Transcriptome Analysis and Identification of Genes Involved in Moth Sex Pheromone Biosynthetic Pathways

Iowa State University Capstones, Theses and Graduate Theses and Dissertations Dissertations

2019

Transcriptome analysis and identification of genes involved in moth sex pheromone biosynthetic pathways

Xiaoyi Dou Iowa State University

Follow this and additional works at: https://lib.dr.iastate.edu/etd

Part of the Entomology Commons

Recommended Citation Dou, Xiaoyi, "Transcriptome analysis and identification of genes involved in moth sex pheromone biosynthetic pathways" (2019). Graduate Theses and Dissertations. 17671. https://lib.dr.iastate.edu/etd/17671

This Dissertation is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital Repository. It has been accepted for inclusion in Graduate Theses and Dissertations by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected].

Transcriptome analysis and identification of genes involved in moth sex pheromone biosynthetic pathways

Xiaoyi Dou

A dissertation submitted to the graduate faculty

in partial fulfillment of the requirements for the degree of

DOCTOR OF PHILOSOPHY

Major: Entomology

Program of Study Committee: Russell A. Jurenka, Major Professor Ryan C. Smith Joel R. Coats Amy L. Toth Hua Bai

The student author, whose presentation of the scholarship herein was approved by the program of study committee, is solely responsible for the content of this dissertation. The Graduate College will ensure this dissertation is globally accessible and will not permit alterations after a degree is conferred

Iowa State University

Ames, Iowa

2019

DEDICATION This study is dedicated to my family who have been my source of inspiration and give me strength. To my friends who give advice and encouragement to this study, thanks for your support.

iii

TABLE OF CONTENTS

Page LIST OF FIGURES ...... vi LIST OF TABLES ...... viii ACKNOWLEDGMENTS ...... ix ABSTRACT ...... xii CHAPTER 1. INTRODUCTION ...... 1 Moths ...... 1 Sex Pheromone ...... 3 Sex Pheromone Biosynthesis and Regulation ...... 5 Control Strategies using Sex Pheromones ...... 9 Dissertation Organization ...... 11 References ...... 13 CHAPTER 2. TRANSCRIPTOME COMPARISON OF PHEROMONE GLAND- OVIPOSITOR AND TARSI IN THE CORN EARWORM MOTH, Helicoverpa zea ...... 20 Abstract ...... 20 Introduction ...... 21 Materials and Methods ...... 23 Insects and tissue collections ...... 23 RNA isolation, cDNA library construction and Illumina sequencing ...... 23 Assembly of short reads and gene annotation ...... 24 Differential expression analysis ...... 25 Identification of candidate genes involved in moth sex pheromone biosynthesis and chemosensation ...... 25 Phylogenetic analysis ...... 26 Results ...... 26 Illumina sequencing, sequence assembly and gene annotation ...... 26 Differentially expressed genes ...... 27 Candidate genes involved in sex pheromone biosynthesis ...... 28 Candidate genes involved in chemosensation and signal transduction ...... 31 Discussion ...... 33 Acknowledgements ...... 37 References ...... 37 CHAPTER 3. IDENTIFICATION OF FATTY ACYL REDUCTASE IN PHEROMONE GLAND AND TARSI OF THE CORN EARWORM, Helicoverpa zea ...... 53 Abstract ...... 53 Introduction ...... 54 Materials and Methods ...... 55 iv

Insects ...... 55 Chemicals ...... 56 Cloning of the fatty acyl-CoA reductases ...... 56 RT-PCR ...... 57 Functional Assay ...... 57 RNAi ...... 58 Quantitative real-time PCR ...... 59 Phylogenetic construction ...... 60 Results ...... 60 Fatty acyl-CoA reductases cloning ...... 60 RT-PCR indicates the three fatty acyl-CoA reductases are present in tarsi ...... 61 Yeast expression and functional assay indicates fatty acyl-CoA reductases 1 produces fatty alcohols ...... 62 RNAi indicates that FAR1 is involved in 16: Ald production in male tarsi ...... 62 Discussion ...... 63 Acknowledgments ...... 65 References ...... 65 CHAPTER 4. COMPARISON OF PINK BOLLWORM PHEROMONE GLAND TRANSCRIPTOME IN TWO POPULATIONS: LABORATORY AND FIELD ...... 77 Abstract ...... 77 Introduction ...... 78 Materials and Methods ...... 79 Insect collection and pheromone gland extraction and analysis ...... 79 RNA isolation, cDNA library construction and Illumina sequencing ...... 80 De novo assembly of short reads and gene annotation ...... 81 Expression abundance analysis ...... 81 Identification of candidate genes involved in sex pheromone biosynthesis ...... 82 Relative expression of several candidate genes by qPCR ...... 82 Phylogenetic analysis ...... 82 Statistics ...... 83 Results ...... 83 Illumina sequencing and sequence assembly ...... 83 Differential expressed gene analysis ...... 84 Putative genes related to sex pheromone biosynthesis ...... 85 Identification of putative genes related to receptors ...... 91 Identification of putative carrier proteins ...... 92 Discussion ...... 93 Acknowledgements ...... 97 References ...... 97

CHAPTER 5. IDENTIFICATION OF RNA VIRUSES FROM THE TRANSCRIPTOME OF PHEROMONE GLAND IN THE PINK BOLLWORM MOTH ...... 114 Abstract ...... 114 Introduction ...... 115 Materials and Methods ...... 117 Insect collection and Pheromone gland extraction ...... 117 RNA isolation and Illumina sequencing ...... 118 De novo Assembly of Short Reads and Gene Annotation ...... 118 Sequence and Phylogenetic analysis ...... 118 Results ...... 119 Viruses found in the transcriptome ...... 119 Positive-sense single-strand viruses ...... 120 Negative-sense single-strand RNA viruses ...... 122 Discussion ...... 123 Acknowledgements ...... 125 References ...... 125 CHAPTER 6. IDENTIFICATION OF AN ACETYLTRANSFERASE IN CABBAGE LOOPER ...... 138 Abstract ...... 138 Introduction ...... 138 Materials and Methods ...... 141 Insect ...... 141 Cloning of the FAT ...... 141 Tissue specificity studies by qPCR ...... 142 Functional Assay in insect and yeast cells ...... 143 RNAi ...... 144 Results ...... 145 Cloning of biosynthetic FAT homolog candidates ...... 145 Tissue distribution of FAT ...... 146 Functional expression in insect and yeast cells ...... 147 RNAi ...... 147 Discussion ...... 147 References ...... 151 CHAPTER 7. CONCLUSIONS ...... 158 APPENDIX A. ADDITIONAL FIGURES AND TABLES FOR CHAPTER 2 ...... 164 APPENDIX B. ADDITIONAL FIGURES AND TABLES FOR CHAPTER 4 ...... 177 APPENDIX C. ADDITIONAL FIGURES AND TABLES FOR CHAPTER 5 ...... 185

LIST OF FIGURES

Page

Figure 2. 1. Volcano plots for differentially expressed genes between tissues...... 46

Figure 2. 2. Number of differentially regulated genes in different tissues, grouped by gene ontology ...... 47 Figure 2. 3. Comparison of expression level of candidate genes involved in sex pheromone biosynthesis between tissues based on FPKM values...... 48

Figure 2. 4. Phylogenetic analysis of identified desaturases...... 49

Figure 2. 5. Phylogenetic analysis of identified fatty acyl-CoA reductase...... 50

Figure 2. 6. Phylogenetic analysis of gustatory receptor...... 51

Figure 2. 7. Heat map of selected genes related to sex pheromone production and signal transduction based on gene expression...... 52

Figure 3. 1. Multiple amino acid sequence alignments of HzeaFAR1, HzeaFAR4, HzeaFAR5 with other funnctional FARs from different moths ……………………………….. 67

Figure 3. 2. RT-PCR of three HzeaFARs in female PGs, female tarsi and male tarsi...... 71

Figure 3. 3. Phylogenetic tree of the fatty acyl reductase from different Lepidoptera species .... 72

Figure 3. 4. GC-MS analysis of HzeaFAR1 expressed in yeast cells without additional substrates...... 73

Figure 3. 5. GC-MS result of HzeaFAR1 expressed in yeast cells with various substrates ...... 74

Figure 3. 6. RNAi knockdown of FAR1 in pheromone gland ...... 75

Figure 3. 7. RNAi knockdown of FAR1 in male tarsi ...... 76

Figure 4. 1. Pink bollworm sex pheromone biosynthetic pathway …………………………… 102

Figure 4. 2. Number of differentially regulated genes in each population, grouped by gene ontology ...... 108

Figure 4. 3. Sequence alignments of DES2, DES6 and DES8 with other desaturases from Lepidoptera...... 110

Figure 4. 4. Relative expression of selected genes between Lab and Field population ...... 111 vii

Figure 4. 5. Phylogenetic relationship of DESs from Lepidoptera constructed using amino acid sequences as described in Experimental Methods ...... 112

Figure 4. 6. Phylogenetic relationship of FARs from Lepidoptera ...... 113

Figure 5. 1. Genome organization and amino acid alignment of PBWV1 and PBWV4 ………125

Figure 5. 2. Phylogenetic tree of PBWV1 and PBWV4 with other iflavirus ...... 132

Figure 5. 3. Genome organization and amino acid alignment of PBWV2 and PBWV3 ...... 133

Figure 5. 4. Phylogenetic tree of PBWV2 and PBWV3 with other (-ss) RNA...... 137

Figure 6. 1. Transcript expression of TniFAT in tissues from T. ni …………………………... 155

Figure 6. 2. Florencence comparison of the EGFP and FAT expression in insect Sf9cells ...... 156

Figure 6. 3. RNAi treatment of pheromone glands...... 157

viii

LIST OF TABLES

Page

Table 2. 1. Assembly results ...... 43

Table 2. 2. Top 20 differentially expressed genes between female PG and female tarsi ...... 44

Table 2. 3. Top 20 differentially expressed genes between male tarsi and female tarsi ...... 45

Table 3. 1. Primers used in this study ……………………………………………………...... 68

Table 3. 2. Percent ratio of alcohols produced in the yeast assay with and without added substrates ...... 69

Table 4. 1. Ratios and amounts of pheromone found in glands collected from the two populations …...…………………………………………………………………….102

Table 4. 2. Analysis of sequencing results...... 102

Table 4. 3. Putative biosynthesis related genes in PBW pheromone glands and the first BLASTp hit in GenBank …………………………………………………………. 103

Table 4. 4. Comparison of candidate transcripts involved in the sex pheromone biosynthetic pathway...... 105

Table 5. 1. Number of viruses found in the transcriptome of pheromone gland in two populations …...…………………………………………………………………… 129

Table 5. 2. Comparison of ssRNA viruses between two populations of the PBW ...... 129

ACKNOWLEDGMENTS

The first person I would like to thank is my PI Dr. Russell Jurenka, who is a very knowledgeable, outstanding and kind mentor. You always take a lot of time to discuss my work and give me the right direction when I face problems. I was very impressed that when I set about learning bioinformatics, you encouraged me and ordered the server for me. Most importantly is that you proposed some ideas to let me practice my bioinformatics. Also, you always revise my manuscripts for publication which helps developed my writing ability. Besides that, you are also a professional teacher. The course insect physiology was difficult for me at first, but with your supervision, I learned a lot. Not only the research under your supervision but also my teaching experience. Thanks for your trust in my teaching abilities. I am really thankful for the word ’Your Ph.D. is yours’, that develops the ability of independently thinking, troubleshooting, and time organization. I am extremely honored to be your intellectual descendant and enrolled in your lab.

I wish to thank the members of my Program of Study committee. I participate in Dr. Ryan Smith’s lab meeting for over one year and presented my research several times. You are a great mentor for my presentation and my research. With your help, I learned many things about mosquito immunity which gives me a new sight in my future work. Dr. Joel Coats taught me a lot of bioassays and the applications of insecticide in the course of insect toxicology. Dr. Hua Bai helped me realize my strength in job hunting and my future post-doc position is really related to my strength. Dr. Amy Toth gave suggestions on my bioinformatics study to help me develop a new perspective for my research. All members gave me a lot of advice for my research, study, and life. Thank all of you so much.

Since I have a minor in bioinformatics, I enrolled in many computer-related courses. Thank you, Dr.

Sijun Liu, the research scientist in Dr. Bonning’s lab. He is the mentor for my bioinformatics study. Thanks for your patience and kindness, teaching me the basic skills in bioinformatics and helping in the development of my projects. Thanks for the project about the identification of viruses. You pointed out the drawbacks and helped me revise the manuscript. You always give me your experience in research and life x

in the United States. Also, thanks for teachers in my bioinformatics classes, Dr. Carolyn Lawrence-Dill,

Dr. Iddo Friedberg, Dr. Xiaoqiu Huang, Dr. Julie Dickerson, and my friend Vishnu and Hamid. You guys are the first bioinformatics persons I met when I started my bioinformatics. Thanks for your kindness, experience, and knowledge to make me confident in my studies.

I need to thank the current postdoc Hyeog-Sun Kwon in Dr. Smith’s lab and previous postdoc Yuting

Chen in Dr. Bonning’s lab. Thanks for your patience, ability, and knowledge. You told me your experiences in Ph.D. study and help me troubleshoot when I had problems. You always let me use your device or kit when it was an emergency. Thanks for your feedback on my research problems that led to the finish of the dissertation. Also, thank you, Rebekah Reynolds, the first and best American friend. You help me develop my English and give me your opinions as an American. You are the person that made me feel that I am not a foreigner in the United States.

Thanks to all members of the Department of Entomology. You make the department feel like a big family. I always feel free asking for help. Thanks to Kelly Kyle who helped me figure out the university policy and solve many problems in my campus life. Thanks for the Entomology graduate student organization. I have made innumerable and indelible friendships here, and I really enjoyed the events and food in our department.

I could not have survived the Ph.D. without the love and support of my family. Thanks to my parents.

Although you are living in China far away from here, you always support me and let me take care of myself.

It was always a relaxing time talking with you. I extend my gratitude to my wife Ling Xiang. Thank you for being in the USA for me, both in the good times as well as the bad times. You sacrifice your own time to accomplish with me here just let me know I am not alone. Thanks for encouraging me even at the hardest time in my Ph.D. study. You always believe in me and help me to be better. Also, thanks for giving me a little daughter Mandy in my life. To my daughter: you are the cutest one I have ever seen. Your smile is impressive and always cheer me up in the long and stressful days. I am greatly looking forward to watching xi

you grow up and cannot wait for the future that you will be. You and your mom are the power and impetus to make me progress. xii

ABSTRACT

Sex pheromones play important roles in chemical communication, especially in the mating behavior of moths. Usually, sex pheromones are a blend of several compounds, mostly fatty acid derivatives, with C10-C18 carbon chain, with 0-4 double bonds, and an oxygenated functional group that could be alcohol, aldehyde, or acetate ester. The biosynthesis has been described and enzymes involved in the pathway include: the introduction of double bonds by fatty acyl desaturases, limited chain shortening by β-oxidation enzymes, functional group modification by fatty acyl-CoA reductases, fatty alcohol acetyltransferases and alcohol oxidases. Some genes have been functionally characterized in various moth species, but the genes encoding fatty alcohol acetyltransferases and alcohol oxidases have not been identified. I conducted the first study involving the identification of genes encoding fatty acyl-CoA reductase in Helicoverpa zea and fatty alcohol acetyltransferase in Trichoplusia ni. We screened the genes from the transcriptome of pheromone gland and tarsi in H. zea and designed experiments that included RT-PCR, qPCR, gene expression and functional assay in vitro, and RNAi in vivo to investigate the function. One gene encoding fatty acyl-CoA reductase was found involved in sex pheromone biosynthesis in pheromone gland and hexadecanal production in tarsi. In addition, this gene is distributed in tarsi as well, but is different from the pheromone gland specific FARs from other moths. The candidate acetyltransferase was selected due to homology with wax synthase. Unfortunately, it was present in all tissues and we did not find the function of acetylation to produce fatty acetate esters. More candidate genes need to be investigated and future work needs to be conducted to identify the acetyltransferase.

