al., 2011). Especially for the use of RNAi for research purposes, this could be a disadvantage.

5.4.2. Microinjection

Microinjection was already used by Fire and Mello in the early days of RNAi (Fire et al., 1998) as a way to introduce dsRNA into C. elegans. Afterwards, some easier and cheaper methods such as soaking and feeding have found their way into nematode and insect RNAi experiments, but microinjection is still a widely used and very efficient research tool to introduce dsRNA into organisms in vivo, not only for nematodes but also for arthropods. The first microinjection experiments reported in insects were with D. melanogaster. In 1998, soon after the first publication of RNAi in nematodes, Kennerdell & Carthew succeeded in downregulating the frizzled and frizzled 2 gene by intracellular RNAi. Another model organism in the insect class is T. castaneum, a stored product pest insect of worldwide importance. Injection of dsRNA, both in larvae and adults, is a widely used technique in functional genomics for this organism, not in the least because, contrary to D. melanogaster, T. castaneum shows a robust systemic RNAi response (Tomoyasu et al., 2004). In the case of larvae, injections are usually carried out dorsally on or between segments, while in adults, the

- 53 -

Chapter I tissue under the wings is the easiest location to inject the organism. RNAi experiments have, for example, been used to investigate important members of metamorphosis and development pathways, such as the nuclear receptors (Tan & Palli, 2008) and Broad (Parthasarathy et al., 2008). For Tribolium and Drosophila, perhaps the two most popular insects as far as these RNAi experiments are concerned, a number of microinjection protocols have been published setting a standard for this kind of work.

Also in Lepidoptera, dsRNA delivery by injection has proven to be successful, albeit with more difficulties compared to many insects from other classes. Most successful experiments have been achieved with B. mori and Manduca sexta but also species from other families have been successfully used in RNAi microinjection experiments. Notably members of the Saturniidae family seem to be sensitive to RNAi using hemocoel injection as the delivery method, contrary to many other lepidopteran species. An extensive overview of RNAi experiments by microinjection in Lepidoptera has been given by Terenius et al. (2011). Also in the honeybee A. mellifera, economically a hugely important insect and a model organism belonging to the Hymenoptera, injection of dsRNA has been proven successful to cause a knockdown effect (Farooqui et al., 2003; Gatehouse et al., 2004; Aronstein & Saldivar, 2005). Besides these model organisms, injection is also used in other species and orders as an important technique for introducing dsRNA into the organism to elicit an RNAi response. Successes have been published for the pea aphid Acyrthosiphon pisum (Jaubert-Possamai et al., 2007), the cockroach Blattella germanica (Martin et al., 2006; Bellés, 2010; Huang & Lee, 2011), the cricket Gryllus bimaculatus (Nakamura et al., 2008; Moriyama et al., 2008) and also in some arachnids such as the tick Ixodes scapularis (Narasimhan et al., 2004) and the two-spotted spider mite Tetranychus urticae (Khila & Grbic, 2007; Grbic et al., 2011).

The microinjection technique has both its advantages and disadvantages compared to the other methods of dsRNA delivery. One important advantage is that it allows researchers to get the dsRNA immediately to the tissue of choice or into the hemolymph and thus avoiding possible barriers such as the integument or the gut which could be a problem in feeding or soaking experiments. Another advantage is that the exact amount of dsRNA brought into an organism is known, in contrast to delivery by soaking or in some cases by feeding. However, there are some important disadvantages as well with this method. The work itself is more delicate and time-consuming than the alternatives, and it also requires some optimization. Factors like needle choice, injection volume and place of injection are very - 54 -

Introduction and general background important and differ greatly between organisms. These factors should be carefully optimized before starting any experiment.

5.5. RNAi in Hemiptera and aphids

A large amount of information and research data has already been published on the use of RNAi in Diptera (mainly Drosophila), Lepidoptera (Bombyx, Manduca) and Coleoptera (Tribolium). For Hemiptera however, the data is still rather scarce. Li et al. have recently published a review on RNAi in Hemiptera which summarizes the reported successes and discusses the different factors that influence RNAi efficiency in these sap-sucking insects (Li et al., 2013). Since Bacillus thuringiensis (Bt)-based insecticides are insufficiently effective against sap-sucking insects (Elaine et al., 2005; Liu et al., 2010; Walker & Allen, 2011; Gatehouse & Price, 2011), the development of RNAi-mediated crop protection against this group of insects could prove very useful.

One of the first successful RNAi experiments in these sap-sucking insects was the work on the Hox genes Deformed (Dfd), proboscipedia (pb) and Sex combs reduced (Scr) in the milkweed bug Oncopeltus fasciatus (Hughes & Kaufman, 2000). To examine the role of these Hox genes, the authors used microinjection as a way to get the dsRNA in the eggs and examined the embryos nine days later. The same group later published several other successful RNAi experiments on the same insect (Liu & Kaufman, 2004; Angelini & Kaufman, 2005; Angelini et al., 2005; Panfilio et al., 2006). Subsequently, successful RNAi experiments in other hemipteran insects were reported as well, such as the triatomine bug Rhodnius prolixus (Araujo et al., 2006), the whitefly Bemisia tabaci (Ghanim et al., 2007) and the tarnished plant bug Lygus lineolaris (Walker & Allen, 2010). In all of these cases, the method of delivery was microinjection and for R. prolixus also ingestion was used.

The first RNAi experiments on aphids were reported in 2006, when Mutti et al. reported that knockdown of the salivary transcript C002, involved in plant-insect interactions, caused mortality in A. pisum that are feeding on plants (Mutti et al., 2006). Soon afterwards, Jaubert- Possamai et al. published their results on the knockdown of two genes in the pea aphid. One of them was the gene coding for a ubiquitinously expressed calreticulin and the other gene expressed the gut-specific cathepsin-L (Jaubert-Possamai et al., 2007). In 2009, Shakesby et al. managed to silence an aquaporin transcript in A. pisum. In these three reported successes, - 55 -

Chapter I the dsRNA was delivered through injection immediately into the haemocoel. The first report of uptake of dsRNA through ingestion was the publication of Whyard et al. in 2009 who observed mortality after feeding the aphids vATPase dsRNA through the artificial diet. Recently, Pitino et al. (2011) also reported on the knockdown of C002 in the polyphagous pest M. persicae, the green peach aphid.

- 56 -

Chapter II: Diversity of nuclear receptors in the animal kingdom

Chapter II

Parts of this chapter have been published in:

Christiaens, O., Iga, M., Velarde, R., Rougé, P. and Smagghe, G. (2010). Halloween genes and nuclear receptors in ecdysteroid biosynthesis and signaling in the pea aphid. Insect Molecular Biology, 19 (Suppl. 2), 187-200. Contributions: All work on the pea aphid nuclear receptors; annotation, sequence analysis, phylogenetics

Shigenobu, S., Bickel, R.D., Brisson, J.A., Butts, T., Chang, C-c., Christiaens, O. et al. (2010). Comprehensive survey of development genes in the pea aphid, Acyrthosiphon pisum: frequent lineage-specific duplications and losses of developmental genes. Insect Mol Biol 19 (Suppl. 2), 47-62. Contributions: Annotation of all pea aphid NRs

The International Aphid Genomics Consortium (2010). Genome sequence of the pea aphid Acyrthosiphon pisum. PLoS Biol 23, e1000313. Contributions: Annotation and analysis of all pea aphid NRs

The International Spider mite Genome Consortium (2011). The genome of Tetranychus urticae reveals herbivorous pest adaptations. Nature 479, 487-492. Contributions: Annotation and analysis of all spider mite NRs

- 58 -

Diversity of nuclear receptors in the animal kingdom

1. Introduction

Over the past two decades, nuclear receptors from many species have been identified, characterized and their sequences have been determined. For species that have had their genome sequenced and annotated such as some model insects and vertebrates, the list of nuclear receptors (NRs) is relatively complete. For many other species however, our current knowledge is fragmentary and the characterized nuclear receptors of most species are usually limited to the EcR and/or USP/RXR since they form the receptor of an important class of pesticides, the ecdysteroid agonists. At the start of this research, the only insect genomes that had been sequenced and annotated were holometabolic insects. Diptera, such as the model insect Drosophila melanogaster and the mosquito Anopheles gambiae were among the first to have the complete genome published (Adams et al., 2000, Holt et al., 2002) and more dipteran insects followed soon. Not long after, the first lepidopteran and coleopteran insects, the silkmoth Bombyx mori and red flour beetle Tribolium castaneum respectively, also followed (Xia et al., 2004; Tribolium Genome Sequencing Consortium, 2008). However, genome information and thus also a complete list of NRs of non-holometabolic insects remained lacking.

Looking at the other arthropod groups, the available data, sequential or functional, is more scarce. The only full set of nuclear receptors that is available and described is that of the water flea, Daphnia pulex, which had its genome sequenced more recently (Thomson et al., 2009). The EcR and RXR had already been cloned in some crustaceans such as the brown shrimp Crangon crangon, the lobster Homarus americanus and the crab Uca pugilator (Verhaegen et al., 2010; Tarrant et al., 2011; Durica et al., 2002) but this was the first full set of crustacean NRs that has been described. Also in the field of the chelicerates, very little research has been done on the nuclear receptors. The EcR and/or RXR has been described for some such as the scorpion Liocheles australasiae (Nakagawa et al., 2007) and the ticks Amblyomma americanum and Ornithodoros moubata (Guo et al., 1997, 1998; Horigane et al., 2007, 2008) but apart from these, the presence of nuclear receptors is still largely unknown in chelicerates. In this annotation research, the NR sets of three arthropod species were identified and investigated, namely the hemimetabolous pea aphid Acyrthosiphon pisum, the holometabolous buff-tailed bumblebee Bombus terrestris and the chelicerate two-spotted spider mite Tetranychus urticae. Until now, no full set of NRs was available for hemimetabolous insects or for arthropods belonging to the Chelicerata. The main objective was to identify these NRs

- 59 -

Chapter II and conduct a comparative study of NRs between a holometabolic insect, a hemimetabolic insects as well as a chelicerate species

2. Material and methods

2.1. Acyrthosiphon pisum

2.1.1. Annotation of the NR genes

The 1.0 release of the A. pisum genome was used as a basis for bioinformatic analysis. NR protein sequences from D. melanogaster, T. castaneum and A. mellifera were used in TBLASTN searches against the Acyr 1.0 assembly of the pea aphid genome. After identification and localization in the genome, automatically predicted gene models were examined and compared with available EST data and also homologous sequences in other species. If necessary, they were manually edited using the Apollo Genome Annotation Curation Tool (Lewis et al., 2002). This editing was done based on the available predicted models (Gnomon, Augustus, Genscan and GeneID) together with EST data and alignments with D. melanogaster, T. castaneum and A. mellifera orthologs. Phylogenetic trees were constructed afterwards to confirm the identity of the different NRs that were found in the genome.

2.1.2. Confirmation of transcription of the NR genes

Presence of these transcripts in the A. pisum RNA was examined by RT-PCR. The pea aphids were taken from a continuous colony in the Laboratory of Agrozoology at Ghent University. A mixture of different stages of A. pisum and a collection of newborn-only aphids were used to extract total RNA using the TRI Reagent (Sigma, Belgium), based on the single-step liquid phase separation method reported by Chomczynski and Sacchi (1987). Then, cDNA was synthesized from 1 µg of this RNA in a 20 µl reaction using the First Strand cDNA synthesis kit (Roche, Germany) according to the manufacturer’s instructions. Both cDNA samples (newborn only and a mixture of stages) were used in these RT-PCR experiments. Primers were designed using Primer3 software (Rozen et al., 2000) and are listed in Table 2.1. Basic amplification conditions for most NRs were 20s at 94°C, 30s at 55°C and 90s at 72°C for 35 cycles, preceeded by a 2min annealing step at 94°C and finalized with a 7min elongation step at 72°C. For HR3, the conditions were 20s at 94°C, 30s at 57°C and 60s at 72°C for 32 cycles, - 60 -

Diversity of nuclear receptors in the animal kingdom for the EcR fragment 20s at 92°C, 30s at 54°C and 150s at 72°C for 32 cycles and for USP 20s at 92°C, 30s at 52°C and 180s at 72°C for 32 cycles. Finally, PCR products were loaded on a 1.5% agarose gel.

Product Forward Reversed Fragment size

E75 (REVERB) CACTCTTGTGAGGGATGCAA CAACGTGTTGAGCGTCCTTA 960 E78 GGAGTTCGCAAAAAGAGTGC ATAACGGTGGCAACTTCAGC 512 HR3 (ROR) AAAGCGGGAAGTTTCGAGTT TGCCTGAGTATCGGGATCTT 443 ECR CAGTTGCTGTGTCACGGTCT CACCCTGTCAGCGACATC 1394 (LXR/FXR) HNF4 AGCTGAACGACCAAGTTTCG TCTCCTGTAAAAGCGGTTCG 498 USP (RXR) TATAGAGCGTTCGGCGATTTT CGTCATGTGACGATGATTCAA 1441 HR78 TCACCCAGTCCTAAATACGC CGGTGTTCATCATTGCATTC 491 TLL (TLX) GTATGCTTTCGGGCATGAAC CTTTGAACACGGCGATTTGT 519 HR51 (PNR) GGACAAGGCACATCGTAACC AAGTTTTTCGCCCACTTGAC 552 DSF CAGCAACACCACCATCATCT TTTGGAAAGGGTCTGAAACG 441 SVP (COUP- CCAGTGTATGCAACCGAACA CAATGACCTGCGAGCTGAC 542 TF) ERR (ERR) CCTCTTCGTCGTCAAATGGT GACAATGATGCTGCTGTTGG 618 HR38 (NURR1) TCAGTATCACGAGCCAAAACC CCAGGTCTTCCAGTTTCAGG 593 FTZ-F1 (SF1) TGGCAGAGTTCACTGTTTGG TTGAGAGTTGGCTGTGATGC 504 HR39 ATGCCTGGTGTGAGCTACTG CTTTGCCCTTCTCCTTCCTT 430 HR4 (GCNF1) CCAGCGAGGTCGTTTATCTC GAGGGTGACGTGTGTGATCTT 542 KNRL-1 TCGGACCTACAACAACCTGA CTGGAGCAAGCAATGGATCT 225 EG GGACGAACCTACAACAACCTG GCAGCAGACAGTGGATCTTG 187

Table 2.1. List of primers used in the transcription confirmation of pea aphid NRs.

2.1.3. Cloning of ApEcR, ApUSP, ApE75 and ApHR3 genes

Complete coding sequence (CDS) fragments of EcR and RXR/USP, as well as partial fragments of E75 and HR3 fragments were amplified by PCR. Same A. pisum cDNA and primers, as well as amplification conditions were used for the initial PCR reactions as for the confirmation (2.1.2). The PCR products were then purified using the Cycle Pure kit (Omega Bio-Tek, USA) and were ligated into a pGEM-T vector (Promega, USA) according to the manufacturer’s conditions. This vector system is a convenient and reliable system to clone PCR products, sequence the inserts and use them directly as templates for dsRNA synthesis reactions later on. Afterwards, plasmids were transformed in competent Escherichia coli XL- 1 Blue Cells by heat shock and then plated out on a carbenicillin-containing LB agar plate. After 16 h incubation, formed colonies were checked for positive transformation by colony PCR and several of these positive colonies were subsequently purified using the Plasmid mini

- 61 -

Chapter II prep kit (Omega Bio-Tek) and sent for sequencing (Agowa, Germany). Finally, these sequences were submitted in Genbank. Accession numbers are ACR45972, ACR45970 and ACR45971 for E75, USP and EcR respectively.

2.2. Bombus terrestris and Bombus impatiens

2.2.1. Annotation of the nuclear receptor genes

Known NR sequences from multiple insects such as D. melanogaster, T. castaneum, A. pisum and the closely related honeybee A. mellifera were used in TBLASTN searches against the B. terrestris and Bombus impatiens genome databases (Bter_1.0_Scaffolds and Bimp_2.0_Scaffolds). The possible NR candidates and gene models were collected, examined by comparing them with available EST evidence and homologous sequences. Where necessary, the gene models were corrected by manual editing using the Apollo Genome Annotation Curation Tool (Lewis et al., 2002). Phylogenetic trees were constructed afterwards to confirm the identity of the different NRs that were found in the genome.

2.3. Tetranychus urticae

2.3.1. Annotation of the nuclear receptor genes

The T. urticae genome was assembled by JGI Genome Portal. Nuclear receptor protein sequences from Homo sapiens, D. melanogaster, D. pulex and C. elegans were used with the Basic Local Alignment Search Tool (BLAST) to search all the possible NRs in the T. urticae genome. A TBLASTN approach was used and after identification and localization in the genome, the automatic annotation of these genes, which was performed by EuGene3.4, GenomeThreader and SpliceMachine was examined more closely. Predicted protein sequences were aligned by CLUSTALW2/CLUSTALX2 (Larkin et al., 2007) and were examined and compared with their respective homologs. Based on these alignments and comparisons and also based on EST and RNA read data, manual corrections were made where necessary using Genome View editor or Artimini. Phylogenetic trees were constructed afterwards to confirm the identity of the different NRs that were found in the genome.

- 62 -

Diversity of nuclear receptors in the animal kingdom

2.3.2. Confirmation of transcription of the NR genes

Presence of these transcripts in the T. urticae RNA was examined by RT-PCR. cDNA was provided to us by Dr. ir. Thomas Van Leeuwen from the Laboratory of Agrozoology at Ghent University. The same procedures for cDNA synthesis, PCR and primer design that were used for the A. pisum NR expression confirmations were used here as well.

Product Forward Reversed Fragment size

E75 CCGCAATGCAGAAGGTAAAT GGTTATCGGAATGGTTGGAA 693 E78 AGCCAGAGTACAGTTCATCC GCTCGGAACCAGGATAAATG 607 HR3 CATGGGATATGGATGAAGAGG GGCAGGTTTGGGTGAAATTA 588 ECR TTCGTCCAACTTTTCCAGATG ATCACCACAGACAAGGCAAA 563 HR96a GCTGCTGAAGGTAACAATAGGC GTTGGAACTGGATTAGCTGAGG 749 HR96b CTTCATCGCCAACATCTACACC AGCTTCAGGAGACTTGAGACCA 800 HR96c TGACACATACAAGGGCAACC CTGGTAATGATCGGCATGAG 787 HR96d TGACGAATAGACTGCCATGC GACCACTTTGGGATCATTCG 797 HR96e CACTTCAACGGGTTTCATTC AGCATCGGGAGAAAAGGATT 706 HR96f GCCATCGGAAGATACTTTGG CATTGGAAGCCTGAGTTTTG 488 HR96g CAGGACTGGCGATGAATTAAC CAAGGAGGAGAACGGAGTTG 418 HR10 GATGGGTCGCAAAATAATGA ATGCTATTGGCTTTGGATGC 369 HNF4 GCCATTATTGCCCCATTG GCCCAGGTAACCAAATTCAA 704 USP/RXR 1 TTCGGGGAAGACAAGTATGTG TCGTAAGCAACCAAGTTCAGT 881 USP/RXR 2 CAGTTGCGAGGGTTGTAAGG CGCTTCGGGATTGAAAAGTA 661 HR78 AATGGGAATGCGAGCTGAC CAGTGCTGGGCGGAGTTA 824 TLL TCATCAGGCAAACATTACGG GGCAAAAGGGTGAGAAACTG 448 HR51 GCTTGCTTTTCATGGCTGTG GCTGGATTATGACCCTTTTCA 445 DSF AAGCTCAAGGGGAAGCTCAT GGGTTTACTTGTGGTGCCATT 596 SVP AACCTCTGAGCCGGACAAGA TGTTTCGTGCCCAGTCAAGG 497 PNR-like ACCTCCAACCTCATCAGCAT CAGCAACATCTTTTATCCTCCA 395 ERR AGGTTGTGCCTTGTTTGTGGTG TGCGTCCTGTATGCAAGACTGAGTGA 593 HR38 ACAGCAACAATCCCAGCAAT TTTGATGCCAGTGGTGATTC 438 HR39 CAACTTTAGGCATTGGTTTTGG TTTGGCACAACAGCTTGATT 817 HR4 GGACGAAATTCTGGAGCTGT TCTTTTCTCACTTGTCATCATGG 1620

Table 2.2 List of primers used in the transcription confirmation of Tetranychus urticae NRs.

Amplification conditions that were used for these PCRs are 20s at 92°C, 30s at 56°C and 60s at 72°C for 35 cycles, preceeded by a 2min annealing step at 94°C and finalized with a 7min elongation step at 72°C. Primers that were used for PCR are listed in Table 2.2. Where necessary, PCR products were also sequenced to clear up any doubts concerning the correct

- 63 -

Chapter II annotation of the gene. The PCR products were therefore purified using the Cycle Pure kit (Omega Bio-Tek, USA) and were then sent for sequencing (Agowa, Germany).

2.4. Phylogenetic analysis

Whole amino acid sequences, as well as ligand binding domain (LBD)- and DNA-binding domain (DBD)-only sequences for the nuclear receptors of a large number of insect species were collected from the Genbank database to be compared with the sequences of the annotated NRs in the pea aphid, spider mite and bumblebees. The chosen NR sequences were then aligned by CLUSTALW2/CLUSTALX2 (Larkin et al., 2007) and the phylogenetic trees were made by the neighbor-joining algorithm and confirmed by Maximum Parsimony trees using MEGA4 software (Tamura et al., 2007). Bootstrap analysis with 1,000 replicates for each branch position was used to assess support for nodes in the tree (Felsenstein, 1985).

3. Results and Discussion

3.1. Nuclear receptors in Acyrthosiphon pisum

3.1.1. A. pisum genome encodes 20 nuclear receptor genes

All available Gene prediction sets (Gnomon, Augustus, Genscan, GeneID) and all available A. pisum sequence data were used to identify the NRs in the pea aphid genome. The in silico detection of NRs in the genome is greatly facilitated by the strongly conserved DBD and LBD regions that characterize these NRs. Blast searches were performed using peptide sequences of all known NRs from D. melanogaster, A. mellifera and T. castaneum. As a result, an initial 20 NR sequences were identified in the A. pisum genome, representing all of the seven NR subfamilies. Predicted mRNA sequences and gene models were also manually edited if necessary using the Apollo Genome Annotation Curation Tool (Lewis et al., 2002). After further analysis, two NR0 sequences turned out to be duplicated Knirps-like (Kni-like) genes, which bring the total set of unique NR genes to 19. RT-PCR was used to confirm the presence of the transcripts in the transcriptome of the pea aphid. PCR samples were loaded onto a 1.5% agarose gel and clear bands with the expected fragment size were obtained for all nuclear receptors, except for the hr83 gene. We did not manage to get a conclusive result for this nuclear recepor despite using several different primer pairs and testing different PCR

- 64 -

Diversity of nuclear receptors in the animal kingdom conditions. These results proved that, except for this HR83, the complete set of NRs found in the A. pisum genome is expressed correctly and none of them are pseudogenes. Table 2.3 presents all the pea aphid orthologs for each of the previously annotated D. melanogaster (King-Jones and Thummel, 2005), A. mellifera (Velarde et al., 2006) and T. castaneum (Tan et al., 2008) NRs. Similar numbers of NRs (19-22) were found in other insect genomes.

3.1.2. Nuclear receptors are conserved within Insecta

All NRs are also structurally very similar to their orthologs. All of them possess a DBD and LBD, except for the NR0 subfamily which only contains a DBD. The table also shows the amino acid identity percentages between the pea aphid NR sequences and those of the other insect species. As could be expected based on previous analyses of NRs, pairwise alignments of the conserved domains of D. melanogaster and A. pisum NRs showed a very high (71-99% identity) convergence for DBDs while the LBDs were more divergent (26-97%; with 77% identity for HR39 being the second highest). The most divergent NRs were HR83 (NR2E5) and TLL (NR2E2), while SVP (NR2F3) showed the least divergence.

In general, these results prove that NRs have a very strong conservation among insects and that little difference can be seen between the holometabolic and hemimetabolic insects, both in sequence similarity and in the repertoire of nuclear receptors they possess, despite the clear differences that exist in development of both groups of insects. Even though we expected bigger differences based on the evolutionary distances of these species, all pea aphid NRs show identity percentages for its ortholog which are similar to those that were reported in earlier NR annotation publications where the NRs in T. castaneum and A. mellifera were compared with the NRs in D. melanogaster.

The NRs that are part of the 20E regulatory cascade, the ‘early’ gene E75 (NR1D3) and the ‘early-late’ genes HR3 (NR1F4), HR4 (NR2A4), HR38 (NR4A4), E78 (NR1E1) and FTZ-F1 (NR5A3) are all present in the pea aphid as well and also show the same kind of convergence as reported with other species. One remarkable observation was the extremely high conservation of the SVP-LBD among insects, much more than for the other NRs (97%, 96% and 99% compared to D. melanogaster, T. castaneum and A. mellifera orthologs, respectively).

- 65 -

AphidBase Dm/Ap Tc/Ap Am/Ap NuReBASE Name Product Drosophila Tribolium Apis ID Refseq identity % identity % identity % DBD LBD DBD LBD DBD LBD NR1D3 Ecdysone-induced protein 75 E75 NP_524133 TC_12440 GB11364 ACYPI007773 XM_001946050 95 58 95 74 95 77 NR1E1 Ecdysone-induced protein 78 E78 NP_524195 TC_03935 GB30226 ACYPI002307 XM_001952697 96 51 94 60 93 60 NR1F4 Hormone receptor like in 46 HR3 NP_788303 TC_08909 GB10650 (LOC100162388) 97 53 100 71 100 72 NR1H1 Ecdysone receptor ECR NP_724456 TC_12112 GB30298 ACYPI001692 XM_001942632 88 63 97 75 99 74 NR2A4 Hepatocyte nuclear factor 4 HNF4 NP_476887.2 TC_08726 GB11424 ACYPI009409 XM_001946893 88 74 92 78 88 NA NR2B4 Ultraspiracle USP/RXR NP_476781 TC_14027 GB16648 ACYPI005934 XM_001947055 92 53 95 72 95 75 NR2D1 Hormone receptor like in 78 HR78 NP_524203 TC_04598 GB18358 ACYPI004234 XM_001948276 87 33 92 44 93 43 NR2E2 Tailless TLL NP_524596 TC_00441 GB20053 ACYPI009360 XM_001945880 83 26 83 26 85 26 NR2E3 Hormone receptor 51 HR51 NP_725457 TC_09378 GB10077 ACYPI007601 XM_001948835 96 67 94 76 94 75 NR2E4 Dissatisfaction DSF NP_477140 TC_01069 GB14217 (LOC100161040) 93 74 91 82 84 62 NR2E5 Hormone Receptor 83 HR83 NP_649647 TC_10460 GB17656 ACYPI48102 71 - 84 14 76 28 NR2F3 Seven up SVP NP_731681 TC_01722 GB17100 ACYPI005513 XM_001943986 96 97 94 96 96 99 NR3B4 Estrogen-related receptor ERR NP_729340 TC_09140 GB11125 ACYPI009262 XM_001948964 92 52 95 57 95 58 NR4A4 Hormone receptor like in 38 HR38 NP_477119 TC_13146 GB17814 ACYPI003909 XM_001944676 99 73 97 74 96 71 NR5A3 Fuchi tarazu transcription factor 1 FTZ-F1 NP_730359 TC_02550 GB16873 ACYPI003708 XM_001945429 99 67 99 77 99 74 NR5B1 Hormone receptor like in 39 HR39 NP_476932 TC_14986 GB11634 ACYPI006350 XM_001946992 83 77 88 82 89 85 NR6A1 Hormone receptor 4 HR4 NP_001033823 TC_00543 GB16863 ACYPI008092 XM_001945691 92 58 91 76 92 74 NR0A2 Knirps-like-1 KNRL-1 NP_788552 TC_03413 GB13710 ACYPI49096 91 - 95 - 91 - Knirps-like-2 KNRL-2 “ “ “ (LOC100168450) 91 - 95 - 91 - NR0A3 Eagle EG NP_524206 TC_03409 GB18215 ACYPI48166 87 - 95 - 92 -

NR1J1 Hormone receptor 96 HR96 NP_524493 TC_10645 GB10331 not present not present NR2E6 Photoreceptor specific NR ApPNR not present TC_13148 GB17775 not present not present NR0A1 Knirps KNI NP_524187 not present not present not present not present

Table 2.3. Nuclear receptors in A. pisum. Genbank IDs of D. melanogaster and A. mellifera orthologs are shown, along with the BeetleBase IDs of the T. castaneum NRs. On the right, identity percentages between the DBD and LBD of A. pisum nuclear receptors and the DBD and LBD of D. melanogaster, T. castaneum and A. mellifera orthologs are also presented. Identity percentages for NR0 LBDs and D. melanogaster HR83 LBD are missing since these receptors lack a LBD. NA: Not available. - 66 -

Diversity of nuclear receptors in the animal kingdom

The latter phenomenon may suggest that the structure of SVP, the insect ortholog of the vertebrate chicken ovalbumin upstream transcription factor (COUP-TF), is very critical to its function and is under strong selective pressure against amino acid replacements in the LBD of the receptor. In D. melanogaster, where two isoforms of this protein are expressed, SVP has multiple reported functions. It is required for the development of four of the eight photoreceptors that develop in ommatidia of the eye (Hiromi et al., 1993; Begemann et al., 1995; Kramer et al., 1995), it is a key component in the control of cell proliferation in Malpighian tubules (Kerber et al., 1998) and it also has an important role as a regulator in the development of neuroblasts by acting upon the Hunchback/Krüppel switch necessary for neuroblast differentiation (Kanai et al., 2005). In Aedes aegypti, this protein also has an effect on the vitellogenesis by acting as a negative regulator in the ecdysone receptor complex- mediated transactivation in the fat body (Miura et al., 2002).

3.1.3. Missing nuclear receptors in A. pisum

Three NRs which were previously found in other insect species seemed to be missing in the A. pisum genome: namely the NR1 subfamily member HR96, the NR0 subfamily member Knirps (Kni), and PNR-like NR (NR2E6). HR96 is an orphan receptor belonging to the NR1 subfamily (NR1J1). It is related to the vertebrate vitamin D-receptor (VDR), PXR and CAR, both of which bind a wide variety of xenobiotics (Laudet, 1997). HR96 is proven to play a role in the response of D. melanogaster and D. pulex to xenobiotics (King-Jones et al., 2006; Karimullina et al., 2012), but a function regarding development or metamorphosis has not been identified yet. This NR is sometimes considered a part of the 20E signaling cascade and it is known that this ecdysteroid-induced NR can bind to the hsp27 20E response element. This suggests that HR96 can compete with EcR-Usp for binding to a common set of target sequences (Fisk, 1995). However, since this NR has no known hormone ligand, it is difficult to speculate on its actual function regarding the ecdysteroid cascade. The absence of this gene suggests its role in this signaling pathway as well as a possible role in the response to xenobiotic stress has become redundant in aphids or being taken over by one or more other proteins. A member of the NR2 group that was initially identified in the honey bee, the NR2E6, and an ortholog for vertebrate photoreceptor cells specific nuclear receptors (PNRs), is also missing in the A. pisum genome. This gene is missing in the Drosophila genomes as well, although it

- 67 -

Chapter II has been identified in most of the other sequenced insect genomes at this moment, including T. castaneum and B. mori. Therefore, the absence of NR2E6 in the A. pisum and Drosophila genomes seems to be a secondary loss in these lineages. A function for NR2E6 in the development of the compound eye has been proposed based on “in situ” localizations in the A. mellifera developing compound eyes. In order to distinguish the different NR2 subfamily genes found in the genome, phylogenetic trees were constructed for this entire subfamily. The NR2 subfamily tree (Fig. 2.1) clearly confirms the position of HR51, which could be mistaken for PNR-like, given both receptors contains a PNR-domain. This confirms that the PNR-like receptor found in T. castaneum and A. mellifera is missing in the pea aphid.

The third NR missing in Acyrthosiphon is Knirps (NR0A1). Most insect species, except the muscomorphan Diptera to which Drosophila belongs, seem to be missing this Knirps NR. In Apis, a third NR0 gene was found, but this seems to be a more recent duplication of knirps- like or eagle, rather than a homolog of this Drosophila knirps gene. The two other receptors in the NR0 group knirps-like (Knrl) and eagle are present in the A. pisum genome. Moreover, we identified two paralogs of knirps-like which are recently duplicated, similar as in A. mellifera and Pediculus humanus, another species belonging to the Paraneoptera, the superorder comprising of all lice (Phthiraptera), true bugs (Hemiptera) and thrips. In Drosophila, knirps has been characterized as a transcriptional repressor important for the segmentation pathway (Nauber et al., 1988). Analysis of these genes in the honey bee suggested no direct involvement during segmentation (Dearden et al., 2006), as is the case in Drosophila. However, in the case of T. castaneum, Knrl has been characterized as having specific functions during head patterning (Cerny et al., 2008).

- 68 -

Diversity of nuclear receptors in the animal kingdom

Figure 2.1 Phylogenetic tree representing the NR2 subfamily members with NR2A, NR2B, NR2D, NR2E and NR2F subfamily members from Drosophila melanogaster (Dm), Tribolium castaneum (Tc), Apis mellifera (Am) and Acyrthosiphon pisum (Ap). Sequences that were the result from this annotation work are underlined in red. This tree was constructed with the neighbor-joining method performed with the full-length protein sequences of NR2 subfamily members. The PNR-like NRs found in T. castaneum and A. mellifera are also added, although no ortholog in the pea aphid could be found. Bootstrap values as percentage of a 1000 replicates >50 are indicated on the tree. On the right, gene groups with official gene nomenclature (Nuclear Receptors Nomenclature Committee 1999) are listed.

- 69 -

Chapter II

Besides the phylogenetic analysis of the NR0 and NR2 subfamilies, the NRs from the 5 other different subfamilies (NR1-NR6) were also examined by phylogenetic analysis and compared to their homologs from several different species representing the major insect orders such as Lepidoptera, Diptera, Hymenoptera, Coleoptera, and also from Crustacea and Arachnida. This phylogenetic analysis demonstrated that many of the NRs of A. pisum show close relationship with the NRs of the human louse (P. humanus). This has been observed in the genome-based phylogeny analysis as well (International Aphid Genomics Consortium, 2010) and is not unexpected since both insects are part of the Paraneoptera superorder. For a number of NRs, the T. castaneum sequences showed high convergence with both the A. pisum and P. humanus NRs as well. The trees for EcR and RXR are shown in Figure 2.2. Here, we noticed that A. pisum NRs generally seem to show a much smaller evolutionary distance to the T. castaneum and A. mellifera orthologs than to the Diptera and Lepidoptera NRs, which often cluster together in a separate branch, in some cases even branching off before the Crustacea and Arachnida. This deviation from the normal topology for some NRs is due to a long branch attraction caused by an acceleration of evolutionary rate in the Mecopterida line (Diptera + Lepidoptera). This is consistent with the earlier findings of Bonneton et al. (2008) who have discovered that a number of NRs in Mecopterida species (Diptera + Lepidoptera), including EcR, USP, E78, HR78 and HR83 have undergone a large increase in evolutionary rate. This acceleration is especially clear in the trees for EcR and USP (Fig. 2.2). Other examples can be seen in Fig. 2.1 as well where the more conserved Tll, HNF4 and SVP show a normal phylogeny while the more rapidly evolving HR83 and HR78 exhibit a similar deviation from normal topology as EcR and USP.

3.1.4. Nuclear receptors involved in the 20E regulatory cascade in A. pisum

The 20E signaling cascade, as mentioned earlier in this work, is involved in moulting/metamorphosis and development. A possible role in insect polymorphism, which is one of the aphid’s most remarkable features, has been proposed as well, but little information on their involvement is known at this point (Hartfelder & Emlen, 2012). The NRs play a very important role in this ecdysteroid signaling pathway. Binding of 20E to the EcR-USP heterodimer is the start of this signal.

- 70 -

Diversity of nuclear receptors in the animal kingdom

Figure 2.2 Phylogenetic tree of the EcR (A) and RXR (B) for a number of arthropod species, including the Acyrthosiphon pisum EcR and RXR from this study (underlined in red), clearly showing the atypical phylogeny due to the long branch attraction caused by the increase in evolutionary rate in the Mecopterida group. Support for the branches is indicated as a percentage of 1000 bootstrap replicates of neighbor-joining. The Homo sapiens sequence used in the EcR tree is the farnesoid X receptor, the mammalian ortholog of the insect EcR.

