Immune evolution in the Immigrans-Tripunctata clade of Drosophila

(1)

Immune evolution in the Immigrans-Tripunctata clade of Drosophila

by

Mark Hanson

B.Sc. Biology, University of Victoria, 2013

A Thesis Submitted in Partial Fulfillment of the Requirements for the Degree of

MASTER OF SCIENCE in the Department of Biology

(2)

ii

Supervisory Committee

Immune evolution in the Immigrans-Tripunctata clade of Drosophila

by Mark Hanson

B.Sc. Biology, University of Victoria, 2013

Supervisory Committee

Dr. Steve Perlman, (Department of Biology)

Supervisor

Dr. Louise Page, (Department of Biology)

Departmental Member

Dr. Caroline Cameron, (Department of Biochemistry/Microbiology)

(3)

iii

Abstract

Supervisory Committee

Dr. Steve Perlman, (Department of Biology)

Supervisor

Dr. Louise Page, (Department of Biology)

Departmental Member

Dr. Caroline Cameron, (Department of Biochemistry/Microbiology)

Outside Member

Drosophila melanogaster has been integral to unravelling the mechanisms of animalian immunity. Diverse species of Drosophila with sequenced genomes have been used to characterize how immune systems respond to natural selection. However, Drosophila is an incredibly speciose lineage, especially so in the subgenus Drosophila. Of the 12 genomes sequenced in 2007, ushering in the era of Drosophila comparative genomics, only three were subgenus Drosophila flies, and none were from the lesser-characterized Immigrans-Tripunctata clade. Recently, multiple Immigrans-Tripunctata clade Drosophila have been sequenced, including the transcriptome of Drosophila neotestacea. I investigated the realized immune responses of D. neotestacea to

characterize the immune repertoire of this divergent lineage. The signalling pathways of D. neotestacea were largely conserved, though there were interesting patterns of

evolution in antimicrobial peptide genes (AMPs). One of these AMPs, a diptericin, was highly dissimilar to diptericins in D. melanogaster, and conserved in other subgenus Drosophila flies. This prompted me to characterize the evolution of the diptericin gene family in Drosophila. I found that Drosophila diptericins have evolved under positive selection, and display intriguing differences in net charge to well-conserved diptericin domains. I assessed the expression profile of this divergent D. neotestacea diptericin, and found that it did not respond to Serratia bacterial challenge, unlike diptericin in D. melanogaster. I also highlight a potential novel drosocin-like AMP conserved throughout the subgenus Drosophila. These results agree that signalling pathways are highly

conserved in diverse insects, including Drosophila. However seemingly-conserved effectors of the Drosophila immune response (such as AMPs) may have previously unappreciated variation in expression and function.

(4)

iv

List of Tables

Table 2.1: Amino acid sequence similarities of diptericin consensus sequences ... 40!

Table 2.2: Net charges of DptA or DptC domains from Drosophila diptericins ... 41!

Table S2.1: Distribution of diptericins in non-Drosophila Diptera ... 47!

Table S2.2: Quantitatve PCR primers used in this thesis ... 48!

Table 3.1: Degenerate primers used to amplify IM4 from D. neotestacea ... 77!

Table 3.2: Genes absent from the D. neotestacea transcriptome and their presence in Immigrans-Tripunctata flies ... 78!

(6)

vi

List of Figures

Figure 1.1: The five major signaling pathways regulating the insect humoral immune

response, as found in D. melanogaster ... 16!

Figure 1.2: Drosophilid maximum likelihood phylogeny ... 17!

Figure 2.1: The subgenus Drosophila encodes a distinct diptericin ... 42!

Figure 2.2: Synteny of diptericins in the genomes of sequenced drosophilids ... 43!

Figure S2.3: Summary phylogeny of Branch-site REL (BSR) analyses using Drosophila diptericin G domains ... 44!

Figure 2.4: Relative expression of diptericins in D. melanogaster and D. neotestacea ... 45!

Figure S2.1: Branch-site REL analysis of Drosophila diptericins ... 49!

Figure S2.2: Alignments of the amino acid sequences of Gly22-Asp45, or Asn46-on regions of Drosophila diptericin G domains ... 51!

Figure S2.3: The Serratia strain used for immune challenges is an oral pathogen of D. neotestacea, and is closely related to the common soil pathogen Serratia marcescens ... 53!

Figure 3.1: Drosophilid maximum likelihood tree of representative lineages of sequenced Drosophila (same as Figure 1.2, but with number of described species added) ... 80!

Figure 3.2: The immune repertoire of D. neotestacea, relative to sequenced Drosophila 81! Figure 3.3: Maximum likelihood phylogeny of Drosophila IM4 ... 82!

Figure 3.4: Maximum likelihood phylogeny of PGRP-SC protein translations ... 83!

Figure 3.5: Proline-rich AMP alignment of translated nucleotide sequences, including drosocin-like sequences from subgenus Drosophila flies ... 84!

(7)

vii

List of Acronyms

AIC – Akaike information criterion AMP – antimicrobial peptide

AttB – attacin B

BLAST – basic local alignment search tool BSR – branch-site REL

DptA – diptericin A DptB – diptericin B DptC – diptericin C

GNBP – Gram-negative binding protein HMM – hidden Markov model

IM – immune-induced molecule

MAFFT – Multiple Alignment using Fast Fourier Transform Mtk – metchnikowin

MUSCLE – MUltiple Sequence Comparison by Log-Expectation

NF-κB – nuclear factor kappa-light-chain-enhancer of activated B cells ORF – open reading frame

PAMP – pathogen associated molecular pattern PCR – polymerase chain reaction

PGRP – peptidoglycan recognition protein PhyML – Phylogenetics by Maximum likelihood PO – phenoloxidase

PPO – prophenoloxidase

PRR – pattern recognition receptor qPCR – quantitative PCR

REL – random-effects likelihood ROS – reactive oxygen species RpL28 – ribosomal protein L 28 RpL32 – ribosomal protein L 32 serpin – serine protease inhibitor TEP – thioester-containing protein

(8)

viii

Acknowledgments

My thanks go out to Dr. Steve Perlman for the wonderful experience that this Masters degree has been. His tutelage and insights have shaped this research, and shaped myself as both a researcher and human being in more ways than I can count. I am a better person for having had the opportunity to study under Steve, and will reflect fondly on my time spent in the Perlman lab. I would also like to acknowledge and thank my committee members, Dr. Louise Page and Dr. Caroline Cameron, for their time and advice during my degree. A big thank you to my coffee buddies over the last two years, the members of the Perlman lab. Special thanks go out to Finn Hamilton, who helped me greatly with his advice, friendship, and patience.

I would not have had this opportunity without the support of my parents, Harris Hanson and Shelley Goldstein, to whom I am eternally grateful. Finally, I would like to thank my other half, Hannah Westlake, who provided insights even after the nine to five work day had concluded.

(9)

Chapter I: Introduction to the Drosophila immune system

The innate immune system is ancient, and important

Selective pressures define organismal defences as hosts adapt to their co-evolving pathogens or parasites (natural enemies). These dynamic interactions play major roles in evolution, and ancient defence mechanisms known as the innate immune response are conserved amongst all animals. Vertebrates have since evolved an adaptive immune system capable of immunological memory, allowing greater specificity in responding to reoccurring natural enemies. However the vertebrate adaptive immune system

fundamentally relies on the contributions of innate immune responses (Kimbrell and Beutler, 2001). Thus, our understanding of immunity necessitates a comprehension of the innate immune system. Invertebrate models are used to elucidate innate immune

mechanisms in the absence of interference from the vertebrate adaptive immune

response. Insects, as invertebrates, offer one such model system, and the powerful genetic tools available to Drosophila researchers have made Drosophila one of the

best-characterized model organisms for the study of innate immunity.

Types of innate immune responses

There are two categories of innate immune responses in invertebrates: the cellular response and the humoral response. The cellular response, as implied by its title,

describes cell-mediated defences enacted by free-floating and sessile blood cells called haemocytes (Lemaitre and Hoffmann, 2007). A variety of haemocytes have been described from diverse hexapod orders such as Lepidoptera, Diptera, Orthoptera, Blattaria, Coleoptera, Hymenoptera, Hemiptera, and Collembola (Lavine and Strand, 2002).