Mating disruption is a successful approach to control the world wide pest insect the pink bollworm moth utilizing the artificial sex pheromone, since it is environmentally friendly and xiii

effective. However, in Israel, the mating disruption technique is failing in some fields likely due to females producing a different ratio of sex pheromone components. We compared the transcriptome of pheromone gland from two populations, the lab which is never exposed to mating disruption and field populations which was exposed to mating disruption. We found some genes encoding desaturases, fatty acyl-CoA reductases, and putative acetyltransferases that were highly expressed in both populations, indicating they may be involved in the sex pheromone biosynthesis.

However, none of them showed significant difference in abundance except two acetyltransferases.

More research on acetyltransferases is needed to identify the ones involved in pheromone biosynthesis.

Also, we utilized the next generation sequencing technology to identify viruses. We used the transcriptome data from the pink bollworm and found some new single-stranded negative sense and positive sense RNA viruses. One iflavirus was present in pink bollworm from Israel and the

USA in different life stages.

The research presented in this dissertation provides significant results using bioinformatics and molecular techniques to help understand pheromone biosynthesis in moths. It presents two transcriptome studies on the pheromone glands of two moths that produce acetate esters and aldehydes as the main pheromone components. It also presents an investigation into a fatty acyl-

CoA reductase in the pheromone gland and tarsi both of which produce aldehyde pheromones.

This research will provide deeper understanding on the molecular mechanism of sex pheromone biosynthesis.

CHAPTER 1. INTRODUCTION

Moths

Lepidoptera is one of the most widely spread and recognizable insect orders in the world. It is the second largest order, just smaller than Coleoptera. Lepidoptera comprises both moths and butterflies. Most lepidopteran insects are moths, and it is thought to have 160,000 species of moth (Smithsonian

Institution, 2012). Some lepidopterans are useful in commerce, such as the silkworm, Bombyx mori. Some moths, especially the larvae (caterpillars), are recognized as major agricultural pests around the world, such as the gypsy moth, Lymantria dispar (Linnaeus), causing damages to forests, the diamondback moth, Plutella xylostella (Plutellidae), a global pest of crucifer crops.

Three moth species used in our research were the corn earworm, Helicoverpa zea (Noctuidae), the pink bollworm, Pectinophora gossypiella (Gelechiidae), and the cabbage looper, Trichoplusia ni

(Noctuidae).

Heliothine moths (Noctuidae) contain some species which are among the most devastating pests in the world. Approximately $3 billion to $7 billion was used in the control of the most prevalent agricultural pest species per annum. 70% of the known host associations are oligophagous, while the remaining are polyphagous (Cho et al., 2008). In North American, there are 148 heliothine species described in 14 genera (Knudson et al., 2003). H. zea and Heliothis virescens are the major heliothine pests in North and South America (Bergvinson, 2005), and these moths were recorded feeding on 235 plant species, including corn, tomato, artichoke, asparagus, cabbage etc. The corn earworm, H. zea is considered to be the most costly crop pest in North

America. Many strategies have been developed to control H. zea. Natural enemies are ineffective in causing corn earworm mortality (Archer and Bynum, 1994). Entomopathogenic nematodes suppress the development of larvae when they are applied to corn silk (Purcell et al., 1992).The 2

most widely and effectively used method is the transgenic plant with Bacillus thuringiensis (BT) toxin (Delannay et al., 1989). However, it has been reported that the corn earworm evolved resistance to Bt toxin expressed in the transgenic sweet corn (Dively et al., 2016). So new control strategies need to be considered.

The pink bollworm (PBW) (Gelechiidae) is a worldwide pest of cotton. It is hard to control since the larvae feed inside flower buds and bolls, where they are protected from the traditional insecticides. In Arizona, Bt cotton showed a long-term of suppression to pink bollworm through a

10 years study in 15 regions (Carrière et al., 2003). However, in India, it was reported that the pink bollworm developed the resistance to transgenic Bt cotton which expresses a single Bt toxin

(Bagla, 2010). Mating disruption, which utilizes artificial sex pheromone was also applied in many areas. It was successful in management of this pest in Arizona, Egypt, Israel and other areas for decades. However, recently it was reported that mating disruption has failed in many cotton fields of Israel and could be due to a change in the pheromone produced by females. In order to begin to figure out how females have changed their pheromone, we compared the transcriptome between a lab population and a population originating from a field in which mating disruption failed.

The cabbage looper (Noctuidae), widely distributed throughout North America, is a serious pest of cruciferous crops, such as broccoli, brussel sprouts, cauliflower, and cabbage. Many natural enemies have been described with various effectiveness as control agents, such as Voria ruralis

(Diptera), Copidosoma truncatellum (Hymenoptera), Trichoplusia ni NPV and an unknown fungi.

Like the two moth species described above, transgenic plants expressing Bt toxin have long been used as a method to control the cabbage looper. However, the cabbage looper has also developed

Bt toxin resistance in the agricultural field (Janmatt and Myers, 2003). 3

Taken altogether, these three moths have unique and various features. Since they are hard to control and they have developed resistance to some control tactics, new approaches need to be considered. As far as we know, mating disruption using sex pheromones is a successful method to control some moths in the long term. A deep understanding of the sex pheromones and their biosynthesis would be important to the application of mating disruption in the field. These moths utilize different sex pheromone components, such as aldehydes in H. zea., fatty acetate esters with two double bonds in PBW and with one double bond in T. ni. Their sex pheromone biosynthetic pathways are similar with slightly different enzymes involved. All the genes involved in the pathway have not been identified totally from these species. So my studies focused on the sex pheromone biosynthetic pathways.

Sex Pheromone

A pheromone is a chemical factor utilized by diverse organisms in members of the same species for communication. Several types of pheromone have been recognized based on the behavior or physiology impact, such as alarm pheromone, sex pheromone, aggregation pheromone, etc. Sex pheromones are released to attract the opposite sex to mate. Even some organisms mimic this chemical so as to lure prey. For example, predatory bolas spiders emit the same sex pheromone signals to ensnare the moths (Gemeno et al., 2000). Mate finding behavior is dependent in almost all moth lineages. However, numerous studies have shown moth pheromone variations between and within individuals (Allison and Cardé, 2008), populations (Löfstedt et al., 1986) and species

(Byers, 2006). These variations are caused by the methodological, environmental, and geographic effects. The first sex pheromone was identified in silkworm moth, Bombyx mori, as (E, Z) -10, 12 hexadecadien-1-ol, bombykol (Butenandt et al., 1959). Since then, sex pheromones from more 4

than 600 species and about 23 superfamilies of Lepidoptera have been identified

((http://web.tuat.ac.jp/~antetsu/LepiPheroList.htm); http://www.pherobase.com). However, the exception are butterflies (super-family: Papilionioidea), which do not produce sex pheromones.

Fit to their diurnal habits, butterflies evolved a different mating system, based on visual signals.

The site of pheromone production varies among insects. In the house fly, a sex pheromone,

(Z)-9-tricosene had been isolated from the female cuticle and feces (Carlson, et al., 1971). In

Coleoptera, the site is usually located in the abdomen (Tillman et al., 1999). In moths, the biosynthesis of sex pheromone occurs in the pheromone glands (PGs) which are located between the 8th and 9th abdominal segments of the female, except for Theresimima ampelophaga

(Zagaenidae), where the gland is located on the dorsal part of abdominal segments 3 to 5 (Jurenka,

2004). Also, in some moths which utilize hydrocarbons as sex pheromone, they produce the molecules in oenocytes when are then transported by lipophorin to the PG for release into the environment.

Usually, a female releases a sex pheromone that consists of a mixture of compounds with a specific ratio to attract conspecific males. Based on the structure of these compounds, a sex pheromone could be classified as type I, type II and miscellaneous type (Löfstedt et al., 2016).

Type I compounds are mostly fatty acid derivatives, with C10-C18 carbon chain, with 0-4 double binds, and oxygenated functional group, including alcohols, aldehydes, and acetate esters. Type II compounds usually consist of hydrocarbons and their epoxides. The hydrocarbons are produced in oenocytes and then transported to the PG where they are epoxidized (Jurenka, 2004).

In general, female moths will not release sex pheromone until they are reproductively competent. Males can detect the pheromone of a female over considerable distance-from tens to 5

perhaps hundreds of meters upwind, and then the male navigates a course along the pheromone plume to her side. Sex pheromones released by adult female insects are detected by narrowly tuned olfactory neurons in the conspecific male antennae. Once the olfactory neuron detects the pheromone, an action potential is generated and travels down the axon through the antenna to the antennal lobe of the brain. In the antennal lobe, there are several glomeruli which are areas of neuropile that are ball-like. These glomeruli are areas that have first-order processing of the neural signal coming from the sensilla on the antennae (Nakagawa et al., 2005). The whole olfactory system is dependent on the receptors expressed on the olfactory sensory neurons. The major proteins involved in the reception of odorants are odorant-binding protein (OBP), chemosensory protein (CSP), odorant receptors (ORs), ionotropic receptors (IRs), sensory neuron membrane protein (SNMPs), and odorant-degrading enzymes (ODEs) (Leal et al., 2013). In insects, the odorant-gated ion channel complex is formed by an OR and co-receptor (Orco) to detect odorants.

To date, ORs specific for sex pheromone components have been identified as pheromone receptors in various moths (Sakurai et al., 2004; Miura et al., 2010; Wang et al., 2011; Xu et al., 2014). The molecules of pheromone are bound and solubilized by OBP and activate membrane-bound ORs to have responses. The molecules will be degradated by ODEs, particularly antennae-specific esterases (Durand et al., 2011; Ishida and Leal, 2005).

Sex Pheromone Biosynthesis and Regulation

The biosynthesis of sex pheromone has been study extensively and well documented in many moth species (Blomquist et al., 2005; Tillman et al., 1999). The biosynthesis is different between type I and type II sex pheromone compounds. Type II sex pheromone components are derived from linoleic and linolenic acid which are obtained in the diet. The acids are then modified through 6

various enzymatic reactions including chain elongation, decarboxylation and epoxidation (Jurenka,

2004). The type I sex pheromone component biosynthesis follows a circadian rhythm with little production during the day and increases during the calling period.

The precursors for the biosynthesis of type I sex pheromone components are saturated fatty acid, such as palmitic acid (16 carbons) and stearic acid (18 carbons) which are produced through fatty acid synthesis with the enzymes acetyl-CoA carboxylase (ACC) and fatty acid synthase

(FAS). These enzymes are similar in most organisms. The fatty acid precursors are modified by several enzymes to produce the final compounds, such as the introduction of double bonds by fatty acyl desaturases (DESs), limited chain shortening by β-oxidation enzymes, functional group modification by fatty acyl reductases (FARs), fatty alcohol acetyltransferases (FATs) and alcohol oxidases (Blomquist and Richard., 2011; Jurenka et al., 2004; Rafaeli et al., 2005; Tillman et al.,

1999;).

Desaturases are key enzymes in the biosynthesis of unsaturated fatty acid. They are categorized into two groups, acyl-acyl carrier protein mainly found in plastids of higher plants and membrane-bound fatty acid desaturase primarily in eukaryotes and bacteria. There are three conserved histidine box regions in the membrane-bound desaturase (Zeng et al., 2019). Insect desaturases play important roles in biosynthesis and perception of semiochemicals (Bousquet et al., 2012), cold tolerance (Kayukawa et al., 2007), feeding behavior (Wang et al., 2016), larval development (Köhler et al., 2009) and defensive fatty acids production (Haritos et al., 2012). There are variety of desaturases have been described that are involved in the sex pheromone biosynthesis in moth, including: ∆5 (Foster et al., 1996), ∆6 (Albre et al., 2012; Wang et al., 2010), ∆9 (Lienard et al, 2008; Rodriguez et al, 2004), ∆10 (Hao et al, 2002), ∆11 (Fujii et al., 2015; Knipple et al.,

1998; Lienard and Löfstedt, 2010; Rosenfield et al., 2001), and ∆14 (Bucek et al., 2015). After 7

the unsaturated fatty acid produced, the precursors are subject to limited chain shortening by β- oxidation, resulting in different chain length precursors. The limited chain shortening by β- oxidation contains several enzymes, including acyl-CoA oxidase, enoyl-CoA hydratase, L-3- hydroxyacyl-CoA dehydrogenase and thiolase.

Specific chain length precursors are reduced to the corresponding alcohols by FARs which have been widely identified in a variety of organisms. In humans and mice, FARs are involved in the production of lipid components in sebum, mainly catalyzing the fatty acyl-CoA precursors into fatty alcohols by the NADPH-dependent reduction (Honsho et al., 2010). In plants, FARs act in epicuticular wax synthesis (Rowland et al., 2006), and storage of wax esters in seeds (Metz et al.,

2000). In moth, the first FAR was identified in B. mori (Moto et al., 2003) and subsequently in

Yponomeuta spp. (Lienard et al., 2010), Ostrinia spp. (Lassance, et al 2010; 2013), Helicoverpa spp. and Heliothis spp. (Hagstrom et al., 2012), and Spodoptera spp (Antony et al., 2016; Carot-

Sans et al., 2015). Some pheromone gland specific FARs reduce multiple saturated or unsaturated fatty acids to corresponding fatty alcohol in sex pheromone biosynthesis. The last step for many moths utilizing acetate ester as sex pheromone is the acetylation of fatty alcohol by acetyl-CoA: fatty alcohol acetyltransferases. In yeast, ATF1 acetylates the fatty alcohol with different chain lengths with various degrees of efficiency. A plant-derived diacylglycerol acetyltransferase also transformed fatty alcohol sex pheromone compounds to acetate esters (Ding et al., 2016). However, no gene encoding an enzyme involved in this reaction has been identified in any insect species.

Similarily, the gene encoding an alcohol oxidase has not been described at the molecular level. A cuticular oxidase has been described from the pheromone glands of some moths utilizing alcohols and aldehydes (Teal and Tumlison, 1988; Fang et al., 1995). A salicyl alcohol oxidase has been identified that catalyzes the formation of salicylaldehyde in chrysomelid leaf beetles (Michalski et 8

al., 2008), but it is not known if it is involved in formation of aldehyde pheromones. The key genes involved in pheromone biosynthesis have been analyzed in several pheromone gland transcriptome studies (Antony et al., 2015; Cha et al., 2017; Ding and Löfstedt, 2015; Gu et al., 2013; Li et al.,

2015).

Pheromone production is regulated by the neuroendocrine system in many insects. In cockroaches, beetles and flies, juvenile hormone and ecdysone are the important factors controlling pheromone production. In most moths, pheromone biosynthesis is regulated by a neuropeptide pheromone biosynthesis activating neuropeptide (PBAN) which is released from the subesophageal ganglion, except in the cabbage looper (Tang et al., 1989). The first PBAN found in moths was in H. zea and has 33 amino acids and a core sequence FXPRLamide at the C- terminus that is required for activity (Raina et al., 1987). The function of PBAN-like neuropeptides varies amongst the insects. In the cockroach, Leucophaea maderae, leucopyrokinin, the first member of this family stimulates hindgut contraction (Holman et al., 1986). In white shrimp,

Penaeus vannamei, it induces myotropic activity (Torfs et al., 2001). In some moths, PBAN also has some other functions, such as the induction of embryonic diapause in B. mori (Imai et al., 1991) and melanization in larvae (Matsumoto et al., 1990).