- 71 -

Chapter II

This complex, after binding to the hormone, will immediately start inducing expression of a number of ‘early’ genes, including Broad and the nuclear receptor E75. The products of these ‘early’ gene will then be responsible for the upregulation of a set of ‘early-late’ genes, including HR3, HR4, E78 and HR39. Through FTZ-F1, this signal will then be passed on to induce expression of the ‘late’ genes. So far, most attention in this field has gone to holometabolous insects, which undergo a pupal metamorphosis stage. No extensive set of NRs for a hemimetabolous insect had been identified until now. Even though the pea aphid still undergoes several larval stages, one could wonder if there would not be any differences between the molting processes of hemi- and holometabolous insects.

Except HR96, all of the NRs involved in the 20E regulatory cascade proved to be present, not only in the genome of pea aphid, but also in its transcriptome, indicating they are expressed correctly. Only the hr96 gene, which was discussed above, is missing from this set of ecdysone-inducible NRs. Since its function in the molting/metamorphosis processes is still unclear, speculation about the implications for the entire pathway are very difficult to make. For EcR, two isoforms were found, EcR-A and EcR-B. Both products differ in the A/B- domain, as is the case with most insect EcRs including Drosophila for example.

3.2. Nuclear receptors in Bombus

3.2.1. B. terrestris and B. impatiens contain 22 NRs

The nuclear receptors in two Bombus species were identified and presented here. In both bumblebee genomes, a total of 22 NRs was discovered (Table 2.4). Very little difference was found between both sets of nuclear receptors. Even at the nucleotide level, sequences for NRs in both species were almost identical. Therefore, the focus here will be on B. terrestris NRs. The set of NRs present in bumblebees is similar to that in the honeybee A. mellifera, annotated by Velarde et al. (2006). All NRs identified in the honeybee are also present in the bumblebee genome. Furthermore, sequence identity between both species was relatively high compared to other insects. Average identity percentages between A. mellifera and B. terrestris are 99.4% for the DBD (Min 95% ; Max 100%) and 97.8% for the LBD (Min 87%; Max 100%). In comparison, average sequence identity between B. terrestris and Drosophila NRs was 91% (Min 75% ; Max 100%) and 64% (Min 29%; Max 99%) for DBD and LBD sequences, respectively. - 72 -

Diversity of nuclear receptors in the animal kingdom

NuReBASE name Product Drosophila Apis Dm/Bt Am/Bt DBD LBD DBD LBD NR1D3 Ecdysone-induced protein 75 E75 NP_524133 GB11364 100 60 100 100 NR1E1 Ecdysone-induced protein 78 E78 NP_524195 GB30226 91 58 100 98 NR1F4 Hormone receptor like in 46 HR3 NP_788303 GB10650 97 65 100 100 NR1H1 Ecdysone receptor ECR NP_724456 GB30298 89 68 100 96 NR1J1 Hormone receptor 96 HR96 NP_524493 GB10331 76 67 100 97 NR2A4 Hepatocyte nuclear factor 4 HNF4 NP_476887.2 GB11424 96 76 NA NA NR2B4 Retinoid X Receptor RXR NP_476781 GB16648 92 56 100 99 NR2D1 Hormone receptor like in 78 HR78 NP_524203 GB18358 93 29 100 100 NR2E2 Tailless TLL NP_524596 GB20053 85 42 99 87 NR2E3 Hormone receptor 51 HR51 NP_725457 GB10077 99 67 100 100 NR2E4 Dissatisfaction DSF NP_477140 GB14217 84 58 100 97 NR2E5 Hormone Receptor 83 HR83 NP_649647 GB17656 75 - 97 - NR2E6 Photoreceptor specific NR PNR-like not present GB17775 NP NP 100 94 NR2F3 Seven up SVP NP_731681 GB17100 97 99 100 99 NR3B4 Estrogen-related receptor ERR NP_729340 GB11125 96 56 100 99 NR4A4 Hormone receptor like in 38 HR38 NP_477119 GB17814 96 76 100 99 NR5A3 Fuchi tarazu transcription factor 1 FTZ-F1 NP_730359 GB16873 100 75 100 99 NR5B1 Hormone receptor like in 39 HR39 NP_476932 GB11634 88 80 97 100 NR6A1 Hormone receptor 4 HR4 NP_001033823 GB16863 97 60 100 99 NR0A2 Knirps-like KNRL NP_788552 GB13710 88 - 99 - NR0A3 Eagle EG NP_524206 GB18215 86 - 96 - NR0A4 AmKnirps-like AmKni-like GB15945 NP - 100 -

Table 2.4 Nuclear receptors in B. terrestris and B. impatiens. Genbank IDs of D. melanogaster and Ensembl IDs of A. mellifera homologs are shown. On the right, identity percentages between the DBD/LBD of B. terrestris NRs and the DBD/LBD of D. melanogaster and A. mellifera orthologs are also presented. Identity percentages for NR0 LBDs and D. melanogaster HR83 LBD are missing since these receptors lack a LBD. NP: Not present; NA: Not available

One of the most remarkable discoveries in the Apis genome was the presence of a NR which had not been found in Drosophila, namely a photoreceptor-cell specific nuclear receptor- related protein (PNR-like; NR2E6). Since its first discovery in the honeybee, it has been found in other insects as well, such as T. castaneum (Tan et al., 2008) and B. mori (Cheng et al., 2008) and also in other Diptera species such as Anopheles and Aedes. In the B. terrestris genome, a homolog of this NR2E6 gene was found as well.

3.2.2. Three E75 isoforms found in B. terrestris

Three different E75 isoforms (E75C, E75D and E75E) were identified in this NR set. In the honeybee, no information about possible E75 isoforms is available while in A. aegypti three isoforms are described (A, B and C), in D. melanogaster, Manduca sexta, and B. mori there are four (A, B, C and D) and the cockroach Blattella germanica contains five different - 73 -

Chapter II isoforms (A, B, C, D and E) (Bialecki et al., 2002; Dubrovskaya et al., 2004; Matsuoka & Fujiwara, 2000; Pierceall et al., 1999; Swevers et al., 2002b; Mané-Padros et al., 2008).

For two Bombus E75 isoforms, BterE75C and BterE75E, the difference is confined to the A/B domain. The third isoform however, BterE75D, shows an alternative A/B-domain, lacks the DBD domain and splices right into the hinge region, rendering this NR incapable of binding to DNA. Alignments with E75 sequences from Drosophila and Blattella helped us identifying the different B. terrestris isoforms. Indeed, DmE75D and MsE75D, which have been described by Dubrovskaya et al. (2004) as well as BgE75D (Mané-Padros et al., 2008) are also missing their DBD and show high conservation in the isoform-specific N-terminus with BterE75D (Fig. 2.3). We did not manage to find Bombus-isoforms similar to the A- and B- isoforms.

The role of E75 has been investigated into detail first in Drosophila and later also in the hemipteran Blattella. Isoform-specific null mutants in the fruit fly revealed distinct functions during development and specific roles in the ecdysteroid signaling cascade. DmE75A mutants die during development between embryo and pharate adults, and some of these larval mutants showed very low titers of ecdysteroids, suggesting that this receptor works as a positive feedback mechanism on ecdysteroid production. DmE75C mutants die as pharate adults or soon after adult emergence while DmE75B mutants are viable (Bialecki et al., 2002). In the hemipteran cockroach Blattella, the five different E75 isoforms show a specific 20E responsiveness as well, although the expression patterns during embryonic and nymphal development are a bit more complex compared to Drosophila. There seem to be complex repressive interactions between the different isoforms (Mané-Padros et al., 2008; Martin, 2010). However, the expression patterns of the different isoforms are more complex compared to Drosophila. In the honeybee A. mellifera, E75 was found to be involved in reproduction physiology as well, yet no different isoforms have been described so far (Paul et al., 2006).

- 74 -

Diversity of nuclear receptors in the animal kingdom

Figure 2.3 Amino acid alignment showing the 5’ end of the E75 protein isoforms in Drosophila melanogaster (Dm), Bombus terrestris (Bter), Manduca sexta (Ms) and Blattella germanica (Bg). (A) shows the specific N-terminal end of the E75D isoforms and (B) shows part of the A/B domain, the complete DBD and the start of the hinge region, including the common part of all E75 isoforms. Zinc fingers are indicated. Amino acid letters are coloured according to amino acid properties (BioEdit default colour scheme).

- 75 -

Chapter II

3.2.3. B. terrestris possesses three NR0 genes

A challenging task was to elucidate the relationships between three discovered NR0 receptors. While so far only the brachyceran flies such as the fruit fly and the house fly are regarded as possessing three different NR0 genes (Knirps, Knirps-like and Eagle), the honeybee has three NR0 genes in its genome as well. In the original honeybee NR annotation paper, the authors thought the set of NR0 genes was the same as the three Drosophila NR0 genes. However, recent genome sequencing of other insects species has revealed that most insects only possess two NR0 genes. The current consensus is that in the brachyceran flies and in the honeybee lineages, two independent gene duplications have given rise to a set of three NR0 nuclear receptors. Indeed, as the phylogenetic tree in Figure 2.4 points out, this third hymenopteran NR0 protein, named AmKni and BterNR0A here, does not seem to be very closely related to DmKni and branches off somewhere between the the Eagle- and Knrl-group.

Figure 2.4 Phylogenetic tree of the NR0 subfamily, containing all NR0 receptors of D. melanogaster (Dm), T. castaneum (Tc), Apis (Am) and B. terrestris (Bter). The sequences that were identified in the B. terrestris genome are underlined in red. The human NR0 sequences were used as an outgroup. Support for the branches is indicated as a percentage of 1000 bootstrap replicates of neighbor-joining.

- 76 -

Diversity of nuclear receptors in the animal kingdom

Additionally, the lack of LBD in this particular case makes phylogenetic analysis even more difficult as well since the presence of a moderately conserved domain often provides a good resolution in phylogenetics. One other possibility is that this third NR0 gene is a pseudogene, as was suggested by one of the automatic gene prediction algorithms. Moreover, all three genes are located very close to each other on the same scaffold, as is the case in Apis as well. Confirmation of expression by RT-PCR for example could give some clarity on this matter.

3.2.4. Annotation of HR83

A nuclear receptor that was particularly difficult to annotate was HR83. The automatic predictions gave rise to two possible gene models (HR83-RA and HR83-RB). HR83-RA contained an extra exon around the DBD while HR83-RB contained an intron further downstream. Examining the DBD of both models, it was clear that the extra exon which was present in HR83-RA was a necessary one, since it contained the highly conserved T/A-box of the DBD. However, there was EST evidence for both models and the fact that the LBD of HR83 is notoriously unconserved did not make things easier. Therefore, also since no occurrence of isoforms is reported in other species for this NR, both models were merged. Figure 2.5 gives a schematic representation of both the original models (RA- and RB- transcripts), as well as the final proposal (HR83-RD). RT-PCR will have to confirm this model.

Figure 2.5 Schematic representation of the two original gene models (RA and RB in blue), as predicted by automatic prediction algorithms and the final model (RD, in blue), chosen based on alignment and domain structure data, as well as EST support (yellow).

- 77 -

Chapter II

3.3. Nuclear receptors in Tetranychus urticae

3.3.1. T. urticae has 30 nuclear receptors

In this T. urticae genome, a total of 30 NRs were identified, representing all 7 of the known nuclear receptor subfamilies (Table 2.5). Most of these NRs were until now never identified in chelicerates. In this genome search, all available gene prediction sets (EuGene3.4, GenomeThreader and SpliceMachine) and all available T. urticae genomic databases were used to identify the NRs for this two-spotted spider mite. tBLASTN searches were performed using the available protein sequences of known NRs in H. sapiens, D. melanogaster, D. pulex and C. elegans and as a result, a set of 30 nuclear receptors was identified. Two of these NRs (HR38a and HR38b, NR4A4) were found in close proximity as tandem repeats, leaving us with 29 unique NRs. Multiple sequence alignments were made to compare these sequences to other arthropod, mammal or nematode NR sequences and to check the automatic annotations retrieved by the gene prediction software. When necessary, corrections were made to the gene models. For some nuclear receptors, fragments were also amplified by reverse transcriptase PCR (RT-PCR) and subsequently sequenced for confirmation.

When comparing this set of nuclear receptors with those found in most insect species, we notice first of all that homologs of most insect nuclear receptors are also present in this chelicerate species, suggesting that these particular nuclear receptors could be present throughout the arthropod phylum. A significant difference however between this chelicerate set of nuclear receptors and those of insects is the total number of nuclear receptors. The higher number of T. urticae NRs (30) compared to insects (19-22) is mainly the result of an expansion of the NR1 family. This is mainly due to an increase in HR96-like proteins. Additionally, a homolog for the recently discovered Dappu-HR10 (NR1M) NR in D. pulex (Thomson et al., 2009) was found in this arthropod as well.

3.3.2. Conservation of nuclear receptors within arthropods is high

Examining the sequence similarity (Table 2.5), we can see that the DBD/LBD sequence identity percentages between Drosophila and Tetranychus NRs (DBD: min 49%; max 98%; average 78.8% and LBD min 17%; max 80%; average 47%) is not so different compared to those between Daphnia and Tetranychus (DBD: min 47%; max 99% ; average 78.6% and LBD min 20% ; max 76%; average 48.9%) but somewhat lower than those found between - 78 -

Diversity of nuclear receptors in the animal kingdom two insect species such as Drosophila and Acyrthosiphon (DBD: min 71%; max 99%; average 90.7% and LBD min 26%; max 97%; average 61%). Nevertheless, these similarities prove that these domains are highly conserved throughout the arthropod phylum and even between this chelicerate species and human homologs (DBD: min 60%; max 89%; average 80.2% and LBD min 30%; max 77%; average 49.2%) the NRs show a high degree of conservation.

3.3.3. The high number of NRs is caused by a NR1 subfamily expansion

The difference in the number of total NRs between this spider mite and other arthropods lies in an expansion of the NR1 subfamily. In this group, a cluster of 8 HR96-like NRs (NR1J group) was identified, whereas insects usually have one HR96 product. In the water flea D. pulex, a similar expansion in the NR1 group was discovered as well, albeit a smaller one. Thomson et al. (2009) found five new NR1 genes previously not found in other arthropods. Two of these NRs showed no relation to any known nuclear receptor in the arthropod world (HR10 and HR11) and three others seemed distantly related to the HR96 receptors but were considered divergent enough to be considered novel receptors. They were called HR97a, HR97b and HR97g.

It is not completely clear whether all these extra HR96-like receptors in Tetranychus should really be considered HR96 orthologs or whether at least some of these warrant a new type of NR1 receptor or even a new subgroup in this NR1 subfamily. The divergence between these NRs was found to be significant and phylogenetic analysis did not offer complete clarity. Neighbour joining trees using full sequences and LBD-only sequences of the NR1 receptors suggested all eight HR96 candidates group together in one big clade with the insect HR96s, but none of them were positioned particularly close to the HR96s in Daphnia or insects (Fig. 2.6). Significant evolutionary distances between each other and also to the insect HR96s were observed Additionally, identity percentages were relatively low. The tree did however suggest that they should not be considered as homologs of the Dappu-HR97 group. At this point, based on the knowledge available, we propose that all eight extra NRs can be considered to be part of the HR96 group, but a clearer view on the position of these receptors and on the evolutionary history of this group could be provided by more available data from other chelicerate and crustacean species.

- 79 -

Chapter II

Figure 2.6 Neighbour Joining tree for the NR1 subfamily. Compressed subtrees contain sequences from D. melanogaster, A. mellifera, T. castaneum, D. pulex and the T. urticae. Sequences that were the result of this annotation work are underlined in red.. The NR4-member HR38 was used as an outgroup. On the right, nuclear rececptor subfamilies are indicated. Bootstrap values as percentage of 1000 bootstrap replicates >50 are indicated on the tree.

- 80 -

NuReBASE name Product Drosophila Daphnia Tetur# Dm/Tu Dp/Tu DBD LBD DBD LBD NR1D3 Ecdysone-induced protein 75 E75 NP_524133 442814 tetur19g00650 98 54 99 65 NR1E1 Ecdysone-induced protein 78 E78 NP_524195 442769 tetur07g04810 95 56 98 51 NR1F4 Hormone receptor like in 46 HR3 NP_788303 442731 tetur03g08440 92 42 96 40 NR1H1 Ecdysone receptor ECR NP_724456 319648 tetur01g15140 88 62 95 67 NR1J1 Hormone receptor 96a HR96a NP_524493 442778 tetur34g00750 68 42 68 43 NR1J1 Hormone receptor 96b HR96b " " tetur30g01210 59 28 53 30 NR1J1 Hormone receptor 96c HR96c " " tetur17g03630 69 42 67 44 NR1J1 Hormone receptor 96d HR96d " " tetur01g07820 56 25 56 35 NR1J1 Hormone receptor 96e HR96e " " tetur20g01820 51 28 48 28 NR1J1 Hormone receptor 96f HR96f " " tetur04g03100 54 31 50 34 NR1J1 Hormone receptor 96g HR96g " " tetur11g01960 54 19 47 20 NR1J1 Hormone receptor 96h HR96h " " tetur36g00260 49 - 49 - NR1M1 Hormone Receptor 10 HR10 - 442777 tetur01g10970 NP NP 21 55 NR2A4 Hepatocyte nuclear factor 4 HNF4 NP_476887.2 442738 tetur05g04280 86 68 82 59 NR2B4 Retinoid X Receptor RXR NP_476781 442727 tetur31g01930 92 53 92 75 " " " " tetur01g09240 95 51 94 70 NR2D1 Hormone receptor like in 78 HR78 NP_524203 442757 tetur30g02190 78 33 82 60 NR2E2 Tailless TLL NP_524596 442885 tetur08g01210 81 - 80 - NR2E3 Hormone receptor 51 HR51 NP_725457 442739 tetur01g11040 86 61 84 54 NR2E4 Dissatisfaction DSF NP_477140 442884 tetur01g02690 84 57 86 53 NR2E5 Hormone Receptor 83 HR83 NP_649647 - tetur01g07700 69 - NP NP NR2E6 Photoreceptor specific NR-like PNR-like - - tetur03g02550 NP NP NP NP NR2F3 Seven up SVP NP_731681 442743 tetur04g01460 97 80 93 76 NR3B4 Estrogen-related receptor ERR NP_729340 442810 tetur28g00490 93 52 96 47 NR4A4 Hormone receptor like in 38 HR38 NP_477119 442749 tetur10g04690 92 55 95 50 tetur10g04710 92 56 95 51 NR5A3 Fuchi tarazu transcription factor 1 FTZ -F1 NP_730359 442811 tetur08g06490 96 63 97 65 NR5B1 Hormone receptor like in 39 HR39 NP_476932 442817 tetur11g04570 85 58 89 59 NR6A1 Hormone receptor 4 HR4 NP_001033823 442822 tetur07g00140 85 34 92 48 NR0A1 Knirps KNI NP_524187 - not present NP NP NP NP NR0A2 Knirps-like-1 KNRL-1 NP_788552 290673 tetur34g00430 81 - 88 - NR0A3 Eagle EG NP_524206 - not present NP NP NP NP

Table 2.5 Nuclear receptors in T. urticae. Genbank IDs of D. melanogaster are shown, along with the Wfleabase IDs of the D. pulex NRs. On the right, identity percentages between the DBDs and LBDs of T. urticae NRs and the DBDs and LBDs of D. melanogaster and D. pulex orthologs are also presented. Missing identity percentages for the LBD of some NRs are due to the fact that these receptors have no canonical LBD. NP: Not present. - 81 -

Chapter II

It is difficult to speculate on the biological meaning of this NR1J expansion. As mentioned earlier, HR96 is closely related to the vitamin D receptor (VDR), PXR and CAR. These NRs play an important role in regulating expression of genes involved in xenobiotic/drug metabolism in vertebrates (Pascussi et al., 2008). In Drosophila and Daphnia, an involvement in the response to xenobiotics has been observed as well (King-Jones et al., 2006; Karimullina et al., 2012) and very recently, Dermauw et al. (2013) found that one of these HR96 genes (HR96h; Tetur36g00260) is involved in the adaptation of the two-spotted spider mite to a new host plant. Since the two-spotted spider mite is known for its extreme adaptability to its environment, this extra set of HR96-like receptors could be one of the many adaptations this mite has developed during its evolution to cope with various problems. Indeed, spider mites are known for their extreme polyphagous nature but also for developing resistance to pesticides very easily and rapidly. Gene expansions like these could play a role in this potential of the spider mite to adapt itself. The expansion could for example indicate that these nuclear receptors have developed more specialized functions that have been integrated in more broadly acting NRs in insects for example. On the other hand, the fact that also in Daphnia the presence of a group of similar NRs is observed could mean that this is a trait that is shared by most non-insect arthropods and that has been lost in the order of Insecta. Only more data and full sets of NRs from other non-insect arthropods will provide the necessary information to clear this up.

A remarkable observation in Daphnia was the presence of two new NRs that do not associate to any of the NR1 groups known until now. They were coined HR10 and HR11, and were assigned to new NR1M and NR1N groups, respectively. In Tetranychus, a receptor was found which clustered together with the Daphnia HR10 (Fig. 2.6). Blast searches pointed towards the vertebrate RAR (NR1B subfamily) as the closest known NR for these HR10s, but phylogenetic analysis clearly showed they form a separate clade and the evolutionary distances between these were relatively large (Thomson et al., 2009). Even both HR10 sequences of Daphnia and Tetranychus seem to be quite divergent, showing only 24% identity in total and 55/21% for DBD/LBD, respectively.

3.3.4. T. urticae possesses two distinct RXR paralogs

Another remarkable finding was the presence of two different RXR paralogs (NR2B4), whereas insects, crustaceans and also mammals have only one RXR/USP product. In the - 82 -

Diversity of nuclear receptors in the animal kingdom

Ixodid tick Amblyomma americanum, the presence of two RXR products had also been reported (Guo et al., 1998; Palmer et al., 1999). The fact that the spider mite also possesses two RXR products suggests this is a shared feature at least in the Acari subclass. Both RXRs will be analyzed in more detail in Chapter III, where the ecdysone receptor will be discussed.

3.3.5. HR83 and Tll LBDs exhibit a low evolutionary constraint

Two NRs were found which at first sight seem to be missing a LBD based on domain searches against the pfam database. Sequence similarity of their DBD points to two NR2 subfamily members; a tailless-like (Tll) product and a Hormone Receptor 83-like product (HR83). Even though these NRs do not seem to possess the canonical structure or the typical LBD-domain, we have decided to categorise them as Tll and HR83 due to the fact that the DBD is sufficiently conserved. In Drosophila and Tribolium, no typical LBD was identified for HR83 as well and the LBD for both genes is notoriously unconserved throughout the insect class. In Vertebrata, the Tll homolog TLX is considered a true orphan, meaning it does not bind to a ligand. Whether this is the case in insects as well still remains to be seen, but it could be a possible explanation for this fast evolving LBD in this receptor. For HR83, which is an arthropod-specific NR, and has no vertebrate ortholog, nothing is known about possible ligands. But whether or not these two NRs are true orphans, the loss of this possible ligand binding capacity is difficult to determine based on sequence data alone. In the NR0 group, only one NR was found in Tetranychus. Phylogenetic data suggested it is the knrl gene, which means that the Eagle NR, which is present in all insects and in the crustacean Daphnia, is missing in this chelicerate.

3.3.6. Confirmation of transcription of NR genes

In order to confirm expression of these NR genes, RT-PCR was performed. Of all the 30 NR genes found in this genome, 25 genes were picked up by PCR and showed clear bands with the expected fragment size on agarose gel (data not shown). Despite several attempts using different primer pairs and a wide range of PCR conditions, we did not manage to pick up four receptors for expression confirmation, namely HR96h, FTZ-F1, KNRL and HR83. In the case of FTZ-F1 and KNRL, there is enough EST/cDNA evidence that shows both receptors are expressed in the spider mite. Furthermore, HR96h has been shown to be expressed as well in a large microarray study (Dermauw et al., 2013). For HR83 however, EST data is missing as

- 83 -

Chapter II well. This could mean that this gene is not expressed, but it could also be due to the fact that its expression is very low and/or very stage- or tissue-specific. Another possibility is that this gene is only expressed in certain conditions or after certain events. In the case of HR83, this seems to be a shared trait in many arthropods. In Acyrthosiphon, the HR83 could not be detected by RT-PCR and the same was reported for Drosophila as well (Sullivan and Thummel, 2003).

In some cases where the annotation was problematic or where there was doubt concerning the correct annotation, fragments were also sequenced to confirm the gene predictions and exon/intron structures. This was the case for TuDSF, TuEcR, TuERR, TuHR4 and TuE78. In the case of HR4 for example, a predicted protein was found which is very different from its insect homologs. Especially in the A/B-domain the divergence was relatively high and some regions that are typically conserved in insect HR4s exhibited deleterious mutations in the Tetranychus HR4 receptor. The DBD seemed to be intact but the LBD contained some large insertions and was very divergent from insect HR4-LBDs altogether (Fig. 2.7). Sequence identity between TuHR4 and DmHR4 LBDs was 34%, which is much lower than what is normally seen for this receptor between insect species. In order to confirm this surprising divergence in the sequence of BterHR4, the transcript was sequenced.

Surprisingly, the sequencing confirmed the automatic annotation, including the large insertions in the LBD, which is otherwise well conserved between TuHR4 and orthologs in other arthropods. What the implications are towards the function of this receptor remains to be seen. In insects, HR4 has been identified as a crucial element in the ecdysteroid signaling cascade. In Blattella, it is vital for the normal development during the nymphal stages (Mané- Padros et al., 2012). It is involved in downregulating some early genes such as e75 and hr3 and is required for the transcriptional activation of the late gene encoding for FTZ-F1 (King- Jones et al., 2005; Martin et al., 2010). Whether HR4 has a ligand or whether it is a true orphan receptor is not known at this time and this divergence in LBD could suggest that ligand binding is not a strict requirement for the activation of this receptor. On the other hand, in most insects a significant degree of conservation has been observed in this domain, which could suggest that the HR4-LBD is under evolutionary constraints. Also, one must be careful with such hypotheses based on sequential data alone. The 3D-structure of the ligand binding domain of the protein could tell us much more on that account.

- 84 -

Figure 2.7 Partial amino-acid alignment of the HR4-LBD region. Residues which are identical in 4 or 5 sequences are colour-shaded to indicate conserved regions. HR4 sequences used in this alignment are from T. urticae (TuHR4), D. melanogaster (DmHR4), T. castaneum (TcHR4), A. mellifera (AmHR4) and A. pisum (ApHR4). The alignment shows two regions (residues 1228 – 1360 and residues 1500-1527) which are very conserved in insects but not in T. urticae. Amino acid letters are coloured according to amino acid properties (BioEdit default colour scheme). Residues for which at least 4 of the 5 sequences are conserved have their background coloured.

- 85 -

Chapter II

In the case of EcR, the automatic annotation was very dubious and it was not immediately clear how to manually correct this model. Two automatic gene prediction models were created which possessed some elements of the EcR. It was clear that both predictions had to be combined in order to puzzle together the complete receptor. Amplification of the receptor and subsequent sequencing helped to finish this annotation with certainty. Also for E78 and ERR, which both showed some remarkable differences compared with insects (deletion in DBD and a 10AA-long insertion in the conserved LBD, respectively), sequencing was performed in order to confirm these remarkable differences with insect homologs. In the case of E78, sequencing proved that the automatic prediction and the predicted deletion in the DBD was incorrect. The insertion in the ERR LBD however was confirmed.

4. Conclusion

The annotation results presented in this chapter, which consisted for a large part of ‘in silico’ work, provided a large volume of new information about nuclear receptors in the Arthropoda phylum and delivered a basis on which further research, for example into the functionality of these genes, can be based on. We investigated the differences between holometabolous and hemimetabolous NR sets, as well as that from a chelicerate and found interesting differences. Figure 2.8 gives a general overview of the NRs found in these investigated species, as well as in some other model insects or other taxonomic groups within the Arthropoda.

One of the first general conclusions that can be drawn from these results is that on one hand the conservation in nuclear receptors at the amino acid level between distantly related species can be high, especially for the DBD and LBD, while on the other hand arthropods have created significant variation in this superfamily during evolution through lineage-specific gene losses and gene expansions. One of the most striking examples of the latter is the NR1 gene expansion in T. urticae. Lineage-specific gene expansions are considered to be important means of creating functional diversity and adaptability (Lespinet et al., 2002) and given the enormous adaptability of the two-spotted spider mite towards xenobiotics, it might not be a big surprise that this HR96-group, which is implied in the response towards xenobiotics in Drosophila and in the crustacean Daphnia (King-Jones et al., 2006; Karimullina et al., 2012), is expanded. Also in Daphnia, an extra NR1 group was found, somewhat similar though smaller to the one in Tetranychus. Moreover, one of these HR96 genes has already been

- 86 -

Diversity of nuclear receptors in the animal kingdom shown to be involved in the spider mite’s adaptation arsenal when it was observed to be upregulated more then twofold when adapting to a new plant (Dermauw et al., 2013). Whether or not we are dealing with lineage-specific expansions in some Chelicerata and Crustacea groups, or with secondary losses in the Insecta class is difficult to determine without more information from all Arthropoda groups. If these expansions are seen throughout the Chelicerata and Crustacea, it could imply that the common arthropod ancestor had multiple of these NR1J/L/M/N receptors, and somewhere during the evolution from the common bilaterian ancestor to this common arthropod ancestor, there was a duplication event of these NRs.

Figure 2.8 Schematic representation of the NR distribution in insects. Holometabolous (Holo) lineages are highlighted in pink; Para: Paraneoptera. On the bottom, a tree indicates the evolutionary relationships between the 22 groups of insect nuclear receptors. For each nuclear receptor, a colored box indicates its presence in a taxonomic group. Questions marks: no sequence data. Red cross: absence of a nuclear receptor in a sequenced genome. *: For the sake of simplicity, the groups NR1M and NR1N, as well as the three Daphnia HR97 genes are shown here as duplicates of HR96. Modified after Bonneton & Laudet (2012).

- 87 -

Chapter II

Another question was whether or not there would be large differences in the sets of NRs between holometabolic and hemimetabolic insects, especially in the ecdysteroid signaling cascade. Based on the results of this A. pisum search, no such differences were found. No NRs were found in Acyrthosiphon which were previously not identified in most of the holometabolic species as well. The only differences were two gene losses (HR96 and PNR- like). The loss of the former could have implications on moulting and metamorphosis, since it can interfere in the ecdysteroid signaling cascade, but its exact role or effect there is still unknown, making it difficult to speculate on this. At the moment, no complete NR sets from other hemimetabolous species have been fully annotated and described. Therefore, it is impossible to declare with certainty that these are secondary losses in the aphid or maybe even hemipteran lineages. Interestingly, the similarities between hemi- and holometabolic species were observed not only for the NRs, but throughout all developmental gene families. One of the important factors in pupal formation for example, broad, proved to be present in A. pisum as well, despite the fact that this insect has no pupal stage. Also for the genes involved in complex eye development for example, which is different for holometabolic insects compared to hemimetabolic species, an identical set of genes was found as in Drosophila (Shigenobu et al., 2010). This might imply that the differences in development between both insect groups are regulated more subtly than just as a result of a different repertoire of nuclear receptor genes.

Another intruiging story is the case of the two different RXR proteins found in Tetranychus. Apart from a species of ticks, which also belong to the subclass of Acari, no other arthropods are known at this time that possess two different RXR genes. Whether this is a trait that is shared amongst all chelicerates, or whether it is a more recent duplication in the Acari lineage is an interesting question because presence throughout the chelicerates could imply that the common arthropod ancestor had two rxr genes and that one was lost in the large Mandibulata clade comprising Crustacea and Insecta. Both RXR proteins will be discussed into more detail in the next chapter.

While this analysis answered some questions, many new ones concerning the evolution of nuclear receptors can be raised based on these results. Most of these can in their turn be answered based on more information from other taxonomical groups for which little is known at this time concerning NRs. The rapid progression in the field of genome sequencing will

- 88 -

Diversity of nuclear receptors in the animal kingdom make sure that in the next decade or so, many more of these gaps will be filled and more permanent conclusions can be drawn.

- 89 -

Chapter II

- 90 -

Chapter III: The ecdysone receptor heterodimer

Chapter III

Parts of this chapter have been published in:

Christiaens, O., Iga, M., Velarde, R., Rougé, P. and Smagghe, G. (2010). Halloween genes and nuclear receptors in ecdysteroid biosynthesis and signaling in the pea aphid. Insect Molecular Biology, 19 (Suppl. 2), 187-200. Contributions: All work on the pea aphid ecdysone receptor and ultraspiracle; annotation, sequence analysis, phylogenetics and modeling

- 92 -

The ecdysone receptor heterodimer

1. Introduction

The functional ecdysone receptor is a heterodimer formed by two nuclear receptors, EcR and USP/RXR. These two are by far the best described nuclear receptors (NRs), mainly because an important class of insecticides, the dibenzoylhydrazines (DBHs), targets this receptor. These compounds are known for their strong activity against many Lepidoptera while many species in the Coleoptera or Diptera seem to show a much lower susceptibility (Wurtz et al., 2000). One of the reasons for this selectivity, at least in some insects, seems to be the fact that the ecdysone receptor does not facilitate binding with these non-steroid ecdysone-agonists. Indeed, despite the strong conservation of the EcR ligand binding domain (LBD), small changes in the amino acid sequence can be enough to affect the size and shape of the ligand binding pocket (LBP) (Kasuya et al., 2003). In this chapter, we will have a closer look at both NRs that form this heterodimer and in particular EcR. The annotation, characterization and modeling of the receptor of both the insect Acyrthosiphon pisum, as well as the chelicerate Tetranychus urticae will be presented. Additionaly, in silico docking experiments with ligands 20-hydroxyecdysone (20E) and ponasterone A (PonA) will be reported as well.

2. Material and Methods

2.1. Sequence analysis

Sequences for A. pisum, Bombus terrestris and T. urticae EcR and RXR/USP were obtained during the annotation work (Chapter II). They were subsequently used in sequence alignments together with the EcR and RXR/USP sequences of several other species belonging to different Arthropoda orders, retrieved from Genbank. Alignments and the calculation of sequence identity percentages were done using the ClustalX/ClustalW algorithm (Larkin et al., 2007), as in Chapter II.

2.2. Phylogenetic analysis

A large number of EcR and RXR sequences from species throughout the Arthropoda were collected from Genbank and were subsequently aligned by ClustalW/ClustallX (Larkin et al., 2007). Neighbour joining phylogenetic trees were finally created from these alignments using

- 93 -

Chapter III

MEGA4 software. Bootstrap analysis with 1,000 replicates for each branch position was used to assess support for nodes in the tree (Felsenstein, 1985).

2.3. 3D-modeling and ligand docking of the ligand binding pockets of

ApEcR and TuEcR

Modeling and docking experiments were performed in cooperation with Pierre Rougé (Université de Toulouse). Multiple amino acid sequence alignments were carried out with CLUSTAL-X (Larkin et al., 2007) using the Risler’s structural matrix for homologous amino acid residues (Risler et al., 1998). Molecular modeling of the EcR ligand-binding domain from the pea aphid, ApEcR-LBD, and from the two-spotted spider mite TuEcR-LBD, were performed on a Silicon Graphics O2 R10000 workstation, using the programs InsightII, Homology and Discover3 (Accelrys, San Diego CA, USA). The atomic coordinates of Tribolium TcEcR-LBD in complex with Ponasterone A (RCSB Protein Data Bank code 2NXX) (Imewa et al., 2007) were used to build the 3D-model of the receptor. The high percentages of both identity and similarity ApEcR-LBD (~75% and ~90%) and TuEcR-LBD (~63.5% and ~89%) share with the template TcEcR-LBD allowed us to build rather accurate 3D models. Steric conflicts were corrected during the model building procedure using the rotamer library (Ponder et al., 1987) and the search algorithm of Homology program (Mas et al., 1992) to maintain proper side-chain orientation. An energy minimization of the final model was carried out by 150 (ApEcR-LBD) and 200 (TuEcR-LBD) cycles of steepest descent using the cvff forcefield of Discover. PROCHECK (Laskowski et al., 1993) was used to assess the geometric quality of the 3D model. In this respect, about 87% (ApEcR-LBD) and 89% (TuEcR-LBD) of the residues of the modeled EcR-LBD were correctly assigned on the best allowed regions of the Ramachandran plot, the remaining residues being located in the generously allowed regions of the plot except for three residues in ApEcR-LBD (Asn36, Glu39 and Glu42) and three residues in TuEcR-LBD (Pro189, Asp234 and Ile235) which occur in the non-allowed region (result not shown). For TuEcR-LBD, ANOLEA (Melo & Feytmans, 1998) was used to evaluate the model and only 5 residues over 237 (vs 8 over 237 for TcEcR-L-BD (2NXX code) used as a template) exhibited an energy over the threshold value. Molecular cartoons were drawn with PyMol (W.L. DeLano, http://pymol.sourceforge.net). The fold recognition program Phyre (http://www.sbg.bio.ic.ac.uk/phyre/html/index.html) (Bennett-Lovsey et al., 2008) that also - 94 -

The ecdysone receptor heterodimer used 2NXX and structurally-related proteins as templates yielded a readily superposable 3D- model for ApEcR-LBD. However, some discrepancies that essentially deal with the shape of the loops, connecting the α-helical stretches, were observed with our labmade modeled structure. Importantly, these discrepancies occur far from the groove responsible for the binding of ecdysone. Docking was performed with InsightII using Discover3 as a forcefield and we took TcEcR-LBD in complex with PonA as a template for docking. Clipping planes of ApEcR-LBD complexed to PonA were rendered with PyMol. Same procedures were used for docking PonA and 20E to TuEcR-LBD.