Detailed descriptions of insect haemocytes are available for lepidopterans and dipterans, particularly Drosophila. Insect haemocytes adhere to and engulf foreign bodies, store precursors of melanization, and have other yet-uncharacterized functions (Strand, 2008). In Drosophila, Melanogaster group species use haemocytes called

(10)

2 other Drosophila, Obscura group species use a recently described novel haemocyte lineage, termed “pseudopodocytes,” to encapsulate foreign bodies similar to lamellocytes (Havard et al., 2012). Thus, diverse Drosophila may use unique, yet-uncharacterized cellular responses to combat natural enemies.

The humoral response refers to the systemic circulation of potent antimicrobial compounds in the host haemolymph. A hallmark of the humoral response is the synthesis and secretion of antimicrobial peptides (AMPs) by the insect fat body, a major immune responsive tissue and analog of the mammalian liver. The fat body is immersed in the insect haemolymph, providing a powerful organ for the secretion of AMPs into the haemolymph. Localized AMP expression has also been described from surface epithelia, reproductive tissues, respiratory tracheae, and the digestive tract in Drosophila (Lemaitre and Hoffmann, 2007). These humoral defences are evolutionarily ancient, and provide a rapid response to invading microorganisms. The genetically tractable Drosophila melanogaster is a versatile model organism used to unravel the mechanisms regulating humoral immunity (Lemaitre and Hoffmann, 2007).

The humoral immune response

There are five major signaling cascades responsible for regulating the humoral immune response in arthropods (Palmer and Jiggins, 2015); these are the immune signaling pathways Toll, Imd, JAK/STAT, JNK, and the melanization response (Figure 1.1). The humoral immune response can be broken into the following stages: recognition, signal transduction, and the realized defence response. The proteins involved in each of these stages are distinct, and can be characterized as pathogen-sensing pattern recognition receptors (PRRs), signaling cascade intermediates, and effector proteins synthesized in response to pathway activation. Additionally, post-transcriptionally activated enzymes (zymogens) can act as both signaling cascade intermediates, or the terminal proteins responsible for the realized defensive response. For recognition proteins and pathway intermediates, near-perfect orthology of regulatory mechanisms is observed in diverse insects (Lazzaro, 2008), and so I will focus my discussion on mechanisms as found in Drosophila.

(11)

3

Recognition

The first challenge faced by an infected host is distinguishing self from non-self. To do so, insects have evolved sophisticated sensory receptors that bind to external components of invading microorganisms. Drosophila encodes three types of PRRs, peptidoglycan recognition proteins (PGRPs), glucan binding proteins, and

lipopolysaccharide receptors. Peptidoglycan in bacterial cell walls is recognized by PGRPs, and different PGRPs are responsible for the recognition of Lys-type

peptidoglycan of Gram-postive bacteria, and DAP-type peptidoglycan of Gram-negative bacteria. Lipopolysaccharide recognition proteins also function to recognize Gram-negative bacteria. However, the “Gram-Gram-negative binding proteins” (GNBPs, an

unfortunate misnomer) GNBP1 and GNBP2 are utilized by D. melanogaster to detect cell wall components of invading Gram-positive bacteria (Lemaitre and Hoffmann, 2007). Beta-1,3-glucan is a structural constituent of fungal cells walls, and GNBP3 recognizes invading fungi and yeasts in D. melanogaster (Kim et al., 2000); pathogenic fungi can also directly activate the enzyme persephone (psh), to induce the antifungal pathway in D. melanogaster (Ligoxygakis et al., 2002).

Signaling cascade intermediates

Upon recognition of a foreign invader, the next task in mounting an effective defensive response is to transmit that information to immune responsive cells or tissues. This task can be accomplished directly, by PRRs that recruit membrane-bound receptor proteins (as is the case for the Imd, JAK/STAT, and JNK pathways), or indirectly by the activation of serine protease cascades that occur extracellularly, and culminate in the recruitment of membrane-bound receptor proteins (characteristic of the Toll pathway). For Toll and melanization response signaling, serine protease cascades are composed of zymogens that are post-transcriptionally cleaved by upstream pathway constituents. To avoid costly overexpression of these signaling cascades, host-encoded serine protease inhibitors (serpins) regulate pathway activity. In the Toll pathway, the terminal serine protease “spaetzle processing enzyme” (SPE) cleaves the cytokine spaetzle, which then binds as a dimer to Toll membrane-bound receptors. This induces dimerization of the Toll proteins at the cell membrane, leading to recruitment of intracellular signaling

(12)

4 proteins. In the Imd, JAK/STAT, and JNK pathways, PRRs recruit their associated

membrane-bound receptors directly to induce intracellular signaling proteins.

Intracellular signaling culminates in the translocation of NF-κB binding proteins into the nucleus to initiate the transcription of defensive effector proteins by the fat body. In local immune responses, effector proteins are regulated not by the Toll or Imd pathways, but instead by various tissue-specific transcription factors. The melanization response instead occurs extracellularly, and serine protease cascades culminate directly in the processing of prophenoloxidases (PPOs) into phenoloxidases (POs) in the haemolymph; PPOs are secreted into the haemolymph by haemocytes (Lemaitre and Hoffmann, 2007).

Clotting and melanization

An effective defence response needs to not only combat invading natural enemies, but also minimize the damage caused by enemy attack. A variety of stress conditions are associated with wounding, and effector proteins of the stress-responsive JAK/STAT pathway are likely initiators of clotting and subsequent melanization at wound sites (Agaisse and Perrimon, 2004). After injury, rapid clotting is crucial to reduce

haemolymph loss. Clotting occurs by the action of fibers to recruit and trap haemocytes to close the wound. The clot is then hardened by enzymes such as transglutaminase, hemolectin, and PPOs, which are regulated by recruited haemocytes. The melanization response describes the de novo synthesis of melanin to aid in wound healing. Melanin and reactive oxygen species (ROSs) produced as by-products of the melanization response also contribute to defence against natural enemies (Lemaitre and Hoffmann, 2007).

Immune effector proteins

Upon receiving the signal of attack by a natural enemy, immune responsive cells, and especially the insect fat body, synthesize a number of immune effector proteins that act directly on natural enemies, or play other roles in stress tolerance. Drosophila opsonins (e.g. TEP proteins) promote phagocytosis of invading microorganisms, and opsonins have been argued to regulate serpin activity. Drosophila immune-induced molecules (IMs) are uncharacterized proteins that are synthesized upon immune

(13)

5 challenge (Lemaitre and Hoffmann, 2007); recently, many IMs were included in the description of a novel AMP gene family, the bomanin gene family (Clemmons et al., 2015). The JAK/STAT regulated Turandot proteins also do not yet have described functions, but are highly induced by various stresses and especially by septic injury. Lastly, AMPs are host-encoded antibiotics that are directly involved in defence against natural enemies (Lemaitre and Hoffmann, 2007).

Antimicrobial peptides (AMPs): host-encoded antibiotics

!

Antimicrobial peptides can have diverse ranges of antimicrobial activity, and understanding their modes of action is of particular interest in the development of novel antibiotics (Lewis, 2013). While AMPs are primarily produced to disrupt and kill

invading microbes, AMPs have also taken on alternate roles. For instance, a mutation in a canine defensin allows this AMP to bind to pigment receptors resulting in black fur colour in domestic dogs (Candille et al., 2007). Additionally, in Sitophilus weevils, the AMP coleoptericin is important in regulating bacterial endosymbionts. Using RNA interference, Login et al. (2011) silenced coleoptericin and observed Sitophilus

endosymbionts proliferate and escape into previously uncolonized host tissues, implying a potentially important role for AMPs in maintaining endosymbiosis.

AMPs display preferential antimicrobial activity against microbes that induce their respective pathways; that is to say, AMPs regulated by the Imd pathway display antimicrobial activity against Gram-negative bacteria, and the same is true of the Toll pathway and Toll pathway AMPs. For instance, the Toll-regulated AMP drosomycin is particularly effective against pathogenic fungi, the Toll-regulated defensin is most effective against Gram-positive bacteria, and the Imd-regulated diptericin is especially effective against Gram-negative bacteria (Lemaitre and Hoffmann, 2007). However, AMPs may act synergistically by cooperating with co-expressed AMPs to facilitate their mode of action. The hymenopteran AMPs hymenoptaecin and abaecin were recently shown to synergistically kill bacteria by hymenoptaecin’s ability to disrupt bacterial cell membranes, facilitating abaecin’s entry into bacteria to disrupt the bacterial housekeeping gene DnaK (Rahnamaein et al., 2015). As such, observed antimicrobial activities may not reflect the entire range of an AMP’s activity, and in vivo confirmation of AMP functions

(14)

6 await the generation of AMP gene knockouts (Lemaitre and Hoffmann, 2007);

observations of in vivo AMP function have proved difficult as AMP duplications provide functionally redundant genes (Clemmons et al., 2015).