PBAN is released from the subesophageal ganglion and binds to a G-protein coupled receptor

(GPCR) on the pheromone gland, causing signal transduction that stimulates pheromone biosynthesis. Ca2+ and cyclic-AMP are used as secondary messengers in the majority of species.

The influx of Ca2+ is mediated by a store-operated channel (SOC) in B. mori (Hull et al., 2007;

2009). It was shown that the influx of Ca2+ activates both fatty acyl reductase and lipases. In heliothine moths, calcineurin, protein kinase A (PKA) and acetyl-CoA carboxylase (ACC) are key components involved in the signal transduction. The calcium activates an adenylate cyclase to 9

produce cAMP. ACC is activated by the combined action of calcium activating calcineurin (PP2B) that dephosphorylates ACC and by cAMP activating a PKA which inhibits an AMPK that phosphorylates ACC (Du et al., 2017, Jurenka, 2017). However, it is unknown if the PBAN signal in H.zea is mediated by SOC, as found in B. mori.

Control Strategies using Sex Pheromones

Invertebrates cause significant agricultural losses worldwide. Crop losses caused by pests may be the equivalent of food required for over a billion people. Synthetic pesticides have been used widely but have many apparent challenges. Many approaches have been developed to complement, reduce and replace the synthetic pesticide applications.

The sterile insect technique (SIT), which releases sterile males to induce wild female sterility, causes the pest population to decline. There are no regulatory requirements for this technique, and it could combine with other integrated pest management approaches as well. The SIT combined with the Wolbachia-infected mosquito strain is considering as an effective method to control mosquito populations (Zhang et al., 2016). There is another method utilizing cry toxins, produced by the bacteria pathogen Bacillus thuringiensis. It is widely used in transgenic crops for efficient pest control. The Bt toxin is activated in the midgut, and then binds to receptors resulting in the formation of pores in midgut cells, destroying the cells and killing the larvae (Soberón et al., 2009).

However, pest resistance has developed because of the evolution of insects. In order to delay pest adaptation to Bt crops, several new approaches have been used. One is producing two or more Bt toxins against the same pest, and the other one is randomly planting Bt seeds and non-Bt seeds of the same crop in fields to form a refuge. 10

Since problems with resistance, non-target effects, and contamination have developed with insecticides, a direct pheromone-based pest management tactic was developed, including mass trapping, attract-and-kill formulations and mating disruption. Broadcasting sex pheromone in fields in order to disrupt reproduction was started in the early 1970s. It still remains an effective method of control since sex pheromones are one of the most potent stimuli known in terms of sensory reception and processing, and behavioral responses are precise and fast. Also, this method utilizing natural products for pest control is an environmentally friendly manner and could reduce the side effects of insecticides. Mating disruption products have been developed for more than 20 insect species and are applied to more than 750,000 hectares worldwide (Miller and Gut, 2015). In

USA, the first registered product was for the pink bollworm, Pectinophora gossypiella

(Lepidoptera: Gelechiidae) in 1978 (Brooks et al., 1979). With the development of artificial sex pheromone and application in fields and the use of Bt cotton, since 2008 no PBW larvae have been found in southwestern U.S. and northern Mexico. Currently, more than 120 pheromone products are registered with the U.S. Environmental Protection Agency for 18 species. There are more than

24 products for codling moth and oriental fruit moth alone. Generally, mating disruption has minimal non-target effects and lepidopteran pheromones have low toxicity to mammals. It has proven to be an important and effective tool in large scale programs targeting pests.

Several mechanisms have been proposed to elucidate the mechanism of mating disruption by artificial sex pheromones, including peripheral reception and central nervous system effects; competition between natural and synthetic sources; camouflage of the natural plumes; imbalance in sensory input; antagonists to attraction and pheromone mimics (Bartell, 1982). To date, only one instance of mating disruption resistance has been established. It was reported that the smaller tea tortrix moth, Adoxophyes honmai, controlled by the (Z)-11-tetradecenyl acetate (Z11-14: OAc) 11

to disrupt mating was ineffective in Japan after 15 years of application (Mochizuki et al., 2002).

The possible reason for resistance is that the natural blend of sex pheromone in a background of

Z11-14: OAc was again detected by males.

Dissertation Organization

To better understand the molecular mechanisms of the sex pheromone biosynthesis, I designed the experience with transcriptomic comparison and gene identification. Throughout this dissertation, the biosynthesis in various moths were studied. I focused on the corn earworm (H. zea), the pink bollworm (P. gossypiella) and the cabbage looper (T. ni), because the biosynthetic pathway, sex pheromone production or the control method are different from each other.

In Chapter 2 and Chapter 3, I compared the transcriptome between pheromone gland and tarsi, finding lots of differential expressed genes that are related to sex pheromone biosynthesis, including the DESs, FARs, AOs, and genes related to signal transduction and chemosensation.

Also, through gene expression and functional assay, I functionally characterized a gene encoding

FAR that is involved in the production of Z11-16:Ald in female pheromone gland and 16:Ald in male tarsi. Taken together with the bioinformatics and molecular work, I propose that the production of Z11-16:Ald and 16:Ald experience the same biosynthetic pathway with the same enzymes.

Chapter 4 of this dissertation focuses on the transcriptomic comparison of pink bollworm pheromone glands between lab and field populations which have different pheromone component ratios. Since these two populations are the same species, few differentially expressioned genes 12

were found, especially the key genes involved in the biosynthetic pathway. Through the amino acid sequence alignment, I found some potential mutations in DESs and FATs in some replicates of the two populations. According to the results, I propose that the difference in component ratios could result from the differential expression of some key enzymes.

Within Chapter 5 I identified several RNA viruses in the pink bollworm. Since the pink bollworm is an invasive pest in USA and first found in Egypt, I compared the transcriptome of the pink bollworm from Israel and USA, in different life stages and tissues. I found one iflavirus that commonly exists in these two populations. Besides that, I found one RNA virus with the same conserved domains but located at a different terminus. Also, I found two negative sense single stranded RNA viruses, one belongs to the family of Bunyaviridae with three segments which is common in the member of bunyavirus, and the other one is not classified because only one segment was found. Altogether, I propose that the iflavirus exists in all pink bollworm population and in different life stages or tissues, and it should be transmitted vertically.

In Chapter 6, I attempted to identify the gene encoding FAT in the cabbage looper. The experiments I designed include: qPCR to check the expression level of FAT in different tissues; gene cloning to get the full length of FAT; RNAi to knockdown the expression of FAT; gene expression in insect and yeast cells; functional assay to figure out the function. I did not find the candidate FAT expressed specifically or very high in pheromone glands, or it is involved in the acetylation of alcohol to acetate ester or the production of wax ester. Taken altogether, I assumed

RNAi is not working efficiently in cabbage looper, and the candidate gene is possibly involved in other functions. 13

References

Albre, J., Liénard, M.A., Sirey, T.M., Schmidt, S., Tooman, L.K., Carraher, C., et al. (2012) Sex pheromone evolution is associated with differential regulation of the same desaturase gene in two genera of leafroller moths. PLoS Genet 8: E1002489.

Allison, J.D. and Cardé, R.T. (2008) Male pheromone blend preference function measured in choice and no-choice wind tunnel trials with almond moth, Cadra cautella. Anim Behav 75: 259-266.

Antony, B., Ding, B.J., Moto, K., Aldosari, S.A., Aldawood, A.S. (2016) Two fatty acyl reductases involved in moth pheromone biosynthesis. Sci Rep 6: 29927.

Antony, B., Soffan, A., Jakše, J., Alfaifi, S., Sutanto, K.D., Aldosari, S.A., Aldawood, A.S., Pain, A. (2015) Genes involved in sex pheromone biosynthesis of Ephestia cautella, an important food storage pest, are determined by transcriptome sequencing. BMC Genomics 16:532.

Archer, T.L., Bynum, Jr. E.D. (1994) Corn earworm (Lepidoptera: Noctuidae) biology on food corn on the High Plains. Environ Entomol 23: 343-348.

Bagla, P. (2010) Hardy cotton-munching pests are latest blow to GM crops. Science 327: 1439- 1439.

Bartell, R.J. (1982) Mechanisms of communication disruption by pheromone in the control of Lepidoptera: A review. Physiol Entomol 7: 353-364.

Bousquet, F., Nojima, T., Houot, B., Chauvel, I., Chaudy, S., Dupas, S., et al. (2012) Expression of a desaturase gene, DESAT1, in neural and nonneural tissues separately affects perception and emission of sex pheromones in Drosophila. Proc Natl Acad Sci USA 109: 249–254.

Brooks, T.W., Doane, C.C., Staten, R.T. (1979) Experience with the first commercial pheromone communication disruptive for suppression of an agricultural pest. In: Ritter, F.J., eds. Chemical ecology: odor communication in animals. Elsevier, Amsterdam, Netherlands, pp. 375-388.

Bucek, A., Matouskovam, P., Vogel, H., Sebesta, P., Jahn U., Weißflog, J., Svatoš, A., Pichová, I. (2015) Evolution of moth sex pheromone composition by a single amino acid substitution in a fatty acid desaturase. Proc Nat Acad Sci USA 112: 12586-12591.

Butenandt, A., Beckmann, R., Stamm, D., Hecker, E. (1959) Über den Sexuallockstoff des Seidenspinners Bombyx mori. Reindarstellung und Konstitution. Z. Naturforsh 14b: 283- 284. 14

Blomquist, G.J., Jurenka, R.A., Schal, C., Tittiger, C. (2005) Biochemistry and molecular biology of pheromone production. In: Gilber, L.I., Latrou, K., Gill, S.S., eds. Comprehensive molecular insect science. LondonL Elsevier, pp. 705-51.

Blomquist, G.J. and Richard, V. (2011) Pheromone production: biochemistry and molecular biology. In: Gilbert, L.I., eds. Insect endocrinology. London: Academic doi:10.1016/B978- 0-12-384749-2.10015-9.

Byers, J.A. (2006) Pheromone component patterns of moth evolution revealed by computer analysis of the Pherolist. J Animal Ecol 75: 399-407.

Carlson, D.A., Mayer, M.S., Silhacek, D.L., James, J.D., Beroza, M., Bierl, B.A. (1971) Sex attractant pheromone of the house fly: isolation, identification and synthesis. Science 174: 76-78.

Carot-Sans, G., Munoz, L., Piulachs MD, Piulachs, M.D., Guerrero, A., Rosell, G. (2014) Identification and characterization of a fatty acyl reductase from a Spodoptera littoralis female gland involved in pheromone biosynthesis. Insect Mol Biol 24: 82-92.

Carrière, Y., Ellers-Kirk, C., Sisterson, M., Antilla, L., Whitlow, M., Dennehy, T.J. (2003) Long- term regional suppression of pink bollworm by Bacillus thuringiensis cotton. Proc Natl Acad Sci USA 100: 1519-1523.

Cha, H.W., Kim, W., Jung, J.K., Lee, D.W. (2017) Putative pheromone biosynthesis pathway in Maruca vitrata by transcriptomic analysis. J Asia Pac Entomol 20: 165-173.

Delannay, X., LaVallee, B.J., Proksch, R.K., Fuchs, R.L., Sims, S.R., Greenplate, P.G., et al. (1989) Field performance of transgenic tomato plants expressing the Bacillus thuringiensis var. kurstaki insect control protein. Biotechnology 7: 1265-1269.

Ding, B-J., Lager, I., Bansal, S., Durrett, T.P., Stymne, S., Löfstedt, C. (2016) The yeast ATF1 acetyltransferase efficiently acetylates insect pheromone alcohols: implications for the biological production of moth pheromones. Lipids 51: 469-475.

Ding, B.J. and Löfstedt C. (2015). Analysis of the Agrotis segetum pheromone gland transcriptome in the light of sex pheromone biosynthesis. BMC Genomic 16:711.

Dively, G.P., Venugopal, P.D., Finkenbinder, C. (2016) Field-evolved resistance in corn earworm to Cry proteins expressed by transgenic sweet corn. PLoS One 12: e0183637.

Du, M., Liu, X., Ma, N., Liu, X., Wei, J., Yin, X., Zhou, S. et al. (2017) Calcineurin- mediated dephosphorylation of acetyl-coA carboxylase is required for pheromone biosynthesis activating neuropeptide (PBAN)-induced sex pheromone biosynthesis in Helicoverpa armigera. Mol Cell Proteomics 16: 2138-2152. 15

Durand, N., Carot-Sans, G., Bozzolan, F., Rosell, G., Siaussat, D., et al. (2011) Degradation of pheromone and plant volatile components by a same odorant-degrading enzyme in the cotton leafworm, Spodoptera littoralis. PLoS ONE 6: e29147.

Fang, N., Teal, P.E.A and Tumlinson, J.H. (1995) Characterization of oxidase(s) associated with the sex pheromone gland in Manduca sexta (L.) females. Arch Insect biochem Physiol 29: 243-257.

Foster, S.P. and Roelofs, W.L. (1996) Sex pheromone biosynthesis in the tortricid moth, Ctenopseustis herana (Felder and Rogenhofer). Arch Insect Biochem Physiol 33: 135–147.

Fujii, T., Yasukochi, Y., Rong, Y., Matsuo, T., Ishikawa, Y. (2015) Multiple Delta11-desaturase genes selectively used for sex pheromone biosynthesis are conserved in Ostrinia moth genomes. Insect Biochem Mol Biol 61: 62-68.

Gemeno, C., Yeargan, K.V., Haynes, K.F. (2005) Aggressive chemical mimicry by the bolas spider Matophora hutchinsoni: identification and quantification of a major prey’s sex pheromone components in the spider’s volatile emissions. J Chem Ecol 26: 1235-1243.

Hagstrom, A.K., Lienard, M.A., Groot, A.T., Hedenstrom, E., Löfstedt, C. (2012) Semi-selective Fatty acyl reductases from four Heliothine moths influence the specific pheromone composition. PLoS ONE 7: e37230.

Hao, G., Liu, W., O’Connor, M., Roelofs, W.L. (2002) Acyl-CoA Z9 and Z10-desaturase genes from a New Zealand leafroller moth species, Planotortrix octo. Insect Biochem Mol Biol 32: 961–966.

Haritos, V.S., Horne, I., Damcevski, K., Glover, K., Gibb, N.; Okada, S., et al. (2012) The convergent evolution of defensive polyacetylenic fatty acid biosynthesis genes in soldier beetles. Nat Commun 3: 1150.

Honsho, M., Asaoku, S., Fujiki, Y. (2010) Posttranslational regulation of fatty acyl-CoA reductase 1, FAR1, controls ether glycerophospholipid synthesis. J Biol Chem 285: 8537-8542.

Holman, G.M., Cook, B.J., Nachman, R.J. (1986) Primary structure and synthesis of a blocked myotropic neuropeptide isolated from the cockroach, Leucophaea maderae. Comp Biochem Physiol 85: 219-224.

Hull, J.J., Kajigaya, R., Imai, K., Matsumoto, S. (2007) Sex pheromone production in the silkworm, Bombyx mori, is mediated by store-operated Ca2+ channels. Biosci, Biotechnol Biochem 71: 1993-2001. 16

Hull, J.J., Lee, J.M., Kajigaya, R., Matsumoto, S. (2009) Bombyx mori homologs of STIM1 and Orai1 are essential components of the signal transduction cascade that regulates sex pheromone production. J Biol Chem 284: 31200-31213.