3. Results and Discussion

3.1. EcR

The EcR of A. pisum, B. terrestris and T. urticae were identified, analyzed and compared with the EcR sequences of other insect species. In general these receptors are very conserved and exhibit the strongly conserved domain structure which is typical for NRs: a poorly conserved A/B transactivating domain, a highly conserved DNA-binding domain (DBD), a variable hinge region, and a well conserved ligand-binding domain (LBD). Figure 3.3 represents the amino acid alignment of the conserved EcR-LBD for a number of arthropod species comprising examples from several insect orders, as well as some crustacean and chelicerate representatives. The Hemiptera are represented by the Sternorrhyncha (Myzus persicae, A. pisum and Bemisia tabaci), the Achenorrhyncha (Nilaparvata lugens) and the Heteroptera (Nezara viridula and Orius laevigatus) suborder, the Hymenoptera by the honeybee Apis mellifera and the bumblebee B.terrestris and the Chelicerata by the spider mite T. urticae, the scorpion Liocheles australasiae and the ticks Ixodes scapularis and Amblyomma americanum.

3.1.1. A. pisum

3.1.1.1. Sequence analysis Two isoforms of EcR were found in A. pisum, namely EcR-A and EcR-B. Both isoforms differ in the 5’ end of the A/B domain, which is typical for EcR isoforms in other insect species as well. Figure 3.1 schematically indicates the difference between the two isoforms. EcR-A exhibits a longer A/B domain (226AA) compared to EcR-B (161AA). The end of the

- 95 -

Chapter III

A/B-domain, which exhibits some conservation among arthropods is the same for both isoforms.

Figure 3.1 Schematic representation of both ApEcR isoforms, highlighting the alternative A/B domains.

Besides this difference in the A/B-domain, both isoforms are identical in the rest of their sequence. ApEcR has the typical conserved DBD and LBD, similar to its orthologs in other insect species. The EcR-DBD shows the typical C4 zinc finger domains in this protein including the highly conserved P-box, the D-box and A/T-box. These highly conserved stretches in the DBD, as well as the zinc finger cysteine residues are indicated on the DBD- alignment (Fig 3.2).

Figure 3.2 Sequence alignment of the ecdysone receptor DNA-binding domains (EcR-DBD) of a number of Arthopoda species, including the three studied in this research. P-box, D-box and T/A-boxes are highlighted and zinc finger cysteine residues are indicated by red dots. Amino acid letters are coloured according to amino acid properties (BioEdit default colour scheme)

- 96 -

Figure 3.3 Sequence alignment of the ecdysone receptor ligand-binding domains (EcR-LBD) of several Arthopoda species, including the three EcRs studied in this research. Red dots indicate the amino acids involved in the ligand binding in the EcR-LBD (Kasuya et al. 2003). Green dots indicate hydrophilic, hydrophobic or aromatic residues involved in the ligand anchorage into the pocket. The twelve alpha-helices are also indicated on the alignment. Amino acid letters are coloured according to amino acid properties (BioEdit default colour scheme). Bars on the left indicate the insect orders (Yellow: Diptera; Green: Lepidoptera; Purple: Coleoptera; Orange: Hymenoptera; Red: Hemiptera) or other arthropod clades (Blue: Crustacea; Black: Chelicerata) Chapter III

Some minor differences were observed in the ApEcR-DBD compared to other arthropods, such as the Ser residue at position 398 in this alignment where Ala is found in most arthropods. At the end of the D-box, ApEcR contains two Asn residues and thus follows the normal D-box sequence for non-Mecopterida. The Mecopterida typically contain a His-Ala or a Arg-Ala couple in that position. Besides these small differences, the sequence was found to be largely identical to those in other insects, especially other non-Mecopterida. Identity percentages for the DBD were 88%, 97% and 99% with D. melanogaster, T. castaneum and A. mellifera orthologs respectively.

For the EcR-LBD, we scored a strong conservation with 63%, 75% and 74% identity compared with the D. melanogaster, T. castaneum and A. mellifera orthologs, respectively. As expected, ApEcR-LBD exhibits a strong sequence conservation towards orthologs in the Hemiptera order. Moreover, comparison with the closely related aphid Myzus persica even revealed a 100% identical LBD on amino acid level. A closer look at the LBD showed us that for the pea aphid, some generally very conserved residues which are important in ligand- receptor interactions nonetheless showed a mutation (e.g. Gln16, Tyr53, Met227, Thr228 en Val235).

3.1.1.2. Modeling and docking

The obtained sequence data for ApEcR were used to perform an in silico reconstruction of the LBD and to try and predict the interactions between the receptor and the ecdysteroid ligand. The 3D-model built for ApEcR-LBD exhibits the canonical structural scaffold of the EcR- LDBs, made of twelve α-helices associated to a short hairpin of two antiparallel β-strands (Fig. 3.4A). In addition, docking of PonA into the hormone-binding groove of ApEcR-LBD revealed a binding scheme similar to that found for other EcR-LBD (e.g. from the beetles Leptinotarsa decemlineata, Tenebrio molitor and Anthonomus grandis) (Billas et al., 2003; Soin et al., 2009). Upon docking, the alkyl chain of the hormone becomes inserted in one of the two pockets located at the bottom of the hormone-binding groove (Fig. 3.4B). A network of nine hydrogen bonds connects the hormone to residues Glu20, Met56, Thr57, Ala 112 and Tyr122, forming the binding groove (Fig. 3.4C). Additionally, stacking interactions with aromatic residues Phe111 and Trp238 help to complete the interaction.

- 98 -

The ecdysone receptor heterodimer

Figure 3.4 A: Ribbon diagram of the modeled ApEcR-LBD. The twelve α-helices and the two β-strands forming the 3D-structure are labeled and differently colored. N and C indicate the N-terminal and C- terminal ends of the polypeptide chain, respectively. B: Clipping plane across the ecdysone-binding groove showing the insertion of the alkyl chain of ponasterone A (PonA; represented in pink stick) in one () of the two pockets located at the bottom of the groove. C: Network of hydrogen bonds (black dotted lines) anchoring PonA (pink stick) to amino acid residues forming the hormone-binding groove of ApEcR-LBD. Aromatic residues involved in stacking interactions with PonA are colored orange.

3.1.2. B. terrestris

Two isoforms of BterEcR were discovered (BterEcR-A and BterEcR-B). Unsurprisingly, these two isoforms differed in the unconserved A/B-domain. The isoforms have a similar structure compared to the EcR isoforms found in the pea aphid and many other insects, where the A/B-domain of both isoforms was found to be different for the most part (Fig. 3.1). Only about 50AA near the end of this domain, close to the DBD, are shared by both isoforms. The

- 99 -

Chapter III

DBD of B. terrestris EcR shows no unexpected elements and is almost identical compared to other non-Mecopterida insects (Fig. 3.2). Sequence identity for the DBD was 89% and 100% compared to D. melanogaster and A. mellifera respectively. The B. terrestris EcR LBD is virtually identical to the A. mellifera EcR LBD (96% sequence identity on aminoacid level). Sequence identity compared to D. melanogaster was 68% (Fig. 3.3).

3.1.3. T. urticae

3.1.3.1. Sequence analysis The TuEcR ORF contains 483 amino acids, which is similar to the EcR in insect species. The sequence shows a high degree of conservation in general compared to its insect orthologs, both in the DBD and the LBD. Figure 3.2 shows a highly conserved TuEcR-DBD with very few mutations at the amino acid level. The only minor differences that were found in comparison with other arthropod EcR-DBDs were situated in or around the A/T-box (Asn453, Ile469, Ser473 and Arg475) which is generally the region where most mutations in the DBD are observed. The zinc finger structures, P-box and D-box are identical compared to most arthropod EcR sequences. The LBD unsurprisingly shows a much higher degree of divergence (Fig. 3.3). Besides the changes in regions in the LBD where the evolutionary constraints are known to be weaker, we also found a number of residues which are usually conserved in the Arthropoda and which are critical in ligand-binding or for anchoring the ligand in the binding pocket are substituted in the two-spotted spider mite (Leu16, Met50, Ala57, Cys101 and the residues between 132-137). Helix 12 exhibits the typical non-insect arthropod sequence, ending in WDIQE.

3.1.3.2. Modeling and docking

Whether the above mentioned changes in residues which are considered to be involved in ligand binding have any implications on the structure or even the correct functionality of the LBP cannot be determined from sequence data alone. In silico prediction models however can give a first indication of the structure of the receptor, as well as the ligand-binding possibilities.

- 100 -

The ecdysone receptor heterodimer

Figure 3.6 A. Ribbon diagram representation of the overall three-dimensional structure of the modeled TuEcR-LBD. The twelve -helices building the three-dimensional fold of the receptor are differently colored and numbered H1-H12; the two short strands of -sheet are colored purple and numbered 1 and 2. N and C correspond to the N- and C-terminus of the polypeptide chain, respectively. B. H-bonding network (yellow dashed lines) connecting PonA (pink stick representation) to the amino acid residues forming the ecdysteroid-binding groove of TuEcR-LBD. Aromatic residues Tyr109 and Tyr120 (orange stick representation) create some stacking interaction with the hormone. C. Clip into the ecdysteroid- binding groove of TuEcR-LBD (white dotted line) showing the docking of ponA (pink stick representation) into the groove. Residues participating in the H-bonding (Gu22, Thr55, Thr58, Ala110, Tyr120) and stacking interactions (Tyr109, Tyr120) with PonA are colored blue and orange, respectively. H-bonds are in dashed red lines. The pocket harboring the aliphatic chain of PonA is indicated by a star (). D. Clip into the ecdysteroid-binding groove of TuEcR-LBD (white dotted line) showing the docking of 20-hydroxyecdysone (20E) (pink stick representation) into the groove. The O atom at C25 of 20E protruding out of the groove is blue colored (). Hydrophobic residues (Met93, Leu228, Leu232, Trp236) bording the pocket harboring the aliphatic chain of ecdysteroid are colored orange.

- 101 -

Chapter III

According to the limited amino acid changes observed between the amino acid sequences of TcEcR-LBD and TuEcR-LBD, the latter also consists of twelve -helices tightly packed around a ligand-binding groove that specifically anchor PonA and other ecdysteroids (Fig. 3.6A). Nine amino acid residues of TuEcR-LBD participate in the binding of PonA through a network of 8 hydrogen bonds (Glu22, Thr55, Thr58, Ala110 and Tyr120) and stacking interactions (Tyr109, Tyr120 and Trp236) (Fig. 3.6B). This binding scheme is similar to that observed with EcR-LBD of other insects.

Upon binding, the aliphatic chain of PonA becomes anchored to a pocket located at the extremity of the groove (Fig. 3.6C). In spite of a bulkiness equivalent to that of PonA, 20E cannot be properly accommodated by the ecdysteroid-binding groove of TuEcR-LBD. Docking of 20E to the ecdysteroid-binding groove resulted in a severe steric hindrance due to the narrow character of the pocket harboring the extremity of the aliphatic chain of the hormone (Fig. 3.6D). The O-atom of the hydroxyl group linked to the C25 of 20E protrudes out of the pocket. In addition, the O-atom is faced to highly hydrophobic residues (Met93, Leu228, Leu232, Trp236) susceptible to prevent the anchoring of the OH-containing aliphatic chain. As a result, no binding of 20E is suspected to occur to TuEcR-LBD. The fact that TuEcR seems to have problems binding 20E is very surprising and had not been reported earlier for other species. It could suggest T. urticae EcR uses another ecdysteroid as ligand than 20E.

3.1.4. Phylogeny

A phylogenetic tree was built based on sequences from a large number of arthropod EcRs which are known at this point (Fig 3.5). The A. pisum EcR branches off just before most other Hemiptera NRs and other non-Mecopterida insects. B. terrestris EcR branches together with A. mellifera, as expected. The T. urticae EcR remarkably branches off before all other known Arachnida while it belongs to the Acari, just like Amblyomma, Ixodes and Ornithodoros.

- 102 -

The ecdysone receptor heterodimer

Fig 3.5. Neighbour Joining tree for EcR sequences. Compressed subtrees contain sequences from most known Diptera and Lepidoptera EcR sequences. Human NR1H receptors were used as an outgroup. Sequences that were the result from this annotation work are underlined in red. Support for the branches is indicated as a percentage of 1000 bootstrap replicates of neighbor-joining.

- 103 -

Chapter III

3.2. RXR/Ultraspiracle

The retinoid-X-receptor (RXR), or Ultraspiracle (USP) is well known for its role as a heterodimerisation partner for many nuclear receptors. The best known partnership is with EcR, forming the functional ecdysteroid receptor. However, interactions with many other NRs have been described as well, mainly in Vertebrata. One of the most important regions that play a role in this dimerization is the region between helices 7 and 11 in the LBD of NRs. Helices 9 and 10 are the most important helices for dimerization but some residues on H7 and H11 are also involved. The DBD, and in particular the zinc fingers, play a role in the dimer stability as well.

Whether or not RXR/USP is ligand-binding is not clear and might depend on the phylogenetic position of the insect. As mentioned in Chapter II, Diptera and Lepidoptera (Mecopterida) have undergone an increase in evolutionary rate and recently it was shown that EcR and USP showed an even higher increase during the early divergence of this clade (Bonneton et al., 2003; 2006). In USP, the divergence is so considerable that the USP of taxonomical groups close to the Mecopterida, such as beetles (Coleoptera) and bees (Hymenoptera) exhibit a sequence which is more similar to the vertebrates than the Mecopterida. This can be seen throughout the complete sequences but is best exemplified by the two large inserts between helices 1 and 3 and helices 5 and 6 (Fig. 3.8) in the mecopteridan species. RXR has recently been shown to be able to bind 9-cis retinoic acid in the more basal insect Locusta migratoria (Orthoptera) (Nowickiy et al., 2008), as is the case in vertebrates (Oro et al., 1990). Contrary, in more evolved insect clades such as bugs and beetles, structural data from the USP shows that the LBP is missing, implying that USP is an orphan receptor and works in a ligand- independent manner in these species. In Mecopterida however, a large LBP has been discovered suggesting USP might have a yet unknown ligand in that clade (Iwema et al., 2007). The fact that this receptor has two different names within the Insecta is a consequence of this increase in evolutionary rate as well. The insect NRs were for the mostpart first identified in Mecopterida, mainly Drosophila and Bombyx, and this early research showed a RXR homolog which was thought to be an orphan receptor in insects for a long time. Hence, the USP was first considered to be a related, though non-orthologous, receptor and was given another name. Over the years, some have suggested keeping the name USP for all insects while others suggested it should be used in Mecopterida-only given their sequence differences - 104 -

The ecdysone receptor heterodimer with other insects, but no consensus has been agreed. As a result, both names are often used for the same receptor in the same species. We decided to use USP for the insect receptor and RXR for the chelicerate RXR ortholog. The RXR/USP for the three species was identified and characterized. Figures 3.7 and 3.8 represent amino acid sequence alignments of the DBD and LBD regions, respectively, of these RXR/USP sequences compared to some orthologs in other arthropod species.

Figure 3.7 Sequence alignment of the ecdysone receptor DNA-binding domains (EcR-DBD) of a number of Arthopoda species, including the three studied in this research. P-box, D-box and T/A-boxes are highlighted and zinc finger cysteine residues are indicated by red dots. Amino acid letters are coloured according to amino acid properties (BioEdit default colour scheme).

3.2.1. A. pisum

The ApUSP-DBD shows 92%, 95% and 95% sequence identity compared to D. melanogaster, T. castaneum and A. mellifera orthologs respectively. The A-, D- and T-boxes as well as the other zinc fingers are very well conserved, especially compared to other non-Mecopterida insects. The ApUSP-LBD, while also conserved (53% 72% and 75% identity compared to D. melanogaster, T. castaneum and A. mellifera respectively), does show some mutations in otherwise very conserved residues over all arthropods (Leu325, Arg328, Asp456, Asp458, Glu466 and Gly502).

- 105 -

Figure 3.8 Sequence alignment of the retinoid-X-receptor ligand-binding domains (RXR-LBD) of several Arthopoda species, including the three RXRs studied in this research. The twelve alpha-helices are indicated on the alignment. Amino acid letters are coloured according to amino acid properties (BioEdit default colour scheme). - 106 -

The ecdysone receptor heterodimer

Especially the substitution of the acidic polar residues Asp and Glu by the nonpolar Gly502 is interesting since this position is usually very conserved and is considered as a contact point in the dimerization surface. Whether or not this has any implications towards the possible structure of this heterodimer is impossible to say based on sequence data alone.

3.2.2. B. terrestris

The Bombus USP shows a high degree of identity with that of A. mellifera (100% and 99% identity for DBD and LBD, respectively) which has been identified first by Velarde et al. (2006). While the role of the liganded USP/EcR in moulting, metamorphosis and other juvenile development processes is well documented, they found that the USP is expressed in the adult brain of the honeybee and that the expression of this receptor changes, both in quantity and localization, during the development of foraging. The authors suggested that this receptor could therefore also play a role in social behavioural development of the adult worker bee. Similar to the pea aphid, the BterUSP-DBD shows only minor changes compared to the sequence in other arthropods (e.g. Ser228). In the LBD, only one remarkable mutation was found at a position where the conservation is usually very high (Gly525).

3.2.3. T. urticae

One of the most remarkable findings of the annotation work was the presence of two RXRs in the chelicerate two-spotted spider mite T. urticae. To our knowledge, only one case of two RXRs in an arthropod species had been reported so far. In 1998, Guo et al. reported the presence of two RXRs in the ixodid tick A. americanum. This implies that the presence of two RXRs could be a trait which is shared at least in the subclass of Acari. Whether or not Arachnida or Chelicerata also possess two different RXRs remains to be seen.

Figure 3.9 Schematic representation of both Tetranychus RXR proteins and the different NR domains, indicating the identity percentages in the A/B-domain, DBD and LBD of both receptors.

- 107 -

Chapter III

A search in the Genbank database also revealed the presence of two different RXRs for Rhipicephalus pulchellus, another ixodid tick, which were retrieved from transcriptome sequencing. However, these two seem to be different isoforms rather than different receptors. Both TuRXRs did not show a very high conservation between them. The DBD, LBD showed 93% and 81% sequence identity, respectively, but the unconserved A/B-domain and the variable hinge domain showed very little conservation, resulting in 68% sequence identity for the full sequence (Fig. 3.9). Both genes were also found far from each other in the genome. The divergence at sequence level was comparable to that which is observed between completely different NRs. This divergence rules out recent gene duplications and could suggest that the gene duplication which gave rise to both RXRs happened much earlier in the Chelicerata lineage. In order to confirm this hypothesis, more information on RXRs in other chelicerates is necessary though. Both RXRs showed much more similarity to the vertebrate RXRs (77% and 74%) than to the D. melanogaster USPs (53% and 51%). The same was the case for the A. americanum RXRs (Guo et al., 1998; Palmer et al., 1999). This is another example of the increase in evolutionary rate that the Diptera have undergone. In fact, it has been documented that this increased evolutionary rate especially affects the USP in Mecopterida, resulting in a significant degree of divergence (Iwema et al., 2009).

When comparing the sequences of the DBD for both RXRs, a few minor differences in the least conserved parts of this region were observed which are highlighted in Figure 3.9. In the LBD, the divergence between both sequences is much higher. Although most of the substitutions in this domain are located in the least conserved regions of the LBD, some of them (e.g. Cys459 and Asn567 in TuRXR1 and Ala412 and His505 in TuRXR2 in Figure 3.8) are located in normally very conserved parts.

- 108 -

The ecdysone receptor heterodimer

Figure 3.9 Sequence alignment of the two T. urticae RXRs. In the bottom sequence (RXR2), only the amino acid changes compared to the upper sequence (RXR1) are mentioned. Identical residues are indicated by a dot. Amino acid letters are coloured according to amino acid properties (BioEdit default colour scheme).

Comparing both T. urticae RXRs to those in A. americanum, we did not notice an exceptionally high degree of conservation between these two chelicerate RXRs (Table 3.1). The alignment also confirms this observation (data not shown). Furthermore, there were no similarities in the alignment which could help us determine which A. americanum RXR is the ortholog of which T. urticae RXR. The sequence identity percentages of DBDs and LBDs indicate that the AamRXR1 is more closely related to both TuRXRs than AamRXR2 is to both TuRXRs, although the difference is not that big. The A/B-domain does not give any clear indication. The fact that no obvious orthology within both couples can be found is a bit unexpected since the gene duplication which gave rise to these RXRs possibly happened early in the Chelicerate lineage.

- 109 -

Chapter III

TuRXR1 - DBD TuRXR2 - DBD AamRXR1 - DBD 93% 92% AamRXR2 - DBD 89% 91%

TuRXR1 - LBD TuRXR2 - LBD AamRXR1 - LBD 75% 70% AamRXR2 - LBD 72% 67%

Table 3.1 Identity percentages between the DBD and LBD of Tetranychus and Amblyomma RXRs

The residues which are considered to be part of the heterodimerization interface hydrophobic core of the Heliothis EcR/USP dimer (positions 538-539 and 542-543) were found to be conserved throughout Arthropoda, suggesting these are critical for heterodimerization and are under strong evolutionary constraint. The three residues which are responsible for the USP L8-9/EcR H7 interaction in the same Heliothis dimer showed more divergence. The first two (positions 502 and 504 in Figure 3.8) seemed to be rather conserved in the type of amino acid, showing only substitutions to other structurally similar amino acids. The third residue however, which is relatively conserved within the Mecopterida lineage (residue 509 in Figure 3.8), showed significant divergence in other arthropod sequences. Other interesting residues are Ser576 and Glu576 in RXR1 and TuRXR2 respectively. These residues are part of the important 9 amino acid long AF-2 sequence and are considered critical for transactivation. In most insects, this position is usually occupied by a glutamic acid residue but in other chelicerates as well as most Crustacea, this is not the case.

Together with the genome sequencing, RNAseq data was provided for each gene model for different stages as well as different host plants (http://bioinformatics.psb.ugent.be/orcae/). The expression patterns for both RXRs showed some differences as well. Both RXR1 and RXR2 are expressed in all four stages (embryo, larva, nymph and adult) but the data clearly showed that the expression of RXR1 is higher than that of RXR2 in all stages except the larval stage. The biggest difference in expression between both genes was seen in the embryo, where RXR1 shows a 4-fold higher expression than RXR2. Also, when the different stages for one receptor were compared, we noticed that the expression of RXR1 was highest in the embryo and the larva while the expression of RXR2 was mainly seen in larvae and nymphs.

- 110 -

The ecdysone receptor heterodimer

3.2.4. Phylogeny

Fig 3.10. Neighbour Joining tree for RXR/USP. Sequences that were the result from this annotation work are underlined in red. Compressed subtrees contain sequences from most known Diptera and Lepidoptera RXR/USP sequences. Support for the branches is indicated as a percentage of 1000 bootstrap replicates of neighbor-joining.

- 111 -

Chapter III

A large number of RXR sequences from a wide range of arthropods and vertebrates were collected and aligned with each other. Based on this alignment, a phylogenetic tree was built using the neighbor joining algorithm (Fig 3.10). The A. pisum and B. terrestris USPs group together with the hemipteran Nezara viridula and the hymenopteran A. mellifera, respectively, and thus show the expected phylogeny. T. urticae RXRs however form a separate clade with the spider Agelena silvatica within the clade of Chelicerata while they would be expected to group together with the A. americanum RXRs. The evolutionary distances between RXRs/USPs in the Chelicerata and Insecta were found to be longer than between the different vertebrate RXRs. Since the split of Teleostomi species into the Osteichthyes (bony fishes) and Tetrapoda about 400 million years ago, the RXRs have undergone very little divergence in this group while in insects, much more divergence is observed.

4. Conclusions

This chapter provided a deeper analysis of the EcR and RXR/USP sequences from A. pisum, B. terrestris and T. urticae. We have examined these proteins at sequence level and also provided predicted models for the pea aphid and spider mite EcR. For A. pisum and B. terrestris, two isoforms for the EcR were found, similar to the isoforms found in most insects for this receptor. No major changes for the three species were found compared to the EcRs in related species. The protein modeling and subsequent docking did however suggest that TuEcR could be incapable of binding 20E since the OH group at C25 does not seem to fit in the ligand binding pocket. In addition, the O-atom was found to be faced towards highly hydrophobic residues which further interferes with the anchoring of the OH-containing aliphatic chain of the steroid. This suggests that T. urticae might have a different moulting hormone than 20E.

The USP of both insect species did not show great differences with those of other insects. In the spider mite however, two distinct RXR proteins were discovered. These two RXRs did not exhibit a high conservation between each other and comparison with the two known A. americanum RXRs also revealed a high degree of divergence. The function of these two RXRs and how they cooperate is unknown. Regions in the LBD which are important for dimerization were found to be well conserved in both TuRXRs however, suggesting that they both have retained their function as heterodimerization partners.

- 112 -

Chapter IV: Ecdysone biosynthesis in Acyrthosiphon pisum and Tetranychus urticae

Chapter IV

Parts of this chapter have been published in:

Christiaens, O., Iga, M., Velarde, R., Rougé, P. and Smagghe, G. (2010). Halloween genes and nuclear receptors in ecdysteroid biosynthesis and signaling in the pea aphid. Insect Molecular Biology, 19 (Suppl. 2), 187-200. Contributions: All work on the pea aphid Halloween genes; annotation, analysis, phylogenetics and modeling

- 114 -

Ecdysone biosynthesis in Acyrthosiphon pisum and Tetranychus urticae

1. Introduction

In the previous chapters, we have taken a closer look at the nuclear receptors in arthropods and the ecdysone receptor in particular. One of the best understood pathways in which a number of these receptors are involved, is the ecdysteroid signaling cascade which regulates processes such as moulting and metamorphosis. The onset of this pathway is the binding of the functional ecdysone receptor to its ecdysteroid ligand. In insects, this hormone is usually 20-hydroxy-ecdysone (20E). In this chapter, we will have a closer look at the synthesis of these ecdysteroids in arthropods and the genes that are involved in the transformation of cholesterol to the functional ecdysteroid, a group of Cytochrome P450 (CYP) enzymes called the Halloween genes. The names of these genes originates from the fact that embryos carrying mutations in these genes showed a failure in cuticle formation and were said to ‘resemble ghosts’. The CYPs can be divided into four major clades (Mito CYP2, CYP, CYP3 and CYP4) and the Halloween genes can be found in two of them, the mitochondrial CYP clan, comprising all CYPs that are found in the mitochondrial genome, and the CYP2 clan.

Figure 4.1 Summary of the biosynthesis of 20-hydroxyecdysone (20E) presenting the different Halloween genes (boxed) and their function in the ecdysteroid biosynthesis pathway. The intermediate products and their structure are also presented. The changes for which the enzyme is responsible are highlighted in yellow.

- 115 -

Chapter IV

These enzymes, well known for their monooxygenase activity, constitute one of the largest families and are distributed throughout a wide variety of living organisms, from bacteria to mammals (Werck-Reichhart et al., 2000). To date, four P450 enzymes, namely CYP306A1 (Phantom, Phm), CYP302A1 (Disembodied, Dib), CYP315A1 (Shadow, Sad) and CYP314A1 (Shade, Shd), involved in the ecdysteroid biosynthesis have been identified and characterized. As shown in Figure 4.1, the products of phm, dib and sad sequentially convert the precursor of E, 2,22,25-trideoxyecdysone (ketodiol), into 22,2-dideoxyecdysone (ketotriol), 2-deoxyecdysone and ecdysone (E) (Chavez et al., 2000; Warren et al., 2004; Niwa et al., 2004, 2005; Rewitz et al., 2006b). Further on, the product of shd mediates the last step of the conversion from E into 20E (Petryk et al., 2003; Rewitz et al., 2006a; Maeda et al., 2008). In addition to these, CYP307A1 (Spook, Spo), CYP307A2 (Spookier, Spok), the paralog gene of Spo, and CYP307B1 (Spookiest, Spot) are identified. They are all involved in the initial conversion process from 7-dehydrochoresterol into ketodiol, but their biochemical functions are not well understood (Namiki et al., 2005; Ono et al., 2006). Spok has thus far only been identified in Drosophila, while spot is identified in mosquitoes (Aedes aegypti and Anopheles gambiae), honey bees (Apis mellifera) and red flour beetles (Tribolium castaneum). Together, they are called the Halloween genes. Halloween genes have been identified/predicted in multiple insect species (Niwa et al., 2004, 2005; Warren et al., 2004; Sieglaff et al., 2005; Rewitz et al., 2006a,b, 2007) and the function of these genes is characterized in the fruitfly (D. melanogaster), the silkmoth (Bombyx mori) and the tobacco hornworm (Manduca sexta). In addition to insects, the Halloween genes are also identified in the crustacean genome of Daphnia pulex (Rewitz and Gilbert, 2008), suggesting a high conservation for ecdysteroid biosynthesis in the Arthropoda phylum.

2. Material and methods

2.1. Annotation of Halloween genes

Putative Halloween gene sequences were searched and obtained by TBLASTN, using the known orthologs from D. melanogaster, A. mellifera and T. castaneum against the complete scaffold collection of the A. pisum and T. urticae genomes. Whole amino acid sequences for the Halloween gene orthologs in A. mellifera, T. castaneum, D. melanogaster, A. aegypti, A. gambiae, M. sexta and B. mori were collected from the Genbank database and were used in

- 116 -

Ecdysone biosynthesis in Acyrthosiphon pisum and Tetranychus urticae the annotation of these predicted genes. The chosen sequences were aligned by CLUSTALW2/CLUSTALX2 (Larkin et al., 2007) and corrections in the predicted sequences were made where necessary based on these alignments, on EST- and RNA Seq support and based on the information that is known about the conserved domains in these proteins. Phylogenetic trees, created to confirm the identity of the different CYPs, were constructed using the same methods as in Chapter II.

2.2. Confirmation of expression

Similar to the work on the nuclear receptors (Chapter II), gene expression was confirmed by reverse transcriptase PCR. The primers used in these confirmations are listed in Table 4.1. In some cases where the annotation corrections were difficult, sequencing was done as well.

Product Forward Reversed Fragment size ApSpo1 TGTCCGTCGTGCTCCTGATAC CGCAGGACTAGTGTTGATTTCG 1235 ApSpo2 TTGTTCCCCGAAAGTGTATTCA TCCTGAGCCTTTGCGTAACTG 402 ApSpo3 CGCCGTATTATTGTTCCTGATTC CTTATATCCCATGCACGACCG 1312 ApPhm GTTTTGGATCATCGGCGTAATACT CGGACTGAGCGTTATGCCA 1441 ApDib ATCGACACGGCGACGTATTC TTCACTGTCCAGCTTTGGTCC 516 ApSad CTGCTGGTGGTGGTCGTAAGT TAAGGAAGGGTTGCAAATGGC 1105 ApShd1 GGCGCCTGTTGCATAGTAGTC GTATAAAATTCGAGTCAACGGGTGT 1504 ApShd2 CGCCTGTTGCATAGTAATCGC CCACCTTTCCGGTCTGTACG 1300 ApShd3 TTCGAAGCCATGTTTCTCTGC CCGATGTGAAATTGCTTGACC 1217

TuSpo ATGCATTGGCTTTGTGTGACTG GGCTTAGGAATCGAGATGGTTG 945 TuDib TGGTCCATTGGTTCGAGAAG TGCTGCCAATTCAGTAGCCA 502 TuSad GGAGCCGCACATCTACATGAATA AGCAGCAGCCAGAAACAAATCTAC 807 TuShd CTGTCAACTGGCGGGTCTTC GGATGAATTGAGCGACGAGC 731

Table 4.1 List of primers used for the transcription confirmation of pea aphid NRs.

2.3. Identification of PonA as a major ecdysteroid in T. urticae

Spider mite extracts were collected and subjected to biochemical analysis using HPLC– enzyme immunoassay and liquid chromatography/mass spectrometry. These analyses were done by the Amber Lab, that is part of ENVOC at the Faculty of Bioscience Engineering, UGent. These data are published in the spider mite main genome paper (Grbic et al., 2011).

- 117 -

Chapter IV

3. Results and Discussion

3.1. Halloween genes in Acyrthosiphon pisum

Three candidates for A. pisum spook (spo1, spo2 and spo3), one candidate for phantom (phm) and disembodied (dib) and also three candidates for shade (shd1, shd2 and shd3) were found in the A. pisum genome after TBLASTN searches (Table 4.2). The expression of these predicted sequences was confirmed by RT-PCR. Alignments of the different A. pisum Halloween gene candidates with those of other insect orders (Lepidoptera, Diptera, Hymenoptera and Coleoptera) show high conservation of the typical insect P450 motifs (helix-C, helix-I, helix-K, PERF-motif and heme-binding domain). Two clear Spo-like homologs, Spo1 and Spo2, were detected in the pea aphid genome. Sequence comparison at amino acid level shows that both Spo1 and Spo2 are similar (85% identity) to each other.

Name D. melanogaster A. mellifera T. castaneum AphidBase ID Spook (Spo) / AF484415 / CYP307A1/2 spookier (Spok) NM_001110990 - AAJJ01000951 ACYPI001519 ACYPI002012 CYP307B1 Spookiest (Spot) - AADG05005080 AAJJ01001163 ACYPI000716 CYP306A1 Phantom (Phm) AF484413 XM_391946 XM_963384 ACYPI006623 CYP302A1 Disembodied (Dib) AF237560 XM_001122832 XM_969159 ACYPI006729 CYP315A1 Shadow (Sad) AY079170 XM_395360 XM_965029 ACYPI000973 CYP314A1 Shade (Shd) AF484414 DQ244074 XM_967606 ACYPI008228 ACYPI006755 ACYPI009813

Table 4.2 Halloween genes identified in A. pisum. Genbank IDs for D. melanogaster, A. mellifera and T. castaneum orthologs are also presented. In the case of T. castaneum CYP307A1/2 and CYP307B1, contigs were given on which the gene is found. Idem for A. mellifera CYP307B1.

A third Spo-candidate (Spo3) was found as well. This Spo3 was very different from Spo1 and Spo2 however, showing only 38% and 39% identity, respectively. When compared with other Halloween genes from other species, Spo3 showed 33-42% identity with Spo/Spok orthologs, and 44-47% identity with Spot orthologs, suggesting Spo3 might be a Spot ortholog rather than a Spo ortholog. Sequence alignments of the Spo/Spok/Spot showed a closer relationship between Spo3 and the Spot-group than the Spo-group. However, this was not the case in all - 118 -

Ecdysone biosynthesis in Acyrthosiphon pisum and Tetranychus urticae regions or even in all conserved domains. In the Heme-binding domain for example, the Spo3 candidate showed an intermediate sequence somewhere between Spo and Spot. In the Helix-C domain however, Spo3 clearly showed a typical Spot-like sequence (Fig. 4.2). In order to confirm the identity of the annotated Halloween genes and the position of Spo3 in the Spo/Spok/Spot group in particular, a phylogenetic tree was built containing all the Halloween genes in A. pisum, together with the orthologs from a number of other insects (Fig. 4.3). This phylogenetic analysis confirmed the hypothesis that the Spo3 gene is a Spot ortholog rather than a Spo ortholog.

Figure 4.2 Partial amino acid alignments representing the regions in which the Helix-C and the Heme- binding domains are located for the Spo/Spok/Spot orthologs from A. pisum (Ap), T. castaneum (Tc), A. mellifera (Am), A. aegypti (Aa), A. gambiae (Ag), D. melanogaster (Dm), B. mori (Bm) and Manduca sexta (Ms).

Three Shd candidates were identified in A. pisum. Two of them, Shd1 and Shd2 show a high conservation (94% identity), suggesting these could be recently duplicated genes, while another Shd candidate, Shd3, only shows 66% identity with both Shd1 and Shd2. Surprisingly, Shd3 seems to be lacking most of the heme-binding domain (Fig. 4.4) which means the protein may not be functional at all. However, RT-PCR showed that the protein is expressed in the pea aphid (Fig 4.5). This leads to the hypothesis that the protein might have other functions or another role in the pea aphid than the ones attributed to the Shd proteins so far. The Shd orthologs that are responsible in the 20E biosynthesis are probably Shd1, Shd2, or both.