Indeed, the mode of action employed by many AMPs is not well understood. For instance, diptericin is a canonical readout of Imd pathway activation (Lemaitre and Hoffmann, 2007). Yet in vitro assays have found that while the intact diptericin molecule exhibits antibacterial activity, its constituent domains do not display antibacterial activity when synthesized independently and introduced in combination. Moreover, this lack of activity by the independent domains cannot readily be attributed to protein folding induced by binding to the bacterial cell membrane; when immersed in the membrane-mimicking solvent trifluoroethanol, diptericin maintains an unordered linear structure (Cudic et al., 1999).

Evolution of immune system genes in insects

Innate immune mechanisms are ancient, and evolutionary pressures have acted to conserve immune signaling mechanisms amongst diverse animals. Indeed, we observe near-perfect orthology of major signaling cascade intermediates in diverse insects

(Lazzaro, 2008), though in aphids (The International Aphid Genomic Consortium, 2010), kissing bugs (Mesquita et al., 2015), and chelicerate arthropods (Palmer and Jiggins, 2015), Imd pathway constituents have been independently lost. Nonetheless, maintenance of functional immune signaling pathways is essential for an effective defence against natural enemies. This places evolutionary importance on avoiding signaling pathway disruption by coevolving natural enemies. From the natural enemy’s perspective, it is imperative to avoid succumbing to host defences. Pathogens employ a number of general strategies to avoid host defences such as taking up residence in host tissues to avoid circulating defence compounds, or actively suppressing the host immune response (Lemaitre and Hoffmann, 2007).

Pathogens may suppress the host immune response by interfering with conserved members of immune signaling pathways. Indeed, strong signatures of adaptive evolution (positive selection) are observed in signaling pathway components (Lazzaro, 2008). Furthermore, signaling cascade intermediates do not typically undergo genomic

(15)

7 duplications or deletions, possibly due to constraints imposed by potential pleiotropic effects (Palmer and Jiggins, 2015). Amongst Drosophila genomes, signaling genes of the Imd pathway display putatively adaptive amino acid substitutions, particularly at protein domains used by host inhibitors to avoid pathway overexpression (Lazzaro, 2008). It is proposed that pathogens take advantage of these host inhibitory mechanisms to facilitate suppression of the host immune response.

While signaling cascade intermediates show remarkable conservation, effector proteins such as AMPs undergo frequent genomic rearrangement, duplication, and loss. No less than 12 members of the cecropin multigene family are encoded by the silkworm Bombyx mori, occurring in multiple clusters on different chromosomes (Ponnuvel et al., 2010). In the Drosophila-restricted drosomycin gene family, there are six drosomycin genes encoded by Melanogaster group species, one of which (Dro3) has been

independently lost in Drosophila yakuba. Despite this extensive genomic rearrangement, the drosomycin gene family shows no signature of positive selection, suggesting these genes have not co-evolved with natural enemies (Jiggins and Kim, 2005); this is true of most insect AMPs (Lazzaro, 2008).

However, while AMPs do not typically coevolve with natural enemies, AMPs in eusocial insects appear to be the exception to this rule (Vilcinskas, 2013). This has been attributed to low genetic diversity within colonies, leading to defences that more

immediately clear invading pathogens. Eusocial insects also encode fewer total AMPs (Evans et al., 2006), and so adaptive evolution in eusocial insect AMP genes may imply a need for more diverse modes of action in conserved AMP gene families.

To explain the lack of adaptive evolution in most insect AMPs, it is proposed that pathogens have difficulty surviving the battery of AMPs encoded by a host. As such, pathogens with beneficial mutations that tolerate one AMP do not readily persist. Conserved cell wall constituents recognized by PRRs are also of structural importance, and cannot be altered without consequence, imposing a functional restriction on a

pathogen’s ability to avoid detection (Lazzaro, 2008). Thus, inhibiting signaling pathway intermediates appears to be the arena in which host-natural enemy immune coevolution occurs.

(16)

8 Drosophila melanogaster and Drosophila immunity

!

The genetically tractable model D. melanogaster has been highly influential in our understanding of innate immunity. The landmark sequencing of 12 Drosophila genomes in 2007 also offered researchers the ability to study the evolution of immune systems amongst closely related species. However, while immune signaling pathways are highly conserved in diverse Drosophila (Sackton et al., 2007), Drosophila effector proteins (e.g. AMPs) undergo constant genomic rearrangement. Moreover, our knowledge of AMP functions in diverse Drosophila is lacking, as characterization of AMPs has been performed almost entirely using D. melanogaster.

But Drosophila is an incredibly speciose lineage, particularly in two main subgenera, Sophophora and Drosophila (Figure 1.2). While D. melanogaster belongs to the subgenus Sophophora, the vast majority of Drosophila are members of the subgenus Drosophila (Obbard et al., 2012). Moreover, immune evolution studied in Drosophila has used the original 12 sequenced genomes (Clark et al., 2007), which lacks a representative of the Immigrans-Tripunctata clade of the subgenus Drosophila. Recently, newly

available sequence data from divergent species has emerged, allowing insight into this genetically lesser-characterized lineage of Drosophila. Two genomes of Immigrans-Tripunctata flies have now been sequenced, the genomes of Drosophila albomicans (Zhou et al., 2012) and Drosophila guttifera (Koshikawa et al., 2015), while Hamilton et al. (2014) sequenced the transcriptome of Drosophila neotestacea. Taking advantage of this newly available sequence data, I used the D. neotestacea transcriptome, with comparisons to the D. albomicans and D. guttifera genomes, to investigate the realized immune response of an Immigrans-Tripunctata fly.

Drosophila neotestacea, and the D. neotestacea transcriptome

Drosophila neotestacea is a woodland mushroom-breeding fly, and member of the Testacea species group of the subgenus Drosophila. Testacea group flies exhibit varying degrees of susceptibility to their common nematode parasite Howardula aoronymphium, which sterilizes D. neotestacea and Drosophila putrida, but not other Testacea group species (Perlman and Jaenike, 2003). Howardula nematodes infect their fly hosts by using a harpoon-like stylet to pierce the larval cuticle, after which they take

(17)

9 up residence in the host haemocoel, develop inside the maturing host fly, and release infective juvenile nematodes into host haemolymph. Flies then shed these infective juveniles onto mushroom breeding sites, thus renewing the cycle.

However, D. neotestacea is also infected with a defensive bacterial

endosymbiont, Spiroplasma, which rescues fertility in nematode-infected flies. Jaenike et al. (2010) reported the spread of Spiroplasma infection in D. neotestacea populations, highlighting a case study for evolution of defence through symbiosis. Recent interest in the mechanisms by which symbionts protect their host has surged, as inherited symbionts offer an intriguing mechanism to prevent the spread of disease through insect vectors. For instance, a Wolbachia endosymbiont that prevents the replication of dengue virus and Plasmodium falciparum is capable of infecting Aedes and Anopheles mosquitoes (Moreira et al., 2009; Hughes et al., 2011), and is being used to suppress the spread of dengue fever in Australia (Hoffmann et al., 2011).

To examine the mechanism(s) of defence employed by Spiroplasma, Hamilton et al. (2014) sequenced the transcriptome of D. neotestacea infected with Spiroplasma and/or H. aoronymphium. This transcriptome was generated with flies under diverse immune challenges, as Howardula is likely to introduce natural enemies directly into host haemolymph through the nematode wound site. As would be predicted of a parasite that inflicts cuticle damage and sterilizes its host, genes involved in cuticle formation were upregulated by infection, and genes involved in egg production were downregulated. One immune gene that was modestly upregulated in response to infection was a diptericin AMP. Upon further examination, I found this diptericin to be highly divergent from characterized diptericins in sequenced Drosophila.

Objectives of this thesis

There are two objectives to this thesis. First, I sought to investigate and

characterize the divergent diptericin AMP in D. neotestacea. To do this, I assembled the diptericin gene region from a diversity of sequenced Drosophila, and used phylogenetic analysis to understand the evolutionary history of this AMP gene family. My results prompted me to test for evidence of adaptive evolution in the Drosophila branch of the diptericin gene family. Following this, I assessed the expression profile of diptericin in D.