Jurenka, R. (2004) Insect pheromone biosynthesis. Top Curr Chem 239: 97-132.

Jurenka, R. (2017) Regulation of pheromone biosynthesis in moths. Curr Opin Insect Sci 24: 29- 35.

Kayukawa, T., Chen, B., Hoshizaki, S., Ishikawa, Y. (2007) Upregulation of a desaturase is associated with the enhancement of cold hardiness in the onion maggot, Delia antiqua. Insect Biochem Mol Biol 37: 160–1167.

Knipple, D.C., Rosenfield, C.L, Miller, S.J., Liu, W., Tang, J., Peter, W.K., et al. (1998) Cloning and functional expression of a cDNA encoding a pheromone gland-specific acyl-CoA Delta11-desaturase of the cabbage looper moth, Trichoplusia ni. Proc Natl Acad Sci USA 95: 15287-15292.

Köhler, K., Brunner, E., Xue, L.G., Boucke, K., Greber, U.F., Mohanty, S., et al. (2009) A combined proteomic and genetic analysis identifies a role for the lipid desaturase Desat1 in starvation-induced autophagy in Drosophila. Autophagy 5: 980–990.

Imai, K., Konno, T., Nakazawa, Y., Komiya, T., Isobe, M., KOGA, K. (1991) Isolation and structure of diapause hormone of the silkworm, Bombyx mori. Proc Jpn Acad 67: 98-101.

Ishida, Y., Leal, W.S. (2005) Rapid inactivation of a moth pheromone. Proc Natl Aca. Sci USA 102: 14078-79.

Lassance, J.M., Groot, A.T., Liénard, M.A., Antony, B., Borgwardt, C., Andersson, F., et al. (2010) Allelic variation in a fatty-acyl reductase gene causes divergence in moth sex pheromones. Nature 466: 486–9.

Lassance, J.M., Lienard, M.A., Antony, B., Qian, S., Fujii, T., Tabata, J., et al. (2013) Functional consequences of sequence variation in the pheromone biosynthetic gene pgFAR for Ostrinia moths. Proc Natl Acad Sci USA 110:3967-3972.

Leal, W.S. (2013) Odorant reception in insects: roles of receptors, binding proteins, and degrading enzymes. Annu Rev Entomol 58: 379-91.

Li, Z.Q., Zhang, S., Luo, J.Y., Wang, C.Y., Lv, L.M., Dong, S.L., et al. (2015) Transcriptome comparison of the sex pheromone glands from two sibling Helicoverpa species with opposite sex pheromone components. Sci Rep 5: 9324.

Lienard, M.A. and Löfstedt, C. (2010) Functional flexibility as a prelude to signal diversity? : Role of a fatty acyl reductase in moth pheromone evolution. Commun Integr Biol 3: 586- 588. 17

Lienard, M.A., Hagstrom, A.K., Lassance J.M., Löfstedt, C. (2010) Evolution of multicomponent pheromone signals in small ermine moths involves a single fatty-acyl reductase gene. Proc Natl Acad Sci USA 107: 10955-10960.

Lienard, M.A., Strandh, M., Hedenstrom, E., Johansson, T., Löfstedt, C. (2008) Key biosynthetic gene subfamily recruited for pheromone production prior to the extensive radiation of Lepidoptera. BMC Evol Biol 8: 270.

Löfstedt, C., Löfqvist, J., Lanne, B.S., Van Der Pers, J.N.C., Hansson, B.S. (1986) Pheromone dialects in European turnip moth Agrotis segetum. Oikos 46: 250-257.

Löfstedt, C., Wahlberg, N. and Millar, J.G. (2016) Evolutionary patterns of pheromone diversity in Lepidoptera. In by Allison J.D. and Cardé, R.T., eds. Pheromone communication in moths: evolution, behavior, and application. University of California Press, Oakland, California, p. 401.

Matsumoto, S., Kitamura, A., Nagasawa, H., Kataoka, H., Orikasa, C., Mitsui, T., et al. (1990) Functional diversity of a neurohormone produced by the suboesophageal ganglion: Molecular identity of melanization and reddish colouration hormone and pheromone biosynthesis activating neuropeptide. J Insect Physiol 36: 427-432.

Metz, J.G., Pollard, M.R., Anderson, L., Hayes,T.R., Lassner, M.W. (2000) Purification of a jojoba embryo fatty acyl-coenzyme A reductase and expression of its cDNA in highe rucic acid rapeseed. Plant Physiol 122: 635–644.

Mochizuki, F., Fukumoto, T., Nogunchi, H., Sugie, H., Morimoto, T., Ohtani, K. (2002) Resistance to mating disruptant composed of (Z)-11-tetradecenyl acetate in the smaller tea tortrix, Adoxophyes honmai (Yasuda) (Lepidoptera: Tortricidae). Appl Entomol Zool 37: 299-304.

"Moths". Smithsonian Institution. Retrieved 2012-01-12.

Moto, K., Yoshiga, T., Yamamoto, M., Takahashi, S., Okano, K., Ando, T., et al. (2003) Pheromone gland-specific fatty-acyl reductase of the silkmoth, Bombyx mori. Proc Natl Acad Sci USA 100: 9156-9161.

Michalski, C., Mohagheghi, H., Nimtz, M., Pasteels, J., Ober, D. (2008) Salicyl alcohol oxidase of the chemical defense secretion of two chrysomelid leaf beetles. J Biol Chem 283: 19219- 28.

Miller, J.R. and Gut, L.J. (2015) Mating disruption for the 21st century: Mating technology with mechanism. Environ Entomol 44: 427-453.

Miura, N., Nakagawa, T., Touhara, K., Ishikawa, Y. (2010) Broadly and narrowly tuned odorant receptors are involved in female sex pheromone reception in Ostrinia moths. Insect Biochem Molec Biol 40: 64–73. 18

Nakagawa, T., Sakurai, T., Nishioka, T., Touhara, K. (2005) Insect sex-pheromone signals mediated by specific combinations of olfactory receptors. Science 307: 1638-1642.

Purcell, M., Johnson, M.W., Lebeck, L.M., Hara, A.H. (1992) Biological control of Helicoverpa zea (Lepidoptera: Noctuidae) with Steinernema carpocapsae (Rhabditida: Steinernematidae) in corn used as a trap crop. Environ Entomol 21: 1441-1447.

Rafaeli, A. (2005). Mechanisms involved in the control of pheromone production in female moths: recent developments. Entomol Exp Appl 115: 7–15.

Raina, A.K., Jaffe, H., Klun, J.A., Ridgway, R.L., Hayes, D.K. (1987) Characteristics of a neurohormone that controls sex pheromone production in Heliothis zea. J Insect Physiol 33: 809-814.

Rodriguez, .S, Hao, G., Liu, W., Piña, B., Rooney, A.P., Camps, F. et al. (2004) Expression and evolution of delta9 and delta11 desaturase genes in the moth Spodoptera littoralis. Insect Biochem Mol Biol 34: 1315-1328.

Rosenfield, C., You, K.M., Marsella-Herrick, P., Roelofs, W.L., Knipple, D.C. (2001) Structural and functional conservation and divergence among acyl-CoA desaturases of two noctuid species, the corn earworm, Helicoverpa zea, and the cabbage looper, Trichoplusia ni. Insect Biochem Mol Biol 31: 949-964.

Sakurai, T., Nakagawa, T., Mitsuno, H., Mori, H., Endo, Y., Tanoue, S., et al. (2004) Identification and functional characterization of a sex pheromone receptor in the silkmoth Bombyx mori. Proc Natl Acad Sci USA 101: 16653–16658.

Soberón, M., Gill, S.S., Bravo, A. (2009) Signaling versus punching hole: how do Bacillus thuringiensis toxins kill insect midgut cells? Cell Mol Life Sci 66: 1337–1349.

Tang, J.D., Charlton, R.E., Jurenka, R.A., Wolf, W.A., Phelan, P.L., Srend, L., et al. (1989) Regulation of pheromone biosynthesis by a brain hormone in two moth species. Proc Natl Acad Sci USA 86: 1806-1810.

Teal, P.E.A., Tumlinson, J.H. (1988) Properties of cuticular oxidases used for sex pheromone biosynthesis by Heliothis zea. J Chem Ecol 14: 2131-45.

Torfs, P., Nieto, J., Cerstiaens, A., Boon, D., Baggerman, G., Poulos, C. et al. (2001) Pyrokinin neuropeptides in a crustacean: Isolation and identification in the white shrimp Penaeus vannamei. Eur J Biochem 268: 149-154.

Tillman J.A., Seybold, S.J. Jurenka, R.A., Blomquist, G.J. (1999) Insect-pheromones-an overview of biosynthesis and endocrine regulation. Insect Biochem Mol Biol 29: 481-514. 19

Wang, G., Vásquez, G. M., Schal, C., Zwiebel, L. J., Gould, F. (2011) Functional characterization of pheromone receptors in the tobacco budworm Heliothis virescens. Insect Mol Biol 20: 125–133.

Wang, H.L., Lienard, M.A., Zhao, C.H., Wang, C.Z., Löfstedt, C. (2010) Neofunctionalization in an ancestral insect desaturase lineage led to rare Delta6 pheromone signals in the Chinese tussah silkworm. Insect Biochem Mol Biol 40: 742-751.

Wang, Y., da Cruz, T.C., Pulfemuller, A., Grégoire, S., Ferveur, J.F., Moussian, B. (2016) Inhibition of fatty acid desaturases in Drosophila melanogaster larvae blocks feeding and developmental progression. Arch Insect Biochem Physiol 92: 6–23.

Xu, W., Papanicolaou, A., Liu, N-Y., Dong, S-L., Anderson, A. (2014) Chemosensory receptor genes in the Oriental tobacco budworm Helicoverpa assulta. Insect Mol Biol 24: 253- 263.

Zeng, J-M., Ye, W-F., Noman, A., Machado, R.A.R., Lou, Y-G. (2019) The desaturase gene family is crucially required for fatty acid metabolism and survival of the brown planthopper, Nilaparvata lugens. Int J Mol Sci 20: 1369.

Zhang, D., Lee, R.S., Xi, Z, Bourtzis, K., Gilles, J.R.L. (2016) Combining the sterile insect technique with the incompatible insect technique: III-robust mating competitiveness of irradiated triple Wolbachia-infected Aedes albopictus males under semi-field conditions. PloS One 11: e0151864.

CHAPTER 2. TRANSCRIPTOME COMPARISON OF PHEROMONE GLAND-

OVIPOSITOR AND TARSI IN THE CORN EARWORM MOTH, Helicoverpa zea

Modified from a manuscript published in Comparative Biochemistry and Physiology – Part D: Genomics and Proteomics 31: 100604

Xiaoyi Dou, Sijun Liu, Seung-Joon Ahn, Man-Yeon Choi, Russell Jurenka

Abstract

The corn earworm, Helicoverpa zea, utilizes (Z)-11-hexadecenal as the major sex pheromone component. The saturated fatty acid derivative hexadecanal is also found in the pheromone gland and recently a large amount (0.5-1.5 µg) was found in male tarsi with lower amounts (0.05-0.5 µg) in female tarsi. In this study, we compared the transcriptome between female pheromone glands

(including the ovipositor) and female and male tarsi to identify differences between these tissues, particularly the genes involved in sex pheromone biosynthesis and in chemosensation. We found

9 fatty acyl-CoA desaturases, 20 fatty acyl-CoA reductases, 7 alcohol oxidases, some G protein- coupled receptors and many genes involved in signal transduction and pheromone transportation.

Also we found gustatory and olfactory receptors associated with the tarsi and ovipositor.

Differential expression analysis showed that there were many genes differentially expressed between tissues, including the candidate desaturases, fatty acyl-CoA reductases and alcohol oxidases. We discuss how some of these genes produce proteins that could be involved in the biosynthesis of hexadecanal in tarsi and (Z)-11-hexadecenal in the pheromone gland and the possible role of proteins in chemosensation of the tarsi and ovipositor. 21

Introduction

Sex pheromones play an important role in mating communication and reproduction in moth species. Species-specific sex pheromones are biosynthesized and released by the pheromone gland

(PG) which is usually located between the 8th and 9th abdominal segments in female moths

(Jurenka, 2004). The majority of moth sex pheromones consist of multi-component blends of C10-

18 hydrocarbon chains, with one or more double bonds, and alcohols, acetate esters or aldehydes as a functional group.

The biosynthesis of sex pheromones in moth is regulated by pheromone biosynthesis activating neuropeptide (PBAN) which binds to a receptor on PGs to induce pheromone biosynthesis (Jurenka and Rafaeli, 2011; Jurenka, 2017). After biosynthesis of saturated fatty acid, fatty acyl-CoA desaturases introduce double bonds into the fatty acyl chains which may be followed by one or two rounds of chain shortening steps by limited β-oxidation enzymes (Jurenka,

2004). The terminal carboxyl group is modified to form one of the functional groups: alcohol, aldehyde or acetate ester by fatty acyl-CoA reductase (FAR), alcohol oxidase (AO) or fatty acyltransferase, respectively (Jurenka, 2004). Thus far, six types of desaturases and several FARs have been identified (Lofstedt et al., 2017). However, the genes for acetylation of fatty alcohols to acetate esters and oxidation of fatty alcohols to aldehydes have not been identified at the molecular level despite being characterized biochemically in several moth pheromone gland assays (Jurenka,

2004; Jurenka and Roelofs, 1989; Teal and Tumlinson, 1986; Fang et al., 1995; Luxová and Svatoǎ,

2006). The key genes involved in pheromone biosynthesis have been analyzed in several moth PG transcriptomic studies (Antony et al., 2015; Cha et al., 2017; Chen et al., 2017; Ding and Löfstedt

2015; Gu et al., 2013; He et al., 2017; Li et al., 2015, 2018; Vogel et al., 2010; Xia et al., 2015).

The present study is the first to compare transcriptomes between PG-ovipositors and tarsi from 22

males and females to assess production of aldehyde pheromones and to help in characterizing chemosensory organs that are not located on the antennae or mouthpart.

The heliothine genera Helicoverpa and Heliothis utilize (Z)-11-hexadecenal (Z11-16: Ald) or

(Z)-9-hexadecenal (Z9–16: Ald) as the major pheromone component (Hillier and Baker, 2016).

Moreover, the minor gland component hexadecanal (16: Ald) has been found in several heliothines, but is used as a sex pheromone in only a few species (Hillier and Baker, 2016). Recently, Choi et al. (2016) found a large amount (0.5-1.5 µg) of 16: Ald in the male tarsi and lower amounts (0.05-

0.5 µg) on female tarsi of several heliothines including Helicoverpa zea. Based on the process of the major sex pheromone component biosynthesis, the production of 16: Ald is assumed to be the same but without the action of a desaturase.

Olfactory and gustatory systems play important roles in feeding, mating, and oviposition behaviors. Tarsi are known to contain gustatory receptors (GRs): for example twenty eight GRs were found expressed in the tarsi of Drosophila melanogaster to recognize sweet or bitter compounds (Ling et al., 2014). Ozakim et al. (2011) utilized RNAi to silence one GR from the foreleg tarsal sensilla of the butterfly, Papilio xuthus, indicating its role in identifying host plants for oviposition. The ovipositor also contains both olfactory and gustatory chemosensilla that are usually involved in detecting hosts for oviposition (Klinner et al., 2016; Xia et al., 2015; Yadav and Borges. 2017).