- 119 -

Chapter IV

Figure 4.3 Phylogenetic tree of the Halloween genes. This tree was constructed with the neighbor-joining method performed with the amino acid sequences of the whole sequences. Bootstrap values as percentage of a 1000 replicates >50 are indicated on the tree. Halloween genes from A. pisum (Ap), T. castaneum (Tc), A. mellifera (Am), A. aegypti (Aa), A. gambiae (Ag), D. melanogaster (Dm), B. mori (Bm) and Manduca sexta (Ms) were included in the analysis. The A. pisum sequences that were the result from this annotation work are underlined in red. Mito clan: Mitochondrial CYP clan; 2 clan: CYP2 clan - 120 -

Ecdysone biosynthesis in Acyrthosiphon pisum and Tetranychus urticae

Figure 4.4 Partial amino acid alignment of shade orthologs highlighting the conserved heme-binding domain. Due to a 17 amino acid deletion, this domain is missing from Ap-Sh3. Orhologs are from A. pisum (Ap), T. castaneum (Tc), A. mellifera (Am), A. aegypti (Aa), A. gambiae (Ag), D. melanogaster (Dm), B. mori (Bm) and Manduca sexta (Ms).

Figure 4.5 Analysis of expression for the different shade (shd) gene candidates using RT-PCR. The gel clearly shows that all three shade candidates are expressed, even Ap-shd3, although clearly at a lower level than the other two shd candidates. Lane 0 was loaded with a DNA marker (Mass Ruler DNA Ladder Mix, Fermentas), while lanes 1-3 contain the PCR products of each candidate. Fragment sizes were 1504 (shd1), 1300 (shd2) and 1217 (shd3).

3.2. Ecdysteroid biosynthesis in T. urticae

The survey in this chelicerate genome resulted in the identification of four of the five Halloween genes, namely Spo, Dib, Sad and Shd (Table 4.3). For Dib, two very similar paralogous genes were found in close proximity to each other. There was some difference in the 5’ end of the open reading frame between the two paralogs but overall they are virtually identical (97% identity at nucleotide level). In contrast, Spok or Spot, present in some insects, - 121 -

Chapter IV were missing in the spider mite. These three related P450s are assumed to be involved in the transition from 7-dehydrocholesterol (7dC) to ketodiol, but this part of the ecdysteroid biosynthesis is still poorly understood. Two other non-P450 genes which are nonetheless involved in the ecdysteroid biosynthesis, namely Neverland (Nvd) and Shroud (sro), were also identified. The product of the former is active in the onset of the synthesis pathway, mediating the conversion of cholesterol into 7-dehydrocholesterol (7dC). The latter, Shroud, is assumed to belong to the same so-called ‘black box’ together with the Spo/Spok/Spot group (Fig 4.1).

D. Name A. mellifera T. castaneum Tetur ID melanogaster CYP307A1 Spook (Spo) AF484415 - AAJJ01000951 tetur10g03900 CYP307A2 NM_001110990 not present

CYP307B1 Spookiest (Spok) - AADG05005080 AAJJ01001163 not present CYP306A1 Phantom (Phm) AF484413 XM_391946 XM_963384 not present CYP302A1 Disembodied (Dib) AF237560 XM_001122832 XM_969159 tetur05g02670 tetur05g02550

CYP315A1 Shadow (Sad) AY079170 XM_395360 XM_965029 tetur06g05620 CYP314A1 Shade (Shd) AF484414 DQ244074 XM_967606 tetur03g03020

Table 4.3 Halloween genes identified in T. urticae. Genbank IDs of D. melanogaster, A. mellifera and T. castaneum orthologs are also presented. In the case of T. castaneum CYP307A1/2 and CYP307B1, contigs were given on which the gene is found. Idem for A. mellifera CYP307B1.

Surprisingly, T. urticae lacks Phantom, the biosynthetic C25 hydroxylase involved in the conversion of ketodiol to 2,22-dideoxyecdysone (2,22dE). The absence of this enzyme would implicate that the spider mite uses the ecdysteroid 25-deoxy-20-hydroxyecdysone (Ponasterone A, PonA) as the moulting hormone, instead of the typical arthropod 20E. Both ecdysteroids only differ in the hydroxyl-group at the C25 position (Fig. 4.6).

Figure 4.6 Structure of ecdysteroids 20-hydroxy ecdysone (left) and Ponasterone A (right.)

- 122 -

Ecdysone biosynthesis in Acyrthosiphon pisum and Tetranychus urticae

HPLC–enzyme immunoassay and liquid chromatography/mass spectrometry confirmed the presence of PonA and the absence of 20E in the spider mite. This result was published in the spider mite genome paper (Grbic et al., 2011) and indicates that PonA might be the only or at least the main moulting hormone in this spider mite. In decapod crustaceans, PonA has been found to act as moulting homone, albeit coincidental with 20E (Chung, 2010). This is the first case where PonA is identified as the main moulting hormone without the presence of 20E. The implications towards the ecdysteroid signaling cascade are not necessarily that big since PonA is known to be a very potent ligand for all known EcRs.

Also interesting to note here is that CYP18A1, which encodes a C26 hydroxylase/oxidase involved in 20E catabolism has not been found as well. This gene is clustered together with Phantom in all insect and crustacean genomes that were studied thus far. This indicates that both genes were most likely lost together, affecting both the synthesis and the inactivation pathways of the spider mite moulting hormone. Figure 4.7 gives an overview of the different enzymes in Tetranychus involved in the ecdysteroid biosynthesis and their point of action on the steroid structure.

Figure 4.7 Overview of the different enzymes involved in ecdysteroid biosynthesis indicating the conversion they are responsible for. Elements in black are present in the T. urticae genome, while the two CYPs in red are missing (Grbic et al., 2011).

- 123 -

Chapter IV

4. Conclusion

In A. pisum, all Halloween genes proved to be present, except Spok, which is a recent duplication of Spo in Drosophila. However, the pea aphid turned out to possess also two Spo paralogs which are the result of a recent independent duplication event. Additionally, three different Shade paralogs were found as well in the pea aphid, which have never been reported before in other insects. The implications of these duplications are unknown at this moment.

In T. urticae, one of the Halloween genes, namely Phantom, was not present. This gene encodes the CYP enzyme responsible for the hydroxylation at the C25 position which means that 20E, the moulting hormone in most arthropods, cannot be synthesized. Instead, Ponasterone A would be synthesized. Biochemical analysis via mass spectrometry confirmed this hypothesis and proved that PonA is the main moulting hormone in this chelicerate.

- 124 -

Chapter V: RNAi

Chapter V

1. Introduction

In this chapter, we will report the attempts to use RNA interference (RNAi) as a functional genomics tool, to add some functionality to the nuclear receptors which were discovered in the pea aphid Acyrthosiphon pisum. In a first part, the development of the protocol which was developed for the injection of aphids with dsRNA will be described. In a second part, the results of the RNAi experiments themselves which were conducted, targeting EcR and USP will be reported and discussed. In the final part, we will discuss the factors which can influence RNAi efficiency as well as the results of some experiments in the pea aphid in which we investigated some of these factors.

2. Development of a microinjection protocol

For Tribolium and Drosophila, useful protocols are available for microinjection experiments. Information on needles, injection procedure, position on the insect to inject, immobilization strategies are all described for these model insects. For aphids however, little information can be found, so a new protocol has been devised. Here, the development of this protocol and the experiences during optimization of all these different factors are reported.

2.1. Microinjection setup

The basis is the microinjection setup (Fig. 5.2). Since only amounts ranging in the nanoliters can be injected in aphids, injections could not be performed using normal micropipettes. Therefore, a nanoinjector (Femtojet, Invitrogen) was used as a pressure source. Injected volumes are determined by the combination of a certain injection pressure and injection time. The injections were performed under a Leica stereomicroscope with the help of a Narishige micromanipulater. This is a device in which the needle can be placed that allows us to carefully move this needle in three dimensions. This helped with the precise movements needed during these procedures. This microinjection setup can be used for the injection of other insects as well.

- 126 -

RNAi

Figure 5.2 Microinjection setup consisting of a microscope, a micromanipulator and a microinjector. (Photo: O. Christiaens)

2.2. Sedation and immobilization

To facilitate injection, the aphids had to be immobilized. Using any type of glue-based method proved difficult given the fragility of these species, especially the limbs. Materials such as glue and sticky tape caused amputated legs and the agarose plates containing little gutters that are used for the injection of beetle larvae also caused significant damage to the aphid. Therefore, we went looking for a method to sedate the aphids temporarily. For many insects, including Tribolium, cool temperatures, for example by keeping them on ice, helps to calm them down and even immobilize the insects enough for these kind of procedures. However, in the case of aphids this proved insufficient. Long periods of time on ice had no or little effect on the activity of the aphids. Subsequently, the effect of diethylether, a common method of anaesthesia in insects, was tested on aphids. The sedation effect proved sufficient and, importantly, the procedure was also non-toxic for the aphids. In order to prevent direct physical contact between the aphid and ether, a 50mL falcon was modified by placing a small cage inside to create some sort of ‘gas chamber’ in which the aphids could be placed (Fig. 5.3). Only 20-30 seconds inside the falcon were enough to knock out the aphids for several minutes, which is more than enough time to be injected. The sedated aphids were placed - 127 -

Chapter V against a vertical agar surface which prevented the aphids sliding over the surface at the moment of injection.

Figure 5.3 (A) Picture of the falcon used for sedation. Some cotton wool is placed at the bottom of the falcon and placed above the cotton wool there was a modified syringe which was closed on one side by a fine mesh. A small amount of ether is then poured onto the cotton wool and aphids are placed inside the modified syringe. (B) The agarose plate on which the aphids were placed for injection. Agarose was poured in a glass petri dish and a plastic, grooved dish was floated on the agarose. After the agarose solidified, the plastic floater was taken off. (Photos: O. Christiaens)

2.3. Microcapillaries

Another important factor are the needles that are used to inject the aphids. At first, some commercially available glass microcapillary needles, such as Eppendorf’s Femtotip needles, were tried. Injection of aphids with these needles proved rather difficult however, since the tip of the needle was not really equipped to penetrate the tough cuticle of the aphid easily. Needles would bend or even break when attempting to pierce through the cuticle. A needle with a much shorter, yet stronger, tip was needed and these were found when attempting to create self-made needles using a needle puller. Glass capillaries (BLAUBRAND IntraMARK, 50µL) were used to create these needles with a two-step pulling procedure on the Narishige PC-10 needle puller (temperature settings 80 and 83 for step 1 and step 2, respectively). The needle with the shortest end of the two was used succesfully in injection experiments. One of the main drawbacks of the self-made needles is of course the lack of uniformity compared to commercially available needles and the need to calibrate the volume with each newly made needle.

- 128 -

RNAi

2.4. Needle insertion

The angle at which the needle penetrates the cuticula is also important. Not only because it determines the ease by which the needle goes through the cuticle, but more importantly, it is vital not to damage anything internally in the aphid. Especially damage to the gut could have detrimental effects to the fitness of the aphids. In theory, a very small angle, as close as possible to a parallel position along the dorsal or ventral side is best to avoid this damage. However, it is easier to insert the needle in a perpendicular way through the cuticle. A solution was found by injecting in one of the segmentation grooves on the thorax of the aphid. In practice, the angle by which to insert the needle will also differ depending on the morphology of each aphid individually because some aphids are rounder than others. The few aphid-injection papers that are published (Mutti et al., 2006; Jaubert-Possamai et al., 2007; Shakesby et al., 2009) report injecting both ventrally or dorsally. In the experiments performed in this thesis, injections were performed dorsally.

2.5. Volume

A critical factor is the injected volume. Jaubert-Possamai et al. reported (2007) that for the injection of L3 nymphs of A. pisum 23nL could be injected with limited mortality, while a 46nL injection caused a significant mortality rate (29-45%). Since we were planning to inject L2 nymphs and adult aphids as well, some injection volumes were tested in advance on their mortality. For adults, we found that volumes up to 200nL did not cause any mortality and for L2 nymphs, 20nL could be injected without any problems.

2.6. Toxicity of injected water, buffer and dye

When the technical aspects of the injection procedure were optimized, some tests were set up to investigate whether the injection of the buffers or the water in which the dsRNA is solubilized had any detrimental effects on the aphids. Also, a red food colorant dye was tested for its toxicity. This dye can be mixed with the sample that is injected in order to have a visual confirmation of the successful injection. While this dye was not used in the RNAi experiments itself, it is very useful in the optimization of these experiments and in the training of people who want to use this technique. Figure 5.4 represents an injection of a water sample mixed with this food colorant into an adult aphid. You can clearly see the colorant being - 129 -

Chapter V spread throughout the thorax and part of the head through the hemolymph. These little tests proved that both water and the Elution Buffer from the Ambion MEGAscript kit did not cause any mortality when injected. Also, the dye proved completely harmless as well.

Figure 5.4 Series of pictures taken during the injecting of an adult aphid with a sample containing a food colorant (Photos: O. Christiaens)

3. RNAi experiments

3.1. Material and Methods

3.1.1. Cloning of C002 and vATPase

Fragments for ApEcR and ApUsp were already available in a pGEM-T vector after the cloning work reported in Chapter II and a vector containing a GFP-sequence was present in the lab as well. Vectors with C002 and vATPase sequences however still had to be made. Pea aphid RNA extraction and subsequent cDNA synthesis were done in the same way as explained in Chapter II. cDNA fragments for Coo2 and vATPase were then amplified by PCR. Amplification conditions were 20s at 94°C, 30s at 59°C and 30s at 72°C for 32 cycles, preceeded by a 2min annealing step at 94°C and finalized with a 7min elongation step at 72°C. These fragments were subsequently ligated into a pGEM-T vector and cloned following the same procedures used in Chapter II. This pGEM-T vector system provides an easy way of - 130 -

RNAi cloning PCR inserts and using them as a template for dsRNA synthesis afterwards. The primers used for the initial PCR amplification are listed in Table 5.1.

Fragment Fragment Forward Reversed size vATPase TTAGCCAACACTGGAATAAACGTC CCAAACAGTCCATGCATATTATT 185 Coo2 TCTCGTCGTGTATCCAGTGC GTAGGCGAGTGAGATTCCAA 799

dsEcR1 *TGCCGACTGTACCTTACGTG *TTTCAATGGCATCTCCCAAC 423 dsEcR2 *GACCTCCCGAAGAGCTTTGT *TAGGCGTGCCCATTATCATT 390 dsEcR3 *CACGTCGCACATATCCTACAC *TTGCCGTACTTGCACTGGTA 435 dsUSP *CAATGGGTCCTCAGTCACCT *TGCACGGCTTCTCTTTTCAT 412 dsvATPase * TTAGCCAACACTGGAATAAACGTC *CCAAACAGTCCATGCATATTATT 185 dsCoo2 *GGGAAGTTACAAATTATACG *CTCCCATAGCCATCTTG 620 dsGFP *TACGGCGTGCAGTGCT *TGATCGCGCTTCTCG 455

Table 5.1 Primers used for fragment amplification before cloning and for dsRNA synthesis. * Only gene-specific parts of the primer are listed in the table. These are preceeded by the T7 adaptor TAATACGACTCACTATAGGG for dsRNA synthesis.

3.1.2. Preparation of dsRNA

After cloning, the inserts were sequenced for verification and then used as a template for dsRNA synthesis. This synthesis was done with the Ambion MEGAScript kit using plasmids containing the target sequences as a template and following the manufacturer’s instructions. Primers containing T7-adapters which were used for the dsRNA synthesis are listed in Table 5.1. For EcR, three fragments were created, spanning different regions of the gene. EcR1 was a fragment including an isoform A specific part, while EcR2 and EcR3 were not isoform- specific (Fig. 5.5).

Figure 5.5 Schematic representation of the position of the three dsEcR fragments used in these experiments

- 131 -

Chapter V

Transcription incubation was done overnight (16h) and elution from the purification columns was performed either with the Elution Buffer delivered with the kit, or with nuclease-free water. Subsequently, the quality and quantity of the extracted RNA were examined by gel electrophoresis and spectrophotometry using a NanodropTM ND-1000 (Thermo Fisher Scientific, Asse, Belgium) respectively. If necessary, the dsRNA was eventually evaporated using a SpeedVac, to obtain a more concentrated sample.

3.1.3. Feeding assays

The feeding assay that was used in these experiments is based on the one which was developed by Sadeghi et al. (2009). Cages of approximately 2 cm diameter are made where the artificial diet is placed between parafilm layers and up to 15 aphids can be placed in one cage. Aphids can pierce through the upper parafilm layer and can suck up the diet. Insects can be monitored easily during the experiment. (Fig. 5.6). The artificial diet, which is based on formulation A of Prosser and Douglas (1992), was mixed with the dsRNA in a maximal ratio of 30%/70% dsRNA/diet. As controls, dsGFP and also elution buffer or water, depending on the medium in which the dsRNA was solubilized, were used.

3.1.4. Microinjection assays

Microinjections were performed using Blaubrand IntraMARK green colour-code 50µL microcapillaries from BRAND GMBH (Wertheim, Germany) which were self-pulled using the two-step pulling procedure (heat setting 80 and 83 for step 1 and step 2, respectively) with a Narishige PC-10 (London, United Kingdom). These capilaries were subsequently connected to a Femtojet (Invitrogen) and a Narishige micromanipulator was used to assist with injections. Aphids were immobilized using diethylether and were placed on an agarose surface for injection. The insects were injected dorsally, just behind the head. After injection, aphids were placed on fresh faba bean (Vicia faba) leaves in order to monitor for phenotypic effects or to collect them for sampling for expression analysis by qPCR. Leaves were replaced on a daily basis. Aphids that did not survive the injection procedure were removed from analysis. Pictures were taken with a Leica DFC295 camera mounted onto the microscope.

- 132 -

RNAi

Figure 5.6 Feeding assay procedure (Sadeghi et al., 2009). The feeding apparatus was made using a plexiglass cylinder (a4) on which a layer of parafilm was placed (a1). The artificial diet was pipetted on the parafilm (b) and a second, thinner layer of parafilm was placed on top of that (c). This was then held together by a plastic ring (a3, d). Afterwards, aphids are taken from the fava bean leaves with a fine brush and placed in the feeding cage (e-f). Subsequently, the small petri dish (a2) which contains a fine mesh for ventilation, is placed on top of the feeding cage (g). These cages are then placed upside down into the holders in which a small amount of wet paper is placed in order to maintain humidity (h).

3.1.5. dsEcR feeding assay

In the first RNAi experiment, feeding as a delivery method was chosen. dsEcR was mixed with the artificial diet in a 1:3 ratio. The final concentration of dsRNA in the diet was 200 ng/µL. Two dsEcR fragments were tested, namely EcR2 and EcR3. As a negative control, the Ambion MEGAscript kit’s elution solution was added to the diet in the same concentration as the dsEcR in the treated cages. Additionally, dsGFP was used as another control in order to make sure observations were not due to a non-specific reaction to dsRNA. Four cages of each treatment were set up and 10 neonate aphids (less than 24h old) were placed in each cage. Neonates were chosen because this allowed the observation of possible silencing effects on moulting/development during several juvenile stages. Aphids were kept on the diet - 133 -

Chapter V continuously for 72 hours and mortality as well as development through the number of moults were observed in each group.

3.1.6. dsEcR/dsUSP microinjection experiment

Microinjection was also used as a delivery method. In order to be able to monitor effects on development, juvenile stages have to be used in these experiments. Since the injection of L1 nymphs proved to be challenging, L2 nymphs, which are much easier to be injected without harming them, were eventually used in these injection experiments. To get a synchronized population of L2’s, adults were placed on fresh Vicia faba plants and were taken off 12h later. The neonates stayed on the plants and were all L2 after 48h. These insects were then used in the injection experiments. 20nL of 4 µg/µL dsEcR2 and dsUSP solutions were injected into each aphid. For each treatment (dsEcR2, dsUSP and dsGFP) 40 aphids were injected and afterwards, the aphids were placed on faba bean leaves in order to monitor development (moulting) and mortality or for sampling. Fifteen aphids were used for the phenotypic monitoring and 25 aphids were used for realtime qPCR sampling. For each time point, 5 aphids were pooled together for RNA extraction.

3.1.7. dsC002 microinjection experiment

For the dsC002 injection experiment, the experimental procedures of Mutti et al., 2006 were followed. 30nL of a 4 µg/µL solution of dsC002 was injected into 15 adult aphids and the mortality was observed over 10 days. After injection, aphids were placed back onto faba bean leaves. Injections with H2O and with dsGFP were used as controls.

3.1.8. vATPase feeding assay experiment

Feeding bioassays were set up with vATPase dsRNA according to Whyard et al., 2009. We tested two concentrations. The first one was 0.34 µg/100mL diet, which corresponds to the LC50 reported in the paper. Additionally, a 1µg/100mL diet assay was set up. The elution buffer from the dsRNA synthesis kit, in which the dsRNA was eluted, was used as a control. After four days, the diet was replaced with fresh diet. Mortality was evaluated daily over 7 days.

- 134 -

RNAi

3.1.9. Quantitative real-time PCR analysis of EcR and USP

Real time qPCR was used for expression analysis. For each sample, 5 aphids were taken from the leaves and total RNA was isolated using the QIAGEN RNeasy Mini Kit following the manufacturer’s instructions. Afterwards, 1µg of RNA was used for cDNA synthesis using Superscript II reversed transcriptase (Invitrogen, Merelbeke, Belgium) and oligodT-primers. Three reference genes were selected, namely tubulin, actin and rpL7. Realtime qPCR was performed with the Bio-Rad CFX96 Touch ™ (Bio-Rad Laboratoties, Hercules, California, USA) using SYBR Green as a fluorescent dye. The data analysis was performed using the Bio-Rad CFX Manager. Amplification conditions were 3min at 95°C followed by 39 cycles of 10s at 95°C and 30s at 58°C. Primers used in qPCR analysis are qEcR_F (CAACTGTCATTCAGTCGGTTT), qEcR_R (TTTTCTCCACTTTCCAACCA), qTub_F (TGGACAATCAGGTGCTGGA) and qTub_R (CTCGGCTTCTTTCCTCACAA). No-RT controls were included in the analysis.

3.2. Results and discussion

3.2.1. Silencing EcR and Usp in A. pisum

3.2.1.1. dsEcR feeding assay

RNAi feeding experiments administering dsEcR to aphid nymphs showed no effect on moulting or mortality. A small, though non-significant, difference in moulting could be observed between the negative control and the treatments, but no difference was found between the dsEcR- or dsUSP-fed groups compared to those that were fed on dsGFP containing diet.

3.2.1.2. DsEcR/dsUSP microinjection experiment

Feeding might be an easy way of delivery, for several species it has proven to be a non- efficient way of delivery as well. In order to investigate whether or not the delivery method might be the cause of the lack of effect seen in the previous experiments, a microinjection experiment was set up as well. DsEcR2 was injected straight into the haemocoel of L2 aphids and afterwards the mortality and moulting were observed. Additionally, the expression of EcR after injection with dsEcR2 was examined by real-time qPCR. - 135 -

Chapter V

Fig 5.7 Effect of dsEcR and dsUSP feeding on Acyrthosiphon pisum. Aphids were placed on an artificial diet containing dsRNA for EcR and USP (final concentration of 200ng/µL diet). The number of moults accumulated over three days by 15 aphids is represented in the figure. Water and dsGFP were used as controls. Differences between control and treatment were not significant. The bar graph on the right represents the situation after 72 hours, including the standard deviation over the three repetitions.

16 2,5 14

2 12

1,5

8 aphid 6 1 GFP GFP

Survival Survival (# ofaphids) 4 EcR 0,5 EcR

2 USP USP Accumulative Numberofmoults per 0 0 0 2 4 6 0 2 4 6 A Days B Days

Figure 5.8 Effect of dsEcR injection on Acyrthosiphon pisum. The amount of 80ng dsRNA for EcR and USP was injected into L2 nymphs. An equal amount of dsGFP was injected as a control. Afterwards, aphids were placed on faba bean leaves and phenotypic effects were observed over 5 days after injection: (A) represents the observed survival (Y-axis, number of aphids); (B) shows the average accumulative number of moults for each group of 15 aphids. - 136 -

RNAi

Similar to the feeding experiments, no clear phenotypic effect on either survival or development could be seen in any of the injection experiments that were carried out with dsEcR or dsUSP (Fig. 5.8). Differences in survival between the treatment and control groups were small and likely not linked to the administration of dsRNA. On the moulting rate of the aphids, only small differences were observed between the three groups. After day 1, a small retardation can be seen for EcR moulting, but they catch up after day 2, suggesting the aphids in the EcR group were early stage 2 when injected while those in the GFP and USP group were a bit older. Since the synchronization of these aphids is done over 24 hours, age differences of up to a day between individuals can take place. Repetitions or perhaps experiments with larger test groups could neutralize these effects.

In order to examine if there is a silencing effect to be seen on transcript level, real time qPCR experiments were conducted. Samples were taken daily during the course of the experiment and the expression of EcR was examined for each day after injection (Fig. 5.9).

1,8 1,6

1,4 1,2 1 0,8 Control 0,6 EcR

Relative Relative expression 0,4 0,2 0 Day 1 Day 2 Day 3 Day 4 Day 5 Time

Fiure 5.9 Relative expression of EcR in Acyrthosiphon pisum after injection with 80ng dsEcR. After injection, aphids were placed on faba bean leaves and 5 aphids were taken off the plants daily and pooled for RNA extraction with the RNeasy kit (Qiagen). Afterwards, cDNA was synthesized and the expression of EcR was analyzed by real-time qPCR, using tubulin as a reference gene.

On day 1, surprisingly, we see a small increase in the expression of EcR compared to the control. On day 2, no difference was seen while on day 3 the EcR mRNA levels were slightly lower compared to the control. While there are some differences in the EcR expression levels, these are rather small and not what would be expected as a clear sign of RNAi-mediated gene

- 137 -

Chapter V silencing. We therefore do not believe that these differences are linked to the injection of dsEcR but rather that these fluctuations have to do with the natural variation that can be seen for EcR expression between different phases within a stage. For example, EcR expression is usually strongly upregulated before a moult, while expression can be relatively low just after ecdysis.

Both feeding and injection experiments aimed at silencing EcR or USP showed no clear effects, either phenotypically or at the level of mRNA. There are several possible reasons why RNAi might not work in certain cases. One possible factor could be the dsRNA-fragment selected in the target gene itself since these have been observed to play a role in RNAi efficiency. Therefore, experiments with several dsEcR fragments spanning most of the open reading frame were performed. However, no difference was observed between the different dsEcR fragments. The target genes themselves could also be an important factor. Effects on moulting would logically require silencing of the EcR and USP in the epidermal tissues and it is known that silencing genes by RNAi in these tissues has proven to be difficult (Terenius et al., 2011). However, both nuclear receptors should be expressed in multiple tissues other than the epidermis, especially USP which is an important heterodimerization partner for many nuclear receptors. Additionally, expression analysis on whole body level showed no silencing whatsoever. Another element which could affect RNAi efficiency is the amount of dsRNA that is introduced into the insect. In our experiments however, the amounts were already relatively high and comparable to µg dsRNA/body weight ratios used in successful RNAi experiments with other insects (Terenius et al., 2011).

3.2.2. Testing dsvATPase and dsCoo2 as positive controls

In order to test whether or not the aphids from our culture were susceptible for RNAi, an attempt was made to reproduce two successful pea aphid RNAi publications. A first research which reported successful RNAi in Acyrthosiphon was published by Mutti et al. (2006) who silenced a salivary transcript c002, which is necessary for the survival of the pea aphid on fava beans. The same primers were used to synthesize the dsRNA-fragment and the dsC002 was injected into adult aphids. dsGFP and dsH20 injections were used as controls. Afterwards, the aphids were placed back on fava bean leaves and mortality was monitored.

- 138 -

RNAi

While we did see a small difference between the c002 silencing treatment and the control (dsGFP) since mortality is observed a bit earlier (Fig. 5.10), the difference is very small compared to the results reported by Mutti et al. (2006). They found that dsC002 injection caused 100% mortality in half the time as the dsGFP control did (8 days compared to 16 days). Even though C002 is thought to be a protein necessary for feeding on plants, the effect of dsC002 administration by feeding through an artificial diet was also tested, but no effects were seen there as well (data not shown).

14 GFP 12 H20 10 Coo2 8 6

4 Survival Survival (# ofaphids) 2 0 Day 0 Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 Day 7 Day 8 Day 9 Day 10 Time Figure 5.10 Effect of dsC002 injection on adult aphids. The amount of 120ng dsC002 was injected dorsally into the haemocoel of adult aphids and afterwards aphids were placed on faba bean leaves in order to investigate the effects on survival (X-axis, in number of aphids).

Another successful Acyrthosiphon RNAi experiment was reported by Whyard et al. (2009). These researchers wanted to compare the RNAi efficiency between insects from different orders and one of the insects tested was the pea aphid. The target gene used in this research was the E-subunit of the vATPase gene. Whyard et al. (2009) reported high mortality rates with very small concentrations of dsRNA. The LC50 they observed was 3.4 µg/g diet, which is relatively low. In order to try to reproduce this experiment, feeding assays were set up administering vATPase dsRNA to neonate aphids and observing the mortality rate over 6 days. Survival after 6 days was still 90% and no difference was seen between the control and both treatments (Fig. 5.11). Injection experiments with much higher amounts of dsRNA administered to the aphids did also not result in any mortality (data not shown).

- 139 -

Chapter V

100

90 Control

0,34 µg dsRNA /100µL diet

80 Survival Survival (%) 1µg dsRNA /100µL diet

70 1 2 3 4 5 6 7 Days

Figure 5.11 Effects of dsvATPase feeding in Acyrthosiphon pisum juveniles. Two concentrations of dsRNA (0.34 µg/100mL diet and 1µg/100mL diet) were tested. The elution buffer from the dsRNA synthesis kit, in which the dsRNA was eluted, was used as a control. After four days, the diet was replaced with fresh diet. Mortality was evaluated daily over 7 days.

4. RNAi efficiency in the pea aphid

The results presented in this chapter clearly show a very low efficiency or even a complete lack of any RNAi silencing response to the delivery of dsRNA. While this is an observation shared by RNAi experiments in many insect species (Bellés, 2010; Terenius et al., 2011), it is still surprising given the fact that a number of publications had reported successful RNAi experiments in the pea aphid over the last five years. However, the fact that other laboratories that are trying to use RNAi in aphids, are also having similar problems (personal communication) point out that this is not merely a problem at our end. In what follows, we will have a look at some of the possible reasons behind these observed problems regarding RNAi in the pea aphid and discuss some other factors influencing RNAi efficiency.

4.1. Material and Methods

4.1.1. DsRNA degradation experiments

To test the stability of dsRNA in the artificial diet used for the bioassays, feeding cages were set up containing 130 µL artificial diet, to which 40ng of dsEcR2 was added. In three cages, aphids were placed while three other cages were left untouched. After 84 hours, the dsRNA was purified with the RNeasy kit (Qiagen) and this dsRNA was subsequently loaded on an 1.5% agarose gel. - 140 -

RNAi

Haemolymph was collected from about 500 adult aphids by amputating one or more legs and collecting the droplets of haemolymph coming out using microcapillaries. Haemolymph was collected in a phenylthiourea (PTU) buffer in order to avoid melanisation and kept on ice. Once enough haemolymph was collected, haemocytes were removed by centrifuging at 1,000g for 8 minutes and collecting the cell-free supernatant. An amount of 500ng dsRNA was incubated in either 3.5µL of RNase-free water or 3.5µL cell-free hemolymph at 23°C and 35% relative humidity, the same conditions at which the aphids used in the RNAi experiments were kept. Samples were collected after 1 hour, 3 hours and 12 hours. DsRNA was recovered using the RNeasy kit (Qiagen) and run on a 1.5% agarose gel.

4.1.2. Analysis of the RNAi machinery core genes

Sequences for DcR2, Ago2, R2D2, Eri1 and Sid-1 orthologs in the pea aphid were identified and collected from Genbank and primers were designed using the same Primer3 biotool as used in Chapter II. L2 nymphs were injected with 50ng dsEcR (25nL 2µg/µL) and for five days, samples were taken daily in order to examine gene expression for these five RNAi- related genes. Control animals were injected with H2O. The RNAi extraction and real-time qPCR experiments were performed in the same way as for the EcR and USP expression. For RNA extraction, 8 aphids were used per sample and for these qPCRs, three reference genes were used, namely tubulin, actin and rpL7. Primers are listed in Table 5.2. No-RT controls were included in the analysis.

Fragment Forward Reversed Fragment size

qDcR2 AACCCATCCAACCTACCAAT CAGTTATTTCACCAGGAGTTTTGTG 137 qAgo2 AGAGATGGTGTAAGCGAAGG TGCCAGAAGGGACATTAGAAA 199 qR2D2 CAAAAATCGCCTTGTTCCTC CCACATGCTTGGCTTCTTTT 140 qEri1 ATGGCTCGGTTTCTTTATGG GAGGGTTGCCTTCAAATTCC 187 qSid1 TTATGCAATGGGATCAGCAC CAAATGCCAAAACACCAAAC 197 qActin CGTGAAAAGATGACCCAAATC CCAGAGTCCAAAACGATACCA 122 qTubulin TGGACAATCAGGTGCTGGA CTCGGCTTCTTTCCTCACAA 100 qRPL7 TTGAAGAGCGTAAGGGAACTG TATTGGTGATTGGAATGCGTTG 76

Table 5.2 Primers used in RNAi-related genes qPCR experiments

- 141 -

Chapter V

4.2. Results and Discussion

4.2.1. Degradation of dsRNA by the pea aphid

There are multiple factors that can explain a lack of RNAi sensitivity in certain species. One very important factor is the fate of dsRNA after introduction in the aphid. Since dsRNA processing and subsequent targeting and degradation of mRNA are intracellular processes, the dsRNA should remain stable and intact in the hemolymph or the digestive tract sufficiently long to be taken up by the cells. Once taken up into the cell, RNAi usually happens, as can be seen in experiments where dsRNA is transfected into cells or experiments with expression of hairpin RNAs by transgenes. In these experiments, the success rate is high (Terenius et al., 2011).

RNAi after dsRNA delivery by feeding has been observed to be troublesome in many insects, possibly due to degradation by ribonucleases in the digestive tract of the insects. In the hemipteran tarnished plant bug, Lygus lineolaris, for example, it has been demonstrated that saliva is able to degrade dsRNA quickly (Allen and Walker, 2012). In the lepidopteran Bombyx mori, gut extracts proved very efficient in degrading the dsRNA as well (Liu et al., 2013). Besides degradation in the gut or by saliva, fast degradation in haemolymph has been reported in multiple insects as well. Liu et al. (2013) reported dsRNA degradation in haemolymph of the RNAi-insensitive silkworm Bombyx mori and Garbutt et al. (2013) compared the degradation in the haemolymph of two insects, the RNAi-sensitive cockroach B. germanica and the less sensitive lepidopteran M. sexta. They observed no degradation in the former while the dsRNA was degraded quickly in M. sexta. After one hour, smearing could be seen on gel, indicating the dsRNA was being degraded and after no more than three hours, the entire dsRNA band had disappeared. These experiments show that a difference in stability of the dsRNA in the insect body could, at least partially, be responsible for the differences that are seen in the sensitivity between different species.

In order to investigate the possible degradation of dsRNAs in the aphid body, we set up two experiments. One experiment examined the effect of aphid feeding on a dsRNA-containing artificial diet on the stability of the dsRNA and another experiment investigated the degradation of dsRNAs in the haemolymph of the aphids. In the first experiment, dsRNA was added to the diet and aphids were placed into the cages to feed on the diet.

- 142 -

RNAi

Figure 5.12 Ex vivo degradation of dsRNA by Acyrthosiphon pisum salivary products. 40ng of dsEcR2 was added to 130µL artificial diet and the stability of the dsRNA in the diet on which aphids were feeding was compared to that in the diet on which no aphids were feeding by analyzing both samples on agarose gel.

After 84h, the dsRNA in diet on which no aphids were feeding showed no degradation what so ever. Contrary, the dsRNA which was in the diet on which aphids had been feeding showed a clear smear on agarose gel, suggesting that the salivary secretions from the aphids while feeding cause some breakdown of the dsRNA in the diet (Fig. 5.12).