(18)

10 neotestacea upon immune challenge with Serratia bacteria and Beauveria fungus, and compared this expression profile with D. melanogaster. Finally, I determined the net charge of diptericin protein domains from distinct diptericin clades, and discuss my results in relation to our previous understanding of diptericin’s potential mode of action.

Second, I annotated the immune repertoire of D. neotestacea, and searched for novel immune genes in this divergent lineage of Drosophila. To annotate immune genes in the D. neotestacea transcriptome, I conducted an extensive literature review to

assemble a list of immune genes involved in Drosophila immune signaling pathways. I then used BLAST and phylogenetic analysis to search for orthologues of D.

melanogaster immune genes in the D. neotestacea transcriptome. I also searched the D. neotestacea transcriptome using immune gene profile Hidden Markov Models (profile HMMs) as a more sensitive search algorithm to extract putative novel immune genes from the D. neotestacea transcriptome. I then searched for orthologues of D. neotestacea putative novel immune genes from other Drosophila, and predicted their mature peptides. Finally, I determined the net charge of a putative novel AMP, and discuss my results in relation to previous endeavours to characterize Drosophila immunity, with particular emphasis on the conservation of AMPs.

Broader impact

Insects have major impacts in diverse fields of biology, and are also of incredible economic and medical importance. Drosophila has established itself as one of the most influential model organisms in our understanding of innate immunity and the evolution of innate immune systems (Nobelprize.org, 2015). As a result of the monumental body of work performed over the last two decades characterizing the innate immune system of D. melanogaster, we are now equipped with the tools to probe at the underlying genetics of disease susceptibility in other insects, and especially other Drosophila. The current study takes advantage of the characterized immune system of D. melanogaster to contribute to our knowledge of immune evolution in a divergent lineage of Drosophila.

Additionally, the spread of antibiotic-resistant bacteria demands new platforms for antibiotic discovery (Lewis, 2013). Insect antimicrobial peptides have been proposed as promising new platforms for antibiotic discovery (Plunkett et al., 2009; Tseng et al.,

(19)

11 2011). This study contributes to this field both by searching for novel immune genes, including novel AMPs, and can offer insight into the mode of action of diptericin and related compounds, by highlighting conserved motifs in the diptericin protein.

Finally, insects are both pests and vectors of disease in a wide array of biological systems. There has been recent interest in establishing a Drosophila model of nematode infection (Yadav et al., 2015). The Testacea species group is naturally infected by Howardula nematodes, and exhibits varying degrees of susceptibility to nematode infection. Thus, Testacea group species offer a natural host-parasite system to study the basis of nematode susceptibility in a Drosophila model. This study gives first insights into the immune signalling pathways induced upon nematode infection.

(20)

12

Literature Cited

Agaisse, H., and Perrimon, N. (2004). The roles of JAK/STAT signaling in Drosophila immune responses. Immunological Reviews, 198(1), 72-82. doi:10.1111/j.0105-2896.2004.0133.x

Candille, S. I., Kaelin, C. B., Cattanach, B. M., Yu, B., Thompson, D. A., Nix, M. A. … and Barsh, G. S. (2007). A β-defensin mutation causes black coat color in domestic dogs. Science, 318(5855), 1418-1423. doi:10.1126/science.1147880 Clark, A. G., Eisen, M. B., Smith, D. R., Bergman, C. M., Oliver, B., … Markow, T. A.

(2007). Evolution of genes and genomes on the Drosophila phylogeny. Nature, 450(7167), 203-218. doi:10.1038/nature06341

Clemmons, A. W., Lindsay, S. A., and Wasserman, S. A. (2015). An effector peptide family required for Drosophila toll-mediated immunity. PLoS Pathogens, 11(4), e1004876. doi:10.1371/journal.ppat.1004876

Cudic, M., Bulet, P., Hoffmann, R., Craik, D. J., and Otvos, J., L. (1999). Chemical synthesis, antibacterial activity and conformation of diptericin, an 82-mer peptide originally isolated from insects. European Journal of Biochemistry / FEBS, 266(2), 549.

Evans, J. D., Aronstein, K., Chen, Y. P., Hetru, C., Imler, J., and Jiang, H. (2006). Immune pathways and defence mechanisms in honey bees Apis mellifera. Insect Molecular Biology, 15(5), 645-656. doi:10.1111/j.1365-2583.2006.00682.x Hamilton, P. T., Leong, J. S., Koop, B. F., and Perlman, S. J. (2014). Transcriptional

responses in a Drosophila defensive symbiosis. Molecular Ecology, 23(6), 1558-1570. doi:10.1111/mec.12603

Havard, S., Doury, G., Ravallec, M., Brehélin, M., Prévost, G., and Eslin, P. (2012). Structural and functional characterization of pseudopodocyte, a shaggy immune cell produced by two Drosophila species of the Obscura group. Developmental and Comparative Immunology, 36(2), 323. doi:10.1016/j.dci.2011.05.009 Hughes, G.L., Koga, R., Xue, P., Fukatsu, T., and Rasgon, J.L. (2011). Wolbachia

infections are virulent and inhibit the human malaria parasite Plasmodium falciparum in Anopheles gambiae. PLoS Pathogens, 7, e1002043.

Hoffmann, A.A., Montgomery, B. L., Popovici, J., Iturbe-Ormaetxe, I., Johnson, P. H., Muzzi, F., and O'Neill, S. L. (2011). Successful establishment of Wolbachia in Aedes populations to suppress dengue transmission. Nature, 476(7361), 454-457. doi:10.1038/nature10356

(21)

13 Jaenike, J., Unckless, R., Cockburn, S. N., Boelio, L. M., and Perlman, S. J. (2010).

Adaptation via symbiosis: Recent spread of a Drosophila defensive symbiont. Science, 329(5988), 212-215. doi:10.1126/science.1188235 Jiggins, F.W., and Kim, K.W. (2005). The evolution of antifungal peptides in

Drosophila. Genetics. 171: 1847-1859.

Kim, Y.S., Ryu, J.H., Han, S.J., Choi, K.H., Nam, K.B., Jang, I.H., Lemaitre, B., Brey, P.T., and Lee, W.J. (2000). Gram-negative bacteria- binding protein, a pattern recognition receptor for lipopolysaccharide and β-1,3-glucan that mediates the signaling for the induction of innate immune genes in Drosophila melanogaster cells. Journal of Biological Chemistry. 275:32721–72

Kimbrell, D. A., and Beutler, B. (2001). The evolution and genetics of innate immunity. Nature Reviews Genetics, 2(4), 256-267. doi:10.1038/35066006 Koshikawa, S., Giorgianni, M. W., Vaccaro, K., Kassner, V. A., Yoder, J. H., Werner, T.,

and Carroll, S. B. (2015). Gain of cis-regulatory activities underlies novel

domains of wingless gene expression in Drosophila. Proceedings of the National Academy of Sciences, 112(24), 7524-7529. doi:10.1073/pnas.1509022112 Lavine, M. D., and Strand, M. R. (2002). Insect hemocytes and their role in

immunity. Insect Biochemistry and Molecular Biology, 32(10), 1295-1309. doi:10.1016/S0965-1748(02)00092-9

Lazzaro, B. P. (2008). Natural selection on the Drosophila antimicrobial immune system. Current Opinion in Microbiology, 11(3), 284-289. doi:10.1016/j.mib.2008.05.001 Lewis, K. (2013). Platforms for antibiotic discovery. Nature Reviews. Drug

Discovery, 12(5), 371. doi:10.1038/nrd3975

Lemaitre, B., and Hoffmann, J. (2007). The host defense of Drosophila melanogaster. Annual Review of Immunology, 25, 697-743.