In this study, we compared the transcriptome results between female PG-ovipositors and female and male tarsi of H.zea. We found 9 DESs, 20 FARs, 7 AOs, some G protein-coupled receptors (GPCRs) and many other transcripts encoding proteins involved in signal transduction and chemosensation. Differential expression analysis showed that there were many genes 23

significantly differentially expressed between tissues, including the candidate desaturases, FARs, and AOs. These genes may be involved in the biosynthesis of 16 Ald in PG and tarsi. In addition, the chemosensory genes found in PG-ovipositor and tarsi may indicate the roles these tissues play in food and host seeking behaviors.

Materials and Methods

Insects and tissue collections

Pupae of the corn earworm, H. zea, were purchased from Frontier Agricultural Sciences

(Newark, DE, USA) and maintained at 25 °C±1 under a photoperiod of L:D 15:9 hr until adult emergence. The terminal 8th and 9th abdominal segments containing the pheromone gland and ovipositor were removed from ten 2-day-old female adults. The tarsi from these same females and from 2-day-old males were also removed. Total RNA was separately isolated from all three tissue samples using the Invitrogen PureLinkTM RNA Mini Kit according to the manufacturer’s instructions. Total RNA from three biological replications was obtained and used for the transcriptome analysis.

RNA isolation, cDNA library construction and Illumina sequencing

The cDNA libraries were prepared using a TruSeq Stranded Total RNA Library Prep Kit

(Illumina, California, USA) and sequencing was performed by the center for Genome Research and Biocomputing at Oregon State University using the Illumina HiSeq 2000 platform. Briefly, ribosomal RNA was removed from total RNA and the remaining RNA was purified, fragmented and primed for cDNA synthesis. The first strand cDNA was synthesized by priming the RNA fragments with random hexamers and by transcribing with reverse transcriptase. RNA template of 24

the first strand cDNA was replaced by incorporating dUTP in place of dTTP to generate second strand cDNA. The cDNA then underwent an end repair process, adenylation of 3’ end and subsequent ligation of the adapter. The adaptor-ligated libraries were purified and enriched with

PCR to create the final cDNA library.

Assembly of short reads and gene annotation

The raw sequence reads generated by Illumina sequencing were checked by FastQC for quality control and trimmed by Trimmomatic tool (ver. 032) (Bolger et al., 2014) to remove adaptors and low-quality reads. Overlapping high-quality reads were de novo assembled to create longer contiguous fragments (contigs) using Velvet/OASES, SoapDenovo-Trans and Trinity

(version r2014-07-17) with the default parameters (Grabherr et al. 2011). All reads were submitted to the SRA database of NCBI under the accession number “SRP194183”. The resulting contigs were locally searched against the NCBI non-redundant (nr) protein database and Swiss-prot database using the BLASTx program, to obtain protein annotations of the assembled contigs. Gene

Ontology terms were performed by the Blast2GO program (Conesa et al., 2005) and the GO functional classification was obtained using WEGO software (Ye et al., 2006). Selected transcripts were compared to PG transcriptomes published from three other heliothine moths, Helicoverpa armigera, Helicoverpa assulta, and Heliothis virescens. The data from H. armigera and H. assulta were obtained from BLASTp searches against the assembled and translated sequences obtained from the sequence read archive (SRR1565435 and SRR1570898) in the NCBI database (Li et al.,

2015). The H. virescens comparison was made by tBLASTn against the NCBI databases: expressed sequence tags and transcriptome shotgun assembly limiting the search to H. virescens

(Vogel et al., 2010). 25

Differential expression analysis

Expression abundance of the transcripts was estimated using the RSEM (RNA-Seq by

Expectation-Maximization) method and followed by the edgeR package in R to analyze for the differential expression (McCarthy et al., 2012). The FPKM (fragments per kilobase per million reads) value were used as the abundance level. The differential expression between tissues was measured by multiple comparison with FDR (false discovery rate) less than 0.05 and fold change larger than 4. We compared the expression abundance of transcripts from female pheromone gland to female tarsi or male tarsi to female tarsi (Supplementary Table 2.4).

Identification of candidate genes involved in moth sex pheromone biosynthesis and chemosensation

Based on the reported pheromone biosynthesis pathways in heliothine moths (Choi et al.,

2002; 2005), the target genes were selected as follows: genes involved in sex pheromone biosynthesis were acetyl-CoA carboxylase (ACC), fatty acid synthase (FAS), desaturase, FAR, and AO; genes involved in chemosensation were odorant binding proteins (OBP), chemosensory proteins (CSP), olfactory receptors (OR), gustatory receptors (GR), ionotropic receptors (IR) and sensory neuron membrane protein (SNMP); genes involved in signal transduction were adenylate cyclases (ADC), cAMP-dependent protein kinase (PKA), inositol triphosphate receptors (IP3R), phospholipase C (PLC), 5’-AMP-activated protein kinases (AMPK), Ca2+/calmodulin-dependent protein kinase II (CaMKII), calcium release-activated calcium channel protein (Orai), serine/threonine-protein phosphatase 2B (PP2B), stromal interaction molecules (STIM) and calmodulin. Also, we selected some receptors that are related to a hormonal regulation of sex pheromone biosynthesis: pheromone biosynthesis activating neuropeptide receptor (PBANR), 26

diapause hormone receptor (DHR), ecdysis-triggering hormone receptor (ETHR), octopamine receptor (OctoR), and sex peptide receptor (SPR).

Phylogenetic analysis

Phylogenetic analysis was conducted with three candidate genes families, desaturase, FAR and GR which are involved in the sex pheromone biosynthetic pathway and the gustatory system.

For comparison, we imported 66 desaturases sequences, 70 identified reductases and 137 GRs from other species and the genes we found in the H.zea transcriptome. The phylogenetic trees were constructed using the neighbor-joining method implemented in MEGA7 with default setting and

1000 bootstrap replicates.

Results

Illumina sequencing, sequence assembly and gene annotation

Illumina sequencing of cDNA libraries of each tissues were conducted with three replicates.

We assembled all of the reads to one transcriptomic file. After removing the adapters and low quality regions, the clean reads were assembled, resulting in 338,754 contigs with average length of 730 bps. To assess the completeness of the assembled data, the transcripts were analyzed using the BUSCO (Benchmarking Universal Single-Copy Orthologs) program using a database of arthropod genes (Simão et al., 2015) (Table 2.1). These results suggest that the quality of the sequencing assembly was acceptable. 27

Differentially expressed genes

Differential expression was analyzed using genes with full open reading frames, with the edgeR program at the gene level with the absolute value of log2 [fold change] larger than 2 and

FDR less than 0.05 (Figure 2.1). In a comparison of female PGs and female tarsi (Figure 2.1A),

1,049 genes were differentially expressed, while 600 genes were upregulated in female PG and

449 genes were downregulated. In a comparison of female tarsi and male tarsi (Figure 2.1B), 101 genes were differential expressed, with 36 genes upregulated in male tarsi and 65 genes downregulated. The top 20 differentially expressed genes between female PG and female tarsi, and between male tarsi and female tarsi are shown in Table 2.2 and Table 2.3. In comparison of the female PG and male tarsi with female tarsi, there were 4 upregulated genes and 34 downregulated genes in common (Supplementary Table 2.1 and Table 2.2). One transcript encoding a protein involved in an elongation of very long chain fatty acids was found in the common upregulated genes, indicating there could be more fatty acid produced in female PGs and male tarsi than female tarsi. Also there were 8 transcripts that encode proteins with unknown functions.

In the comparison of female PG and female tarsi, we found a transcript encoding a facilitated trehalose transporter (HzeaTret) that was significantly differentially expressed between PGs and female tarsi (Tret2). However, in total we found 14 HzeaTret transcripts (Supplementary Table

2.3) that could produce proteins with 12 transmembrane domains which is a typical structure for sugar transporters. This is important because it has been shown that PGs use trehalose as a carbon source to produce pheromone (Foster and Johnson, 2010). Most of the Tret proteins showed around 99% identity to the Tret proteins found in the genome of H. armigera. The most abundant ones were Tret5 (FPKM = 227) in PGs, Tret6 (FPKM = 310) in female tarsi, and Tret9 (FPKM = 28

128) in male tarsi. In the comparison of PGs and female tarsi all of the Tret transcripts were significantly different in abundance levels except Tret3, Tret12, and Tret14. Of the remaining Tret transcripts, 6 were upregulated in PGs while 5 were downregulated. Comparing female and male tarsi only 1 was differentially expressed (Tret6).

The differentially expressed genes were classified by gene ontology (Figure 2.2). Within the selected GO term, there were more genes related to oxidation-reduction process, biosynthetic process and protein binding up regulated in female PG than female tarsi (Figure 2.2A). However, comparison between male and female tarsi indicated that the numbers of up or down regulated genes in each category were not significantly different (Figure 2.2B).

Candidate genes involved in sex pheromone biosynthesis

The compounds found in the pheromone gland of H. zea are Z11-16: Ald, Z9-16: Ald, Z7-16:

Ald and 16: Ald with the ratio of about 90:2:1:7 (Pope et al., 1984). The enzymes involved in biosynthesis include ACC, FAS, desaturase, FAR, and AO. According to the gene annotation results, we found 1 ACC, 3 FASs, 9 desaturases, 20 FARs and 8 AOs (Supplementary Table 2.3).

The expression levels of desaturases, FARs and AOs are shown in Figure 2.3. Phylogenetic trees of desaturases and FARs were constructed for comparison with those in other insects (Figures 2.4 and 2.5).

The sex pheromone precursors are made through the synthesis of fatty acid, which includes the carboxylation by ACC, followed by the synthesis of saturated fatty acid by FAS. We found 1 transcript encoding ACC and 3 transcripts encoding FAS, respectively (Supplementary Table 2.3).

ACC was upregulated in the female PG compared to female tarsi. While only one FAS was upregulated, others were downregulated in the female PG. There was no significant difference 29

between male and female tarsi (FDR >0.05). All four heliothine species had similar ACC and FAS transcripts expressed in their PGs.

Desaturases introduce double bonds into the fatty acid chain at specific positions. The desaturases were identified based on homology to other insect desaturases that have three histidine boxes with eight histidine residues that are involved in creating essential metal complexes

(Rosenfield et al., 2001). We found transcripts encoding 9 desaturases in all the tissues

(Supplementary Table 2.3). According to the four amino acid motif described for insect fatty acyl-

CoA desaturases (Knipple et al., 2002; Li et al., 2017), 8 of the desaturases were classified into subgroups with the following names: LPAQ, PDSN, NPAE, GATD, QPVE, NPVE, KPSE, and

KSVE. The 9th desaturase (DES9) was similar to a sphingolipid ∆4 desaturase. The most abundant desaturase in PGs was HzeaLPAQ (FPKM > 8000), which has been identified previously as a ∆11- desaturase (Rosenfield et al., 2001). Also, HzeaNPVE and HzeaKPSE are ∆9-desaturases that were moderately expressed in PGs (Figure 2.3A). However, since the saturated 16:Ald is produced in tarsi, none of these desaturases are involved in the production of aldehyde in tarsi. The phylogenetic analysis of desaturases with other moth desaturases is shown in Figure 2.4.

HzeaGATD, HzeaNPVE, and HzeaKPSE were clustered as ∆9-desaturases, while HzeaPDSN and

HzeaQPVE clustered with ∆14-desaturases. All four heliothine species PGs have the ∆11- desaturase LPAQ as the desaturase that produces Z11-16-acyl-CoA. They also have a ∆9- desaturase NPVE that is also found in other tissues.

FARs catalyze the reduction of fatty acyl-CoA to the corresponding fatty alcohol (Lassance et al. 2010). We found transcripts encoding 20 FARs in the H. zea transcriptome (Supplementart

Table 2.3). The most abundant one in PGs was FAR1 (FPKM = 1500) in which the protein sequence was 99% identical to H. assulta FAR (Protein ID: ATJ44516), indicating that FAR1 is 30

most likely involved in sex pheromone biosynthesis in H. zea PGs. FAR1 was also the most abundant FAR in the PGs of the other heliothines. Some FARs showed significantly different expression levels in the three tissues (Figure 2.3B). FAR5, FAR7, FAR11, FAR14 and FAR15 were expressed higher in female and male tarsi. They all showed over 99% identical protein sequences to FARs from heliothine moths, so these could be candidate genes encoding a FAR involved in biosynthesis of 16: Ald in tarsi. FAR4 is 99% similar to H. armigera FAR (Protein ID: AKD01770), and it is moderately expressed in female PG and female tarsi, but lower in male tarsi. The phylogenetic tree (Figure 2.5) showed that FAR1 forms a clade with the identified FARs found in

PGs of Helicoverpa (Hagstrom et al., 2012). FAR8 and FAR 10 were clustered with the Ostrinia

FARs (Lassance et al., 2013), while the remaining FARs clustered with the FARs identified from the PG transcriptome data of Spodoptera (Zhang, et al. 2015, 2017) and Agrotis (Gu et al., 2013;

Ding and Löfstedt, 2015).

Alcohol oxidases (AO) modify fatty alcohols to form fatty aldehydes. AO2, AO5, AO6 and

AO7, among 8 found in this study, were differentially expressed between tissues (Figure 2.3C).

However, only AO7 was highly expressed in PGs (FPKM = 733.9). The amino acid sequence has

91% identity to an AO from H. armigera (Protein ID: ATJ44473). AO8 was expressed

(FPKM≈300) in all three tissues, so AO7 and AO8 maybe the candidate genes involved in the production of Z11-16:Ald in PG and 16:Ald in tarsi. These putative AOs were also found in the other heliothine PGs transcripts. However, since no AOs have been identified at the molecular level in any insect, at this time it is unknown if they encode the AO proteins involved in sex pheromone biosynthesis.

In pheromone gland cells there is a constant turnover of the pheromone aldehydes (Foster and

Anderson, 2011). We found 4 aldehyde dehydrogenase (ALDH) transcripts that produce enzymes 31

converting aldehydes to fatty acid. ALDH1 was expressed at high level in all three tissues and is also found in the other heliothine PGs. We also found transcripts encoding 7 odorant degrading enzymes (ODE) that are esterases: two esterases (ODE1 and ODE3) were expressed at the highest levels in all three tissues and were also found in the other heliothine PGs.

Candidate genes involved in chemosensation and signal transduction

In the H. zea transcriptome, we found transcripts that encode 13 OBPs, 12 CSPs, 18 ORs, 9

GRs, 7 IRs and 1 SNMP (Supplementary Table 2.3). The differential expression is shown as a heat map in Figure 2.7. Compared with female PG-ovipositors and female tarsi, 6 OBPs were downregulated in PGs, while the other 6 were not significantly different (FDR >0.05). The most abundant OBP in PG-ovipositors was OBP9 (FPKM = 347.3), but there was no statistical difference between tissues. OBP13 was highly expressed in tarsi (FPKM = 16347 and 7443 in female and male, respectively) which was differentially expressed compared to PG-ovipositors

(FDR = 0.003). Eight out of 12 CSPs were downregulated in PGs and one was upregulated (CSP8).

Four ORs were differentially expressed between female PGs and female tarsi, but they had very low FPKM values (<5). OR6 and OR15 were moderately expressed in all tissues and were not statistically different. Four GRs were downregulated in female PGs although their expression level was very low (FPKM <10). The phylogenetic tree (Figure 2.6) showed GR8 and GR9 may be bitter receptors, and GR1, GR3, GR4, GR5, GR6 clustered together with the sugar receptors. GR7 is closely related to D. melanogaster GR43a which is a fructose receptor (Ling et al., 2014). Also,

GR2 clustered as a CO2 receptor. The ligand-gated ion channels, inotropic receptors (IRs) had moderate or low expression levels in all tissues. Two IRs were downregulated in female PGs with a low abundance level. Interestingly, we found one SNMP with high identity to SNMP2 in H. 32

armigera that was highly expressed in all tissues and significantly different between PG- ovipositors and female tarsi and male tarsi. Comparison of expression levels between male tarsi and female tarsi indicated that all of the genes encoding chemosensory proteins were not significantly different. The OR and GR transcripts were not found to a great extent in the other heliothines because the ovipositors were not included when PGs were extracted for mRNA analysis in the other studies (Li et al., 2015; Vogel et al., 2010).