One consequence of this result is that RNAi experiments using artificial diets where a constant supply of dsRNA is wanted, should be designed carefully and the dsRNA-containing diet should be replaced regularly. Whether or not the degradation by these saliva enzymes is strong or rapid enough to have an effect on the stability of dsRNA in the digestive tract of the aphids is difficult to say based on these results. Even after 84h, there is still a considerable amount of dsRNA intact and also non-intact fragments should in theory be able to elicit a silencing effect if they are taken up by the cell. However, the concentration of the nucleases which are responsible for the degradation in the artificial diet is probably relatively low, so it is difficult to speculate on the speed of intestinal dsRNA degradation in aphids, where a much higher concentration of nucleases must be present. Experiments in other species do suggest that some insects exhibit a very strong dsRNA-degradation in their digestive system (Allen and Walker., 2012; Liu et al., 2013). Whether this is mainly the result of nuclease activity or whether other factors, like the pH, could play a role as well is unknown and should be investigated further. - 143 -

Chapter V

The second degradation experiment (Fig. 5.13) shows a rapid and strong degradation of dsRNA in the haemolymph of aphids. The gel clearly shows that after 1h, a considerable smear is already visible indicating dsRNA degradation. After 3 hours, the band was completely gone and only a smear of smaller fragments were left. After 12 hours, all dsRNA was completely gone. This indicates that (foreign) dsRNA can only stay intact in the haemolymph for a short amount of time. The time for degradation in haemolymph that is observed here is in line with reports from other insects as well. In M. sexta, 200ng dsRNA was degraded within 2-3 hours (Garbutt et al., 2013).

Figure 5.13 Ex vivo degradation of dsRNA by Acyrthosiphon pisum haemolymph. An amount of 500ng dsEcR1 was incubated in either 3.5µL of RNase-free water or 3.5µL cell-free hemolymph, collected from adult aphids. Afterwards, dsRNA was recovered using the RNeasy kit (Qiagen) and run on a 1.5% agarose gel.

Not much is known about nucleases with dsRNase activity in insects. In Bombyx mori, a dsRNase has been described (Arimatsu et al., 2007; Liu et al., 2012) which could contribute to the rapid and strong degradation that is observed in the digestive system of some insects. This protein was originally thought to be expressed in midgut epithelial cells only, but Liu et al. have demonstrated that this protein is present in other tissues as well, including epidermis, fat body, thoracic muscles, Malpighian tubules, brain and silk glands of 5th instar larvae. As far as we know, nothing is known on dsRNases in the haemlolymph of insects. Garbutt et al. (2013) have performed a first analysis of the possible dsRNase present in M. sexta haemolymph. They found that pre-heating the plasma to 100°C for 10 minutes strongly inhibited the ability of the haemolymph to degrade dsRNA. Furthermore, they also - 144 -

RNAi demonstrated that EDTA has the ability to inhibit the degradation as well. This demonstrates that a metal-dependent enzyme is responsible for the degradation of dsRNA in the haemolymph plasma of M. sexta (Garbutt et al., 2013)

In C. elegans, a ribonuclease has been described which specifically seems to target siRNAs (Kennedy et al., 2004). Mutations in this gene caused an enhanced RNA interference phenotype and was subsequently called Eri-1. This gene encodes for a 3′ to 5′ exonuclease of the DEDDh superfamily of RNase T exonucleases. DsRNAs which proved ineffective in triggering RNAi in the wild-type animals did cause a robust silencing effect in nematodes which were devoid of this nuclease. There seem to be Eri-1 related genes at least in some insects. In Tribolium and Drosophila for example, putative homologs for this nuclease have been discovered. However, it is unclear whether these nucleases found in insects have a similar function as Eri-1 in C. elegans. Phylogenetic analysis points out that this Eri-1-like nuclease does not belong to the Eri-1/3’hExo subclass but is part of the so-called Snipper (Snp) subclass. Although Snp can cleave RNA as well as DNA molecules, it seems to play no role in RNAi in Drosophila (Kupsco et al., 2006).

4.2.2. RNAi-related genes

Another possible cause of RNAi insensitivity was proposed by Bellés (2010) who suggested that a low response (upregulation) of core RNAi genes after dsRNA treatment could also play a role. Since the RNAi machinery has an important function in immunity against viruses, one would expect that the introduction of dsRNA would elicit some response. However, at the moment, little is known about the effects of dsRNA intake on the expression of these genes. We do know that at least in some species, the introduction of dsRNA does have an effect on the RNAi machinery, or at least some of the genes that are thought to be involved. In the fish Gobiocypris rarus for example, dicer mRNA levels were found to be upregulated significantly during infection with the Grass carp reovirus, a dsRNA virus (Su et al., 2009). The expression reached a peak within 24h and returned back to its normal level 48h after injection. Also in the prawn Litopenaeus vannamei, dicer-2 expression levels were found to be elevated following the introduction of viral particles, as well as synthetic dsRNA analogues (Chen et al., 2011). In 2012, the first similar experiment in insects was published when Garbutt and Reynolds presented their research on the effect of dsRNA treatment on RNAi core machinery gene expression in M. sexta (Garbutt & Reynolds, 2012). They found - 145 -

Chapter V that injecting dsRNA into the caterpillars caused a clear upregulation of a number of these core genes, including Dcr-2 and Ago2. This lepidopteran is interesting in relation to aphids since for this species, very variable results concerning RNAi were observed as well. Succesful RNAi experiments were reported (Levin et al., 2005; Eleftherianos et al., 2009) while many researchers were experiencing difficulties trying RNAi in this lepidopteran. Terenius et al., who conducted a large study into published and unpublished RNAi results, reported that in less than half the RNAi experiments conducted in M. sexta, a high degree of silencing was observed (Terenius et al., 2011). Finally, also in the silkmoth B. mori, Liu et al. (2013) have demonstrated that injecting dsGFP caused a significant upregulation of DcR-2 and Ago2 in the gut.

4.2.2.1. Identification of RNAi-related genes in the genome

Jaubert-Possamai et al. (2010) have recently studied the presence of RNAi core genes in the pea aphid. They found that the miRNA pathway in Acyrthosiphon shows a considerable expansion, counting two miRNA-specific dcr-1 and ago1 genes, as well as four copies of pasha. For the siRNA pathway, only one copy of each gene was discovered (dcr-2, ago2 and r2d2), which is similar to the situation in Drosophila. Yet a molecular or comparative genomic comparison for the genes of the siRNA pathway was not provided in this publication.

We collected the A. pisum ortholog sequences of the three core elements of the RNAi pathway from Genbank (DcR-2, Ago2 and R2D2) Additionally, we searched the pea aphid genome for orthologs of the Eri1-like nucleases and Sid-1 membrane proteins and identified these as well. Analysis of ApDcR-2 showed two amino-terminal DExH helicase domains, responsible for unwinding the dsRNA, a Dicer-DSRBD (double stranded RNA binding domain), a PAZ-domain which is thought to be involved in dsRNA binding, two RNase III domains, responsible for cleaving the target RNA and finally a carboxy-terminal DSRBD. This is the architecture that corresponds to Dcr-2 in most other insect species, as well as CeDcr-1. Only in TcDcr-2, a different domain architecture has been observed since the DSRBD was found missing. Overall however, this protein seems closer related to TcDcr-2 than to DmDcr-2. This observation is supported by the fact that core RNAi machinery genes in another aphid, Aphis glycines, which were described recently (Bansal & Michel, 2013), also showed more similarity to the Tribolium set of RNAi genes than to the set in Drosophila.

- 146 -

RNAi

In the pea aphid, we have found one Sid-1-like (sil) gene. Comparison with homologs from other species revealed no remarkable features. For some time, researchers have speculated whether the lack of a Sid-1-like (Sil) protein in Drosophila could be partly the reason why systemic RNAi was not observed in this dipteran insect. Indeed, in C. elegans, Sid-1 is the best characterized protein involved in systemic RNAi and is responsible for trans-membrane channel-mediated uptake of dsRNA (Winston et al., 2002; Feinberg & Hunter, 2003). The presence of three Sils in Tribolium in combination with its strong response to RNAi added support to this theory. However, in recent years it has become evident that the number of Sils should not be directly correlated with a strong systemic RNAi response. Indeed, in some insects where multiple Sils genes are found, there is no robust systemic RNAi response. An example is B. mori which carries three different Sils, yet does not show a strong systemic RNAi response. Additionally, in some species where multiple Sils are found and systemic RNAi is observed, the Sils have been shown to be unnecessary for this response to happen (Tomayasu et al., 2008; Luo et al., 2012). Furthermore, insect Sils show more similarity to the nematode Tag-30 protein, which has been shown to be unnecessary for systemic RNAi (Tomoyasu et al., 2008). These discrepancies raise the question whether the presence of these Sils can be linked to the presence or absence of systemic RNAi in insects and even whether or not the Sils have the same function in insects as they do in nematodes.

4.2.2.2. Quantitative expression analysis of RNAi-related genes In order to examine whether or not the introduction of dsRNA into the species can elicit a response on the level of expression of these genes, an experiment was set up where aphids were injected with dsEcR and expression levels of dcr-2, ago2, r2d2, eri1 and sid-1 were monitored at certain intervals after injection (6h and 24h). These time points were chosen based on the observation that injecting dsRNA in caterpillars of Manduca caused an upregulation of these genes almost immediately after injection with a peak after 6 hours and a return to normal expression levels after 24 hours (Garbutt & Reynolds, 2012). Contrary to Manduca, no significant response of the expression of the RNAi machinery core genes DcR-2, Ago2 and R2D2 was seen in the pea aphid (Fig. 5.14A-C). Also the nuclease Eri1 showed no difference in expression pre- and post-injecion (Fig. 5.14D). Only for Sid-1 a very small difference in expression could be observed after 6 hours (Fig. 5.14E). However, it is unlikely that this difference has any biological relevance. The upregulations that were seen in other species for some of these genes were from a much higher magnitude. DcR-2 for - 147 -

Chapter V example showed an 80-fold and 8-fold increase in expression, respectively, in the M. sexta fat body 6 hours after injection. Expression in other tissues, such as the hemocytes and the midgut were even more strongly upregulated (362/395-fold higher, respectively). For Ago2, the observed upregulations 6 hours after injection in M. sexta were 8-, 22- and 27-fold in the fat body, hemocytes and midgut respectively. These kinds of responses were not observed in the pea aphid.

Figure 5.13 Relative expression of RNAi-related genes in Acyrthosiphon pisum after injection with 50ng dsEcR. dsGFP was injected in the control population. After injection, the L2 nymphs were placed on faba bean leaves and samples were taken daily for RNA extraction (5 individuals) with the RNeasy kit (Qiagen). Afterwards, cDNA was synthesized and the expression of EcR was analyzed by real-time qPCR, using tubulin, RPL7 and actin as reference genes.

- 148 -

RNAi

While a lack of upregulation of these RNAi machinery genes could be seen as a cause for the observed insensitivity of RNAi in some insects, it could also be seen as a symptom of some other problems, such as a lack of intracellular uptake or lack of a robust systemic transport system. Indeed, if dsRNA cannot reach the cytoplasm of the cells, it is probably also unable to elicit a response at the level of these RNAi core genes. One of the insects in which RNAi has the highest success rate is Tribolium. This beetle is known to have a robust systemic response following dsRNA introduction. In Drosophila however, systemic RNAi does not seem to occur (Roignant et al. 2003). The group of Yoshinoro Tomoyasu at Kansas State University has done research trying to figure out what exactly makes Tribolium respond so well to RNAi. In 2008, they searched the Tribolium genome for RNAi related genes and looked for differences between the sets of RNAi genes between this beetle and the sets in other species, including C. elegans and Drosophila (Tomoyasu et al., 2008). They found that even though the RNAi mechanism itself is conserved among species, the number and degree of conservation of its core genes can vary significantly. Even between insect species, considerable differences were found, resulting from lineage-specific losses and duplication events which can possibly explain the variation in RNAi efficiency between different insect species. Another conclusion was that the systemic response that is observed in Tribolium works through different mechanisms than in C. elegans since orthologs of certain important elements are missing in Tribolium, and in insects in general. RdRPs for example, which are responsible for the amplification of dsRNA in the nematode (Pak and Fire, 2007; Sijen et al., 2007; reviewed in Röther and Meister, 2011), are not present in Tribolium.

Since the inventories of genes involved in systemic RNAi and amplification did not show clear differences which could explain these observed differences in RNAi response between Tribolium and Drosophila, the authors also proposed the hypothesis that the core machineries in different species could work with different efficiencies. Genome-wide analyses of core genes showed several differences between both species. In Drosophila, only one Ago2 is expressed, while Tribolium has two, due to a lineage-specific duplication. Both TcAgo2s also seem to be involved in the RNAi pathway. Since the availability of different Ago proteins are known to affect RNAi efficiency in C. elegans (Yigit et al., 2006), the copy number of Ago2 could allow a more efficient RNAi response in Tribolium compared to Drosophila. Another important difference between both species was the domain architecture of the Dcr-1 and Dcr- 2 proteins. DmDcr-1 for example is missing the conserved N-terminal helicase domain while

- 149 -

Chapter V

TcDcr-1 does possess this domain and hence is more similar to the C. elegans Dcr-1 gene. The authors hypothesize that TcDcr-1 could, in addition to TcDcr-2, be involved in the siRNAi pathway while in Drosophila, only DmDcr-2 is involved in the siRNA pathway and DmDcr-1 is a miRNA-specific element. Dcr-2 also exhibits some domain architecture differences in both species. DmDcr-2 exhibits a strongly reduced PAZ-domain while TcDcr2 is missing the C-terminal dsRBD domain. The analysis of the RNAi core genes in A. glycines indicated that while the miRNA pathway in the pea aphid might have gone through a considerable expansion, the siRNA pathway shows a high degree of conservation and evidence of purifying selection (Bansal & Michel, 2013). This observed conservation could be related to their function in viral defense as well. Aphids have undergone considerable gene losses in the immunodeficiency (IMD) pathway (The International Aphid Genomics Consortium, 2010; Gerardo et al., 2010), a signaling pathway involved in viral immune response (Avadhanula et al., 2009; Costa et al., 2009). Aphids are notorious vectors for plant viruses and are therefore constantly exposed to viral material which could mean that the RNAi machinery is under strong evolutionary pressure. Interestingly, the fact that aphids have such a high exposure to viruses might be an explanation for the observed lack of RNAi effect as well as the observation that the RNAi- related genes seem to have a relatively high basal expression level, as if the genes are constantly ‘turned on’.

Virus-mediated effects could in fact also explain the difference in sensitivity observed between different experiments in different laboratories, regardless of the experimental setup, the target genes or the delivery methods. Indeed, recently it came to light that latent, non- pathogenic viruses can have an inhibitory effect on the the RNAi machinery either by the production of viral inhibitors of RNAi or through the accumulation of viRNAs that could saturate the RNAi machinery and prevent the access to Argonaute proteins of other siRNAs from exogenous origin (Li et al., 2002; Havelda et al., 2005; Flynt et al., 2008; Nayak et al., 2010). .

- 150 -

RNAi

5. Conclusion

Initial attempts to silence EcR and USP turned out unsuccessful. No phenotypic effects could be observed and expression analysis on EcR after administration of dsEcR showed no effect on transcript level either. While RNAi efficiency has proven to be highly variable in many insects (Bellés, 2010; Terenius et al., 2011) this was nonetheless a bit of a surprise, given that other laboratories had already reported successful application of RNAi in the pea aphid (Mutti et al., 2006; Jaubert-Possamai et al., 2007; Whyard et al., 2009; Shakesby et al., 2009) Surprisingly, even attempts to reproduce two of these publications were unsuccessful. Other laboratories however appear to be struggling with the same problems as we were and are also having problems achieving RNAi in the pea aphid (personal communication, Angela Douglas; James Carolan).

There are several factors which could influence RNAi efficiency in insects. First of all, the choice of target gene can be very important. Indeed, targeting genes which are mainly expressed in tissues such as the epidermis, which are known to be RNAi-insensitive, could prove to be difficult to achieve silencing (Terenius et al., 2011). Also, after making a choice concerning the target gene, the dsRNA-fragment design itself, especially the length, is important as well. More and more data seem to suggest that dsRNAs which are too short can cause a lack of silencing (Saleh et al., 2006; Mao et al., 2007; Huvenne & Smagghe, 2010; Bolognesi et al., 2012; Miller et al., 2012). Bolognesi et al. (2012) found that for silencing the DvSnf7 gene in Diabrotica the dsRNA fragment length had to be longer than 60bp. Interestingly, it turned out to be sufficient that 21bp of those 60 or more basepairs were target gene-specific. This suggests that the importance of the fragment length is not linked to the RNAi machinery itself but might be linked for example to the uptake of the dsRNA by the cells.

A very important question is what happens with the dsRNA after delivery in insects. Is the dsRNA taken up quickly into the cells after delivery? Is there considerable degradation before the uptake happens? This depends first of all on the delivery method itself. Research thus far seems to indicate that dsRNA delivery by feeding is difficult in many species. Especially for non-gut genes, RNAi by oral dsRNA delivery seems to be very often ineffective. Another reason could be a very rapid degradation of dsRNAs in the digestive tract by nucleases or for example by chemical properties of the gut environment. Cellular uptake itself can also be an - 151 -

Chapter V important barrier. Two uptake mechanisms of dsRNA have been proposed in insects: a transmembrane channel-mediated uptake similar to the Sid-1 mechanism in the nematode C. elegans and a endocytosis-based uptake mechanism (Huvenne & Smagghe, 2010). We have found a possible homolog for the CeSid-1 gene in A. pisum, similar to the ones found in Tribolium, Bombyx, Spodoptera, Apis and Nilaparvata but whether or not these are involved in dsRNA uptake in insects is still debated (Winston et al., 2002; Gordon & Waterhouse, 2007; Xu & Han, 2008; Tian et al., 2009; Zha et al., 2011).

Our research also demonstrated that pea aphids are capable of dsRNA breakdown, both in saliva and in the haemolymph. A feeding bioassay was set up and the stability of the dsRNA in the artificial diet was investigated. After three days, it was clear that the aphids feeding on the artificial diet cause degradation of the dsRNA, likely due to nucleases which are present in the secreted salivary products during feeding. These results add support to those reported by other groups suggesting that the digestive tract of insects could be a dsRNA-unfriendly environment (Allen & Walker, 2012, Liu et al., 2013). It is known that the presence of nucleases, as well as gut pH are very variable over different species. In Lepidoptera for example, the gut is known to be very alkaline and the pH can go up to 12 in the midgut of caterpillars (Dow, 1992). Since RNA is known to be unstable in very alkaline environments, this could be a factor as well in the degradation of dsRNA after feeding in many insects. While not as extreme as in lepidopterans, the aphid midgut is alkaline as well, having a pH ranging from 7,5 to 8,5 (Cristofoletti et al., 2003).

Another experiment, examining degradation of dsRNA in haemolymph demonstrated a rapid breakdown of dsRNA. After no more than three hours, almost all dsRNA was degraded. These results are similar to what has been discovered in the lepidopteran M. sexta where a similar dsRNA-degradation in haemolymph was observed as well as a variable RNAi efficiency (Garbutt et al., 2013). Observing such a rapid breakdown of dsRNA, the question we have to ask is whether or not enough dsRNA can be taken up by the cells in order to see an RNAi effect.

In order to investigate whether or not the absence or lack of (upregulation of) expression of some core genes of the RNAi machinery could be linked to the observed lack of silencing, the expression of three of these core genes (DcR-1, Ago2 and R2D2) was examined after dsRNA

- 152 -

RNAi injection. No up- or downregulation of these genes was observed after dsEcR injection, contrary to observations in M. sexta and B. mori (Garbutt and Reynolds, 2012; Liu et al., 2013). However, we did notice that the basal expression of these genes was relatively high compared to the reference genes which were used (RPL7, actin, tubulin) which would suggest that these genes are constantly sufficiently expressed. Since the siRNA pathway is considered to be an important line of defense against viruses (Galiana-Arnoux et al., 2006; van Rij et al., 2006; Wang et al., 2006), this might not come as a big surprise, especially given the high exposure to (plant-)viruses to which aphids are subjected. Additionally, more and more evidence indicates that viruses can also have inhibiting effects on the RNAi machinery, either by the production of viral inhibitors of RNAi or through the accumulation of viRNAs that could saturate the RNAi machinery and prevent the access to Argonaute proteins of other siRNAs from exogenous origin (Li et al., 2002; Havelda et al., 2005; Flynt et al., 2008; Nayak et al., 2010). More research towards the effects of these non-pathogenic viruses on the RNAi machinery, also in other species than Drosophila will have to be conducted.

- 153 -

- 154 -

Chapter VI: General Conclusions and future perspectives

Chapter VI

1. Nuclear receptors: diversity and conservation in Arthropoda

The nuclear receptor (NR) protein superfamily is involved in a large range of critical biological processes in the animal kingdom. Since their discovery in Vertebrata during the 1950s and 1960s, a lot of research has been done on their structure and function. Most of this research has been conducted towards vertebrate NRs however, given their importance as potential drug targets. With the Drosophila melanogaster genome project and the subsequent first identification of the NRs in an invertebrate genome, it became apparent that insects have a set of NRs which is distinct from vertebrates. Some elements, such as the DNA-binding domain and the ligand-binding domain sequences, as well as the 3D-structure, the mechanisms of dimerization and the mode of action of most NRs show a strong conservation in the entire animal kingdom. Nonetheless, diversity has been observed as well. Insects generally possess less than half the number of NRs compared to Vertebrata. In addition, they have developed a number of insect-unique NRs. This diversity has arisen mainly due to lineage-specific gene losses and duplications. Most of what was known about arthropod NRs at the start of this project was limited to the holometabolic insects. Information on NRs in hemimetabolic or non-insect arthropods was very scarce and usually limited to the NRs which form the ecdysone receptor heterodimer, namely EcR and RXR/USP. With the Acyrthosiphon pisum and Tetranychus urticae NRs, we presented the first complete sets of NRs for a hemimetabolic insect and chelicerate arthropod.

The NRs in A. pisum, Bombus terrestris and T. urticae exhibit a high degree of evolutionary constraint in the conserved domains. At the same time, diversity between these taxonomic groups within the Arthopoda seems to be created by gene losses and duplications as well. The Acyrthosiphon NR set did not show that much difference compared to the holometabolic NR sets, including the NRs which are involved in the ecdysteroid cascade. This proves that within Insecta, the set of NRs is well conserved. The T. urticae NRs on the other hand did exhibit some remarkable differences. First of all, a large expansion of the NR1 group was observed due to multiple duplication events of HR96-like receptors, as well as the presence of a homolog of the recently discovered HR10 in Daphnia pulex. Making conclusions towards the implications of this expansion would be speculative, but the HR96 receptor seems to be involved in the xenobiotic response in Drosophila as well as in Daphnia (King-Jones et al., 2006; Karimullina et al., 2012). One of these HR96s, namely HR96h/Tetur36g00260, has already been observed to be implicated in the adaptation of spider mites to a new host plant - 156 -

General conclusions and future perspectives

(Dermauw et al., 2013). Functional research towards the other HR96s in T. urticae will have to elucidate whether they also have a role in the xenobiotic response. If this is the case, this expansion of HR96s could be linked to the extreme adaptation potential that these spider mites possess. It would also be interesting to investigate if HR96 plays a role in the ecdysteroid signaling cascade. Its capability to compete with EcR-USP in binding to the hsp27 response element has been documented in Drosophila (Fisk, 1995), but further information about its role in this important pathway is not available. Interestingly, while a HR96-like expansion was observed in T. urticae, HR96 was found missing altogether in the pea aphid. Where exactly during evolution of this hemipteran insect HR96 was lost and whether its function had become redundant or its role has been taken over by other proteins or NRs will have to be investigated as well.

The results of the annotation work, which was mainly in silico research, form the foundations for further functional research towards these NRs in insects. This includes research to elucidate the role of these NRs in key metabolic and developmental processes, to find out how they play their specific part in certain pathways and also how they cross-talk between different pathways. Most of what is known about NRs in insects comes from research on Drosophila. But given the divergence between the Mecopterida, to which Drosophila belongs, and other insects due to an increase in evolutionary rate, NRs could have different functions and properties in non-Mecopterida compared to Drosophila. For one NR, namely RXR/USP, it has already been demonstrated that this divergence can have important consequences when it became clear that many non-Mecopterida insects have no ligand binding pocket (LBP) while the Mecopterida USP contains a large LBP. Further research is also needed towards orphan receptors. Indeed, most NRs are still categorized as orphan receptor because the ligand is simply not known at this time. Finding ligands for these orphans could lead to new ways of thinking in NR research and could perhaps even deliver possible new pest control targets.

2. Ponasterone A is the main ecdysteroid in T. urticae

EcR and USP, which form the functional ecdysone receptor heterodimer are by far the most extensively described NRs. The main reason for this is that they are targets for an important class of insecticides, the diacylhydrazines. In this research, we have taken a closer look at both receptors in the three test species and presented the predicted structure of ApEcR and

- 157 -

Chapter VI

TuEcR, as well as the docking of the models with ponasterone A (PonA) and 20- hydroxyecdysone (20E). The most remarkable discovery in this context is the fact that two distinct RXR proteins are present in the two-spotted spider mite. This had thus far only been reported in the ixodid tick Amblyomma americanum (Guo et al., 1998). Whether this trait is specific to Acari, Arachnids or perhaps all Chelicerata will have to be investigated further. The divergence between both RXRs could suggest that the gene duplication happened long before the Acari branched off, but information on the RXRs from other Chelicerata are needed to prove this hypothesis. Additionally, functional research towards the role and expression of these two RXRs will have to be done in order to elucidate how these NRs work together and what their specific, unique role is. Since RXR is a heterodimerization partner for many NRs, it functions in many pathways and is thus involved in many biological processes.

The ecdysteroid signaling cascade sets off after binding of the ecdysone receptor with its natural ligand. In most arthropods this hormone has been found to be 20-hydroxyecdysone (20E). In the decapod Crustacea, PonA was recently identified as a major moulting hormone as well, albeit coincidental with 20E (Chung, 2010). We have now demonstrated that PonA is the main moulting hormone in the two-spotted spider mite and docking of 20E with the predicted TuEcR-LBD structure seems to confirm this, as 20E did not seem to fit in the LBP. Whether or not this trait is shared in chelicerates is unknown and will have to be further investigated. But if this is the case, it seems that PonA could have been the original main ecdysteroid hormone of the common arthropod ancestor and has gradually been replaced by 20E during the evolution of Crustacea and later insects.

3. The pea aphid exhibits a low efficiency for RNAi

Once the annotation of NRs was finished, a first attempt was made to start the functional analysis of a number of pea aphid NRs. RNA interference (RNAi) was chosen as the method for this functional genomics analysis since this technique has proven its worth in functional genomics research in many insects. Additionally, RNAi is more and more seen as a promising pest control strategy as well, mainly due to its possibilities for selectivity. Investigating double stranded RNA (dsRNA) delivery by feeding could contribute to that field as well. However, RNAi proved to be troublesome in this hemipteran insect. Both feeding as well as injection experiments with dsEcR, dsUSP, dsC002 and dsvATPase proved ineffective and

- 158 -

General conclusions and future perspectives expression analysis after dsEcR injection also showed no silencing at transcript level. The possible causes for this lack of RNAi effect are many and have been discussed in this work. A number of important factors, such as degradation of the dsRNA by aphids and the presence and expression of the RNAi machinery core genes, have been investigated as well.

The pea aphid haemolymph appears to exhibit rapid dsRNA digestion activity. dsRNA which was incubated in collected cell-free haemolymph was almost completely degraded after three hours. These results are similar to those recently observed in the lepidopteran insect Manduca. Contrary, the RNAi sensitive Blattella germanica showed no degradation of dsRNA in the haemolymph (Garbutt and Reynolds, 2012). Whether or not this degradation could explain a lack of silencing effect in the injection experiments would depend on the speed of dsRNA- uptake by the insect cells and whether or not enough dsRNA molecules reach the cells to elicit a strong RNAi effect. Besides degradation in the haemolymph, degradation of dsRNA in the artificial diet on which aphids are feeding was observed as well, suggesting the aphid’s salivary secretions also contain a strong dsRNA-degrading activity. Whether or not this activity is as strong, or perhaps stronger, than that observed in haemolymph has to be investigated. Incubation experiments with salivary gland as well as gut extracts should be able to tell us whether or not delivery of dsRNA by feeding could be a suitable and efficient delivery method. If dsRNA turns out to be degraded too quickly in a certain species, other ways of oral delivery can be tested, such as feeding bacteria producing dsRNAs, nanoparticle- mediated delivery and also virus-mediated delivery (virus-induced gene silencing; VIGS) (Tian et al., 2009; Zhang et al., 2010; Kumar et al., 2012). Furthermore, the cause of the degradation, the presence of nucleases and/or environmental conditions, should be investigated as well. Possible nucleases in the haemolymph and digestive system can be characterized. It would for example be interesting to see if addition of EDTA to the pea aphid plasma has any effect on the degradation, as it did in M. sexta (Garbutt et al., 2013).

In order to directly investigate the fate and the stability of dsRNA in insects, quantitative PCR can be used (Garbutt & Reynolds, 2012). Total RNA, including the injected non-species specific dsRNA can be extracted and cDNA can be synthesized after a denaturing procedure. In this way, it would be possible to investigate the amount of intact dsRNA that remains present at certain time intervals after injection, both extra- and intracellular. Uptake experiments in cell lines using labeled dsRNA, similar to those conducted by Huvenne &

- 159 -

Chapter VI

Smagghe (2010) could give us an idea whether or not these molecules are being taken up by cells from different tissues. A continuous cell line for aphids is however not yet available as far as we know. While the development of such a cell culture would be an interesting advancement for future research in itself, in the meantime primary cell cultures could perhaps be used for these kinds of experiments.

Five RNAi-related genes in the pea aphid were examined and expression analysis after dsRNA-injection has been investigated, namely the RNAi machinery core genes dcr-2, ago2 and r2d2, as well as the putative eri1 and sid-1 orthologs. All five were found to be present in the pea aphid genome and transcripts were present in the transcriptome. Analysis after injection with dsRNA did not show any change in the expression of these genes, contrary to what was reported in M. sexta recently (Garbutt et al., 2013). A lack of response to the introduction to dsRNA could imply that dsRNA is either not recognized by the aphid, or that these genes are constantly expressed at a significantly high level. The fact that aphids are constantly in contact with (plant)viruses and that the siRNA pathway plays an active role in insect response to viruses could be linked to this. Moreover, viruses could have a direct inhibitory effect on the RNAi machinery as well. Indeed, Li et al., (2002) showed that Flock House Virus (FHV) is capable of inhibiting the RNAi machinery in D. melanogaster. Since then, more reports of viral interactions with the RNAi machinery have been reported and two possible mechanisms have been proposed as well: 1) The production of a large amount of viRNA could cause competition for other dsRNA molecules in the DcR and RISC proteins and 2) viruses could produce proteins that act as inhibitors for the RNAi machinery (Havelda et al., 2005; Flynt et al., 2008; Nayak et al., 2010). More research towards the role of (non- pathogenic) viruses, including plant viruses - for which aphids are a vector – on the RNAi machinery should be conducted. For example, a comparison between the RNAi response in virus-free populations and aphids which are infected with viruses would be very interesting.

DcR-2, Ago2 and R2D2 are not the only proteins involved in the RNAi mechanism. More products have been identified and many are yet unknown as well. Furthermore, since the sets of RNAi-related genes as well as the RNAi response seem to differ so much between insect species, the situation in aphids could be radically different in Tribolium. High throughput analysis of candidate RNAi-related genes should be performed in order to identify the elements that play a role in this process. Tomayasu et al. (2008) developed an assay in

- 160 -

General conclusions and future perspectives

Tribolium with which the involvement of genes in the RNAi pathway can be investigated in vivo. Similar systems could be developed in other insects as well. Furthermore, the factor of the juvenile stage should also be investigated. Bansal & Michel (2013) for example have observed significant differences, up to a two-fold increase, in the expression of some of these core genes during different developmental stages.

4. RNAi as a possible pest control strategy

Because of its high specificity and potential activity, RNAi also offers great promise as a pest control strategy (Baum et al., 2007; Price & Gatehouse, 2008; Huvenne & Smagghe, 2010; Zhang et al., 2013). Finding a suitable delivery method for dsRNA is crucial in this context. The most realistic way to get dsRNA into the targeted pest animals, especially herbivorous insects, would be by feeding. This can be accomplished by the use of dsRNA-producing transgenic plants or for example by spraying the plants with dsRNA-solutions. The main advantage of transgenic plants is that insects would continuously be supplied with dsRNA. Especially given the possibility that dsRNA replication or amplification does not happen in insects contrary to nematodes, this continuous supply of dsRNA could be of great importance. The spraying of dsRNA might be an easier and cheaper method than the use of transgenic plants. However, the limited stability of dsRNA in certain conditions would be something that has to be addressed first. Research towards a good formulation which keeps dsRNA stable for some time in different outdoor circumstances should also be done.

Several research papers have already demonstrated the possibility of enhanced resistance to insect pests in plants by specific dsRNA production (Baum et al., 2007; Mao et al., 2011; Pitino et al., 2011; Zha et al., 2011 ). Acyrthosiphon has been investigated in that context as well, by Whyard et al. (2009) who found that low concentrations of dsvATPase taken up orally were able to cause a high mortality. However, our attempt to reproduce this experiment showed no effect. For the sap-sucking Hemiptera, RNAi could be a useful addition in the scope of pest control methods. With the development of Bt (Bacillus thuringiensis) toxin- based pesticides, an effective method to control lepidopteran and coleopteran pests has been achieved. Sap-sucking insects such as most Hemiptera however remain unaffected by this type of pesticides (Elaine et al., 2005; Liu et al., 2010, Walker & Allen, 2011). Therefore, new approaches to control these economically important phloem sap-sucking insects should

- 161 -

Chapter VI be examined. We believe that the use of RNAi as a method of pest control could be an important novel technique and could be integrated in pest management programs, together with other already existing methods. Many issues still have to be addressed however until we reach that point. The mode-of-action behind RNAi in insects and the reasons for the observed variability in RNAi effects should be cleared up first. Furthermore, the (bio)safety of plants producing these dsRNA molecules should be investigated as well. The choice of target gene as well as the chosen dsRNA fragment, will be of the utmost importance in order to avoid non-specific effects on non-target organisms, as well as effects on humans. Another important hindrance is the use of transgenic crops, which is still a highly controversial issue. Especially in Europe, where there is still a lack of acceptance for this technology from the majority of consumers, farmers and politicians due to a combination of health, biosafety and ethical considerations.