Ligoxygakis, P., Pelte, N., Hoffmann, J.A., and Reichhart, J.M. (2002). Activation of Drosophila Toll during fungal infection by a blood serine protease. Science 297:114–16

Login, F. H., Balmand, S., Vallier, A., Vincent-Monégat, C., Vigneron, A., Weiss-Gayet, M., Rochat, D., and Heddi, A. (2011). Antimicrobial peptides keep insect

endosymbionts under control. Science, 334(6054), 362-365. doi:10.1126/science.1209728

(22)

14 Mesquita, R. D., Vionette-Amaral, R. J., Lowenberger, C., Rivera-Pomar, R., Monteiro,

F. A., …, Oliveira, P.L. (2015). Genome of Rhodnius prolixus, an insect vector of chagas disease, reveals unique adaptations to hematophagy and parasite infection. Proceedings of the National Academy of Sciences, 112(48), 14936.

doi:10.1073/pnas.1506226112

Moreira, L. A., Iturbe-Ormaetxe, I., Jeffery, J. A., Lu, G., Pyke, A. T., Hedges, L. M., Rocha, B.C., Hall-Mendelin, S., Day, A., Riegler, M., Hugo, L.E., Johnson, K.N., Hay, B.H., McGraw, E.A., van den Hurk, A.F., Ryan, P.A., and O'Neill, S. L. (2009). A Wolbachia symbiont in Aedes aegypti limits infection with dengue, Chikungunya, and Plasmodium. Cell, 139, 1268–1278.

Nobelprize.org. 2015. “The Nobel Prize in Physiology or Medicine 2011". Nobel Media AB 2014. Web. 16 Oct 2015.

<http://www.nobelprize.org/nobel_prizes/medicine/laureates/2011/>

Obbard, D. J., Maclennan, J., Kim, K., Rambaut, A., O'Grady, P. M., and Jiggins, F. M. (2012). Estimating divergence dates and substitution rates in the Drosophila phylogeny. Molecular Biology and Evolution, 29(11), 3459.

doi:10.1093/molbev/mss150

Palmer, W. J., and Jiggins, F. M. (2015). Comparative genomics reveals the origins and diversity of arthropod immune systems. Molecular Biology and Evolution, 32(8), 2111-2129. doi:10.1093/molbev/msv093

Perlman S.J., and Jaenike J. (2003). Infection success in novel hosts: an experimental and phylogenetic analysis of Drosophila-parasitic nematodes. Evolution 57: 544-557. Plunkett, R. M., Murray, S. I., and Lowenberger, C. A. (2009). Generation and

characterization of the antibacterial activity of a novel hybrid antimicrobial peptide comprising functional domains from different insect cecropins. Canadian Journal of Microbiology, 55(5), 520-520. doi:10.1139/W09-001

Ponnuvel, K. M., Subhasri, N., Sirigineedi, S., Murthy, G. N., and Vijayaprakash, N. B. (2010). Molecular evolution of the cecropin multigene family in silkworm Bombyx mori. Bioinformation, 5(3), 97.

Rahnamaeian, M., Cytryńska, M., Zdybicka-Barabas, A., Dobslaff, K., Wiesner, J., Twyman, R. M., Zuchner, T., Sadd, B.M., Regoes, R.R., Schmid-Hempel, P., and Vilcinskas, A. (2015). Insect antimicrobial peptides show potentiating functional interactions against Gram-negative bacteria. Proceedings of the Biological Sciences / the Royal Society, 282(1806), 20150293. Doi:

(23)

15 Sackton, T.B., Lazzaro, B.P., Schlenke, T.A., Evans, J.D., Hultmark, D., and Clark, A.G.

(2007). Dynamic evolution of the innate immune system in Drosophila. Nature Genetics.,39:1461–1468.

Strand, M. R. (2008). The insect cellular immune response. Insect Science, 15(1), 1-14. doi:10.1111/j.1744-7917.2008.00183.x

The International Aphid Genomics Consortium. (2010). Genome sequence of the pea aphid acyrthosiphon pisum. PLoS Biology, 8(2), e1000313.

doi:10.1371/journal.pbio.1000313

Tseng, J.-M., Huang, J.-R., Huang, H.-C., Tzen, J. T. C., Chou, W.-M. and Peng, C.-C. (2011). Facilitative production of an antimicrobial peptide royalisin and its antibody via an artificial oil-body system. Biotechnology Progress, 27: 153–161. doi: 10.1002/btpr.528

Vilcinskas, A. (2013). Evolutionary plasticity of insect immunity. Journal of Insect Physiology, 59(2), 123-129. doi:10.1016/j.jinsphys.2012.08.018

Yadav, S., Shokal, U., Forst, S., and Eleftherianos, I. (2015). An improved method for generating axenic entomopathogenic nematodes. BMC Research Notes, 8(1), 461. doi:10.1186/s13104-015-1443-y

Zhou, Q., Zhu, H., Huang, Q., Zhao, L., Zhang, G., Roy, S. W., Vicoso, B., Xuan, Z., Ruan, J., Zhang, Y., Zhao, R., Ye, C., Zhang, X., Wang, J., Wang, W. and Bachtrog, D. (2012). Deciphering neo-sex and B chromosome evolution by the draft genome of Drosophila albomicans. BMC Genomics, 13(1), 109-109. doi:10.1186/1471-2164-13-109

(24)

16 F igu re 1 .1 : T he f ive m aj or s igna li ng pa thw ays re gul at ing t he ins ec t hum ora l i m m une re spons e, a s f ound i n D . m el anogas te r: T ol l, Im d, J A K /S T A T , J N K , a nd t he m el ani za ti on re spons e. T he T ol l pa thw ay i s i nduc ed by G ra m -pos it ive ba ct eri a a nd f ungi . T he Im d pa thw ay i s i nduc ed by G ra m -ne ga ti ve ba ct eri a. T he J A K /S T A T pa thw ay re sponds to a va ri et y of orga ni sm al s tre ss es , a nd m ay tra ns cri be e ff ec tor prot ei ns tha t re gul at e t he m el ani za ti on re spons e. T he J N K pa thw ay a nd Im d pa thw ay s ha re the M A P K K K ki na se dT A K 1, a nd t he J N K pa thw ay i s s ubs eque nt ly i nduc ed by Im d a ct iva ti on. T he J N K pa thw ay a ls o m edi at es c el lul ar a popt os is . T he m el ani za ti on re spons e oc curs e xt ra ce ll ul arl y, a nd re sul ts in t he produc ti on of m el ani n a nd re ac ti ve oxyge n s pe ci es tha t a re us ed t o com ba t na tura l e ne m ie s. ANK$

Toll$

psh$ sphinx1$ sphinx2$ GNBP3$ mo dS P$ G ras s$ Sp iri t$ Sphe roide $ SP E$ Pro <s pätzl e$ My d8 8$ Tube $ Pel le $ D or sal $ Dif $ ne c$ GNBP1$ PG RP <S A$ PG RP <S D $ Spn1 $ Fung i$ Ye as t$ Me tc hn ik ow in $ De fe nsin $ Dr osomy cin $ Toll$ AMP s$ ANK$ D or sal $ Dif $ A$ N$ K$ Bo m an in $

Im

d$

FA DD$ red d$ Re lis h$ PG RP <L C$ PG RP <L E$ G ram <n eg .$b ac te ria$ D IA P2 $ TA B2 $ Ken ny$ IRD 5$ TA K1 $ Re lis h$ Dr osocin $ AR ac in $ Im d$ AMP s $ GNBP2$ Cac tu s$ G ram <p os .$b ac te ria$ PG RP <L F$ PG RP <L B$ PG RP <S B2 $ PG RP <S C2 $ PG RP <S B1 $ PG RP <S C1 $ D ip te ric in $ Se nju $ Cas par $ Dnr 1$ Sic kie $ ANK$ Cac tu s$ Pi rk $ Uev 1a$ be n$ Eﬀ ete $ Pel lin o$ Gp rk2$ Ce cr opin $ Re lis h$

JNK$

WGN$ EG R$ HE P$ Bas ke t$ dT RA Fs $ msn $ slpr $ MK K4 $ puc $ dRO N C$ TA B2 $ TA K1 $ JRA $ KAY $ re ap er $ D IA P1 $ dRIC E$ DCP E1$ dA RK $ Ap op to si s$ Cdk 5Eα$

Me

lan

izaK

on

$

res

po

ns

e$

PPO ’s $ Spn2 8d$ Sp n7 7B a$ Me lan izaK on $ Re sp on se $ upd3 $ upd1 $ _Dome $ Ho p$ ST AM$ Soc s$36E$ Su( var )2E10$ TE P$ IV $ TE P$ II$ TE P$ I$ To tA $ Le cKn $24A $ Slbo $ To tC $ TE P$ III$ et$