Sex pheromone biosynthesis is regulated by PBAN, which is released from the subesophageal ganglion into the hemolymph and binds to PBAN receptors in the cell membrane of PGs in moths.

We found two PBAN receptor transcripts, PBANR_B and PBANR_C encoding proteins highly homologous to PBAN receptors from other moths. They were expressed at low levels in tarsi but the highest in PGs (Figure 2.7). In addition, some GPCRs were found, that show high homology to DHR of H. zea (Protein ID: AGR34305), ETHR of Manduca sexta (Protein ID: AAX19163),

OctoR1 and OctoR2 of H. armigera (Protein ID: XP_021187627, XP_021194395) and SPR of H. armigera (Protein ID: XP_021183461) (Supplementary Table 2.3). The expression comparison showed that ETHR and DHR were expressed at higher level in female PGs than female tarsi (FDR

<0.05) while OctoR2 was expressed at low levels. There was no difference in the expression of these receptors between male tarsi and female tarsi.

In the H. zea transcriptome, 8 ADCs, 3 PKAs, 3 IP3Rs, 3 PLCs, 4 AMPKs, 3 CaMKIIs, 1

Orai, 2 PP2Bs, 2 STIMs, and 3 calmodulin transcripts were found, which are possibly involved in

PBAN signal transduction. There were some differentially expressed genes between the tissues

(Figure 2.7). Comparison between female PG and female tarsi, 3 ADCs, 2 PKAs, 1 IP3 receptor,

3 AMPKs and 1 PP2B were upregulated in female PGs, while 1 ADC, 1 IP3 receptor, 2 PLCs, 1

Orai, 1 CaMKII and 1 calmodulin were downregulated. The rest were not significantly different 33

(FDR >0.05). Comparison of signal transduction genes between male tarsi and female tarsi found no significant differences (FDR >0.05). Most of the PBAN signal transduction transcripts were also found in the other heliothines indicating that the same pathway occurs in these moths to stimulate pheromone biosynthesis.

Discussion

Moth sex pheromones are composed of multiple components in species-specific ratios. The corn earworm, H. zea, utilizes Z11-16:Ald as the major sex pheromone component with several minor components (Pope et al., 1984). One of the minor components, 16:Ald is found in pheromone glands at trace levels but is not required for male attraction, therefore is not considered a sex pheromone in H. zea (Hillier and Baker, 2016). Relative to our study, 16:Ald was found in legs of heliothine moths: a large amount (0.5-1.5 µg) in male tarsi and a lower amount (0.05-0.5

µg) in female tarsi, (Choi et al., 2016). However, the biological role of 16:Ald in the legs of heliothines is still unclear.

Here, we compared the transcriptome between female PGs and female and male tarsi, and found large differences in expression levels between PGs and female tarsi and smaller differences between male and female tarsi, for the key enzymes involved in sex pheromone biosynthesis.

There were more differentially expressed genes in female PGs than in female and male tarsi. Gene ontology enrichment analysis also showed more upregulated genes than downregulated genes in female PGs.

Since H. zea and H. armigera are very closely related, in fact hybridization can occur

(Anderson et al., 2018), most of the top BLASTp hits were to H. armigera proteins encoded in its genome. In fact, most of the transcripts found in the H. zea PG transcriptome also exist in the H. 34

armigera PG transcriptome (Supplementary Table 2.3). The other heliothine PG transcriptomes of

H. assulta and H. virescens also had many similar transcripts indicating the close phylogenetic relationship of these species.

Pheromone glands utilize trehalose as a carbon source to produce pheromone (Foster and

Johnson, 2010) and the Tret1 amino acid sequences were shown to be conserved in insects

(Kanamori et al., 2009). Fourteen transcripts encoding Tret proteins were found that could be involved in transporting trehalose across cell membranes, thereby regulating the concentration of hemolymph trehalose and allowing for the uptake of trehalose as an energy source and source of carbon for pheromone biosynthesis (Kikawada et al., 2007).

Pheromone biosynthesis is initiated by PBAN binding with its receptor on PG cells. Two

PBAN receptor isoforms were more abundant in PGs than tarsi. Removing the source of PBAN by decapitation did not reduce the 16: Ald amounts in tarsi indicating the production of 16:Ald in tarsi is not under PBAN control (Choi et al., 2016). In addition, we found transcripts encoding for four other GPCRs: DHR, ETHR, SPR, and OctoR. It is interesting to note that the DHR and ETHR belong to the same family of receptors as does the PBAN receptor (Jurenka, 2015). The sex peptide has been implicated in termination of pheromone production (Hanin et al., 2012) while octopamine could modulate pheromone production (Rafaeli and Gileadi, 1995).

PBAN binding to the receptor in PGs initiates a signal cascade to stimulate pheromone production in moths. One of the first steps in the cascade is the influx of extracellular Ca2+ that is mediated by a store-operated channel (SOC) as shown in Bombyx mori (Hull et al., 2007, 2009).

In heliothine moths, calcium activates an ADC to produce cAMP. ACC is activated by the combined action of calcium activating calcineurin (PP2B) that dephosphorylates ACC and by cAMP activating a PKA which inhibits an AMPK that phosphorylates ACC (Du et al., 2017; 35

Jurenka, 2017). However, it is unknown if the PBAN signal in H. zea is mediated by SOC as found in B. mori. Our transcriptomic data found many transcripts related to SOC signal transduction: 3

PLCs, 3 IP3 receptors, 2 STIMs, and 1 Orai which are all involved in causing the influx of extracellular Ca2+. This indicates that the influx of extracellular calcium could be derived from the

SOC pathway in H. zea. There were many differentially expressed genes involved in signal transduction. Most of the ACDs, PKAs, PP2B and AMPKs were upregulated in female PGs and all of these genes were also found in tarsi.

The olfactory pathways are involved in mate finding, host and food seeking behaviors and avoidance of predators or parasitoids, while the gustatory pathways are involved in feeding and egg-laying behaviors (Depetris-Chauvin et al., 2015). Odorant perception usually occurs in the antennae and maxillary palps, but in this study, we also found some ORs in PG-ovipositors and tarsi. OR1 had 65% identity to D. melanogaster OR83b (Protein ID: AAT71306) and 88% identity to BmOR1 (Protein ID: NM_001043595), indicating that this could be the homologous OR in H. zea. This is the general odorant receptor coreceptor (ORCO) found in insects that is required for other OR activity and is usually found highly expressed in the antennae (Walter, 2013). OR6 had the highest expression in all three tissues and was homologous to OR31 from H. armigera which was also found expressed in male and female tarsi (Liu et al., 2014).

GRs located in the proboscis, tarsi and ovipositor are involved in the sense of taste (Liman et al., 2014). Many lepidopterans rely on tarsal contact chemoreception using gustatory receptors for discriminating between host and non-host to oviposit eggs (Ozakim et al., 2011; Ramaswamy et al., 1987). Nine GRs and 7 IRs found in PG and tarsi could be involved in host seeking behavior.

We found one GR similar to the CO2 receptor, 2 GRs to bitter receptors, 5 GRs to sugar receptors and one GR similar to DmGr43. GR7 shares 97% identity to the fructose specific HarmGR4 that 36

was found in the distal part of the antennae of H. armigera (Jiang et al., 2015). GR2 has 63% identity to the D. melanogaster CO2 receptor and 98% to HarmGR1 (XP_021185659) which was found in the labial palps of H. armigera. It was demonstrated that HarmGR1 and HarmGR3 is required for CO2 sensing in a heterologous assay (Ning et al., 2016). The other GRs have 29-32% identity to GRs from D. melanogaster that are involved in the detection of fatty acids and sugars.

IR1 had 63% identity to the IR25a from D. melanogaster that is involved in detecting sour carboxylic acids (Chen and Amrein, 2017). IR2 has 35% identity to IR93a from D. melanogaster that is involved in detecting humidity.

SNMP was first identified from Lepidoptera and exhibited high expression in pheromone receptor neurons of the antennae (Rogers et al., 2001). Since then SNMP2 has been found that is expressed in antennae but also other tissues and is thought to be involved in multiple olfactory roles (Sun et al., 2018). The SNMP2 gene was also found in the transcriptome of H. zea PG and tarsi. It is highly expressed in the male tarsi (FPKM = 1329) and moderately expressed in the female tarsi (FPKM = 398.9), and expressed at a lower level in female PG-ovipositors

(FPKM=75.74). Although the role of SNMP2 is unclear, the relatively high expression levels indicates that it might be involved in a chemoreception activity.

Although biochemical factors involved in 16:Ald production in tarsi of H. zea are still unknown, these comparative transcriptome results provide sequence and transcript abundance information that can be used to identify key enzymes involved in pheromone biosynthesis and the perception of chemical signals. Biochemical identification of key enzymes will require expression in a heterologous system followed by a functional assay and/or gene knockdown experiments. 37

Acknowledgements

We thank Kelly Donahue for technical support. This work was supported in part by base funding from USDA-ARS CRIS 2072-22000-040-00D to M,-Y. C and by the United States-Israel

Binational Agricultural Research and Development Fund (BARD#IS-4722-14) to R.J.

References

Anderson, C.J., Oakeshott, J.G., Tay, W.T., Gordon, K.H.J., Zwick, A., Walsh, T.K. (2018) Hybridization and gene flow in the mega-pest lineage of moth, Helicoverpa. Proc Natl Acad Sci USA 115, 5034-5039.

Antony, B., Soffan, A., Jakse, J., Alfaifi, S., Sutanto, K.D., Aldosari, S.A. et al. (2015) Genes involved in sex pheromone biosynthesis of Ephestia cautella, an important food storage pest, are determined by transcriptome sequencing. BMC Genomics 16: 532.

Bolger, A.M., Lohse, M., Usadel, B. (2014) Trimmomatic: A flexible trimmer for Illumina Sequence Data. Bioinformatics btu170.

Cha, W.H., Kim, W., Jung, J.K., Lee, D.W. (2017) Putative pheromone biosynthesis pathway in Maruca vitrata by transcriptomic analysis. J Aisa Pacc Entomol 20: 165-173.

Chen, Y., Amrein, H. (2017) Ionotropic receptors mediate Drosophila oviposition preference through sour gustatory receptor neurons. Curr Biol 27: 2741-2750.

Chen,D.-S., Dai, J.-Q., Han, S.-C. (2017) Identification of the pheromone biosynthesis gene from the sex pheromone gland transcriptome of the diamondback moth, Plutella xylostella. Sci Rep 7: 16255.

Choi, M.Y., Han, K.S., Boo, K.S., Jurenka, R.A. (2002) Pheromone biosynthetic pathways in the moth Helicoverpa zea and Helicoverpa assulta. Insect Biochem Mol Biol 32: 1353-1359.

Choi, M.Y., Groot, A., Jurenka, R.A. (2005) Pheromone biosynthetic pathways in the moths Heliothis subflexa and Heliothis virescens. Arch Insect Biochem Physiol 59: 53-58.

Choi, M.Y., Ahn, S.J., Park, K.C., Meer, R.V., Carde, R.T., Jurenka, R.A. (2016) Tarsi of male heliothine moths contain aldehydes and butyrate esters as potential pheromone componens. J Chem Ecol 42: 425-432.

Conesa, A., Gotz, S., Garcia-Gomez, J.M., Terol, J., Talon, M., Robles, M. (2005) Blast2GO: a universal tool for annotation, visualization and analysis in functional genomics research. Bioinformatics 21: 3674–3676.

Depetris-Chauvin, A., Galagovsky, D., Grosjean, Y. (2015) Chemicals and chemoreceptors: ecologically relevant signals driving bahavior in Drosophila. Front Ecol Evol 3: 41.

Ding, B.J. and Lofstedt, C. (2015) Analysis of the Agrotis segetum pheromone gland transcriptome in the light of sex pheromone biosynthesis. BMC Genomics, 16: 711.

Du, M., Liu, X., Ma, N., Liu, X., Wei, J., Yin, X., et al. (2017) Calcineurin-mediated dephosphorylation of acetyl-coA carboxylase is required for pheromone biosynthesis activating neuropeptide (PBAN)-induced sex pheromone biosynthesis in Helicoverpa armigera. Mol Cell Proteomics 16: 2138-2152.

Fang, N., Teal, P.E.A., Tumlinson, J.H. (1995) Characterization of oxidase(s) associated with the sex pheromone gland in Manduca sexta (L.) females. Arch Insect Biochem Physiol 29: 243–257.

Foster, S.P., Johson, C.P. (2010) Feeding and hemolymph trehalose concentration influence sex pheromone production in virgin Heliothis virescens moth. J Insect Physiol 56: 1617-1623.

Foster, S., Anderson, K. (2011) The use of mass isotopomer distribution analysis to quantify synthetic rates of sex pheromone in the moth Heliothis virescens. J Chem Ecol 37: 1208- 1210.

Grabherr, M.G., Haas, B.J., Yassour, M., Levin, J.Z., Thompson, D.A., Amit, I. et al. (2011) Full- length transcriptome assembly from RNA-Seq data without a reference genome. Nat Biotechnol 29: 644-652.

Gu, S.H., Wu, K.M., Guo, Y.Y., Pickett, J.A., Field, L.M., Zhou, J.J. et al. (2013) Identification of genes expressed in the sex pheromone gland of the black cutworm Agrotis ipsilon with putative roles in sex pheromone biosynthesis and transport. BMC Genomics 14: 636.

Hagstrom, A.K., Lienard, M.A., Groot, A.T., Hedenstrom, E., Lofstedt, C. (2012) Semi-selective fatty acyl reductases from four heliothine moths influence the specific pheromone composition. PLoS One 7: e37230.

Hanin, O., Azrielli, A., Applebaum, S.W., Rafaeli, A. (2012) Functional impact of silencing the Helicoverpa armigera sex-peptide receptor on female reproductive behaviour. Insect Mol Biol 21: 161-167.

He.P., Zhang, Y.F., Hong, D.Y., Wang. J., Wang, X.L., Zuo, L.H. et al. (2017) A reference gene set for sex pheromone biosynthesis and degradation genes from the diamondback moth, Plutella xylostella, based on genome and transcriptome digital gene expression analyses. BMC genomics 18: 219. 39

Hillier, N.K. and Baker, T.C. (2016) Pheromones of heliothine moths. In: Allison, J. and Carde, R. eds. Communication in moths - evolution, behavior, and application. University of California Press, pp. 301-333.

Hull, J.J., Kajigaya, R., Imai, K., Matsumoto, S. (2007) Sex pheromone production in the silkworm, Bombyx mori, is mediated by store-operated Ca2+ channels. Biosci, Biotechnol Biochem 71: 1993-2001.

Jiang, X-J., Ning, C., Guo, H., Jia, Y.Y., Huang, L-Q., Qu, M-J., et al. (2015) A gustatory receptor tuned to d-fructose in antennal sensilla chaetica of Helicoverpa armigera. Insect Biochem Mol Biol 60: 39-46.

Jurenka, R. A. (2004) Insect pheromone biosynthesis. Top Curr Chem 239: 97-132.

Jurenka, R. A. (2015) The PRXamide neuropeptide signalling system. Adv Insect Physiol 49: 123- 170.

Jurenka, R. A. (2017) Regulation of pheromone biosynthesis in moths. Curr Opin Insect Sci 24: 29-35.