- 162 -

References References

Adams, M. D., Celniker, S. E., Holt, R. A., et al. (2000). The genome sequence of Drosophila melanogaster. Science 287, 2185-2195. Ahn, J.E., Guarino, L.A. and Zhu-Salzman, K. (2007). Seven-up facilitates insect counter-defense by suppressing cathepsin B expression. FEBS J 274, 2800-2814. Ahn, J.E., Guarion, L.A. and Zhu-Salzman, K. (2010). Coordination of hepatocyte nuclear factor 4 and seven-up controls insect counter-defense cathepsin B expression. J Biol Chem 285, 6573-6584. Akman Gündüz, E., Douglas, A.E. (2009). Symbiotic bacteria enable insect to use a nutritionally inadequate diet. Proc R Soc B 276, 987-991. Allen, A.K. and Spradling, A.C. (2008). The Sf1-related nuclear hormone receptor HR39 regulates Drosophila female reproductive tract development and function. Development 135, 311-321. Allen, M.L. and Walker, W.B. (2012). Saliva of Lygus lineolaris digests double stranded ribonucleic acids. J Insect Physiol 58, 391-396. Andre, E., Conquet, F., Steinmayr, M., Stratton, S.C., Porciatti, V. and Becker-Andre, M. (1998). Disruption of reinoid-related orphan receptor β changes circadian behavior, causes retinal degeneration and leads to vacillans phenotype in mice. EMBO J 17, 3867-3877. Angelini, D.R. and Kaufman, T.C. (2005). Functional analyses in the milkweed bug Oncopeltus fasciatus (Hemiptera) support a role for Wnt signaling in body segmentation but not appendage development. Dev Biol 283, 409-423. Angelini, D.R., Liu, P.Z., Hughes, C.L. and Kaufman, T.C. (2005). Hox gene function and interaction in the milkweed bug Oncopeltus fasciatus (Hemiptera). Dev Biol 287, 440-455. Arakane, Y., Hogenkamp, D.G., Zhu, Y.C., Kramer, K.J., Specht, C.A., Beeman, R.W., Kanost, M.R. and Muthukrishnan, S. (2004). Characterization of two chitin synthase genes of the red flour beetle, Tribolium castaneum, and alternate exon usage in one of the genes during development. Insect Biochem Mol Biol 34, 291-304. Aranda, A. and Pascual, A. (2001). Nuclear hormone receptors and gene expression. Physiol Rev 81, 1269-1304. Araujo, R.N., Santos, A., Pinto, F.S., Gontijo, N.F., Lehane, M.J. and Pereira, M.H. (2006). RNA interference of the salivary gland nitrophorin 2 in the triatomine bug Rhodnius prolixus (Hemiptera: Reduviidae) by dsRNA ingestion or injection. Insect Biochem Mol Biol 36, 683-693. Arimatsu, Y., Kotani, E., Sugimura, Y. and Furusawa, T. (2007). Molecular characterization of a cDNA encoding extracellular dsRNase and its expression in the silkworm, Bombyx mori. Insect Biochem Mol Biol 37, 176-183. Aronstein, K. and Saldivar, E. (2005). Characterization of a honey bee Toll related receptor gene Am18w and its potential involvement in antimicrobial immune defense. Apidologie 36, 3-14. Ashburner, M. (1973). Sequential gene activation by ecdysone in polytene chromosomes of Drosophila melanogaster. Dev Biol 35:47-61. Astle, J., Kozlova, T. and Thummel, C. S. (2003). Essential roles for the Dhr78 orphan nuclear receptor during molting of the Drosophila tracheal system. Insect Biochem Mol Biol 33, 1201–1209. Avadhanula, V., Weasner, B.P., Hardy, G.G., Kumar, J.P. and Hardy, R.W. (2009). A novel system for the launch of alphavirus RNA synthesis reveals a role for the Imd pathway in arthropod antiviral response. PLoS Pathog 5, e1000582. doi:10.1371/journal.ppat.1000582. - 164 -

References

Ayer, S., Walker, N., Mosammaparast, M., Nelson, J.P., Shilo, B.Z. and Benyajati, C. (1993). Activation and repression of Drosophila alcohol dehydrogenase distal transcription by two steroid hormone receptor superfamily members binding to a common response element. Nucleic Acids Res 21, 1619-1627. Bain, D.L., Heneghan, A.F., Connaghan-Jones, K.D. and Miura, M.T. (2007). Nuclear Receptor structure: implications for function. Annu Rev Physiol 69, 201-220. Baker, K.D., Shewchuk, L.M., Kozlova, T., Makishima, M., Hassell, A., Wisely, B., Caravella, J.A., Lambert, M.H., Reinking, J.L., Krause, H., Thummel, C.S., Willson, T.M. and Mangelsdorf, D.J. (2003). The Drosophila orphan nuclear receptor DHR38 mediates an atypical ecdysteroid signaling pathway. Cell 113, 731-742. Baker, K.D., Beckstead, R.B., Mangelsdorf, D.J. and Thummel, C.S. (2007). Functional interactions between the Moses corepressor and DHR78 nuclear receptor regulate growth in Drosophila. Genes Dev 21, 450-464. Bansal, R. and Michel, A.P. (2013). Core RNAi machinery and Sid1, a component for systemic RNAi in the hemipteran insect, Aphis glycines. Int J Mol Sci 14, 3786-3801. Baum, J.A., Bogaert, T., Clinton, W., Heck, G. R., Feldmann, P., Ilagan, O., Johnson, S., Plaetinck, G., Munyikwa, T., Pleau, M., Vaughn, T. and Roberts, J. (2007). Control of coleopteran insect pests through RNA interference. Nature Biotechnol 25, 1322-1326. Beatty, J.M., Smagghe, G., Ogura, T., Nakagawa, Y., Spindler-Barth, M. and Henrich, V.C. (2009). Properties of ecdysteroid receptors from diverse insect species in a heterologous cell culture system – a basis for screening novel insecticidal candidates. FEBS J 276, 3087-3098. Beckett, D. (2001). Regulated assembly of transcription factors and control of transcription initiation. J Mol Biol 314, 335-352. Begemann, G., Michon, A.M., Van der Voorn, L., Wepf, R. and Mlodzik, M. (1995). The Drosophila orphan nuclear receptor Seven-up requires the Ras pathway for its function in photoreceptor determination. Development 121, 225-235. Bellés, X. (2010). Beyond Drosophila: RNAi In Vivo and Functional Genomics in Insects. Annu Rev Entomol 55, 111-128. Bennett-Lovsey, R.M., Hebert, A.D., Stemberg, M.J.E. and Kelley, L.A. (2008). Exploring the extremes of sequence/structure space with ensemble fold recognition in the program Phyre. Proteins Struct Funct Bioinformatics 70, 611-625. Benoit, G., Malewicz, M. and Perlmann, T. (2004). Digging deep into the pockets of orphan nuclear receptors: insights from structural studies. Trends Cell Biol 14, 369-376. Bertrand, S., Brunet, F.G., Escriva, H., Parmentier, G., Laudet, V. and Robinson-Rechavi, M. (2004). Evolutionary genomics of nuclear receptors: From twenty-five ancestral genes to derived endocrine systems. Mol Biol Evol 21, 1923-1937. Bertrand, S., Thisse, B., Tavares, R., Sachs, L., Chaumot, A., Bardet, P.L., Escriva, H., Duffraisse, M., Marchand, O., Safi, R., Thisse, C. and Laudet, V. (2007). Unexpected novel relational links uncovered by extensive developmental profiling of nuclear receptor expression. Plos Genet 3, e188. Bialecki, M., Shilton, A., Fichtenberg, C., Segraves, W.A. and Thummel, C.S. (2002). Loss of the ecdysteroid- inducible E75A orphan nuclear receptor uncouples molting from metamorphosis in Drosophila. Dev Cell 3, 209-220. - 165 -

References

Billas, I., Moulinier, L., Rochel, N. and Moras, D. (2001). Crystal structure of the ligand-binding domain of the ultraspiracle protein USP, the ortholog of retinoid X receptors in insects. J Biol Chem 276, 7465-7474. Billas, I., Iwema, T., Garnier, J., Mitschler, A., Rochel, N. and Moras, D. (2003). Structural adaptability in the ligand-binding pocket of the ecdysone hormone receptor. Nature 426, 91-96. Billas, I.M.I., Browning, C., Lawrence, M.C., Graham, L.D., Moras, D. and Hill R.J. (2009). The structure and function of ecdysone receptors. In: Ecdysone: Structures and Functions (G. Smagghe ed), pp. 335-360. Springer, Dordrecht. Bolognesi, R., Rameseshadri, P., Anderson, J., Bachman, P. et al. (2012). Characterizing the mechanism of action of double-stranded RNA activity against western corn rootworm (Diabrotica virgifera virgifera LeConte). PLoS ONE 7(10): e47534. Doi:10.1371/journal.pone.0047534 Bonneton, F., Chaumot, A., and Laudet, V. (2008). Annotation of Tribolium nuclear receptors reveals an increase in evolutionary rate of a network controlling the ecdysone cascade. Insect Biochem Mol Biol 38, 416-429. Bonneton, F. and Laudet, V. (2012). Evolution of nuclear receptors in insects. In Insect Endocrinology (ed. Lawrence I. Gilbert). pp. 219–252. New York: Academic Press. Boore, J.L., Lavrov, D.V. and Brown, W.M. (1998). Gene translocation links insects and crustaceans. Nature 392, 667–668. Bourguet, W., Ruff, M., Chambon, P., Gronemeyer, H. and Moras, D. (1995). Crystal structure of the ligand- binding domain of the human nuclear receptor RXRα. Nature 375, 377-382. Bourguet, W., Vivat, V., Wurtz, J.M., Chambon, P., Gronemeyer, H. and Moras, D. (2000). Crystal structure of a heterodimeric complex of RAR and RXR ligand-binding domains. Mol Cell 5, 289-298. Brelivet, Y., Kammerer, S., Rochel, N., Poch, O. and Moras, D. (2004). Signature of the oligomeric behaviour of nuclear receptors at the sequence and structural level. EMBO Rep 5, 423-429. Bridgham, J.T., Eick, G.N., Larroux, C., Deshpande, K., Harms, M.J., Gauthier, M.E.A., Ortlund, E.A., Degnan, B.M. and Thornton, J.W. (2010). Protein evolution by molecular tinkering: Diversitfication of the nuclear receptor superfamily from a ligand-dependent ancestor. PLoS Biol 8: e1000497. Doi:10.1371/journal.pbio.1000497. Broadus, J. and Doe, C.Q. (1995). Evolution of neuroblast identity: seven-up and prospero expression reveal homologous and divergent neuroblast fates in Drosophila and Schistocerca. Development 121, 3989-3996. Browning, C., Martin, E., Loch, C., Wurtz, J.M., Moras, D., Stote, R.H., Dejaegere, A.P. and Billas, I.M.L. (2007). Critical role of desolvation in the binding of 20-hydroxyecdysone to the ecdysone receptor. J Biol Chem 282, 32924-32934. Bruey-Sedano, N., Alabouvette, J., Lestradet, M., Hong, L., Girard, A., Gervasio, E., Quennedey, B., and Charles, J. P. (2005). The Drosophila ACP65A cuticle gene: Deletion scanning analysis of cis-regulatory sequences and regulation by DHR38. Genesis 43, 17–27. Bryant, Z., Subrahmanyan, L., Tworoger, M., LaTray, L., Liu, C-R., Li, M-J., van den Engh, G. and Ruohola- Baker, H. (1999). Characterization of differentially expressed genes in purified Drosophila follicle cells: Toward a general strategy for cell type-specific developmental analysis. Proc Natl Acad Sci USA 96, 5559-5564. Brzozowski, A.M., Pike, A.C.W., Dauter, Z., Hubbard, R.E., Bonn, T., Engstrsm, O., Ohman, L., Greene, G.L., - 166 -

References

Gustafsson, J-A. and Carlquist, M. (1997). Molecular basis of agonism and antagonism in the oestrogen receptor. Nature 389, 753-758. Bunce, C.M. and Campbell, M.J. (eds.) (2010). Nuclear receptors, Proteins and Cell Regulation 8, 455-457. DOI 10.1007/978-90-481-3303-1, Springer Science Business Media BV. Carapelli, A., Lio, P., Nardi, F., Van der Wath, E. and Frati, F. (2007). Phylogenetic analysis of mitochondrial protein coding genes confirms the reciprocal paraphyly of Hexapoda and Crustacea. BMC 7(suppl2), S8. Carlson, G.R., Dhadialla, T.S., Hunter, R., Jansson, R.K., Jany, C.S., Lidert, Z. and Slawecki, R.A. (2001). The chemical and biological properties of methoxyfenozide, a new insecticidal ecdysteroid agonist. Pest Manag Sci 57, 115-119. Carmichael, J.A., Lawrence, M.C., Graham, L.D., Pilling, P.A., Chandana Epa, V., Noyce, L., Lovrecz, G., Winkler, D.A., Pawlak-Skrezecz, A., Eaton, R.E., Hannan, G.N. and Hill, R.J. (2005). The X-ray structure of a hemipteran ecdysone receptor ligand-binding domain: Comparison with a Lepidopteran ecdysone receptor ligand-binding domain and implications for insecticide design. J Biol Chem 280, 22258–22269. Cerny, A.C., Grossmann, D., Bucher, G. and Klingler, M. (2008). The Tribolium ortholog of knirps and knirps- related is crucial for head segmentation but plays a minor role during abdominal patterning. Dev Biol 321: 284-294. Chavez, V.M., Marques, G., Delbecque, J.P., Kobayashi, K., Hollingsworth, M., Burr, J., Natzle, J.E. and O’Connor, M.B. (2000). The Drosophila disembodied gene controls late embryonic morphogenesis and codes for a cytochrome P450 enzyme that regulates embryonic ecdysone levels. Development 127, 4115- 4126. Chen C-K., Kkuhnlein, R.P., Eulenberg, K.G., Vincent, V., Affolter, M., Schuh, R. (1998). The transcription factors KNIRPS and KNIRPS RELATED control cell migration and branch morphogenesis during Drosophila tracheal development. Development 125, 4959-4968. Chen, X., Tian, H., Zou, L., Tang, B., Hu, J. and Zhang, W. (2008). Disruption of Spodoptera exigua larval development by silencing chitin synthase gene A with RNA silence. Bull Entomol Res, 98, 613-619. Chen, Y.H., Jia, X.T., Zhao, L., Li, C.Z., Zhang, S., Chen, Y.G., Weng, S.P. and He, J.G. (2011). Identification and functional characterization of Dicer2 and five single VWC domain proteins of Litopenaeus vannamei. Dev Comp Immunol 35, 661-671. Cheng, D., Xia, Q., Duan, J., Wei, L., Huang, C., Li, Z., Wang, G. and Xiang, Z. (2008). Nuclear receptors in Bombyx mori: Insights into genomic structure and developmental expression. Insect Biochem Mol Biol 38, 1130-1137. Chomczynski, P., and Sacchi, N. (1987). Single-step method of RNA isolation by acid guanidinium thiocyanate- phenol-chloroform extraction. Anal Biochem 162, 156-159. Christopherson, K.S., Mark, M.R., Bajaj, V. and Godowski, P.J. (1992). Ecdysteroid-dependent regulation of genes in mammalian cells by a Drosophila ecdysone receptor and chimeric transactivators. Proc Natl Acad Sci USA 98, 1549-1554. Chung, J.S. (2010). Hemolymphe ecdysteroids during the last three molt cycles of the blue crab, Callinectes sapidus: quantitative and qualitative analysis and regulation. Arch Insect Biochem Physiol 73, 1-13. Clayton, G., Peak-Chew, S., Evans, R. and Schwabe, J. (2001). The structure of the ultraspiracle ligand-binding domain reveals a nuclear receptor locked in an inactive conformation. Proc Natl Acad Sci USA 98, 1549- - 167 -

References

1554. Clemens, J.C., Worby, C.A., Simonson-Leff, N., Muda, M., Maehama, T., Hemmings, B.A. and Dixon, J.E. (2000). Use of double-stranded RNA Drosophila cell lines to dissect signal transduction pathways. Proc Natl Acad Sci USA, 97, 6499-6503. Crispi, S., Giordano, E., D’Avino, P.P., Furia, M. (1998). Cross-talking among Drosophila nuclear receptors at the promiscuous response element of the ng-1 and ng-2 intermolt genes. J Mol Biol 275, 561-574. Cristofoletti, P.T., Ribeiro, A.F., Deraison, C., Rahbé, Y. and Terra, W.R. (2003). Midgut adaptation and digestive enzyme distribution in a phloem feeding insect, the pea aphid Acyrthosiphon pisum. J Insect Physiol 49, 11-24. Cruz, J., Nieva, C., Mané-Padros, D., Martin, D. and Belles, X. (2008). Nuclear receptor BgFTZ-F1 regulates molting and the timing of ecdysteroid production during nymphal development in the hemimetabolous insect Blattella germanica. Dev Dyn 237, 3179-3191. Cook, C.E., Yue, Q. and Akam, M. (2005). Mitochondrial genomes suggest that hexapods and crustaceans are mutually paraphyletic. Proc R Soc B 272, 1295-1304. Costa, A., Jan, E., Sarnow, P. and Schneider, D. (2009). The Imd pathway is involved in antiviral immune responses in Drosophila. PLoS One 4, e7436. doi:10.1371/journal.pone.0007436. Dai, H., Ma, L., Wang, J., Jiang, R., Wang, Z. and Fei, J. (2008). Knockdown of ecdysis-triggering hormone gene with a binary UAS/GAL4 RNA interference system leads to lethal ecdysis deficiency in silkworm. Acta Biochim Biophys Sinica (Shanghai), 40, 790-795. Daniel, A., Dumstrei, K., Lengyel, J.A., and Hartenstein, V. (1999). The control of cell fate in the embryonic visual system by atonal, tailless and EGFR signaling. Development 126, 2945–2954. Davis, M.M., Yang, P., Chen, L., O’Keefe, S.L. and Hodgetts, R.B. (2007). The orphan nuclear receptor DHR38 influences transcription of the DOPA decarboxylase gene in epidermal and neural tissues of Drosophila melanogaster. Genome 50, 1049–1060. Dearden, P.K., Wilson, M.J., Sablan, L., Osborne, P., Havler, M., McNAughton, E. et al. (2006). Patterns of conservation and change in honey bee development genes. Genome Res 16, 1376-1384. Dedryver, C.-A., Le Ralec, A. and Fabre, F. (2010). The conflicting relationships between aphids and men: A review of aphid damage and control strategies. De Mendonça, R.L., Bouton, D., Bertin, B., Escriva, H., Noel, C., Vanacker, J.M., Cornette, J., Laudet, V. and Pierce, R.J. (2002). A functionally conserved member of the FTZ-F1 nuclear receptor family from Schistosoma mansoni. Eur J Biochem 269, 5700-5711. Denlinger, D.L., Yocum, G.D. and Rinehart, J.P. (2005). Hormonal control of diapause. In: Comprehensive Molecular Insect Science, Volume 3: Endocrinology (Eds L.I. Gilbert, K. Iatrou and S.S. Gill), pp. 615- 650. Pergamon Press, New York,. Dermauw, W., Wybouw, N., Rombauts, S., Menten, B., Vontas, J., Grbic, M., Clark, R.M., Feyereisen, R. and Van Leeuwen, T. (2013). A link between host plant adaptation and pesticide resistance in the polyphagous spider mite Tetranychus urticae. Proc Natl Acad Sci USA 110 no2, E113-E122. de Rosny, E., de Groot, A., Jullian-Binard, C., Gaillard, J., Borel, F., Pebay-Peyroula, E., Fontecilla-Camps, J.C. and Jouve, H.M. (2006). Drosophila nuclear receptor E75 is a thiolate hemoprotein. Biochemistry 45, 9727-9734. - 168 -

References de Rosny, E., de Groot, A., Jullian-Binard, C., Borel, F., Suarez, C., Le Pape, L., Fontecilla-Camps, J.C. and Jouve, H.M. (2008). DHR51, the Drosophila melanogaster homologue of the human photoreceptor cell- specific nuclear receptor, is a thiolate heme-binding protein. Biochemistry 47, 13252-13260. Dhadialla, T.S., Carlson, G.R. and Le, D.P. (1998). New insecticides with ecdysteroidal and juvenile hormone activity. Ann Rev Entomol 43, 545-569. Dhadialla, T.S., Retnakaran, A. and Smagghe, G. (2005). Insect growth- and development-disrupting insecticides, in Comprehensive Insect Molecular Science, (eds. Gilbert, L.I., Kostas, I. and Gill, S.). Vol. 6, pp. 55–116. Pergamon Press, New York,

Dhe-Paganon, S., Duda, K., Iwamoto, M., Chi, Y.I. and Shoelson, S.E. (2002). Crystal structure of the HNF4 alpha ligand binding domain in complex with endogenous fatty acid ligand. J Biol Chem 277, 37973- 37976. Dietzl, G., Chen, D., Schnorrer, F., Su, K.C., Barinova, Y., Fellner, M., Gasser, B., Kinsey, K., Oppel, S., Scheiblauer, S., Couto, A., Marra, V., Keleman, K. and Dickson, B.J. (2007). A genome-wide transgenic RNAi library for conditional gene inactivation in Drosophila. Nature 448, 151-156. Dow, J.A.T. (1992). pH gradients in lepidopteran midgut. J Exp Biol 172, 355-375. Dubrovskaya, V.A., Berger, E.M. and Dubrovsky, E.B. (2004). Juvenile hormone regulation of the E75 nuclear receptor is conserved in Diptera and Lepidoptera. Gene 340, 171–177. Dubrovsky, E.B., Dubrovskaya, V.A. and Berger, E.M. (2004). Hormonal regulation and functional role of Drosophila E75A orphan nuclear receptor in the juvenile hormone signaling pathway. Dev Biol 268, 258- 270. Durica, D.S., Wu, X., Anilkumar, G., Hopkins, P.M. and Chung, A.C. (2002). Characterization of crab EcR and RXR homologs and expression during limb regeneration and oocyte maturation. Mol Cell Endocrinol 28, 59-76. Elaine, B.A., Miguel, S.S. and Christopher, R.M. (2005). Mechanisms of hopperburn: an overview of insect taxonomy, behavior and physiology. Ann Rev Entomol 50, 125-151. Eleftherianos, I., Xu, M., Yadi, H., Ffrench-Constant, R.H. and Reynolds, S.E. (2009). Plasmatocyte-spreading peptide (PSP) plays a central role in insect cellular immune defenses against bacterial infection. J Exp Biol 212, 1840-1848. Escriva, H., Delaunay, F. and Laudet, V. (2000). Ligand binding and nuclear receptor evolution. Bioessays 22, 717-727. Escriva, H., Laudet, V. and Robinson-Rechavi, M. (2003). Nuclear receptors are markers of animal genome evolution. J Struct Funct Genomics 3, 177-184. Farooqui, T., Robinson, K., Vaessin, H. and Smith, B.H. (2003). Modulation of early olfactory processing by an octopaminergic reinforcement pathway in the honeybee. J Neurosci 23, 5370-5380. Feinberg, E.H. and Hunter, C.P. (2003). Transport of dsRNA into cells by the transmembrane protein SID-1. Science 301, 1545-1547. Felsenstein, J. (1985). Confidence limits on phylogenies: An approach using the bootstrap. Evolution 39, 783- 791. Fescemeyer, H.W., Masler, E.P., Kelly, T.J. and Lusby, W.R. (1995). Influence of development and - 169 -

References

prothoracicotropic hormone on the ecdysteroids produced in vitro by the prothoracic glands of female gypsy moth (Lymantria dispar) pupae and pharate adults. J Insect Physiol 41, 489-500. Finley, K.D., Taylor, B.J., Milstein, M. and McKeown, M. (1997). Dissatisfaction, a gene involved in sex- specific behavior and neural development of Drosophila melanogaster. Proc Natl Acad Sci USA 94, 913– 918. Finley, K.D., Edeen, P.T., Foss, M., Gross, E., Ghbeish, N., Palmer, R.H., Taylor, B.J. and McKeown, M. (1998). Dissatisfaction encodes a tailless-like nuclear receptor expressed in a subset of CNS neurons controlling Drosophila sexual behavior. Neuron 21, 1363–1374. Fire, A., Xu, S.Q., Montgomery, M.K., Kostas, S.A., Driver, S.E. and Mello, C.C. (1998). Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 391, 806-811. Fisk, G.J., and Thummel, C.S. (1995). Isolation, regulation, and DNA-binding properties of three Drosophila nuclear hormone receptor superfamily members. Proc Natl Acad Sci USA 92, 10604-10608. Fisk, G.J. and Thummel, C.S. (1998). The DHR78 nuclear receptor is required for ecdysteroid signaling during the onset of Drosophila metamorphosis. Cell 93, 543–555. Flynt, A., Liu, N., Martin, R. and Laï, E.C. (2008). Dicing of viral replication intermediates during silencing of latent Drosophila viruses. Proc Natl Acad Sci USA 106, 5270-5275. Forman, B.M., Goode, E., Chen, J., Oro, A.E., Bradley, D.J., Perlmann, T., Noonan, D.J., Burka, L.T., McMorris, T. and Lamph, W.W. (1995). Identification of a nuclear receptor that is activated by farnesol metabolites. Cell 81, 687-693. Forman, B.M., Tzameli, I., Choi, H.-S., Chen, J., Simha, D., Seol, W., Evans, R.M. and Moore, D.D. (1998). Androstane metabolites bind to and deactivate the nuclear receptor CAR-β. Nature 395, 612-615. Friedrich, M. and Tautz, D. (1995). Ribosomal DNA phylogeny of the major extant arthropod classes and the evolution of myriapods. Nature 376, 165–167. Galiana-Arnoux, D., Dostert, C., Schneemann, A., Hoffmann, J.A. and Imler, J.L. (2006). Essential function in vivo for Dicer-2 in host defense against RNA viruses in Drosophila. Nature Immunol 7, 590-597. Gampe, R.T., Montana, V.G., Lambert, M.H., Miller, A.B., Bledsoe, R.K., Milburn, M.V., Kliewer, S.A., Willson, T.M. and Xu, H.E. (2000). Asymmetry in the PPARγ/RXRα crystal structure reveals the molecular basis of heterodimerization among nuclear receptors. Mol Cell 5, 545-555. Garbutt, J.S. and Reynolds, S.E. (2012). Induction of RNA interference genes by double-stranded RNA; implications for susceptibility to RNA interference. Insect Biochem Mol Biol 42, 621-628. Garbutt, J.S., Bellés, X., Richards, E.H. and Reynolds, S.E. (2013). Persistence of double-stranded RNA in insect hemolymph as a potential determiner of RNA interference success: Evidence from Manduca sexta and Blattella germanica. J Insect Physiol 59, 171-178. Gatehouse, H.S., Gatehouse, L.N., Malone, L.A., Hodges, S., Tregidga, E. and Todd, J. (2004). Amylase activity in honey bee hypopharyngeal glands reduced by RNA interference. J Apicult Res 43, 9-13. Gatehouse, J.A. and Price, D.R.G. (2011). Protection of crops against insect pests using RNA interference. Insect Biotechnol 2, 145-168. Gauhar, Z., Sun, I.V., Hua, S., Mason, C.E., Fuchs, F., Li, T.R., Boutros, M. and White, K.P. (2009). Genomic mapping of binding regions for the ecdysone receptor protein complex. Genome Res 19, 1006-1013.

- 170 -

References

Gerardo, N., Altincicek, B., Anselme, C., Atamian, H., et al. (2010). Immunity and other defenses in pea pahids, Acyrthosiphon pisum. Genome Biol 11, R21 doi:10.1186/gb-2010-11-2-r21. Ghanim, M., Kontsedalov, S. and Czosnek, H. (2007). Tissue-specific gene silencing by RNA interference in the whitefly Bemisia tabaci (Gennadius). Insect Biochem Mol Biol 37, 732-738. Gissendanner, C.R., Crossgrove, K., Kraus, K.A., Maina, C.V. and Sluder, A.E. (2004). Expression and function of conserved nuclear receptor genes in C. elegans. Dev Biol 266, 399-416. Glass, C.K., Rosenfeld, M.G. (2000). The coregulator exchange in transcriptional functions of nuclear receptors. Genes Dev 14, 121-141. Gordon, K.H.J. and Waterhouse, P.M. (2007). RNAi for insect-proof plants. Nature Biotech 25, 1231-1232. Goulson, D. (2010). Bumblebees: behaviour, ecology and conservation. Oxford University Press, 317p. Graham, L., Pilling, P., Eaton, R.E., Gorman, J.J., Braybrook, C., Hannan, G.N., Pawlak-Skrzecz, A., Noyce, L., Lovrecz, G.O., Lu, L. and Hill, R.J. (2007). Purification and characterization of recombinant ligand- binding domains from the ecdysone receptors of four pest insects. Protein Expr Purif 53, 309-324. Grbic, M., Van Leeuwen, T., Richard, M.C. et al. (2011). The genome of Tetranychus urticae reveals herbivorous pest adaptations. Nature, 479:487-492. Gu, L., Knipple, D.C. (2013). Recent advances in RNA interference research in insects: implications for future insect pest management strategies. Crop Protection 45, 36-40. Guo, X., Harmon, M.A., Laudet, V., Mangelsdorf, D.J. and Palmer M.J. (1997). Isolation of a functional ecdysteroid receptor homologue from the Ixodid tick Amblyomma americanum (L.). Insect Biochem Mol Biol 27:945-962. Guo, X., Xu, Q., Harmon, M.A., Jin, X., Laudet, V., Mangelsdorf, D.J. and Palmer M.J. (1998). Isolation of two functional retinoid X receptor subtypes from the Ixodid tick, Amblyomma americanum (L.). Mol Cell Endocrinol 139, 45-60. Hartfelder, K. and Emlen, D.J. (2012). Endocrine control of insect polyphenism. In Insect Endocrinology (ed. Lawrence I. Gilbert). pp. 464–522. New York: Academic Press. Havelda, Z., Hornyik, C., Valoczi, A., Burgyan, J. (2005). Defective interfering RNA hinders the activity of a tombusvirus-encoded posttranscriptional gene silencing suppressor. J Virol 79, 450-457. Henrich, V. (2012). The ecdysteroid receptor. In Insect Endocrinology (ed. Lawrence I. Gilbert). pp. 219–252. New York: Academic Press. Hill, D.S. (1987). Agricultural insect pests of temperate regions and their control. Cambridge University Press, 1987, 659p. Hirai, M., Shinoda, T., Kamimura, M., Tomita, S. and Shiotsuki, T. (2002). Bombyx mori orphan receptor, BmHR78: cDNA cloning, testis abundant expression and putative dimerization partner for Bombyx Ultraspiracle. Mol Cell Endocrin 189, 201-211. Hiromi, Y., Mlodzik, M., West, S.R., Rubin, G.M., and Goodman, C.S. (1993). Ectopic expression of seven-up causes cell fate changes during ommatidial assembly. Development 118, 1123-1135. Hiruma, K. and Riddiford, L.M. (2001). Regulation of transcription factors MHR4 and βFTZ-F1 by 20- hydroxyecdysone during a larval molt in the tobacco hornworm, Manduca sexta. Dev Biol 162, 426-437. Holt, R.A., Subramanian, G.M., Halpern, A., et al. (2002). The genome sequence of the malaria mosquito Anopheles gambiae. Science 298, 129-149. - 171 -

References

Horigane, M., Ogihara, K., Nakajima, Y., Shinoda, T. and Taylor, D. (2007). Cloning and expression of the ecdysteroid receptor during ecdysis and reproduction in females of the soft tick Ornithodoros moubata (Acari: Argasidae). Gen Comp Endocrinol 156, 298-311. Horigane, M., Ogihara, K., Nakajima, Y. and Taylor, D. (2008). Isolation and expression of the retinoid X receptor from last instar nymphs and adult females of the soft tick Ornithodoros moubata (Acari: Argasidae). Insect Mol Biol 16, 601-612. Horner, M.A., Pardee, K., Liu, S., King-Jones, K., Lajoie, G., Edwards, A., Krause, H.M. and Thummel, C.S. (2009). The Drosophila DHR96 nuclear receptor binds cholesterol and regulates cholesterol homeostasis. Genes Dev 23, 2711-2716. Hoshizaki, D.K., Blackburn, T., Price, C., Ghosh, M., Miles, K., Ragucci, M. and Sweis, R. (1994). Embryonic fat-cell lineage in Drosophila melanogaster. Development 120, 2489-2499. Huang, J.H. and Lee, H.J. (2011). RNA interference unveils functions of the hypertrehalosemic hormone on cyclic fluctuation of hemolymph trehalose and oviposition in the virgin female Blattella germanica. J Insect Physiol 57, 858-864. Huet, F., Ruiz, C. and Richards, G. (1995). Sequential gene activation by ecdysone in Drosophila melanogaster: the hierarchical equivalence of early and early late genes. Development 121, 1195-1204. Hughes, C.L. and Kaufman, T.C. (2000). RNAi analysis of Deformed, proboscipedia and Sex combs reduced in the milkweed bug Oncopeltus fasciatus: novel roles for Hox genes in the Hemipteran head. Development 127, 3683-3694. Hunter, C.P. (1999). Genetics: a touch of elegance with RNAi. Curr Biol 9, R440-442. Huvenne, H. and Smagghe, G. (2010). Mechanisms of dsRNA uptake in insects and potential of RNAi for pest control: a review. J Insect Physiol 56, 227-235. Hwang, U.W., Friedrich, M., Tautz, D., Park, C.J. and Kim, W. (2001). Mitochondrial protein phylogeny joins myriapods with chelicerates. Nature 413, 154–157. Iga, M. and Smagghe, G. (2009). Identification and expression profile of Halloween genes involved in ecdysteroid biosynthesis in Spodoptera littoralis. Peptides 31, 456-467. Iwema, T., Villas, I., Beck, Y., Bonneton, F., Nierengarten, H., Chaumont, A., Richards, G., Laudet, V. and Moras, D. (2007). Structural and functional characterization of a novel type of ligand-independent RXR- USP receptor. EMBO J 26, 3770-3782. Iwema, T., Chaumot, A., Studer, R.A., Robinson-Rechavi, M., Billas, I.M.L., Moras, D., Laudet, V. and Bonneton, F. (2009). Structural and evolutionary innovation of the hetero-dimerization interface between USP and the ecdysone receptor ECR in insects. Mol Biol Evol 26, 753-768. Jaubert-Possamai, S., Le Trionnaire, G., Bonhomme, J., Christophides, G.K., Rispe, C. and Tagu, D. (2007). Gene knockdown by RNAi in the pea aphid Acyrthosiphon pisum. BMC Biotechnol 7:63. Jaubert-Possamai, S., Rispe, C., Tanguy, S., Gordon, K., Walsh, T., Edwards, O. and Tagu, D. (2010). Expansion of the miRNA pathway in the hemipteran insect Acyrthosiphon pisum. Mol Biol Evol 27, 979- 987. Jensen, E.V. and Jacobson, H.I. (1960). Fate of steroid estrogens in target tissues. In Biological Activities of Steroids in Relation to Cancer, (ed. Pincus, G. and Vollmer, E.P.) pp. 161–74. New York: Academic Press.