JAK/S

TAT

$

STA T$9 2E$ upd2 $ PO 2$ PO 1$ PO3$ Tissue Espe ciﬁc$ se rpins $ Hae m ol ym ph $+$ tr ac he ae $ Tr ac he ae $ Mc o1 $ To tB $ Hm l$ aP S4 $ Str es s$ re sp on se $/ $O ps on izaK on $/ $C lo _n g$ $ Fondue $ BRW D 3$ MP 2$ mo dS P$ GNBP3$ GNBP1$ PG RP ES A$ MP 1$ Hay an $ CG9737$ Ye llow $f$ PG RP EL E$ pr oMP 2$ Sp n2 7a$ Spn5 $ hae m ol ym ph $ Ey es$ dTG $ STA T$9 2E$ STA T$9 2E$ Toll$ Im d$ WGN$ Dome $ et$ Ha emo ly mp h* Cy to pl as m* N ucl eu s* Cy to pl as m* Ha emo ly mp h* N ucl eu s* Eﬀ ecto rs *tr an scr ib ed * by *fa t*b od y* NF /κB $ NF /κB $

(25)

17 F igu re 1. 2: D ros ophi li d m axi m um li ke li hood phyl oge ny of c onc at ena te d A dh, a m d, e ngra il ed a nd gl as s prot ei n s eque nc es , us ing a n L G m ode l of nuc le ot ide s ubs ti tut ion. S upport va lue s re pre se nt 10 0 boot st ra ps . Inc lude d s pe ci es e nc om pa ss m uc h of the di ve rs it y of D ros ophi la , a nd f orm the ba si s of c om pa ra ti ve w ork i n t hi s s tudy. T he se s eque nc es w ere ge ne ra te d f rom F lyBa se c ura te d prot ei n tra ns la ti ons a nd re ce nt ly a va il abl e s eque nc e da ta f rom D . al bom ic ans , D . gut ti fe ra, D . ne ot es tac ea, S. l ebanone ns is , a nd P . v ar ie gat a. T he s ubge ne ra S ophophora a nd D ros ophi la a re e st im at ed t o ha ve di ve rge d 25 -40 M a (O bba rd et al ., 2012). D .#me la no ga ste r# D.# ne ote sta ce a# D.# gu 0f er a# D .a lb omi ca ns # D.# vi ril is # D.# mo ja ve ns is # D.# gr ims ha w i# D.# yak uba # D.# ananassae # D.# ps eu do ob scu ra # D.# w ill is to ni # S.# le banone nsis # P. #v ar ie ga ta # Subge nus( Sophophor a( Subge nus( Dro so ph ila ( Im m ig ran s)* Tr ip un ctata * Rad ia2 on * Gen us( D ro so ph ila * Gen us *S ca pt od ro so ph ila * Gen us *Phor0c a* D.# si mu la ns #

(26)

18

Chapter II: Divergent evolution of the antimicrobial peptide

diptericin in Drosophila

Introduction

Antimicrobial peptides (AMPs) are host-encoded antibiotics that combat invading microorganisms. These short molecules contribute to host defence against diverse natural enemies, and can even contribute to the maintenance of beneficial endosymbionts (Login et al., 2011; Vilcinskas, 2013). Drosophila have been highly influential in our understanding of AMPs as immune responsive genes, and in

characterizing AMP evolution. A variety of gene-editing technologies are available to Drosophila researchers, that have been used to study the function of immune signaling genes. As AMPs are regulated by specific immune signaling pathways, the functions of signaling genes are inferred through both in vitro activity, and the induction of pathway-associated AMPs in vivo (Lemaitre and Hoffmann, 2007).

In Drosophila, AMP expression is regulated by two immune signaling cascades, the Toll and Imd pathways. The Toll pathway responds to Gram-positive bacterial

challenge or fungal infection through a variety of pathogen-specific Pattern Recognition Receptors (PRRs). The Toll signaling cascade culminates in the transport of NF-κB binding proteins into the nucleus, where they initiate the transcription of Toll pathway-induced AMPs such as drosomycin and defensin. The Imd pathway responds to Gram-negative bacterial challenge in a similar fashion, culminating in the expression of Imd-induced AMPs, such as diptericin and attacin. Studies of Drosophila immunity typically use the AMPs drosomycin and diptericin as readouts for expression of the Toll and Imd pathways respectively (Lemaitre and Hoffmann, 2007).

Recently Hamilton et al. (2014) sequenced the transcriptome of Drosophila neotestacea, a woodland mushroom-breeding fly, and member of the Immigrans-Tripunctata clade in the subgenus Drosophila (Figure 1.2). This transcriptome was

generated using flies exposed to a Spiroplasma defensive endosymbiont, and/or a virulent nematode parasite, Howardula aoronymphium. One AMP highlighted for its potentially interesting pattern of expression was a diptericin-like orthologue identified as diptericin B (DptB) by BLAST annotation.

(27)

19 Diptericins are proline and glycine rich AMPs that are strongly up-regulated in response to Gram-negative bacterial challenge. Drosophila melanogaster contains two diptericins, diptericin A (DptA) and DptB. The diptericin molecule contains two protein domains, the proline-rich P domain and glycine-rich G domain, both bearing great

similarity to attacin P and G domains (Hedengren et al., 2000). However, while the whole molecule appears to be necessary to exhibit antibacterial activity, it is thought that only the cationic G domain is responsible for bacterial killing (Cudic et al., 1999). Indeed, cationic protein domains are thought to be required for antimicrobial peptide activity (Lemaitre and Hoffmann, 2007).

Upon inspection, only one diptericin was expressed in the D. neotestacea transcriptome. This D. neotestacea diptericin-like orthologue showed extremely poor identity to both DptA and DptB in D. melanogaster. These two D. melanogaster diptericins can be distinguished in the genomes of sequenced Drosophila as DptA is found upstream of DptB, and DptA has only one exon while DptB has two exons.

Curated diptericins in FlyBase (vFB2015_04) indicate that DptB is universally conserved in sequenced Drosophila, but DptA is restricted to the subgenus Sophophora. By using the D. neotestacea diptericin-like orthologue as a query, I found additional diptericins from a diversity of sequenced Drosophila.

To characterize the evolutionary ancestry of this divergent diptericin, I used phylogenetic analysis and assembled the diptericin gene region from sequenced drosophilids (Clark et al., 2007), as well as the recently sequenced Drosophila albomicans (Zhou et al., 2012), Drosophila guttifera (Koshikawa et al., 2015),

Scaptodrosophila lebanonensis, Phortica variegata (Vicoso and Bachtrog, 2015), and other Brachyceran fly genomes (see Table S2.1). I followed up on this characterization with qPCR assays to look for induction of diptericins in response to immune challenges by a Gram-negative bacteria (Serratia) or an entomopathogenic fungus (Beauveria).

(28)

20

Methods

Phylogenetic analysis of diptericin genes

To test hypotheses regarding defensive symbiosis, Hamilton et al. (2014) sequenced the transcriptome of D. neotestacea infected with either nematodes and/or Spiroplasma bacteria. One immune gene with modest up-regulation in response to infection treatments was a diptericin-like AMP. Alignments between this D. neotestacea diptericin and diptericins from D. melanogaster resulted in poor sequence similarity. To identify the evolutionary origins of this diptericin-like AMP, I extracted annotated diptericins from FlyBase, and used BLAST to search GenBank databases for diptericin genes from a diversity of Diptera. I also used BLAST to search the genomes of D. albomicans, D. guttifera, S. lebanonensis, and P. variegata, to pull out diptericin-like sequences from these flies. The well-conserved G domain of these diptericin nucleotide sequences was then translation-aligned using MUSCLE. I used PhyML to construct a Maximum likelihood phylogeny for these diptericin sequences using the aligned G domains with an AIC-selected best model of nucleotide substitution provided by Datamonkey.org model selection (Delport et al., 2010). Diptericins from Mayetiola destructor, D. ananassae (Dana\GF11125) and D. simulans (Dsim\GD11418) were excluded from this phylogeny due to divergent sequence resulting in long-branch

attraction to diptericins of unrelated flies. These analyses were performed in Geneious 7. Phylogenetic analysis revealed four distinct clades of diptericins, with three clades containing Drosophila diptericins. To better understand the potential evolutionary relatedness of these Drosophila diptericins, I compared the amino acid sequence

similarities of Drosophila sequences from each of these clades. I generated amino acid sequence alignments of concatenated diptericin P domains and G domains (but not signal peptides or propeptides) as aligned with D. melanogaster diptericins from Hedengren et al. (2000). I then assessed pairwise identity of these within-clade alignments to determine baseline expected relatedness for syntenic conserved diptericin orthologues within

Drosophila; this baseline was provided by the universally conserved DptB. So as to avoid noise introduced from a greater number of available genomes in the subgenus

(29)

21 ananassae, D. pseudoobscura, and D. willistoni, to generate amino acid consensus

sequences for Sophophora DptA and Sophophora DptB. To generate subgenus

Drosophila consensus sequences, I used diptericin-like orthologues from D. virilis, D. mojavensis, D. grimshawi, D. albomicans, D. neotestacea, and D. guttifera, and used DptB sequences from D. virilis, D. mojavensis, D. hydei, D. grimshawi, and D.

albomicans (Table 2.1). This change in input species was due to available DptB sequence from D. hydei contributed by this study, and a lack of comparable DptB sequence from D. neotestacea and D. guttifera.