Jurenka, R. A., Rafaeli, A. (2011) Regulatory role of PBAN in sex pheromone biosynthesis of heliothine moths. Front Endocrinol (Lausanne) 2: 46.

Jurenka, R.A., Roelofs, W.L. (1989) Characterization of the acetyltransferase used in pheromone biosynthesis in moths: specificity for the Z isomer in Tortricidae. Insect Biochem 19: 639– 644.

Kanamori, Y., Saito, A., Hagiwara-Komoda, Y., Tanaka, D., Mitsumasu, K., Kikuta, S., et al. (2010) The trehalose transporter 1 gene sequence is conserved in insects and encodes proteins with different kinetic properties involved in trehalose import into peripheral tissues. Insect Biochem Mol Biol 40: 30–37.

Kikawada, T., Saito, A., Kanamori, Y., Nakahara, Y, Iwata, K., Tanaka, D., et al. (2007) Trehalose transporter 1, a facilitated and high-capacity trehalose transporter, allows exogenous trehalose uptake into cells. Proc Natl Acad Sci USA 104: 11585-11590.

Klinner, C.F., König, C., Missbach, C., Werckenthin, A., Daly, K.C., BischKnaden,S., et al. (2016) Functional olfactory sensory neurons housed in olfactory sensilla on the ovipositor of the hawkmoth Manduca sexta. Front Ecol Evol 4: 139.

Lassance, J.M., Groot, A.T., Lienard, M.A., Antony, B., Borgwardt, C., Andersson, F. et al. (2010) Allelic variation in a fatty-acyl reductase gene causes divergence in moth sex pheromones. Nature 466: 486-489.

Lassance, J.M., Lienard, M.A., Antony, B., Qian,S., Fujii,T., Tabata, J. et al. (2013) Functional consequences of sequence variation in the pheromone biosynthetic gene pgFAR for Ostrinia moth. Proc Natl Acad Sci USA 110: 3967-3972.

Löfstedt. C.. Wahlberg. N., Millar. J.G. (2017) Evolutionary patterns of pheromone diversity in Lepidoptera. In: Allison, J.D., Carde R.T. eds. Pheromone communication in moths: Evolution, behavior, and application. University of California Press, Oakland, California. pp. 43-78.

Li, R.T., Ning, C., Huang, L.Q., Dong, J.F., Li, X., Wang, C.Z. (2017) Expressional divergences of two desaturase genes determine the opposite ratios of two sex pheromone components in Helicoverpa armigera and Helicoverpa assulta. Insect Biochem Mol Biol 90: 90-100.

Li, Z.Q., Zhang, S., Luo, J.Y., Wang, C.Y., Lv, L.M., Dong, S.L. et al. (2015) Transcriptome comparison of the sex pheromone glands from two sibling Helicoverpa species with opposite sex pheromone components. Sci Rep 5: 9324.

Li, Z.Q., Ma, L., Yin, Q., Cai, X.M., Luo, Z.X., Bian, L.et al. (2018) Gene identification of pheromone gland genes involved in type II sex pheromonebiosynthesis and transportation in female tea pest Ectropis grisescens. G3 (Bethesda) 8: 899-908.

Liman, E.R., Zhang, Y.V., Montell, C. (2014) Peripheral coding of tast. Neuron 81: 984-1000.

Ling, F., Dahanukar, A., Weiss, L.A., Kwon, J.Y., Carlson, J.R. (2014) The molecular and cellular basis of taste coding in the legs of Drosophila. J Neurosci 34: 7148-7164.

Liu, N.Y., Xu, W., Papanicolaou, A., Dong, S-L., Anderson, A. (2014) Identification and characterization of three chemosensory receptor families in the cotton bollworm Helicoverpa armigera. BMC Genomics, 15: 597.

Luxová, A., Svatoǎ, A. (2006) Substrate specificity of membrane-bound alcohol oxidase from the tobacco hornworm moth (Manduca sexta) female pheromone glands. J Mol Catal B Enzym 38: 37–42.

McCarthy, J.D., Chen, Y., Smyth, K.G. (2012) Differential expression analysis of multifactor RNA-Seq experiments with respect to biological variation. Nucleic Acids Res 40: 4288- 4297.

Ning, C., Yang, K., Xu, M., Huang, L-Q., Wang, C-Z. (2016) Functional validation of the carbon dioxide receptors in labial palps of Helicoverpa armigera moth. Insect Biochem Mol Biol 73: 12–19. 41

Ozakim K., Ryuda, M., Yamada, A., Utoguchi, A., Ishimoto, H., Calas, D. et al. (2011) A gustatory receptor involved in host plant recongnition for oviposition of a swallowtail butterfly. Nat Commun 2: 542.

Pope, M.M., Gaston, L.K., Baker, T.C. (1984) Composition, quantification and periodicity of sex pheromone volatiles from individual Heliothis zea females. J Insect Physiol 30: 943-945.

Rafaeli, A., Gileadi, C. (1995) Modulation of the PBAN-stimulated of pheromonotropic activity in Helicoverpa armigera. Insect Biochem Mol Biol 25: 827-834.

Ramaswamy, S.B., Ma, W.K., Baker G.T. (1987) Sensory cues and receptors for oviposition by Heliothis virescens. Entomol Exp Appl 43: 159-168.

Rogers, M.E., Steinbrecht, R.A., Vogt,R.G. (2001) Expression of SNMP-1 in olfactory neurons and sensilla of male and female antennae of the silkmoth Antheraea polyphemus. Cell Tissue Res 303: 433-446.

Simão, F.A., Waterhouse, R.M., Ioannidis, P., Kriventseva, E.V., Zdobnov, E.M. (2015) BUSCO: assessing genome assembly and annotation completeness with single-copy orthologys. Bioinformatics 31: 3210-3212.

Sun, L., Wang, Q., Zhang, Y., Guo, H., Xiao, Q., Zhang, Y. (2018) Expression patterns and colocalization of two sensory neurone membrane proteins in Ectropis obligua Prout, a geometrid moth pest that uses Type-II sex pheromones. Insect Mol Biol doi: 10.1111/imb. 12555.

Teal, P.E.A., Tumlinson, J.H. (1986) Terminal steps in pheromone biosynthesis by Heliothis virescens and H. zea. J Chem Ecol 12: 353–366.

Xia, Y.-H, Zhang, Y.-N., Hou, X-Q., Li, F., Dong, S.-L. (2015) Large number of putative chemoreception and pheromone biosynthesis genes revealed by analyzing transcriptome from ovipositor-pheromone glands of Chilo suppressalis. Sci Rep 5: 7888.

Yadav, P., Borges, R.M. (2017) The insect ovipositor as a volatile sensor within a closed microcosm. J Exp Biol 220: 1554-1557.

Walter, S.L. (2013) Odorant reception in insects: roles of receptors, binding proteins, and degrading enzymes. Annu Rev Entomol 58: 373-391. 42

Vogel, H., Heidel, A., Heckel, D. and Groot, A. (2010) Transcriptome analysis of the sex pheromone gland of the noctuid moth Heliothis virescens. BMC Genomics 11: 29.

Ye, J., Fang, L., Zheng, H., Zhang, Y., Chen, J., Zhang, Z., et al. (2006) WEGO: a web tool for plotting GO annotations. Nucleic Acids Res 34: W293–297.

Zhang, Y.N., Zhu, X.Y., Fang, L.P., He, P., Wang, Z.Q., Chen, G.et al. (2015) Identification and Expression Profiles of Sex Pheromone Biosynthesis and Transport Related Genes in Spodoptera litura. PLoS One 10: e0140019.

Zhang, Y.N., Zhang, L.W., Chen, D.S., Sun, L., Li, Z.Q., Ye, Z.F. et al. (2017) Molecular identification of differential expression genes associated with sex pheromone biosynthesis in Spodoptera exigua. Mol Genet Genomics 292: 795-809.

Table 2. 1. Assembly results

Average BUSCO result (%) # of contigs Max length length Complete Fragmented Missing

338,754 730 14,225 93.9% 6.3% 0.4%

Table 2. 2. Top 20 differentially expressed genes between female PG and female tarsi

Gene logFC P-Value FDR

Acyl-CoA Delta(11) desaturase (HzeaLPAQ) 7.628 4.24E-66 1.27E-61

Putative fatty acyl-CoA reductase (FAR1) 7.717 1.11E-63 1.67E-59

alcohol oxidase (AO7) 6.985 2.84E-56 8.52E-53

Transmembrane protein 6.851 6.77E-54 1.45E-50

homeobox protein abdominal-B-like 7.389 4.54E-53 7.58E-50

Unknown 5.861 6.25E-48 6.96E-45

ecdysone-responsive G protein-coupled protein-2 6.371 1.16E-46 1.20E-43

Troponin C 7.497 2.94E-46 2.85E-43

cytochrome P450 4V2-like -6.370 2.48E-44 2.26E-41

Facilitated trehalose transporter (Tret2) 6.148 4.11E-44 3.63E-41

Speckle targeted PIP5K1A-regulated poly(A) polymerase 6.960 7.56E-44 6.49E-41

Unknown 7.784 8.87E-44 7.40E-41

Bile salt-activated lipase 6.716 3.60E-41 2.30E-38

cytochrome P450 4V2-like isoform X1 -6.220 8.03E-41 4.82E-38

Unknown -6.300 2.90E-40 1.64E-37

Vitellogenin 5.553 8.40E-40 4.35E-37

Carboxylesterase -5.900 2.17E-39 1.05E-36

General odorant-binding protein (OBP_13) -6.650 4.22E-39 1.95E-36

Glycine receptor -5.390 4.19E-38 1.70E-35

Endocuticle structural glycoprotein SgAbd-5 6.077 5.52E-38 2.18E-35

LogFC = log2 (female PG / female tarsi). LogFC > 0 indicates up regulated in female PG; logFC < 0 indicates down regulated in female PG. FDR = false discovery rate.

Table 2. 3. Top 20 differentially expressed genes between male tarsi and female tarsi

Gene logFC P-Value FDR

Cytochrome P450 CYP341B2 5.63787041 2.94E-60 3.10E-56

Alpha-tocopherol transfer protein-like -5.664287 4.55E-59 3.59E-55

Unknown -6.5170683 4.01E-50 2.53E-46

Cytochrome P450 4C1-like 4.01207307 1.69E-40 6.68E-37

B-cell lymphoma/leukemia 11B -5.5949113 2.33E-40 8.17E-37

nose resistant to fluoxetine protein 6-like -6.4646364 1.31E-37 4.12E-34

Lipase -4.365309 3.47E-35 7.82E-32

Splicing factor 3B -3.3834163 1.83E-30 2.62E-27

Nose resistant to fluoxetine protein 3.61399027 1.29E-28 1.40E-25

Alpha-tocopherol transfer protein-like -3.296347 3.27E-27 3.22E-24

Unknown -3.8397625 9.98E-25 8.29E-22

Presenilin homolog -2.9255116 1.24E-24 1.01E-21

Cytochrome P450 4g15-like 3.13625103 4.26E-24 3.13E-21

Connectin -2.8509878 6.73E-21 3.60E-18

Integrator complex -2.5857153 1.04E-20 5.48E-18

Unknown -5.76572 1.18E-19 5.66E-17

Probable RNA-directed DNA polymerase -2.7722723 1.92E-19 8.91E-17

RNA-directed DNA polymerase 3.09408 1.92E-19 8.91E-17

Unknown -2.3533 3.20E-19 1.46E-16

alpha-tocopherol transfer protein-like isoform X1 -2.427159 3.90E-19 1.76E-16

LogFC = log2 (male tarsi / female tarsi). LogFC > 0 indicates up regulated in male tarsi; logFC < 0 indicates down regulated in male tarsi. FDR= false discovery rate.

Figure 2. 1. Volcano plots for differentially expressed genes between tissues. Full length

ORF genes were selected. Red dot: Significant differential expressed genes with FDR less than

0.05, log2FC larger than 2 or less than -2. Black dot: Non-significant differentially expressed genes. A: Female pheromone gland vs female tarsi. B: Female tarsi vs male tarsi.

Figure 2. 2. Number of differentially regulated genes in different tissues, grouped by gene ontology. All genes are full length ORF with FDR less than 0.05 and LogFC larger than 2 (UP) or less than -2 (DOWN). A. Female pheromone gland compared to female tarsi. B. Male tarsi compared to female tarsi. ** – significant difference, ns – not significant using a Pearson Chi-

Square test in the WEGO program.

Figure 2. 3. Comparison of expression level of candidate genes involved in sex pheromone biosynthesis between tissues based on FPKM values. Error bars show strandard deviations.

Different letters represent significant difference with FDR less than 0.05. No letter means there is no expression in some tissues. A: desaturase, B: fatty acyl-CoA reductase, only those with expression in all tissues shown. C: alcohol oxidase. 49

Figure 2. 4. Phylogenetic analysis of identified desaturases. Each color represents a different type of desaturase found in moths. Numbers in parenthesis are GenBank accession ID.

Desaturases found in this study are were marked with a triangle.

Figure 2. 5. Phylogenetic analysis of identified fatty acyl-CoA reductase. Numbers in parenthesis are GenBank accession ID. Fatty acyl-CoA reductases found in this study are marked with a triangle.

Putative Lepidoptera bitter Fly bitter receptors receptors

DmGr43a-like CO2 receptors

Fly bitter receptors Sugar receptors

Figure 2. 6. Phylogenetic analysis of gustatory receptor. Pink: candidate Lepidoptera bitter receptors, Red: Fly bitter receptors, Green: Sugar receptors, Purple: GR43a-like, Lime: CO2 receptors. All H. zea GRs found in this study are marked with a triangle.

Figure 2. 7. Heat map of selected genes related to sex pheromone production and signal transduction based on gene expression. Genes with an FPKM larger than 4 were selected. 53

CHAPTER 3. IDENTIFICATION OF FATTY ACYL REDUCTASE IN PHEROMONE

GLAND AND TARSI OF THE CORN EARWORM, Helicoverpa zea

Manuscript to be submitted

Xiaoyi Dou, Russell Jurenka

Abstract

Most moths utilize sex pheromones released by the female to attract a mate. Females produce the sex pheromone in the pheromone gland in a biosynthetic pathway which consists of several key enzymes. Fatty acyl-CoA reductase is one of the key enzymes, which catalyzes the conversion of fatty acyl-CoA to the corresponding alcohol, playing an important role in the producing the final proportion of each component. In Helicoverpa zea, (Z)-11-hexadecenal is the major sex pheromone component. A large amount of hexadecanal was also found in female and male tarsi.

In our previous study, we have compared the transcriptome between pheromone glands and tarsi and found 20 fatty acyl-CoA reductases in both tissues. In this study, we functionally characterized three FARs which were expressed at high levels according to the transcriptome of PGs and tarsi.

Fatty acyl-CoA reductase 1 was homologous to other moth pheromone gland specific fatty acyl-

CoA reductases, and it was also present in male tarsi. The functional expression in yeast cells indicates that only fatty acyl-CoA reductase 1 was able to produce fatty alcohols. In addition, a decreased mRNA level of FAR1 in female PGs and male tarsi by RNAi knockdown caused a significant decrease in the production of (Z)-11-hexadecenal in PGs and hexadecanal in male tarsi.