- 172 -

References

Jetten, A.M. (2009). Retinoid-related orphan receptors (RORs): critical roles in development, immunity, circadian rhythm, and cellular metabolism. Nucl Recept Signal 7, e003. Jiang, C., Lamblin, A.F., Steller, H. and Thummel, C.S. (2000). A steroid-triggered transcriptional hierarchy controls salivary gland cell death during Drosophila metamorphosis. Mol Cell 5, 445-455. Jones, G., Jones, D., Teal, P., Sapa, A. and Wozniak, M. (2006). The retinoid-X receptor orthologue, ultraspiracle, binds with nanomolar affinity to an endogenous morphogenetic ligand. FEBS J 273, 4983- 4996. Kageyama, Y., Masuda, S., Hirose, S., and Ueda, H. (1997). Temporal regulation of the mid-prepupal gene FTZ- F1: DHR3 early late gene product is one of the plural positive regulators. Genes Cells 2, 559-569. Kamath, R.S., Martinez-Campos, M., Zipperlen, P., Frasher, A.G. and Ahringer, J. (2000). Effectiveness of specific RNA-mediated interference through ingested double-stranded RNA in Caenorhabditis elegans. Genome Biol 2, research0002. Kanai, M.I., Okabe, M., and Hiromi, Y. (2005). Seven-up controls switching of transcription factors that specify temporal identities of Drosophila neuroblasts. Dev Cell 8, 203-213. Kanginakudru, S., Royer, C., Edupalli, S.V., Jalabert, A., Mauchamp, B., Chandrashekaraiah, Prasad, S.V., Chavancy, G., Couble, P. and Nagaraju, J. (2007). Targeting ie-1 gene by RNAi induces baculoviral resistance in lepidopteran cell lines and in transgenic silkworms. Insect Mol Biol 16, 635-644. Kapitskaya, M.Z., Dittmer, N.T., Deitsch, K.W., Cho, W.L., Taylor, D.G., Leff, T. and Raikhel, A.S. (1998). Three isoforms of a hepatocyte nuclear factor-4 transcription factor with tissue- and stage-specific expression in the adult mosquito. J Biol Chem 273, 29801–29810. Kapitskaya, M.Z., Li, C., Miura, K., Segraves, W. and Raikhel, A.S. (2000). Expression of the early-late gene encoding the nuclear receptor HR3 suggests its involvement in regulating the vitellogenic response to ecdysone in the adult mosquito. Mol Cell Endocrinol 160, 25-37. Karimullina, E., Li, Y., Ginjupalli, G.K. and Baldwin, W.S. (2012). Daphnia HR96 is a promiscuous xenobiotic and endobiotic nuclear receptor. Aguat Toxicol 116-117, 69-78. Doi: 10.1016/j.aquatox.2012.03.005. Kasuya, A., Sawada, Y., Tsukamoto, Y., Tanaka, K., Toya, T., et al. (2003). Binding mode of ecdysone agonists to the receptor: comparative modeling and docking studies. J Mol Model 9, 58-65. Kennedy, S., Wang, D. and Ruvkun, G. (2004). A conserved siRNA-degrading RNase negatively regulates RNA interference in C. elegans. Nature 427, 645-649. Kennerdell, J.R. and Carthew, R.W. (1998). Use of dsRNA-mediated genetic interference to demonstrate that frizzled and frizzled 2 act in the wingless pathway. Cell 95, 1017-1026. Kennerdell, J.R. and Carthew, R.W. (2000). Heritable gene silencing in Drosophila using double-stranded RNA. Nature Biotechnol 18, 896-898. Kerber, B., Fellert, S., and Hoch, M. (1998). Seven-up, the Drosophila homolog of the COUP-TF orphan receptors, controls cell proliferation in the insect kidney. Genes Dev 12, 1781-1786. Khila, A. and Grbic, M. (2007). Gene silencing in the spider mite Tetranychus urticae: dsRNA and siRNA parental silencing of the Distal-less gene. Dev Genes Evol 217, 241-251. King-Jones, K. and Thummel, C.S. (2005). Nuclear receptors - a perspective from Drosophila. Nat Rev Genet 6, 311-323. King-Jones, K., Horner, M.A., Lam, G. and Thummel, C.S. (2006). The DHR96 nuclear receptor regulates - 173 -

References

xenobiotic responses in Drosophila. Cell Metab 4, 37-48. Kostrouchova, M., Krause, M., Kostrouch, Z. and Rall, J.E. (2001). Nuclear hormone receptor CHR3 is a critical regulator of all four larval molts of the nematode Caenorhabditis elegans. Proc Natl Acad Sci USA 98, 7360-7365. Kozlova, T., Pokholkova, G.V., Tzertzinis, G., Sutherland, J.D., Zhimulev, I.F. and Kafatos, F.C. (1998). Drosophila hormone receptor 38 functions in metamorphosis: A role in adult cuticle formation. Genetics 149, 1465-1475. Kozlova, T. and Thummel, C.S. (2003). Essential roles for ecdysone signaling during Drosophila mid-embryonic development. Science 301, 1911-1914. Kramer, S., West, S.R., Hiromi, Y. (1995). Cell fate control in the Drosophila retina by the orphan receptor seven-up: its role in the decisions mediated by the ras signaling pathway. Development 121, 1361-1372. Kumar, P., Pandit, S.S. and Baldwin, L.T. (2012). Tobacco rattle virus vector: a rapid and transient means of silencing Manduca sexta genes by plant mediated RNA interference. PLoS ONE 7, e31347. Kupsco, J.M., Wu, M-J, Marzluff, W.F., Thapar, R. and Duronio, R.J. (2006). Genetic and biochemical characterization of Drosophila Snipper: A promiscuous member of the metazoan 3’hExo/ERI-1 family of 3’ to 5’ exonucleases. RNA 12, 2103-2117. Lam, G.T., Jiang, C. and Thummel C.S. (1997). Coordination of larval and prepupal gene expression by the DHR3 orphan receptor during Drosophila metamorphosis. Development 124, 1757-1969. 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). CLUSTALW and CLUSTALX Version 2. Bioinformatics 23, 2947-2948. Laskowski R.A., MacArthur M.W., Moss D.S. and Thornton J.M. (1993). PROCHECK: a program to check the stereochemistry of protein structures. J Appl Cryst 26, 283-291. Laudet, V. (1997). Evolution of the nuclear receptor superfamily: early diversification from an ancestral orphan receptor. J Mol Endocinol 19, 207-226. Laudet, V., Gronemeyer, H. (2002). The Nuclear Receptor Facts Book. New York: Academic Press. Lavorgna, G., Ueda, H., Clos, J. and Wu, C. (1991). FTZ-F1, a steroid hormona receptor-like protein implicated in the activation of fushi tarazu. Science 252, 848-851. Lespinet, O., Wolf, Y.I., Koonin, E.V. and Aravind, L. (2002). The role of lineage-specific expansion in the evolution of eukaryotes. Genome Res 12, 1048-1059. Levin, D.M., Breuer, L.N., Zhuang, S., Anderson, S.A., Nardi, J.B. and Kanost, M.R. (2005). A hemocyte- specific integrin required for hemocytic encapsulation in the tobacco hornworm, Manduca sexta. Insect Biochem Mol Biol 35, 369-380. Lewis, S.E., Searle, S.M.J., Harris, N., Gibson, M., Iyer, V., Richter, J. et al. (2002). Apollo: a sequence annotation editor. Genome Biol 3, research0082.1-0082.14. Lezzi, M., Bergman, T., Henrich, V., Vögtli, M., Frömel, C., Grebe, M., Przibilla, S. and Spindler-Barth, M. (2002). Ligand-induced heterodimerization between the ligand binding domains of the Drosophila ecdysteroid receptor and ultraspiracle. Eur J Biochem 269, 3237-3245. Li, C., Schwabe, J., Banayo, E. and Evans, R. (1997). Coexpression of nuclear receptor partners increases their solubility and biological activities. Proc Natl Acad Sci USA 94, 2278-2283. - 174 -

References

Li, H., Li, W.X. and Ding, S.W. (2002). Induction and suppression of RNA silencing by an animal virus. Science 296, 1319-1321. Li, Y., Warren, J.T., Boysen, G., Gilbert, L.I., Gold, A., Sangaiah, R., Ball, L. and Swenberg, J.A. (2006). Profiling of ecdysteroids in complex biological samples using liquid chromatography/ion trap mass spectrometry. Rapid Commun Mass Spectrom 20, 185-192. Li, J., Wang, X.-P., Wang, M.-Q., Ma, W.-H. and Hua, H.-X. (2013). Advances in the use of the RNA interference technique in Hemiptera. Insect Sci 20, 31-39. Lindblom, T.H., Pierce, G.J. and Sluder, A.E. (2001). C. elegans orphan nuclear receptor contributes to xenobiotic resistance. Curr Bioll 11, 864-868. Liu, P.Z. and Kaufman, T.C. (2004). Kruppel is a gap gene in the intermediate germband insect Oncopeltus fasciatus and is required for development of both blastoderm and germ band-derived segments. Development 131, 4567-4579. Liu, S., Ding, Z., Zhang, C., Yang, B. and Liu, Z. (2010). Gene knockdown by intro-thoracic injection of double- stranded RNA in the brown planthopper, Nilaparvata lugens. Insect Biochem Mol Biol 40, 666-671. Liu, J., Swevers, L., Iatrou, K., Huvenne, H. and Smagghe, G. (2012). Bombyx mori DNA/RNA non-specific nuclease: expression of isoforms in insect culture cells, subcellular localization and functional assays. J Insect Physiol 58, 1166e1176. Liu, J., Smagghe, G. and Swevers, L. (2013). Transcriptional response of BmToll9-1 and RNAi machinery genes to exogenous dsRNA in the midgut of Bombyx mori. J Insect Physiol, In press. Lum, L., Yao, S., Mozer, B., Rovescalli, A., Kessler, D.V., Nirenberg, M. and Beachy, P.A. (2003). Identification of Hedgehog pathway components by RNAi in Drosophila cultured cells. Science 299, 2039-2045. Luo, Y., Wang, X., Yu, D. and Kang, L. (2012). The SID-1 double-stranded RNA transporter is not required for systemic RNAi in the migratory locust. RNA Biol 9, 663e671. Maeda, S., Nakashima, A., Yamada, R., Hara, N., Fujimoto, Y., Ito, Y. and Sonobe, H. (2008). Molecular cloning of ecdysone 20-hydroxylase and expression pattern of the enzyme during embryonic development of silkworm Bombyx mori. Comp Biochem Physiol 149B, 507-516. Maglich, J.M., Caravella, J.A., Lambert, M.H., Willson, T.M., Moore, J.T. and Ramamurthy, L. (2003). The first completed genome sequence from a teleost fish (Fugu rubripes) adds significant diversity to the nuclear receptor superfamily. Nucleic Acids Res 31, 4051-4058. Mallatt, J.M., Garey, J.R. and Shultz, J.W. (2004). Ecdysozoan phylogeny and Bayesian inference: first use of nearly complete 28S and 18S rRNA gene sequences to classify the arthropods and their kin. Mol Phylogenet Evol 31, 178–191. Mané-Padros, D., Cruz, J., Vilaplana, L., Pascual, N., Belles, X. and Martin, D. (2008). The nuclear hormone receptor BgE75 links molting and developmental progression in the direct-developing insect Blattella germanica. Dev Biol 315, 147-160. Mané-Padros, D., Borran-Castells, F., Belles, X. and Martin, D. (2012). Nuclear receptor HR4 plays an essential role in the ecdysteroid-triggered gene cascade in the development of the hemimetabolous insect Blattella germanica. Moll Cell Endocrinol 348, 322-330.

- 175 -

References

Mao, Y.B., Cai, W.J., Wang, J.W., Hong, G.J., Tao, X.Y., Xue, X.Y., Wang, L.J., Huang, Y.P. and Chen X.Y. (2007). Silencing a cotton bollworm P450 monooxygenase gene by plant-mediated RNAi impairs larval tolerance of gossypol. Nature Biotechnol 25, 1307-1313. Mao, Y.B., Tao, X.Y., Xue, X.Y., Wang, L.J. and Chen, X.Y. (2011). Cotton plants expressing CYP6AE14 double-stranded RNA show enhanced resistance to bollworms. Transgenic Res 20, 665-673. Marco Antonio, D.S., Guidugli-Lazzarini, K.R., do Nascimento, A.M., Simões, Z.L. and Hartfelder, K. (2008). RNAi-mediated silencing of vitellogenin gene function turns honeybee (Apis mellifera) workers into extremely precocious foragers. Biomed Life Sci 95, 952-961. Martin, D., Maestro, O., Cruz, J., Mané-Padros, D. and Bellés, X. (2006) RNAi studies reveal a conserved role for RXR in molting in the cockroach Blattella germanica. J Insect Physiol 52, 410-416. Martin, D. (2010). Functions of nuclear receptors in insect development. In Nuclear Receptors. Current concepts and future challenges (Vol. 8) (Eds. C.M. Bunce & M.J. Campbell): Springer. Mas M.T., Smith K.C., Yarmush D.L., Aisaka K. and Fine R.M. (1992). Modeling the anti-CEA antibody combining site by homology and conformational search. Proteins Struc Func Genet 14, 483-498. Matsuoka, T. and Fujiwara, H. (2000). Expression of ecdysteroid-regulated genes is reduced specifically in the wing discs of the wing-deficient mutant (fl) of Bombyx mori. Dev Genes Evol 210, 120–128. Melo, F. and Feytmans, E. (1998). Assessing protein structures with a non-local atomic interaction energy. J Mol Biol 277, 1141-1152. Miller, S.C., Miyata, K., Brown, S.J., Tomoyasu, Y. (2012). Dissecting systemic RNA interference in the red flour beetle Tribolium castaneum. Parameters affecting the efficiency of RNAi. PLoS ONE 7: e47431. doi: 10.1371/journal.pone.0047431. Minakuchi, C., Ogura, T., Miyagawa, H. and Nakagawa, Y. (2007). Effects of the structures of ecdysone receptor (EcR) and ultraspiracle (USP) on the ligand-binding activity of the EcR/USP heterodimer. J Pestic Science 32, 379-384. Miura, K., Zhu, J., Dittmer, N.T., Chen, L., and Raikhel, A.S. (2002). A COUP-TF/Svp homolog is highly expressed during vitellogenesis in the mosquito Aedes aegypti. J Mol Endocrinol 29, 223-238. Mlodzik, M., Hiromi, Y., Weber, U., Goodman, C.S., and Rubin, G.M. (1990). The Drosophila seven-up gene, a member of the steroid receptor gene superfamily, controls photoreceptor cell fates. Cell 60, 211–224. Moore, J.T., Collins, J.L. and Pearce, K.H. (2006). The nuclear receptor superfamily and drug discovery. Chem Med Chem 1, 504-523. Moran, E. and Jimenez, G. (2006). The tailless order receptor acts as a dedicated repressor in the early Drosophila embryo. Mol Cell Biol 26, 3446-3454. Moras, D. and Gronemeyer, H. (1998). The nuclear receptor ligand-binding domain: structure and function. Curr Opin Cell Biol 10, 384-391. Moriyama, Y., Sakamoto, T., Karpova, S.G., Matsumoto, A., Noji, S. and Tomioka, K. (2008). RNA interference of the clock gene period disrupts circadian rhythms in the cricket Gryllus bimaculatus. J Biol Rhythms 23, 308-318. Motola, D.L., Cummins, C.L., Rottiers, V., Sharma, K.K., Li, t., Li, Y., Suino-Powell, K., Xu, H.E., Auchus, R.J., Antebi, A. and Mangelsdorf, D.J. (2006). Identification of ligands for DAF-12 that govern dauer formation and reproduction in C. elegans. Cell 124, 1209-1223. - 176 -

References

Mutti, N.S., Park, Y., Reese, J.C. and Reeck, G.R. (2006). RNAi knockdown of a salivary transcript leading to lethality in the pea aphid, Acyrthosiphon pisum. J Insect Sci 6, 1-7. Mykles, D. (2011). Ecdysteroid metabolism in crustaceans. J Steroid Biochem Mol Biol 127, 196-203. Nakagawa, Y., Sakai, A., Magata, F., Ogura, T., Miyashita, M. and Miyagawa, H. (2007). Molecular cloning of the ecdysone receptor and the retinoid X receptor from the scorpion Liocheles australasiae. FEBS J 274, 6191-6203. Nakamura, T., Mito, T., Miyawaki, K., Ohuchi, H. and Noji, S. (2008). EGFR signaling is required for re- establishing the proximodistal axis during distal leg regeneration in the cricket Gryllus bimaculatus nymph. Dev Biol 319, 46-55. Namiki, T., Niwa, R., Sakudoh, T., Shirai, K., Takeuchi, H. and Kataoka, H. (2005). Cytochrome P450 CYP307A1/Spook: a regulator for ecdysone synthesis in insects. Biochem Biophys Res Commun 337, 367-374. Narasimhan, S., Montgomery, R.R., DePonte, K., Tschudi, C., Marcantonio, N., Anderson, J.F., Saber, J.R., Cappello, M., Kantor, F.S. and Fikrig, E. (2004). Disruption of Ixodes scapularis anticoagulation by using RNA interference. Proc Natl Acad Sci USA 101, 1141-1146. Nauber, U., Pankratz, M.J., Lehmann, R., Kienlin, A., Seifert, E., Klemm, U. and Jackle, H. (1988). Abdominal segmentation of the Drosophila embryo requires a hormone receptor-like protein encoded by the Gap gene knirps. Nature 336, 489-492. Nayak, A., Berry, B., Tassetto, M., Kunitomi, M., Acevedo, A., Deng, C., Krutchinsky, A., Gross, J., Antoniewski, C. and Andino, R. (2010). Cricket paralysis virus antagonizes Argonaute2 to modulate antiviral defense in Drosophila. Nature Struct Mol Biol 17, 547-554. Niwa, R., Matsuda, T., Yoshiyama, T., Namiki, T., Mita, K., Fujimoto, Y., and Kataoka, H. (2004). CYP306A1, a cytochrome P450 enzyme, is essential for ecdysteroid biosynthesis in the prothoracic glands of Bombyx and Drosophila. J Biol Chem 279, 35942-35949. Niwa, R., Sakudoh, T., Namiki, T., Saida, K., Fujimoto, Y., and Kataoka, H. (2005). The ecdysteroidogenic P450 Cyp302a1/disembodied from the silkworm, Bombyx mori, is transcriptionally regulated by prothoracicotropic hormone. Insect Mol Biol 14, 563-571. Nowickyj, S., Chithalen, J., Cameron, D., Tyshenko, M., Petkovich, M., Wyatt, G., Jones, G. and Walker, V. (2008). Locust retinoid X receptors: 9-Cis-retinoic acid in embryos from a primitive insect. Proc Natl Acad Sci USA 105, 9540-9545. Ohno, C.K. and Petkovich, M. (1993). FTZ-F1 beta, a novel member of the Drosophila nuclear receptor family. Mech Devel 40, 13-24. Ono, H., Rewitz, K.F., Shinoda, T., Itoyama, K., Petryk, A., Rybczynski, R., Jarcho, M., Warren, J.T., Marqués, G., Shimell, M.J., Gilbert, L.I. and O'Connor, M.B. (2006). Spook and Spookier code for stage-specific components of the ecdysone biosynthetic pathway in Diptera. Dev Biol 298, 555-570. Oro, A.E., McKeown, M. and Evans, R.M. (1990). Relationship between the product of the Drosophila ultraspiracle locus and the vertebrate retinoid X receptor. Nature 347, 298–301. Pak, J. and Fire, A. (2007). Distinct populations of primary and secondary effectors during RNAi in C. elegans. Science 315, 241-244. Palanker, L., Necakov, A.S., Sampson, H.M., Ni, R., Hu, C., Thummel, C.S. and Krause, H.M. (2006). Dynamic - 177 -

References

regulation of Drosophila nuclear receptor activity in vivo. Development 133, 3549-3562. Palanker, L., Tennessen, J.M., Lam, G., and Thummel, C.S. (2009). Drosophila HNF4 regulates lipid mobilization and beta-oxidation. Cell Metab 9, 228–239. Palli, S.R., Hormann, R.E., Schlattner, U., and Lezzi, M. (2005). Ecdysteroid receptors and their applications in agriculture and medicine. Vitam Horm 73, 59–100. Palmer, M.J., Harmon, M.A. and Laudet, V. (1999). Characterization of EcR and RXR homologues in the ixodid tick, Amblyomma americanum (L.). Integr Comp Biol 39, 747-757. Panfilio, K.A., Liu, P.Z., Akam, M., Kaufman, T.C. (2006). Oncopeltus fasciatus zen is essential for serosal tissue function in katatrepsis. Dev Biol 292, 226-243. Parthasarathy, R., Tan, A. and Palli, S.R. (2008). Transcription factor broad suppresses precocious development of adult structures during larval-pupal metamorphosis in the red flour beetle, Tribolium castaneum. Mech Develop 125, 299-313. Parvy, J.P., Blais, C., Bernard, F., Warren, J.T., Petryk, A., Gilbert, L.I., O’Connor, M.B. and Dauphin- Villemant, C. (2005). A role for betaFTZ-F1 in regulating ecdysteroid titers during post-embryonic development in Drosophila melanogaster. Dev Biol 282, 84-94. Pascussi, J-M., Gerbal-Chaloin, S., Duret, C., Daujat-Chavanieu, M., Vilarem, M-J. and Maurel, P. (2008). The tangle of nuclear receptors that controls xenobiotic metabolism and transport: Crosstalk and consequences. Annu Rev Pharmacol Toxicol 48, 1-32. Paul, R.K., Takeuchi, H. and Kubo, T. (2006). Expression of two ecdysteroid-regulated genes, Broad-complex and E75, in the brain and ovary of the honeybee (Apis mellifera L.). Zool Sci 23, 1085-1092. Petryk, A., Warren, J.T., Marqués, G., Jarcho, M.P., Gilbert, L.I., Kahler, J., Parvy, J.P., Li, Y., Dauphin- Villemant, C., and O'Connor, M.B. (2003). Shade is the Drosophila P450 enzyme that mediates the hydroxylation of ecdysone to the steroid insect molting hormone 20-hydroxyecdysone. Proc Natl Acad Sci USA 100, 13773-13778. Phelps, C., Gburcik, V., Suslova, E., Dudek, P., Forafonov, F., Bot, N., MacLean, M., Fagan, R.J. and Picard, D. (2006). Fungi and animals may share a common ancestor to nuclear receptors. Proc Natl Acad Sci USA 103, 7077-7081. Pierceall, W.E., Li, C., Biran, A., Miura, K., Raikhel, A.S. and Segraves, W.A. (1999). E75 expression in mosquito ovary and fat body suggests reiterative use of ecdysone-regulated hierarchies in development and reproduction. Mol Cell Endocrinol 150, 73–89. Pignoni, F., Baldarelli, R.M., Steingrímsson, E., Diaz, R.J., Patapoutian, A., Merriam, J.R., and Lengyel, J.A. (1990). The Drosophila gene tailless is expressed at the embryonic termini and is a member of the steroid receptor superfamily. Cell 62, 151–163. Pineda, S., Schneider, M.I., Smagghe, G., Martínez, A.M., Del Estal, P., Viñuela, E., Valle, J. and Budia, F. (2007). Lethal and sublethal effects of methoxyfenozide and spinosad on Spodoptera littoralis. J Econ Entomol 100, 773-780. Pitino, M., Coleman, A.D., Maffei, M.E., Ridout, C.J. and Hogenhout, S.A. (2011). Silencing of aphid genes by dsRNA feeding from plants. PLOS ONE 6, e25709. Ponder J.W. and Richards F.M. (1987). Tertiary templates for proteins. Use of packing criteria in the enumeration of allowed sequences for different structural classes. J Mol Biol 193, 775-791. - 178 -

References

Preitner, N., Damiola, F., Lopez-Molina, L., Zakany, J., Duboule, D., Albrecht, U., Schibler, U. (2002). The orphan nuclear receptor REV-ERB α controls circadian transcription within the positive limb of the mammalian circadian oscillator. Cell 110, 251-260. Price, D.R.G. and Gatehouse, J.A. (2008). RNAi-mediated crop protection against insects. Trends Biotechnol 26, 393-400. Prosser, W.A. and Douglas, A.E. (1992). A test of the hypotheses that nitrogen is upgraded and recycled in an aphid (Acyrthosiphon pisum) symbiosis. J Insect Physiol 38, 93-99. Raikhel, A.S., Miura, K. and Segraves, W.A. (1999). Nuclear receptors in mosquito vitellogenesis. Am Zool 39, 722–735. Raikhel, A.S., Brown, M.R. and Belles, X. (2005). Hormonal control of reproductive processes. In: Comprehensive Molecular Insect Science, Volume 3: Endocrinology (Eds L.I. Gilbert, K. Iatrou and S.S. Gill), pp. 433-491. Elsevier, Pergamon. Rajagopal, R., Sivakumar, S., Agrawal, N., Malhotra, P. and Bhatnagar, R.K. (2002). Silencing of midgut aminopeptidase N of Spodoptera litura by double-stranded RNA establishes its role as Bacillus thuringiensis toxin receptor. J Biol Chem 277, 46849-46851.

Rasmont, P. & Mersch, P. (1988). Première estimation de la dérive faunique chez les bourdons de la Belgique (Hymenoptera: Apidae). Ann Soc R Zool Belg 118, 141-147. Regier, J.C., Shultz, J.W. and Kambic, R.E. (2005). Pancrustacean phylogeny: hexapods are terrestrial crustaceans and maxillopods are not monophyletic. Proc R Soc B 272, 395-401. Regier, J.C., Shultz, J.W., Zwick, A., Hussey, A., Ball, B., Wetzer, R., Martin, J.W. and Cunningham, C.W. (2010). Arthropod relationships revealed by phylogenenomic analysis of nuclear protein-coding sequences. Nature 463, 1079-1083. Reinking, J., Lam, M.M.S., Pardee, K., Sampson, H.M., Liu, S., Yang, P., Williams, S., White, W., Lajoie, G., Edwards, A. and Krause, H.M. (2005). The Drosophila nuclear receptor E75 contains heme and is gas responsive. Cell 122, 195-207. Reitzel, A.M. and Tarrant, A.M. (2009). Nuclear receptors complement of the cnidarian Nematostella vectensis: phylogenetic relationships and developmental expression patterns. BMC Evol Biol 9, 230. Renaud, J.P., Rochel, N., Ruff, M., Vivat, V., Chambon, P., Gronemeyer, H., Moras, D. (1995). Crystal structure of the RARγ ligand-binding domain bound to all-trans retinoic acid. Nature 378, 681-689. Rewitz, K.F., Rybczynski, R., Warren, J.T., and Gilbert, L.I. (2006a). Developmental expression of Manduca shade, the P450 mediating the final step in molting hormone synthesis. Mol Cell Endocrinol 247, 166-174. Rewitz, K.F., Rybczynski, R., Warren, J.T., and Gilbert, L.I. (2006b). Identification, characterization and developmental expression of Halloween genes encoding P450 enzymes mediating ecdysone biosynthesis in the tobacco hornworm, Manduca sexta. Insect Biochem Mol Biol 36, 188-199. Rewitz, K.F., O'Connor, M.B., Gilbert, L.I. (2007). Molecular evolution of the insect Halloween family of cytochrome P450s: phylogeny, gene organization and functional conservation. Insect Biochem Mol Biol 37, 741-753. Rewitz, K.F., Gilbert, L.I. (2008). Daphnia Halloween genes that encode cytochrome P450s mediating the synthesis of the arthropod molting hormone: Evolutionary implications. BMC Evol Biol 8, 60.

- 179 -

References

Riddihough, G. and Pelham, H.R.B. (1987). An ecdysone response element in the Drosophila hsp27 promotor. EMBO J 6, 3729-3734. Risler, J-L., Delorme, M.O., Delacroix, H. and Henaut, A. (1998). Amino acid substitutions in structurally related proteins. A pattern recognition approach. Determination of a new and efficient scoring matrix. J Mol Biol 204, 1019-1029. Robinson-Rechavi, M., Carpentier, A.S., Duffraisse, M., and Laudet, V. (2001). How many nuclear hormone receptors are there in the human genome? Trends Genet 17, 554-556. Robinson-Rechavi, M., Maina, C.V., Gissendanner, C.R., Laudet, V. and Sluder, A. (2005). Explosive lineage- specific expansion of the orphan nuclear receptor HNF4 in nematodes. J Mol Evol 60, 577-586. Roether, S. and Meister, G. (2011). Small RNAs derived from longer non-coding RNAs. Biochimie 93, 1905– 1915. Roignant, J.Y., Carré, C., Mugat, B., Szymczak, D., lepesant, J.A. and Antoniewski, C. (2003). Absence of transitive and systemic pathways allows cell-specific and isoform-specific RNAi in Drosophila. RNA 9, 299-308. Rothe, M., Nauber, U. and Jackle, H. (1998). Three hormone receptor-like Drosophila genes encode an identical DNA-binding finger. Embo J 8(10), 3087-3094. Rozen, S., and Skaletsky, H.J. (2000). Primer3 on the WWW for general users and for biologist programmers. Methods Mol Biol 132, 365-386. Rudolph, K. M., Liaw, G. J., Daniel, A., Green, P., Courey, A. J., Hartenstein, V., and Lengyel, J. A. (1997). Complex regulatory region mediating tailless expression in early embryonic patterning and brain development. Development 124, 4297–4308. Russel, S.R.H., Heimbeck, G., Goddard, C.M., Carpenter, A.T.C. and Ashburner, M. (1996). The Drosophila Eip78C gene is not vital but has a role in regulating chromosome puffs. Genetics 144, 159-170. Sadeghi, A., Van Damme, E.J.M. and Smagghe, G. (2009). Evaluation of the susceptibility of the pea aphid, Acyrthosiphon pisum, to a selection of novel biorational insecticides using an artificial diet. J Insect Sci 9, article 65. Saleh, M.-C., Tassetto, M., van Rij, R.P., Goic, B., Berry, B., Jacquier, C., Antoniewski, C. and Andino, R. (2009). Antiviral immunity in Drosophila requires systemic RNA interference spread. Nature 458, 346- 350. Sanetra, M., Begemann, G., Becker, M.-B. and Meyer, A. (2005). Conservation and co-option in developmental programmes: the importance of homology relationships. Front Zool 2, 15. Schluns, H. and Crozier, R.H. (2007). Relish regulates expression of antimicrobial peptide genes in the honeybee, Apis mellifera, shown by RNA interference. Insect Mol Biol 16, 753-759. Schrick, K., Nguyen, D., Karlowski, W.M. and Mayer, K.F.X. (2004). Start lipid/sterol binding domains are amplified in plants and are predominantly associated with homeodomain transcription factors. Genome Biol 5, R41. Schubert, M., Brunet, F., Paris, M., Bertrand, S., Benoit, G. and Laudet, V. (2008). Nuclear hormone receptor signaling in amphioxus. Dev Genes Evol, DOI10.1007/s00427-008-0251-y. Sengupta, P., Colbert, H.A. and Bargmann, C.I. (1994). The C. elegans gene odr-7 encodes an olfactory-specific member of the nuclear receptor superfamily. Cell 79, 971-980. - 180 -

References

Seol, W., Choi, H.S. and Moore, D.D. (1996). An orphan nuclear hormone receptor that lacks a DNA binding domain and heterodimerizes with other receptors. Science 272, 1336-1339. Shakesby, A.J., Wallace, I.S., Isaacs, H.V., Pritchard, K., Roberts, D.M. and Douglas, A.E. (2009). A water- specific aquaporin involved in aphid osmoregulation. Insect Biochem Mol Biol 39, 1-10. Shea, C., Richer, J., Tzertzinis, G. and Maina, C.V. (2010). An EcR homolog from the filarial parasite, Dirofilaria immitis requires a ligand-activated partner for transactivation. Mol Biochem Parasitol 171, 55- 63. Shigenobu, S., Bickel, R.D., Brisson, J.A., Butts, T., Chang, C-c., Christiaens, O. et al. (2010). Comprehensive survey of development genes in the pea aphid, Acyrthosiphon pisum: frequent lineage-specific duplications and losses of developmental genes. Insect Mol Biol 19 (Suppl. 2), 47-62. Siaussat, D., Mottier, V., Bozzolan, F., Porcheron, P. and Debernard, S. (2004). Synchronization of Plodia interpunctella lepidopteran cells and effects of 20-hydroxyecdysone. Insect Mol Biol 13, 179-187. Siaussat, D., Bozzolan, F., Porcheron, P. and Debernard, S. (2007). Identification of steroid hormone signaling pathway in insect cell differentiation. Cell Mol Life Sci 64, 365-376. Siaussat, D., Porcheron, P. and Debernard, S. (2009). The Ecdysteroids’ effects in the control of cell proliferation and differentiation. In: Ecdysone: Structures and Functions (G. Smagghe ed), pp. 185-204. Springer, Dordrecht. Sieglaff, D.H., Duncan, K.A. and Brown, M.R., 2005. Expression of genes encoding proteins involved in ecdysteroidogenesis in the female mosquito, Aedes aegypti. Insect Biochem Mol Biol 35, 471–490. Sijen, T., Steiner, F.A., Thijssen, K.L., and Plasterk, R.H. (2007). Secondary siRNAs result from unprimed RNA synthesis and form a distinct class. Science 315, 244–247. Sluder, A.E. and Maina, C.V. (2001). Nuclear receptors in nematodes: themes and variations. Trends Genet 17, 206-213. Smagghe, G., Gomez, L.E. & Dhadialla, T.S. (2013). The bisacylhydrazine insecticides for selective pest control. Advances in Insect Physiology 43, 163-249. Smith, H.G. and Hallsworth, P.B. (1990). The effects of yellowing viruses on yield of sugar beet in field trials, 1985 and 1987. Ann Appl Biol 116, 503-511. Soin, T., Iga, M., Swevers, L., Rougé, P., Janssen, C.R. and Smagghe, G. (2009). Towards Coleoptera-specific high-throughput screening systems for compounds with ecdysone activity: development of EcR reporter assays using weevil (Anthonomus grandis)-derived cell lines and in silico analysis of ligand binding to A. grandis EcR ligand-binding pocket. Insect Biochem Mol Biol 39, 523-534. Stone, B.L. and Thummel, C.S. (1993). The Drosophila 78C early-late puff contains E78, an ecdysone-inducible gene that encodes a novel member of the nuclear hormone receptor superfamily. Cell 75, 307-320. Su, J., Zhu, Z., Wang, Y., Zou, J., Wang, N. and Jang, S. (2009). Grass carp reovirus activates RNAi pathway in rare minnow, Gobiocypris rarus. Aquaculture 289, 1-5. Sudarsan, V., Pasalodos-Sanchez, S., Wan, S., Gampel, A. and Skaer, H. (2002). A genetic hierarchy establishes mitogenic signaling and mitotic competence in the renal tubules of Drosophila. Development 129, 935- 944. Sullivan, A.A. and Thummel, C.S. (2003). Temporal profiles of nuclear receptor gene expression reveal coordinate transcriptional responses during Drosophila development. Mol Endocrinol 17, 2125-2137. - 181 -

References

Sung, C., Wong, L.E., Chang Sen, L.Q., Nguyen, E., Lazaga, N., Ganzer, G., McNabb, S.L. and Robinow, S. (2009). The unfulfilled/DHR51 gene of Drosophila melanogaster modulates wing expansion and fertility. Dev Dyn 238, 171-182. Surakasi, V.P., Mohamed, A.A.M. and Kim, Y. (2011). RNA interference of beta 1 integrin subunit impairs development and immune responses of the beet armyworm, Spodoptera exigua. J Insect Physiol 57, 1537- 1544. Suzuki, Y., Truman, J.W. and Riddiford, L.M. (2008). The role of Broad in the development of Tribolium castaneum: implications for the evolution of the holometabolous insect pupa. Development 135, 569-577. Swevers, L. and Iatrou, K. (1998). The orphan receptor BmHNF-4 of the silkmoth Bombyx mori: Ovarian and zygotic expression of two mRNA isoforms encoding polypeptides with different activating domains. Mech Dev 72, 3–13. Swevers, L., Ito, K. and Iatrou, K. (2002a). The BmE75 nuclear receptors function as dominant repressors of the nuclear receptor BmHR3A. J Biol Chem 277, 41637-41644. Swevers, L., Eystathioy, T. and Iatrou, K. (2002b). The orphan nuclear receptors BmE75A and BmE75C of the silkmoth Bombyx mori: hormonal control and ovarian expression. Insect Biochem Mol Biol 32, 1643– 1652. Talbot, W.S., Swywryd, E.A. and Hogness, D.S. (1993). Drosophila tissues with different metamorphic responses to ecdysone express different ecdysone receptor isoforms. Cell 73, 1323–1337. Tamura, K., Dudley, J., Nei, M. and Kumar, S. (2007). MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Mol Biol Evol 24, 1596-1599. Tan, A. and Palli, S.R. (2008). Identification and characterization of nuclear receptors from the red flour beetle, Tr. Castaneum. Insect Biochem Mol Biol 38, 430-439. Tarrant, A.M., Behrendt, L., Stegeman, J.J. and Verslycke, T. (2011). Ecdysteroid receptor from the American lobster Homarus americanus: EcR/RXR isoform cloning and ligand-binding properties. Gen Comp Endocrinol 173, 346-355. Tatchell, G.M. (1989). An estimate of the potential economic losses to some crops due to aphids in Britain. Crop Prot 8, 25-29. Terenius, O., Papanicolaou, A., Garbutt, J.S., Eleftherianos, I., Huvenne, H. et al. (2011). RNA interference in Lepidoptera: An overview of successful and unsuccessful studies and implications for experimental design. J Insect Physiol 57, 231-245. The International Aphid Genomics Consortium (2010). Genome sequence of the pea aphid Acyrthosiphon pisum. PLoS Biol 23, e1000313. Doi: 10.1371/journal.pbio.1000313. Thomas, H.E., Stuttenberg, H.G. and Stewart, A.F. (1993). Heterodimerization of the Drosophila ecdysone receptor with retinoid X receptors and ultraspiracle. Nature 362, 471-475. Thomson, S.A., Baldwin, W.S., Wang, Y.H., Kwon, G. and Leblanc, G.A. (2009). Annotation, phylogenetics, and expression of the nuclear receptors in Daphnia pulex. BMC Genomics 10, 500. Tian, H., Peng, H., Yao, Q., Chen, H., Xie, Q., Tang, B. and Zhang, W. (2009). Developmental control of a Lepidopteran pest Spodoptera exigua by ingestion of bacterial expressing dsRNA of a non-midgut gene. PLoS One 4, e6225.