Synteny of diptericins in sequenced Drosophila genomes

To determine evolutionary relationships of Drosophila diptericins, I assembled the diptericin gene regions of sequenced drosophilid flies. For Drosophila sequenced in Clark et al. (2007), I extracted diptericin gene regions from FlyBase. For the D.

albomicans, D. guttifera, S. lebanonensis, and P. variegata genomes, I used BLAST searches with queries from assembled diptericins, including the D. neotestacea diptericin-like orthologue. I extracted the entire diptericin-containing scaffold and manually

searched for conserved diptericin motifs in this gene region to identify potential additional but divergent diptericins. Due to current genomic scaffold assembly of D. guttifera, the signal peptide and P domain of the D. guttifera copy of the D. neotestacea like orthologue was not available, and thus the N-terminus of this diptericin-like gene is not included in this analysis. Also, the intergenic region between the two D. guttifera diptericins was not fully sequenced, and thus the reported length for this intergenic region represents currently available sequence.

Following the identification of diptericins in the diptericin gene region of drosophilid flies, I generated alignments of DptA, DptB, and orthologues of the D. neotestacea diptericin-like gene to provide an anchor for subsequent gene region alignment. For some species, additional diptericin duplications were present, and I used flanking genes to determine the ancestral gene copy for alignment purposes. The P. variegata genome encodes two DptB orthologues not found in the same genomic scaffolds.

(30)

22 Positive selection on Drosophila diptericins

Phylogenetic analysis found that DptA and the diptericin-like lineage (hereafter referred to as DptC for ease of explanation) form distinct clades; yet these two genes are syntenic orthologues in the genomes of sequenced Drosophila. The DptA gene is

restricted to the subgenus Sophophora, while the DptC gene is restricted to the subgenus Drosophila. This prompted me to investigate rates of synonymous and non-synonymous change (dN/dS) in the conserved G domain of these diptericin clades. Also, as diptericins from numerous Brachyceran flies cluster with the Drosophila DptB clade, the Drosophila DptB gene appears to be the ancestral gene from which DptA and DptC duplicated. Thus, a revised codon alignment of drosophilid diptericin clades was used to assess rates of dN/dS amongst these clades; based on this inferred Drosophila DptB duplication event, DptA and DptC form a monophyletic lineage nested within DptB.

I used Branch-site REL (BSR) (Kosakovsky-Pond et al., 2011) to identify sites evolving under episodic diversifying selection in the G domain of Drosophila diptericins using the datamonkey.org webserver (Delport et al., 2010). I used multiple codon

alignments containing or omitting divergent diptericins to confirm the results of this analysis were robust to included divergent diptericins; this abundance of caution was required as no root clade is specified in BSR analysis, and different root clades impact inferred codon evolution by BSR.

Characterizing ionic charge in DptC

Given the extreme divergence of DptC from DptA, I was interested in potential mutations affecting the net charge of the DptC molecule that resulted from amino acid changes. Antimicrobial peptides are thought to require a cationic net charge for

antimicrobial activity (Lemaitre and Hoffman, 2007). I used protein domains of D. melanogaster DptA and DptB (Hedengren et al., 2000) to generate alignments of DptC protein domains. I then assessed the net charge of each protein domain, and of the whole molecule, at pH 6.5 using Protein Calculator v3.4 (http://protcalc.sourceforge.net/); pH 6.5 is the approximate pH of insect haemolymph (Wyatt et al., 1956).

(31)

23 Drosophila stocks

Fly stocks of D. neotestacea were originally collected from W. Hartford, Connecticut in 2006. The Oregon-R strain of D. melanogaster used in this study was graciously donated by Bruno Lemaitre in autumn 2013, and both species were maintained at 21°C with a 12-hour light:dark cycle on Instant Drosophila Medium (Carolina

Biological Supply); D. neotestacea food vials were supplemented with a small piece of mushroom (Agaricus bisporus). The D. neotestacea strain used in this study was not infected with either Wolbachia or Spiroplasma, bacterial endosymbionts of D. neotestacea. Oregon-R is not infected by bacterial endosymbionts.

Immune challenge with Serratia

My Serratia strain was isolated by extracting bacteria from mycophagous Drosophila cultures, was found to be pathogenic upon oral infection, and is closely related to the soil bacterium Serratia marcescens (Fig. S2). For both D. neotestacea and D. melanogaster, approximately 10 males and 10 females were allowed to lay on ~1/2 tsp Instant Drosophila Medium with 1:1 water; D. neotestacea vials were supplemented with ~0.5g A. bisporus. Newly emerged males were then collected daily and kept in isolation from females for 3-4 days on ~1/2 tsp Instant Drosophila medium with 1:1 water (no mushroom supplement). These males were then lightly anaesthetized on CO2 and

wounded in the left side of the thorax above the wing with a 0.5um tip tungsten needle. For sterile woundings, this tungsten needle was flamed until the tip glowed red hot, and then dipped in 95% EtOH for 5 seconds, then flamed briefly prior to wounding. For septic woundings, this needle was instead flamed until red hot, dipped in 95% EtOH, flamed briefly, and then dipped in OD600 = 0.1 Serratia cultured in Luria-Bertani broth.

Flies were then left to recover in a clean polystyrene vial for 30 minutes prior to transfer to a vial containing ~1/2 tsp Instant Drosophila medium with 1:1 water. Six hours post-wounding, flies were flash frozen in liquid nitrogen and kept at -80°C until RNA was extracted.

(32)

24 Immune challenge with Beauveria bassiana

Beauveria bassiana (hereafter referred to as Beauveria) cultures were graciously provided by Will Hintz and Jon Leblanc in Summer 2015. For both D. neotestacea and D. melanogaster, approximately 10 males and 10 females were allowed to lay on ~1/2 tsp Drosophila food medium with 1:1 water; D. neotestacea vials were supplemented with ~0.5g A. bisporus. Newly emerged males were then collected daily and kept in isolation from females for 3-4 days on ~1/2 tsp Instant Drosophila medium with 1:1 water (no mushroom supplement). These males were then lightly anaesthetized on CO2 and transferred to either a sterile Potato-dextrose agar petri dish, or one

containing sporulating Beauveria culture (strain UAMH 1514). Dishes were then shaken for 30 seconds to cover the flies in fungal spores. After, flies were left to recover in a clean polystyrene vial for 30 minutes prior to transfer to a vial containing ~1/2 tsp Instant Drosophila medium with 1:1 water. Twenty-four hours post-exposure, flies were flash frozen in liquid nitrogen and kept at -80°C until RNA was extracted.

RNA extraction and cDNA synthesis

I extracted RNA from eight individual flies per treatment using Trizol-LS (Invitrogen) with the manufacturer’s protocol, and then eluted RNA into 20μL RNAse-free water. RNA from each fly was then DNAse treated (Thermo Scientific DNAse I) to remove contaminating genomic DNA prior to reverse transcription (Applied Biological Materials, 5X All-In-One RT MasterMix). I confirmed successful RNA extractions using agarose gel electrophoresis of intact total RNA to look for Drosophila 18S/28S rRNA bands; insect 28S rRNA breaks apart upon denaturation resulting in what appears as one band migrating closely to 18S rRNA (Winnebeck et al., 2010).

qPCR for gene expression and data analysis

The qPCR primers used in this study are listed in Table S2.2, and PCR

(33)

25 EvaGreen 2X qPCR MasterMix), using a BioRad CFX96 qPCR system. I assayed the expression of DptA and an additional Imd pathway readout attacin B (AttB) in D. melanogaster and the diptericin-like orthologue and AttB in D. neotestacea for Serratia infections. I assayed respective diptericins in both species for Beauveria infections. These infections were performed once for each pathogen. For all qPCR reactions, target genes were run alongside an endogenous control gene; for D. neotestacea this control was RpL28 (Hamilton et al., 2014), and for D. melanogaster this control was RpL32 (Tsai et al., 2008). Each reaction was run in triplicate, and replicates were considered consistent if the threshold cycle (CT) of each replicate was contained within a 0.5 CT boundary.

Gene expression analysis was performed using the 2-(Delta-Delta CT)method (Livak and Shmittgen, 2001). Two-sample T-tests were used to determine differences in

expression profile using R 3.1 statistical software. The standard deviation of the Delta CT

of the samples within each treatment was used as the start of error propagation measures.

Results

Phylogenetic analysis of diptericin genes

Through BLAST searches of the D. neotestacea transcriptome, a divergent diptericin orthologue was identified with poor amino acid similarity to either of D. melanogaster DptA or D. melanogaster DptB. To understand potential evolutionary relationships of this diptericin, I assembled a diversity of diptericin sequences from GenBank databases and available sequenced genomes. After this, I aligned the well-conserved G domain of these diptericin sequences to establish phylogenetic relatedness of these assembled diptericins (Figure 2.1).

Four diptericin clades emerged from this analysis. One clade contained diptericins from muscoid flies as well as outgroups of Drosophila species. The clade containing D. melanogaster DptB (hereafter referred to as the Drosophila DptB clade) also contained DptB orthologues from other Drosophila flies, from the drosophilid S. lebanonensis; diptericins from the tephritid flies C. capitata, B. dorsalis, and B. cucurbitae also clustered with this clade. The third clade contained D. melanogaster DptA and other subgenus Sophophora DptA orthologues (hereafter referred to as the

(34)

26 DptA clade), but did not contain any subgenus Drosophila diptericins. The last clade contained the D. neotestacea diptericin-like gene, as well as diptericins exclusively from other subgenus Drosophila flies (hereafter referred to as the DptC clade).

To better understand the divergences of these diptericin clades from one another, particularly amongst Drosophila diptericin lineages, I compared amino acid sequence similarities of the consensus sequences of Drosophila diptericins from these diptericin clades (Table 2.1). These comparisons were made with amino acid consensus sequence from the N-terminal P domain and C-terminal G domain of predicted mature diptericin proteins. Similarity of within-clade comparisons (the pairwise identity of the amino acid alignments used to generate consensus sequences for each clade) was 65.1 ±$ 4.9%. However, between clade comparisons yielded much weaker similarity of 39.6 ±$ 0.65%.$

Synteny of diptericins in sequenced Drosophila genomes

As Drosophila diptericins from each diptericin clade were equally dissimilar to one another, I sought to determine the evolutionary origins of these distinct diptericin clades using synteny in sequenced genomes. I assembled the diptericin gene regions of sequenced Drosophila and aligned these gene regions based on the position of DptB in each fly.

What became immediately apparent was that DptA was found upstream of DptB in each member of the subgenus Sophophora, and DptC was similarly upstream of DptB in each member of the subgenus Drosophila (Figure 2.2); the final alignment uses DptA and DptC as anchors to align these diptericin gene regions. Multiple DptC

duplication events have occurred in the subgenus Drosophila flies D. grimshawi, D. mojavensis, and D. virilis. The Drosophila willistoni genome encodes one additional DptB orthologue found 16,853bp downstream of its syntenic DptB, which is not shown in Figure 2.2.

In the D. guttifera genome, the scaffold containing DptC and DptB is not complete, and contains an uncharacterized intergenic region between DptC and DptB. This intergenic region is marked as “> 3000 base pairs” in Figure 2.2, the approximate

(35)

27 length of sequenced nucleotide bases between DptC and DptB. Drosophila guttifera is also unique in that the D. guttifera DptB orthologue has been pseudogenized by multiple mid-exon frameshifts in both the P domain and G domain of the DptB molecule.

Drosophila neotestacea is included in Figure 2.2 despite lacking a sequenced genome. No DptB orthologue was recovered from the D. neotestacea transcriptome. One of the two DptB-like orthologues in S. lebanonensis was found to be syntenic and

conserved upstream of the other, and this orthologue upon further inspection contained only one exon similar to DptA and DptC. As such, it appears one S. lebanonensis DptB-like orthologue is in fact a DptA/DptC orthologue. No DptA/DptC orthologues were observed in P. variegata.

Positive selection on Drosophila diptericins

As DptA and DptC are syntenic orthologues, and DptB is the universally conserved drosophilid diptericin, it is parsimonious to conclude that DptA and DptC resulted from the same DptB duplication event, and have since undergone divergent evolutionary trajectories. Genetic distance implies that these genes are more greatly dissimilar to each other than would be predicted by the relatedness of subgenus Drosophila DptB to subgenus Sophophora DptB (Figure 2.1). The question then becomes: what evolutionary forces have driven this divergent evolution of amino acid sequence (Table 2.1)? To investigate this question, I used Branch-site REL (BSR) to search for episodic diversifying selection (Figure 2.3). Branch-site analyses use

phylogenetic information to infer sites that have undergone episodes of positive selection (dN/dS > 1). Unlike previous branch-site analysis methods, BSR makes no initial

assumptions about which lineages should be evolving under positive selection, and allows dN/dS to vary amongst branches within a lineage (Delport et al., 2010).

The analysis was sensitive to input alignments, as including divergent

diptericins resulted in alternate clades forming the root of BSR neighbour-joining trees. However, in every alignment, the branch leading to DptC was found to have evolved under diversifying positive selection (p < .05). This result was strongest in the diptericin alignment containing all syntenic diptericins except those of Drosophila ananassae (p =

(36)

28 9.3e-4); Drosophila ananassae DptA clusters with D. wilistoni DptA on a long branch in phylogenetic analyses, affecting branch-site dN/dS calculations. The proportion of sites evolving under diversifying selection ranged from 0.221 to 0.303 depending on input alignment, and the Likelihood-ratio test values for positive selection (dN/dS > 1) for this branch ranged from 10.591 to 17.786. There was not strong support for positive selection in the branch between DptA and DptB, except in an alignment that did not include D. ananassae diptericins (p = 0.017, proportion of sites = 0.158, LR = 12.246) (Figure S2.1). Characterizing ionic charge in DptC

Given the extreme dissimilarity in amino acid sequence between the syntenic orthologues DptA and DptC, I was interested in the potential for functional changes in these diptericin molecules. I assessed the net charge of the diptericin P domain and G domain from a variety of sequenced Drosophila at pH 6.5 (Table 2.2).

There are extensive differences in the net charge of DptC domains relative to DptA domains. In general, DptCs had more negative P domains, more positive G

domains, and more positive total net charges. The DptC orthologue in D. albomicans was especially positive, being the only species that had a positive P domain. Due to the incomplete sequence of the D. guttifera genome, I could not recover the P domain of the D. guttifera DptC orthologue; the G domain remained intact, and was the most positively charged domain of any diptericin. Interestingly there is also variation in the net charge of the DptA molecule amongst subgenus Sophophora flies. The DptA molecule in D. anansasae was the only diptericin to have a negatively charged G domain, whilst both D. ananassae and D. pseudoobscura had anionically charged diptericin molecules.

The mode of action by which diptericin kills bacteria is poorly understood, but the Gly22-Asp45 region of the diptericin G domain has been hypothesized as the region responsible for diptericin’s antibacterial activity (Cudic et al., 1999), and shows

conserved homology to the G domain of attacin, another Imd pathway AMP (Hedengren et al., 2000). This region has undergone considerable modification unique to each diptericin clade (Figure S2.2a). However, the Asn46-on region of the G domain shows greater conservation amongst diptericins (Figure S2.2b). Thus, the diptericin G domain

Immune evolution in the Immigrans-Tripunctata clade of Drosophila

Immune evolution in the Immigrans-Tripunctata clade of Drosophila

Supervisory Committee

Immune evolution in the Immigrans-Tripunctata clade of Drosophila

Abstract

Table of Contents

List of Tables

List of Figures

List of Acronyms

Acknowledgments

Chapter I: Introduction to the Drosophila immune system

Toll$

Im

d$

JNK$

Me

lan

izaK

on

$

res

po

ns

e$

JAK/S

TAT

$

Chapter II: Divergent evolution of the antimicrobial peptide

diptericin in Drosophila