- 182 -

References

Timmon, L., Court, D.L. and Fire, A. (2001). Ingestion of bacterially expressed dsRNAs can produce specific and potent genetic interference in Caenorhabditis elegans. Gene 263, 103-112. Timmon, L., and Fire, A. (1998). Specific interference by ingested dsRNA. Nature 395, 854. Tomoyasu, Y., Miller, S.C., Tomita, S., Schoppmeier, M., Grossmann, D. and Bucher, G. (2008). Exploring systemic RNA interference in insects: a genome-wide survey for RNAi genes in Tribolium. Genome Biol 9, R10 doi:10.1186/gb-2008-9-1-r10. Tribolium Genome Sequencing Consortium (2008). The genome of the model beetle and pest Tribolium castaneum. Nature 452, 949-955. Turner, C.T., Davy, M.W., MacDiarmid, R.M., Plummer, K.M., Birch, N.P. and Newcomb, R.D. (2006). RNA interference in the light brown apple moth, Epiphyas postvittana (Walker) induced by double-stranded RNA feeding. Insect Mol Biol 15, 383-391. Uryu, O., Kamae, Y., Tomioka, K. and Yoshii, T. (2013). Long-term effect of systemic RNA interference on circadian clock genes in hemimetabolous insects. Vanden Heuvel, J.P. (2009). Nuclear Receptors: A brief overview. Nuclear Receptor Resource, nrresource.org. Van Leeuwen, T., Vontas, J., Tsagkarakou, A., Dermauw, W. and Tirry, L. (2010). Acaricide resistance mechanisms in the two-spotted spider mite Tetranychus urticae and other important Acari: a review. Insect Biochem Mol Biol 40, 563-572. Van Rij, R.P., Saleh, M.C., Berry, B., Foo, C., Houk, A., Antoniewski, C. and Andino, R. (2006). The RNA silencing endonuclease Argonaute 2 mediates specific antiviral immunity in Drosophila melanogaster. Genes Dev 20, 2985-2995. Velarde, R.A., Robinson, G.E. and Fahrbach, S.E. (2006). Nuclear receptors of the honey bee: annotation and expression in the adult brain. Insect Mol Biol 15, 583–595. Velthuis, H.H.W. and van Doorn, A. (2006). A century of advances in bumblebee domestication and the economic and environmental aspects of its commercialization for pollination. Apidologie 37, 421-451. Verhaegen, Y., Parmentier, K., Swevers, L., Rougé, P., Soin, T., De Coen, W., Cooreman, K. and Smagghe, G. (2010). The brown shrimp (Crangon crangon L.) ecdysteroid receptor complex: cloning, structural modeling of the ligand-binding domain and functional expression in an EcR-deficient Drosophila cell line. Gen Comp Endocr 168, 415-423. Wagner, R.L., Apriletti, J.W., McGrath, M.E., West, B.L., Baxter, J.D. and Fletterick, R.J. (1995). A structural role for hormone in the thyroid hormone receptor. Nature 378, 690-697. Walker, W.B. and Allen, M.L. (2010). Expression and RNA interference of salivary polygalacturonase genes in the tarnished plant bug, Lygus lineolaris. J Insect Sci 10, 173. Walker, W.B. and Allen, M.L. (2011). RNA interference-mediated knockdown of IAP in Lygus lineolaris induces mortality in adult and pre-adult life stages. Entomol Exp Appl 138, 83-92. Walshe, D.P., Lehane, S.M., Lehane, M.J. and Haines, L.R. (2009). Prolonged gene knockdown in the tsetse fly Glossina by feeding double stranded RNA. Insect Mol Biol 18, 11-19. Wang, X.H., Aliyari, R., Li, W.X., Li, H.W., Kim, K., Carthew, R., Atkinson, P. and Ding, S.W. (2006). RNA interference directs innate immunity against viruses in adult Drosophila. Science 312, 452-454. Wang, Y.H. and LeBlanc, G.A. (2009). Interactions of methyl farnesoate and related compounds ith a crustacean retinoid X receptor. Mol Cell Endocrinol 309, 109-116. - 183 -

References

Warren, J. T., Petryk, A., Marques, G., Parvy, J. P., Shinoda, T., Itoyama, K., Kobayashi, J., Jarcho, M., Li, Y., O'Connor, M.B., and Dauphin-Villemant, C. (2004). Phantom encodes the 25-hydroxylase of Drosophila melanogaster and Bombyx mori: a P450 enzyme critical in ecdysone biosynthesis. Insect Biochem Mol Biol 34, 991-1010. Werck-Reichhart, D., and Feyereisen, R. (2000). Cytochromes P450: a success story. Genome Biol 1, reviews 3003.1-9. White, K.P., Hurban, P., Watanabe, T. and Hogness, D.S. (1997). Coordination of Drosophila metamorphosis by two ecdysone-induced nuclear receptors. Science 276, 114-117. Whyard, S., Singh, A.D. and Wong, S. (2009). Ingested double-stranded RNAs can act as species-specific insecticides. Insect Biochem Mol Biol 39, 824-832. Willy, P.J., Umesono, K., Ong, E.S., Evans, R.M., Heyman, R.A. and Mangelsdorf, D.J. (1995). LXR, a nuclear receptor that defines a distinct retinoid response pathway. Genes Dev 9, 1033-1045. Wing, K.-D., Slawecki, R.A. and Carlson, G.R. (1988). RH5849, a nonsteroidal ecdysone agonist: effects on larval Lepidoptera. Science 241, 470-472. Winston, W.M., Molodowitch, C. and Hunter, CP. (2002). Systemic RNAi in C. elegans requires the putative transmembrane protein SID-1. Science 295, 2459-2459. Wurtz, J-M., Guillot, B., Fagart, J., Moras, D., Tietjen, K. and Schindler, M. (2000). A new model for 20- hydroxyecdysone and dibenzoylhydrazine binding: A homology modeling and docking approach. Protein Sci 9, 1073-1084. Wyszomierski, S.L., Yeh, J. and Rosen, J.M. (1999). Glucocorticoid receptor/signal transducer and activator of transcription 5 (STAT5) interactions enhance STAT5 activation by prolonging STAT5 DNA binding and tyrosine phosphorylation. Mol Endocrinol 13, 330-343. Xia, Q., Zhou, Z., Lu, C., Cheng, D., Dai, F., Li, B., Zhao, P., Zha, X. et al. (2004). A draft sequence for the genome of the domesticated silkworm (Bombyx mori). Science 306, 1937-1940. Xu, W.N. and Han, Z.J. (2008). Cloning and phylogenetic analysis of sid-1-like genes from aphids. J Insect Sci 8: 30. Xu, J., Tan, A. and Palli, S.R. (2010). The function of nuclear receptors in regulation of female reproduction and embryogenesis in the red flour beetle, Tribolium castaneum. J Insect Physiol 56, 1471-1480. Yamada, M., Murata, T., Hirose, S., Lavorgna, G., Suzuki, E. and Ueda, H. (2000). Temporally restricted expression of transcription factor betaFTZ-F1: Significance for embryogenesis, molting and metamorphosis in Drosophila melanogaster. Development 127, 5083–5092. Yao, T., Segraves, W., Oro, A., McKeown, M. and Evans, R. (1992). Drosophila ultraspiracle modulates ecdysone receptor function via heterodimer formation. Cell 71, 63-72. Yao, T., Forman, B., Jiang, Z., Cherbas, L., Chen, K., McKeown, M., Cherbas, P. and Evans, R. (1993). Functional ecdysone receptor is the product of EcR and ultraspiracle genes. Nature 366, 476-479. Yigit, E., Batista, P.J., Bei, Y., Pang, K.M., Chen, C.-C.G., Tolia, N.H., Joshua-Tor, L., Mitani, S., Simard, M.J. and Mello, C.C. (2006). Analysis of the C. elegans argonaute family reveals that distinct Argonautes act sequentially during RNAi. Cell 127, 747-757. Yu, Q., Lin, T., Feng, G., Yang, K. and Pang, Y. (2008). Functional analysis of the putative antiapoptotic genes, p49 and iap4, of Spodoptera litura nucleopolyhedrovirus with RNAi. J Gen Virol 89, 1873-1880. - 184 -

References

Zanaria, E., Muscatelli, F., Bardoni, B., Strom, T.M., Guioli, S., Guo, W. Lalli, E., Moser, C., Walker, A.P., McCabe, E.R.B., Meitinger, T., Monaco, A.P., Sassone-Corsi, P. and Camerino, G. (1994). Nature 372, 635-641. Zelhof, A.C., Yao, T.P., Evans, R.M., McKeown, M. (1995a). Identification and characterization of a Drosophila nuclear receptor with the ability to inhibit the ecdysone response. Proc Natl Acad Sci 92, 10477–10481. Zelhof, A.C., Yao, T.P., Chen, C.D., Evans, R.M. and McKeown, M. (1995b). Seven-up inhibits Ultraspiracle- based signaling pathways in vitro and in vivo. Mol Cell Biol 15, 6736-6745. Zha, W., Peng, X., Chen, R., Du, B., Zhu, L. and He, G. (2011). Knockdown of midgut genes by dsRNA- transgenic plant-mediated RNA interference in the Hemipteran insect Nilaparvata lugens. PLoS One, 6, e20504. Zhang, Z., Burch, P.E., Cooney, A.J., Lanz, R.B., Pereira, F.A., Wu, J., Gibbs, R.A., Weinstock, G. and Wheeler, D.A. (2004). Genomic analysis of the nuclear receptor family: new insights into structure, regulation, and evolution from the rat genome. Genome Res 14, 580–590. Zhang, X., Zhang, J. and Zhu, K.Y. (2010). Chitosan/double-stranded RNA nanoparticle-mediated RNA interference to silence chitin synthase genes through larval feeding in the African malaria mosquito (Anopheles gambiae). Insect Molecular Biology 19, 683-693. Zhong, W., Sladek, F.M., Darnell, J.E.J. (1993). The expression pattern of a Drosophila homolog to the mouse transcription factor HNF-4 suggests a determinative role in gut formation. EMBO J 12, 537-544. Zhu, J., Miura, K., Chen, L., Raikhel, A.S. (2003). Cyclicity of mosquito vitellogenic ecdysteroid-mediated signaling is modulated by alternative dimerization of the RXR homologue Ultraspiracle. Proc Natl Acad Sci USA 100, 544-549. Zhu, J.-Q., Liu, S., Ma, Y., Zhang, J.-Q., Qi, H.-S., Wei, Z.-J., Yao, Q., Zhang, W.-Q. and Li, S. (2012). Improvement of pest resistance in transgenic tobacco plants expressing dsRNA of an insect-associated gene EcR. PLoS ONE 7: e38572. Doi:10.1371/journal.pone.0038572.

- 185 -

- 186 -

Summary Summary

The nuclear receptors are an important protein superfamily. They are transcription factors regulating the expression of genes in a large array of biological processes. Their involvement in moulting and metamorphosis, embryonic development, cell differentiation and reproduction of insects is well documented. These proteins generally show a high degree of sequence conservation throughout the Animalia. Especially the DNA-binding domain of the nuclear receptors is highly conserved. At the same time, a large amount of diversity has arisen due to a less conserved ligand-binding domain which gave rise to a whole range of possible ligands and also due to a series of gene duplications and gene losses throughout evolution.

This thesis consists of three main parts. In the first part, we investigated the diversity of nuclear receptors in the Arthropoda by annotating the complete sets of nuclear receptors in three different species, representing three distinct arthropod taxonomic groups. The pea aphid Acyrthosiphon pisum represents the hemimetabolous insects, the bumblebee Bombus terrestris represents the holometabolous insects and the two-spotted spider mite Tetranychus urticae belongs to the Chelicerata. Before the start of this research, no full nuclear receptor sets from hemimetabolous and chelicerate species were reported. In general, the sets of nuclear receptors showed a high degree of conservation within arthropods. Most nuclear receptors were found to be shared throughout this phylum. Nonetheless, some remarkable differences have been found as well. For example, the pea aphid is missing the HR96 and NR2E6 gene, both present in most holometabolic insects. Whether this is a trait shared by most Hemiptera or hemimetabolous species is unsure. In the case of HR96, it could be a recent lineage-specific loss, given that the HR96s are found in Crustacea and Chelicerata as well. Another surprising result was the large number of nuclear receptors (30) found in T. urticae compared to other insects (19-22). This was due to an expansion in the NR1 subfamily, mainly of the HR96 group. These HR96s could possibly be linked to the well documented adaptability towards host plants and pesticides that spider mites possess. Additionally, a homolog for HR10, which was recently discovered in Daphnia pulex and is not present in insects, was also found.

Following the annotation work of all nuclear receptors, two of these receptors were also examined into more detail, namely the EcR and RXR/USP, which form the functional ecdysteroid receptor. The pea aphid EcR and USP showed high similarity to those found in other insect species. Protein modelling of the ligand binding domain revealed the canonical

- 188 -

Summary structure consisting of 12 α-helices and 2 β-sheets. In T. urticae, we found that this chelicerate surprisingly possesses two RXRs and, based on in silico protein modelling and docking experiments, that TuEcR seems unable to bind with the main insect moulting hormone 20- hydroxy-ecdysone (20E) while it does prove to be capable bind Ponasterone A (PonA).

In Chapter 4, we investigated the Halloween genes in the pea aphid and the spider mite. These genes encode cytochrome P450 proteins which are responsible for the biosynthesis of ecdysteroids in Arthropoda. Surprisingly, the two-spotted spider mite seems to be missing one of these genes, namely phantom, encoding for a 25-hydroxylase responsible for the conversion of 2,22,25dE-ketodiol to 2,22dE-ketotriol. The main consequence of this loss would be that, instead of 20E, PonA would be synthesized in the spider mite as the main moulting hormone. Mass spectrometry analysis confirmed this hypothesis and showed that PonA is the main moulting hormone in this chelicerate.

In the final chapter, we attempted to use the novel RNA interference (RNAi) technique as a way to start the functional analysis of these nuclear receptors. However, despite intensive attempts, RNAi in the pea aphid appears to be inefficient. Attempts to silence EcR and USP did not cause a phenotype, and expression analysis via real-time qPCR confirmed this observation. We then attempted to reproduce two experiments where RNAi has been reported to work in the pea aphid. However, no effect was seen here as well. As a result, we conducted a first investigation related to the factors that might determine RNAi success and efficiency. Two important factors were investigated in this thesis. We first took a look at the possible degradation of dsRNA by digestive secretions or in the haemolymph and found that a strong dsRNA-digesting activity was found. Second, we investigated the expression of a number of RNAi-related genes and examined whether administering dsRNA would have an effect on the expression of these genes. We found that the five genes, namely dcr-2, ago2, r2d2, eri1 and sid-1, were expressed at a relatively high level and that administering dsEcR did not have an effect on their expression levels.

Since a large part of this work can be considered fundamental research, the annotation results presented in this thesis are mainly a basis for future research on nuclear receptors in Arthropoda. Further research will have to elucidate the role these nuclear receptors have in key metabolic and developmental processes, find out how they play their specific part in

- 189 -

Summary certain pathways and also how they cross-talk between different pathways. Further research is also needed towards orphan receptors. Indeed, most nuclear receptors are still categorized as orphan receptors because the ligand is simply not known at this time. Finding ligands for these orphans could lead to new ways of thinking in nuclear receptor research and may also deliver possible new pest control targets.

The work on RNAi highlighted that the observed variability in RNAi efficiency throughout the Class of Insecta also counts for the pea aphid. Factors which could be involved in this observed inefficiency, such as nucleases, lack of uptake or perhaps interference by viruses, should be further investigated, not only in the pea aphid, but also in other insects where RNAi has proven to be difficult to achieve. Given its selectivity, RNAi could become an important tool in pest control. Therefore, more research towards this mechanism in pest organisms is needed.

- 190 -

Samenvatting

De nucleaire receptoren vormen een belangrijke eiwit-superfamilie. Het zijn transcriptiefactoren die de expressie reguleren van genen die betrokken zijn in een breed scala aan biologische processen. Hun rol in vervelling en metamorfose, embryogenese, celdifferentiatie, en reproductie zijn goed gedocumenteerd. Deze eiwitten vertonen in het algemeen een hoge graad van conservering in het dierenrijk. Met name het DNA- bindingsdomein van deze receptoren is heel sterk geconserveerd. Tegelijkertijd is er ook heel wat diversiteit te vinden binnen deze groep nucleaire receptoren. Enerzijds doordat hun ligand-bindingsdomein minder geconserveerd is en doorheen de evolutie gezorgd heeft voor een hele reeks aan verschillende liganden. Anderzijds is er ook diversiteit ontstaan door een reeks genduplicaties en -verliezen doorheen de evolutie in het dierenrijk.

Deze thesis kan opgedeeld worden in drie grote delen. In een eerste deel hebben we de diversiteit van nucleaire receptoren binnen het fylum van de Arthropoda onderzocht via de annotatie van deze eiwitten in drie verschillende species binnen deze taxonomische groep. De bonenbladluis Acyrthosiphon pisum (Hemiptera) vertegenwoordigt de hemimetabole insecten, de hommel Bombus terrestris (Hymenoptera) vertegenwoordigt de holometabole insecten, en de spintmijt Tetranychus urticae hoort bij de cheliceraten. Bij de start van dit project was van geen enkele geleedpotig organisme buiten de holometabole insecten een volledige set nucleaire receptoren bekend. Zoals verwacht vertoonden de nucleaire receptoren een hoge graad van conservatie binnen de arthropoden. De meeste nucleaire receptoren worden gedeeld doorheen dit fylum. Anderzijds werden ook een aantal opmerkelijke verschillen opgemerkt. Bij de bladluis bijvoorbeeld ontbreken twee nucleaire receptoren, zijnde HR96 en NR2E6, die wel aanwezig zijn bij de meeste holometabole insecten. Het is op dit moment nog ongekend of dit ook het geval is bij andere Hemiptera of hemimetabole insecten. Maar in het geval van HR96 gaat het waarschijnlijk om een recent verlies aangezien deze nucleaire receptor wel teruggevonden is in Crustacea en Chelicerata. Een ander opmerkelijk resultaat was het grote aantal nucleaire receptoren dat werd gevonden in het T. urticae genoom (30) in vergelijking met insecten (19-22). De oorzaak hiervan was een expansie in de NR1 subfamilie, meer bepaald van de HR96 groep. Deze nucleaire receptoren lijken gelinkt te kunnen worden aan het welbekende vermogen van spintmijten om zich aan te passen naar verschillende waardplanten alsook pesticiden. Naast een expansie van de HR96 groep werd ook een homoloog gevonden van HR10, een nucleaire receptor die onlangs voor het eerst werd

- 192 -

Samenvatting ontdekt in de watervlo Daphnia pulex en dat vooralsnog niet aanwezig blijkt te zijn in insecten.

Na het annoteren werden twee nucleaire receptoren, namelijk EcR en RXR/USP, verder in detail bekeken. Deze twee vormen de functionele ecdysteroïdreceptor en zijn een target voor een belangrijke klasse van insecticiden. De EcR en USP van de bonenbladluis A. pisum bleken een hoge graad van gelijkheid te vertonen met deze in andere insecten. Een in silico 3D-model van het ligand-bindingsdomein van het eiwit werd gemaakt en dit vertoonde de 12 α-helices en 2 β-sheets die typisch zijn voor de structuur van dit domein. Bij T. urticae vonden we verrassend genoeg twee RXRs en ook dat TuEcR niet in staat is om 20-hydroxyecdysone (20E) te binden op basis van in silico eiwit modelering en docking experimenten. Wel bleek de receptor in staat te zijn om te binden met ponasterone A (PonA).

In hoofdstuk 4 hebben we de Halloween genes in de bonenbladluis en de spintmijt onderzocht. Deze genen coderen voor cytochroom P450 eiwitten die verantwoordelijk zijn voor de biosynthese van ecdysteroïden in Arthropoda. Verrassend genoeg bleek dat één van deze genen, namelijk phantom, in de spintmijt niet aanwezig was. Het belangrijkste gevolg hiervan is dat het onmogelijk wordt om 20E te produceren, maar PonA zou in de plaats geproduceerd worden. Massaspectrometrische analyse bevestigde deze hypothese en we hebben kunnen aantonen dat PonA inderdaad het belangrijkste ecdysteroïde hormoon in de spintmijt is. 20E werd niet gevonden.

In het laatste deel van deze thesis werd een poging gedaan om RNA interferentie (RNAi) te gebruiken om een functionele analyse van deze nucleaire receptoren te starten. Ondanks verschillende pogingen slaagden we er echter niet in om RNAi in de bladluis te doen slagen. Pogingen om de expressie van EcR en USP te verlagen veroorzaakten geen duidelijk fenotype en een analyse van de expressie met behulp van real-time qPCR bevestigden deze observatie op het transcriptniveau. Vervolgens hebben we ook getracht om twee reeds gepubliceerde succesvolle RNAi experimenten te herhalen. Ook hier werd echter geen effect gezien. Ten gevolge van deze observaties hebben we een onderzoek gestart naar de factoren die mogelijks het succes en de efficiëntie van RNAi kunnen bepalen. Twee belangrijke factoren werden in deze thesis reeds onderzocht. Eerst hebben we gekeken naar mogelijke afbraak van dsRNA door verteringssecreties of in de hemolymfe, en we vonden een sterke degradatie van het

- 193 -

Samenvatting dsRNA bij allebei. Vervolgens hebben we ook de expressie van een vijftal genen die betrokken zijn bij RNAi bekeken en onderzocht of het toedienen van dsRNA een effect zou hebben op de expressie van deze genen. We vonden dat de vijf genen, namelijk dcr-2, ago2, r2d2, eri1 en sid-1, op een redelijk hoog niveau tot expressie kwamen en dat het toedienen van dsEcR geen effect had op de expressieniveaus van deze genen.

Aangezien dit onderzoek voor een groot deel als fundamenteel onderzoek kan omschreven worden, moeten deze resultaten vooral als een basis dienen voor verder onderzoek naar nucleaire receptoren in Arthropoda. Onderzoek naar de rol van deze nucleaire receptoren in belangrijke metabolische en ontwikkelingsprocessen, alsook onderzoek naar de specifieke rol binnen bepaalde pathways moet gevoerd worden. Verder onderzoek is ook nodig naar de zogenaamde weeskindreceptoren. Van heel wat nucleaire receptoren is het ligand namelijk nog steeds niet gekend. Het ontdekken van liganden voor deze wezen kan leiden tot nieuwe inzichten in de functies van nucleaire receptoren. Daarnaast kan het ook mogelijke nieuwe targets opleveren voor gewasbescherming. Het RNAi-werk toonde aan dat de variabiliteit in de efficiëntie van RNAi die in een groot aantal insecten is geobserveerd, ook geldt voor bladluizen. Factoren die mogelijks betrokken zijn bij deze inefficiëntie, zoals nuclease-degradatie, gebrek aan opname van dsRNA of mogelijks de invloed van virussen moeten nog verder onderzocht worden. Niet enkel in de bladluis, maar ook in andere insecten waar RNAi moeizaam blijkt te zijn. Gezien de selectiviteit die deze techniek biedt heeft ze zeker ook een toekomst in de gewasbescherming. Meer onderzoek naar dit mechanisme in plaagorganismen is daarvoor ook broodnodig.

- 194 -

Curriculum vitae

PERSONAL DETAILS

Name Olivier Christiaens Gender Male Nationality Belgian Place of birth Brugge, Belgium Date of birth August 14 1984

EDUCATION

1996-2002 Mathematics - Sciences at St-Leocollege, Bruges

2002-2008 Master in Bioscience Engineering, with distinction at Ghent University Master thesis: “Insect-plant interacties: focus op insect gemedieerde inductie van het lectine in tabak”

Promotors: Prof. Dr. ir. Guy Smagghe (Laboratory of Agrozoology, Department of Crop Protection, Ghent University) and Prof. Dr. Els Van Damme (Laboratory of Biochemistry and Glycobiology, Department of Molecular Biotechlogy, Ghent University)

WORK EXPERIENCE

2008-2012 Ghent University: Dehousse-scholarship Subject: Key role and diversity of EcR/USP in Arthropoda Place of research: Laboratory of Agrozoology, Department of Crop Protection, Faculty of Bioscience Engineering, Ghent University. Coupure Links, 653, 9000, Ghent.

SCIENTIFIC OUTPUT

A1 peer reviewed publications

1. CHRISTIAENS, O., IGA, M., VELARDE, R., ROUGÉ, P. & SMAGGHE, G. (2010). Halloween genes and nuclear receptors in ecdysteroid biosynthesis and signaling in the pea aphid. Insect Molecular Biology, 19, 187-200.

2. THE INTERNATIONAL APHID GENOME CONSORTIUM (RICHARDS, S., GIBBS, R.A., … CHRISTIAENS, O., … SMAGGHE, G., …. HUNTER, W.) (2010). Genome sequence of the pea aphid Acyrthosiphon pisum. PLoS Biology 8(2):e1000313. doi:10.1371/journal.pbio.1000313.

3. SHIGENOBU, S., BICKEL, R.D., BRISSON, J.A., BUTTS, T., CHANG, C.-C., CHRISTIAENS, O., DAVIS, G.K., DUNCAN, E., FERRIER, D.E.K., IGA, M., JANSSEN, R., LIN, G.-W., LU, H.-L., MCGREGOR, A.P., MIURA, T., SMAGGHE, G., SMITH, J.M., VAN DER ZEE, M., VELARDE, R., WILSON, M.J., DEARDEN, - 196 -

Curriculum vitae

P.K. & STERN, D.L. (2010). Comprehensive survey of developmental genes in the pea aphid, Acyrthosiphon pisum: frequent lineage-specific duplications and losses of developmental genes. Insect Molecular Biology, 19, 47-62.

4. STALJANSSENS, D., KARIMIAN-AZARI, E., CHRISTIAENS, O., BEAUFAYS, J., LINS, L., VAN CAMP, J. & SMAGGHE, G. (2011). The CCK-receptor in the animal kingdom: functions, evolution and structures. Peptides, 32, 607-619.

5. GRBIĆ, M., VAN LEEUWEN, T., CLARK, R.M., ROMBAUTS, S., ROUZÉ, P., GRBIC, V., OSBORNE, E.J., DERMAUW, W., NGOC, P.C., ORTEGO, F., HERNANDEZ-CRESPO, P., DIAZ, I., MARTINEZ, M., NAVAJAS, M., SUCENA, E., MAGALHÃES, S., NAGY, L., PACE, R.M., DJURANOVIC, S., SMAGGHE, G., IGA, M., CHRISTIAENS, O., VEENSTRA, J.A., EWER, J., VILLALOBOS, R.M., HUTTER, J.L., HUDSON, S.D., VELEZ, M., YI, S.V., ZENG, J., PIRES-DASILVA, A., ROCH, F., CAZAUX, M., NAVARRO, M., ZHUROV, V., ACEVEDO, G., BJELICA, A., FAWCETT, J.A., BONNET, E., MARTENS, C., BAELE, G., WISSLER, L., SANCHEZ-RODRIGUEZ, A., TIRRY, L., BLAIS, C., DEMEESTERE, K., HENZ, S.R., GREGORY, T.R., MATHIEU, J., VERDON, L., FARINELLI, L., SCHMUTZ, J., LINDQUIST, E., FEYEREISEN, R. and VAN DE PEER, Y.) (2011). The genome of Tetranychus urticae reveals herbivorous pest adaptations. Nature, 479, 487-492.

6. ZOTTI, M.J., CHRISTIAENS, O., ROUGÉ, P., GRUTZMACHER, A.D., ZIMMER, P.D. & SMAGGHE, G. (2012). Sequencing and structural homology modeling of the ecdysone receptor in two chrysopids used in biological control of pest insects. Ecotoxicology, 21, 906-918.

7. BENGOCHEA, P., CHRISTIAENS, O., AMOR, F., VIÑUELA, E., ROUGÉ, P., MEDINA, P. & SMAGGHE, G. (2012). Ecdysteroid receptor docking suggests that dibenzoylhydrazine-based insecticides are devoid to any deleterious effect on the parasitic wasp Psyttalia concolor (Hym. Braconidae). Pest Management Science, 68, 976-985.

8. ZOTTI, M.J., CHRISTIAENS, O., ROUGÉ, P., GRUTZMACHER, A.D., ZIMMER, P.D. & SMAGGHE, G. (2012). Structural changes under low evolutionary constraint may decrease the affinity of dibenzoylhydrazine insecticides for the ecdysone receptor in non-lepidopteran insects. Insect Molecular Biology, 21, 488-501.

9. AMOR, F., CHRISTIAENS, O., BENGOCHEA, P., MEDINA, P., ROUGÉ, P., VIÑUELA, E. & SMAGGHE, G. (2013). Selectivity of diacylhydrazine insecticides to the predatory bug Orius laevigatus: In vivo and modeling/docking experiments. Pest Management Science, 68, 1586-1594.

10. YU, N., CHRISTIAENS, O., LIU, J., NIU, J., CAPPELLE, K., CACCIA, S. & SMAGGHE, G. (2013). Delivery of dsRNA for RNAi in insets: an overview and future directions. Insect Science 20 (number 1), 5-14.

11. BENGOCHEA, P., CHRISTIAENS, O., AMOR, F., VIÑUELA, E., ROUGÉ, P., MEDINA, P. & SMAGGHE, G. (2013). Insect growth regulators as potential insecticides to control olive fruit fly (Bactrocera oleae Rossi): insect toxicity - 197 -

Curriculum vitae

bioassays and molecular docking approach. Pest Management Science, in press. doi: 10.1002/ps.3350

12. DE WILDE, R., SWEVERS, L., SOIN, T., CHRISTIAENS, O., ROUGÉ, P., COOREMAN, K., JANSSEN, C.R. & SMAGGHE, G. (2013). Cloning and functional analysis of the ecdysteroid receptor complex in the opposum shrimp Neomysis integer (Leach, 1814). Aquatic Toxicology, in press.

Participation at international conferences

8th International Symposium on aphids. Catania, Italy, June 2009.

Christiaens, O., Iga, M. and Smagghe, G. Cloning and characterization of the ecdysone cascade with ecdysone receptor (EcR) and Ultraspiracle (Usp) in the pea aphid. (poster)

Christiaens, O., Iga, M. and Smagghe, G. The Acyrthosiphon genome contains at least 19 nuclear receptors including those involved in the ecdysone cascade. (poster)

Spidermite Genome Consortium meeting. Logroño, Spain, November 2009.

Smagghe, G., Iga, M., Christiaens, O., Liu, J., Van Leeuwen, T. T. urticae hormones: biosynthesis and signaling by ecdysteroids and JH. (oral)

18th International Ecdysone Workshop. Ceske Budejovice, Czech Republic, July 2010

Christiaens, O., Iga, M., Velarde, R., Rougé, P. and Smagghe, G. Analysis of the hemimetabolous pea aphid genome for nuclear receptors involved in the ecdysone signaling cascade. (poster)

Christiaens, O., Iga, M., Velarde, R., Rougé, P. and Smagghe, G. The ecdysone receptor in the hemimetabolous pea aphid: cloning, protein structure modeling and functional analysis by RNAi. (poster)

25th Conference of European Comparative Endocrinologists, Pécs (Hungary) 31 August - 4 Sept 2010

Zotti, M., Christiaens, O., Rougé, P. and Smagghe, G. The ecdysone receptor in a neuropteran (Chrysoperla carnea) and dermapteran insect (Forficula auricularia) used in biological control: sequencing and structural modeling. (poster)

Invertebrate Neuropeptide Conference 2011. Sabah (Malaysia), 13-17 februari 2011.

Smagghe, G., Staljanssens, D., Karimian azari, E., Christiaens, O., Willems, P., De Vos, W.H., Beaufays, J., Lins, L., Nachman, R.J. and Van Camp, J. The CCK(-like) receptor as potential target in insects. (oral)

- 198 -

Curriculum vitae

63rd International Symposium on Crop Protection. Ghent, Belgium. May 2011

Christiaens, O., Huvenne, H. and Smagghe, G. RNAi in the pea aphid (Acyrthosiphon pisum): dsRNA delivery using microinjection and feeding assays. (poster)

Amor, F., Bengochea, P., Christiaens, O., Velazquez, E., Morales, I., Fernandez, M., Smagghe, G. and Viñuela, E. Side effects of modern pesticides on the predatory bug Orius laevigatus (fieber) (Hemiptera: Anthocoridae) under laboratory conditions. (oral)

Third Spider Mite Genome Meeting, Tarragona (Spain), 25-29 September 2011. (orale)

Feyereisen, R., Smagghe, G., Iga, M., Christiaens, O., Liu, J., Dermauw, W. and Van Leeuwen, T. CYP genes in T. urticae: the role of P450 enzymes in host plant adaptation and hormone biosynthesis. (oral)

- 199 -

- 200 -

Dankwoord

Toen ik vier jaar geleden begon aan dit doctoraat had ik geen idee waar het me heen zou leiden en hoe het eindresultaat er zou gaan uit zien. Gedurende die vier jaar twijfel je op sommige momenten zelfs wel eens of dat eindresultaat er ooit nog wel zal komen. Maar uiteindelijk ligt de laatste proefdruk hier toch naast me terwijl ik dit schrijf. Gedurende die vier jaar heb ik heel wat mensen ontmoet die allen een bijdrage hebben geleverd tot dit eindwerk en die ik daar ook even voor wil bedanken.

Eerst en vooral zou ik mijn promotor willen bedanken zonder wiens hulp dit werk nooit afgeraakt zou zijn. Guy, bedankt voor het vertrouwen dat je me een goeie vier jaar geleden schonk. Na het afwerken van mijn Master thesis was ik er in eerste instantie nog niet uit of ik wel nog verder onderzoek wou doen. Na een maandje vakantie begon ik het onderzoek, met name het werk in het labo, al wat te missen en ging ik beseffen dat dit echt wel iets voor mij is. Toen ik die zomer langskwam op het labo was je meteen bereid me een kans te geven om die droom na te streven. Bedankt voor alle advies, de steun en de talloze verbeteringen van allerhande drafts. De meeste tijd gedurende die afgelopen jaren heb ik misschien wel gespendeerd ‘in den bureau’, waar ik heel fijne mensen heb ontmoet. Nick, Maarten, Brecht en Shahnaz, jullie zaten er reeds toen ik begon en hebben me meteen welkom doen voelen. Gedurende de voorbije jaren werkten ook jullie succesvol jullie doctoraat af en werden jullie plaatsen ingenomen door Sara, Dominiek, Na en recent ook Thomas. De goeie sfeer zorgde ervoor dat ik zelden met tegenzin naar het werk vertrok. Daarvoor wil ik jullie allemaal bedanken. Nick, Sara en Maarten (ook Jochem en Thijs mogen hier zeker niet ontbreken), bedankt voor de leuke terrasjes en de quiz-, en voetbalavonden. I also want to thank Na in particular for all the advice and help when setting up experiments, for the Chinese food you brought to the lab every now and then and for the many discussions on RNAi and China ;) Uiteraard waren er ook de vele uren in het labo. Alle mensen van het NR-team (Masatoshi, Thomas, Yves, Ruben, Moises, Fermin, Paloma) en het RNAi team (Hanneke, Na, Kaat, Jisheng, Jinzhi), bedankt voor alle hulp bij allerhande experimenten, het lenen van tips, gesteriliseerde epjes en stiftjes wanneer ik weer eens zonder zat. Alsook voor de input tijdens de talloze NR- en RNAi-meetings  Thomas en Yves wil ik in het bijzonder bedanken voor de begeleiding in het labo toen ik net begon. Ik werd een beetje in het diepe gesmeten en het

was niet altijd even makkelijk werken daar in de kelder van Blok B, maar met jullie hulp liep het uiteindelijk allemaal vrij snel op rolletjes. Jullie hebben me een aantal jaren geleden ook de fietsmicrobe doorgegeven en ook daar ben ik jullie enorm dankbaar voor. Ondertussen hebben we de laatste jaren samen al ettelijke duizenden kilometers afgelegd, inclusief de onvergetelijke fietstrip doorheen de Alpen. Hopelijk volgen er nog veel meer. Fermin, Paloma and Marta, thanks for all the wonderful times I’ve had during your stay here in Belgium and my visits to Spain. I’m already looking forward to the next one! Stay focused! ;) Verder wil ik ook alle andere (oud-)leden van het labo voor Agrozoölogie bedanken voor alle hulp, raad en de vele (al dan niet filosofische) gesprekken en discussies in het labo, op de gang, in de resto, in de tuin, op café etc... Wannes en Thomas, bedankt voor alle hulp bij het mijten-genoomproject en voor de leuke momenten tijdens de Logroño trip. Leen, Rik en Bjorn wil ik ook bedanken om ervoor te zorgen dat alles in het labo altijd op rolletjes bleef lopen. In het bijzonder wil ik zeker Didier bedanken, die ervoor zorgde dat ik altijd bonenplanten had om mijn bladluizen op te kweken, die altijd klaar stond als er iets gerepareerd moest worden en die er ook altijd was voor een aangename babbel. Ook de leden van de examen commissie (Prof. dr. Els J.M. Van Damme, Prof. dr. ir. Peter Bossier, Prof. dr. Dirk de Graaf, Prof. dr. Jozef Vanden Broeck en Prof. dr. Luc Swevers) zou ik graag willen bedanken voor de tijd die ze hebben gestoken in het lezen en beoordelen van deze thesis en natuurlijk voor alle nuttige opmerkingen die deze thesis zeker ten goede zijn gekomen. Dit resultaat zou er natuurlijk nooit geweest zijn zonder de steun van vrienden en familie. Met name mijn ouders wil ik heel erg bedanken voor alles, niet alleen de afgelopen vier jaar, maar ook daarvoor. Zonder jullie steun zou ik nooit staan waar ik nu sta. Wouter, Arne, Mathias, Ben, Bert, Bjorn, Jason, bedankt voor het gezelschap tijdens de voetbalavonden, feestjes, ‘after-work-drinks’, snookerwedstrijdjes, metalconcerten, citytrips, fietsritjes die me regelmatig het werk even deden vergeten. Ook bedankt om gedurende de voorbije vier jaar zonder morren al het gezaag over mijn onderzoek te aanhoren. Uiteraard mag ik hier ook mijn fietskameraden (naast Yves en Thomas ook nog Jeroen, Pieter L, Nick, Pieter VDN) niet vergeten. Niks leuker om je hoofd even leeg te maken dan te gaan uitwaaien op de fiets. Jullie zullen me de komende tijd wel weer wat vaker zien, weer of geen weer 

Olivier

- 202 -