This study is the first to demonstrate the direct function of a fatty acyl-CoA reductase in male tarsi and also confirms its role in sex pheromone biosynthesis in H. zea. 54

Introduction

Sex pheromones in moths play an important role in mate finding. In general, mature females release sex pheromone in the scotophase in a calling behavior to attract conspecific males. The

Type 1 moth sex pheromones usually consist of several components which are fatty acid derivatives, with a functional group that includes fatty alcohols, acetate esters, and aldehydes. In most moths de novo biosynthesis of sex pheromone occurs in specific sex pheromone glands located at the 8th and 9th abdominal segment by a series of enzymes including desaturases, chain- shortening reactions, fatty acyl-CoA reductase, alcohol oxidase and acetyltransferase. Various desaturase families (Albre et al., 2012; Bucek et al., 2015; Foster et al., 1996; Fujii et al., 2015;

Hao et al, 2002; Liénard et al, 2008) and fatty acyl-CoA reductases (FARs) (Antony et al., 2009;

Carot-Sans et al., 2015; Hagström et al., 2012; Liénard et al., 2010; Moto et al., 2003) have been widely studied in the pheromone glands of many moths. However, the acetyltransferase and alcohol oxidase which catalyze the last step of biosynthesis have not been identified at the molecular level in any insect (Jurenka 2004).

The corn earworm, Helicoverpa zea, is a major pest in North America and has a wide host range. H. zea uses two sex pheromone components consisting of (Z)-11-hexadecenal (Z11-16:Ald) and (Z)-11-hexadecenal (Z9-16:Ald) in about a 95:5 ratio (Hillier and Baker, 2016). The biosynthetic pathway begins with the production of palmitic acid through fatty acid synthesis by acetyl-CoA carboxylase and fatty acid synthase (Choi et al., 2002). Then one double bond is introduced into the saturated fatty acid precursor at specific position by a ∆11 desaturase or a ∆9 desaturase. The unsaturated fatty acids are then reduced to the fatty alcohol by FAR and then the alcohol is oxidized by an alcohol oxidase to form the final aldehyde compound. Some minor compounds with different chain length have a β-oxidation process. So far, the FARs from four 55

heliothine moths species have been identified at the molecular level in pheromone glands

(Hagström et al., 2012).

The terminal leg segments (tarsi) are used in walking, but in addition they have other functions.

Many lepidopterans rely on tarsal contact chemoreception using gustatory receptors for discriminating between host and non-host to oviposit eggs (Ozakim et al., 2011; Ramaswamy et al., 1987). Frerot et al. (2013) found that in a day-flying moth, the palm borer, Paysandisia archon, a short-range sex pheromone is produced from the tarsi of males. Recently, a large amount (0.5-

1.5 µg) of hexadecanal (16:Ald) in the male tarsi and lower amounts (0.05-0.5 µg) in female tarsi were found in four heliothine moths, including H. zea. However the exact function of 16:Ald on the tarsi is unknown (Choi et al., 2016).

In our previous study (Dou et al., 2019), we investigated the transcriptomic differences between female pheromone glands and male tarsi. We selected some key genes that could be involved in sex pheromone biosynthesis and conducted the downstream analysis. In this study, we functionally characterized three FARs in vivo and in vitro and found one is involved in sex pheromone production in pheromone glands and 16:Ald formation in tarsi, while the other two genes are not.

Materials and Methods

Insects

Eggs of H. zea were obtained from Frontier Agricultural Sciences (Newark, DE), larvae were reared on an artificial diet (Stonefly Heliothis diet) at 25±1 ℃ under a light:dark cycle 16:8 hr. 56

Pupae were sexed and allowed to emerge separately. A sucrose solution (10%) was provided to adults. Three-day-old virgin adults were used in this study.

Chemicals

Hexadecanoic methyl ester (16:Me), hexadecenoic methyl ester (Z9–16:ME), (Z)

9–oleic methyl ester (9-18: Me) were purchase from Tokyo Chemical Industry (TCI). (Z)–9– tetradecenoic methyl ester (Z9–14:ME) was purchased from Sigma Aldrich. (Z)–11-hexadecenoic acid (Z11-16: Me) was synthesized similar to that described in Jurenka et al. (1994). All compounds used as precursors in the functional assay were ddissolved in 95% ethanol at a 0.02 M stock concentration and diluted to 0.5mM final concentration in the yeast assay.

Cloning of the fatty acyl-CoA reductases

Through the alignment of amino acid sequences of different FARs, specific primers (Table 1) for FAR1 were designed for RACE to get the full length open reading frame. Total RNA was extracted from ten pheromone glands and tarsi with TRIzol reagent (Gibco, Paisley, UK) according to the manufacturer’s instructions and the quantity measured with a Nanodrop 2000/2000c

(Thermo Scientific). One µg of total RNA was used as template for the following PCR amplification. The 5’ RACE and 3’ RACE first-strand cDNA were produced using the SMARTer

RACE cDNA amplification kit (Clontech Laboratories, Inc.) according to the manufacturer’s instructions. The PCR products were gel-purified and cloned into the pGEM-T easy vector

(Promega). Plasmid DNA was collected and sequenced to obtain the full-length FAR1 sequence.

According to the transcriptome obtained previously (Dou et al., 2019), three candidate genes

(FAR1, FAR4, and FAR5) were selected from the transcriptome data, which were highly expressed 57

in PG or tarsi. The full-length sequence of FAR1 was the same as the transcriptome, and the full- length sequences of FAR4 and FAR5 were obtained directly from the transcriptome from three replicates. The sequences were submitted to GenBank under the accession “MN164613 to

MN164615”.

RT-PCR

The first-strand cDNA from female PGs, tarsi and male tarsi was synthesized from 1 µg of total RNA using ProtoScript II First Strand cDNA synthesis kit (NEW ENGLAND BioLabs) in a

20 µl mix according to the protocol. The primers for RT-PCR are shown in Table 1. The PCR conditions were 95 ℃ for 3 min, 30 cycles of 95 ℃ for 30 s, 55 ℃ for 30 s, 72 ℃ for 1 min, and a final extension at 72 ℃ for 10 min. The PCR products were separated by 2% agarose gel electrophoresis.

Functional Assay

The ORFs of three FARs were cloned to pYES 2.1 expression vector (Invitrogen) to form

FAR1-PIB, FAR4-PIB, and FAR5-PIB recombinant DNA. Briefly, the three FARs cDNA were amplified by PCR using the primers in Table 3.1. PCR was performed under conditions at 95 ℃ for 3 min, followed by 35 cycles of 95 ℃ for 30 s, 55 ℃ for 30 s, 72 ℃ for 1 min, and a final extension at 72 ℃ for 10 min. The products were double digested and purified with the gel purification kit (QIAquick gel extraction kit, Qiagen). Then the products were ligated into a

Pyes2.1 vector which was predigested by the same double digestion enzymes, and transformed into JM109 competent cells. Single clones were selected and plasmid DNA was purified with a

QIAprep kit (Qiagen). The plasmids were sequenced in the DNA facility at Iowa State University. 58

Five micrograms of plasmids were transformed into INVSc1 yeast host strain using the S.C.

EasyCompTM kit (Invitrogen) and then selected the transformants on SC-U plates containing 2% glucose for 72 h. Individual colonies were inoculated in 6 mL selective medium ((SC-U)+ 2% glucose) and then incubated for 48 h at 30 ℃ and 200 rpm. The cells were diluted to OD600nm/mL

= 0.4 in a 20 mL SC-U medium with 2% galactose and 0.1% glucose. After incubation for 24 h at

30 ℃ and 200 rpm, the yeast cell culture was dilute to 1: 10 in 2 mL fresh induction medium with

0.5mM methyl-esters as the precursors. The Z9-14:Me, 16:Me, Z9-18:Me, Z9-16:Me, and Z11-

16:Me were used as precursors. All compounds were diluted in 95% ethanol (0.5 M final concentration) with a 0.02 M stock concentration. Cells were incubated for 24 h at 30 ℃ and 200 rpm then pelleted at 2,000 × g and washed with sterile water twice. One mL n-hexane was used to extract the cell pellets and the hexane extracts stored at -20 ℃ until analyzed with gas chromatography-mass spectrometry (GC-MS). A Hewlett Packard 5890 GC coupled to a 5972 mass selective detector was used to determine the production. The column used to separate the extracts was a DB Wax (J&W Scientific, 30mx0.25mm). The GC oven temperature was set as follows: 60 ℃ for 1 min, increase to 230 ℃ by 10 ℃/min, and then held at 230 ℃ for 15 min.

RNAi

The double-stranded RNA was synthesized using MEGAscript RNAi kit (Ambion). The templates contained T7 polymerase at both 5’ and 3’ end. The PCR cycle conditions were set as follows: 95 ℃ for 3 min, 35 cycles of 95 ℃ for 1 min, 60 ℃ for 1 min, 72 ℃ for 1min, and a final elongation at 72 ℃ for 10 min. The purified PCR product was used as templates for dsRNA synthesis. The dsRNA were treated with DNase I and RNase to remove DNA and ssRNA contaminations, followed by purification to remove proteins, free nucleotides and nucleic acid 59

degradation products. The dsRNA was dissolved in elution solution, and the concentrations were measured using Nanodrop 2000/2000c. The enhanced green fluorescent protein (EGFP) dsRNA was used as a control.

Ten µg of dsRNA (2 µg/ul) for EGFP and FAR1 were injected into the 7th and 8th abdominal segments which is close to pheromone glands in newly emerged females The injected females were incubated for 48 h at the regular temperature and then injected with 100 pmol PBAN

(pheromone biosynthesis activating neuropeptide). After 1 h incubation, PGs were removed and extracted into n-hexane containing 150 ng of E4-13: OAc as an internal standard (IS). Similarly, the same amount of dsRNA for EGFP and FAR1 were injected into the abdomen of newly emerged males. After 48 h incubation, tarsi were removed and extracted into n-hexane containing internal standard directly without the injection of PBAN. GC-MS analysis was conducted as described above.

Quantitative real-time PCR

RNAi knockdown effects were measured with quantitative real-time PCR. PGs and tarsi from treated females and males were dissected for total RNA extraction. One µg of total RNA was used for first-strand cDNA synthesis using the same procedures as described above. The cDNA from each treatment was used as template for qPCR. The primers are listed in Table 3.1. qPCR was conducted using SYBR Green Supermix on the Applied Biosystems QuantStudio 3 (Thermo

Fisher Scientific) according to the manufacturer’s protocol. The conditions of thermal cycles were:

95 ℃ for 3 min, 40 cycles of 95 ℃ for 15 s, 60 ℃ for 20 s. Three replicates were used for each sample. The ribosomal protein S7 was used as reference gene. The data were analyzed using the

2-ΔΔCt method (Livak and Schmittgen, 2001). 60

Phylogenetic construction

Sequences used in phylogenetic analysis were based on the Blastp results of FAR1, FAR4 and FAR5 in NCBI database. Multiple sequence alignment was performed by CLUSTALW program, and neighbor joining tree with 1,000 replicates was conducted using MEGA 7 (Kumar et al., 2016).

Results

Fatty acyl-CoA reductases cloning

In the transcriptomic data previously obtained (Dou et al., 2019), we found 20 FARs in female

PGs and male tarsi. We selected FAR1, FAR4, and FAR5 as the candidate genes encoding FAR, because these three genes are highly expressed either in PGs or tarsi. Through the RACE, we identified a FAR which is 100% identical to FAR1 in the transcriptome. We used these three genes

(NCBI GenBank accession numbers MN164613-MN164615) to do the downstream analysis. The open reading frame of FAR1, FAR4 and FAR5 have 1,368 nt, 1,554 nt and 1,566 nt in length, encoding proteins of 456, 518 and 522 aa with predicted molecular weight of 51.63, 59.12 and

59.18 kDa, respectively. FAR1 and FAR4 shared 27.84% aa identity, FAR1 and FAR5 shared

27.78% aa identity, and FAR4 and FAR5 shared 34.64% aa identify. FAR1 shared above 96% aa identical to four other heliothine moths pgFARs while FAR4 and FAR5 were 28% identical to them. Through the prediction of Conserved Domain Database (CDD) in NCBI

(https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi) and InterProtScan

(https://www.ebi.ac.uk/interpro/search/sequence-search), all sequences contained an NAD(P)H- binding region and the C-terminal region of fatty acyl CoA reductase. Also, they have the 61

GXXGXX(G/A) motif at the N-terminus which is similar to the canonical ADP-binding domain involved in the binding of NAD(P)H (Figure 3.1). The amino acid sequences were analyzed using the transmembrane region prediction software TMPRED (Hofmann and Stoffel, 1993) and

TMHMM (Krogh et al., 2001). HzeaFAR1 showed one transmembrane region and HzeaFAR4 and

HzeaFAR5 showed 2 regions (Figure 3.1).

RT-PCR indicates the three fatty acyl-CoA reductases are present in tarsi

The presence of these three FARs in female PGs, tarsi and male tarsi were evaluated by semi- quantitative RT-PCR utilizing the cDNA from the tissues (Figure 3.2). The ribosomal protein S7 which is stably expressed was used as a reference gene (Zhang and Denlinger, 2012). HzeaFAR1,

HzeaFAR4 and HzeaFAR5 were all found in female PGs and tarsi and male tarsi (Figure 3.2), which agrees with the transcriptome data. FAR1 was highly expressed in PG, while FAR4 and

FAR5 were moderately expressed in female tarsi. Male tarsi had moderate expression of FAR1 and

FAR5.

In the phylogenetic tree (Figure 3.3), HzeaFAR1 was clustered together with the FARs from four heliothine moths which could utilize a range of substrates from C8 to C16 (Hagström et al.,

2012). In addition, HzeaFAR1 clustered with the FARs discovered in PGs and involved in pheromone biosynthesis. However, the other two FARs, HzeaFAR4 and HzeaFAR5, were not clustered into the PG specific FARs. 62

Yeast expression and functional assay indicates fatty acyl-CoA reductases 1 produces fatty alcohols

The ORFs of these genes were cloned into pYES2.1 yeast vector and transformed to InvSc1 yeast cells. Empty vector only was used as negative control. Z9-14: Me, Z9-16:Me, Z11-16:Me,

16:Me and Z9-18:Me were used as precursors for the functional assay. No fatty alcohol products were found in the negative control. In yeast expressing FAR1 without the addition of exogenous substrates, various amounts of 10:OH, 12:OH, 14:OH, 16:OH and 9-16:OH were found with

16:OH the highest amount (Figure 3.4). When additional substrates were added as methyl esters increased amounts of Z9-14:OH, Z9-16:OH, and Z11-16:OH were found (Figure 3.5, Table 3.2).

When 16:Me was added, 16:OH did not show a significant increase (Table 3.2). The addition of

Z9-18:Me did not produce the corresponding fatty alcohol. FAR4 and FAR5 expression in the yeast cells did not produce any alcohols either with or without the addition of methyl esters as substrates including 18:Me (data not shown).

RNAi indicates that fatty acyl-CoA reductases 1 is involved in 16: Ald production in male tarsi

Since the expression of FAR1 in yeast cells indicated the reduction of fatty acids to corresponding alcohols, but FAR4 and FAR5 did not, we tested the functional analysis of FAR1 in vivo. We injected dsRNA for FAR1 in newly emerged females and males. The dsRNA for EGFP was injected as a control. The qPCR results indicated that FAR1 expression was significantly decreased after RNAi injection for both female PGs (Figure 3.6A) and male tarsi (Figure 3.7A), compared to the control. Correspondingly FAR1 knockdown also led to a significantly decrease in the amount of Z11-16:Ald in female PGs (Figure 3.6B) and 16:Ald in male tarsi (Figure 3.7B). 63

This showed that FAR1 is involved in the sex pheromone biosynthesis pathway in PG and also in the production of 16:Ald in tarsi. However, the 18:Ald found in male tarsi did not show a significant difference compared to the control (Figure 3.7B), indicating that FAR1 cannot reduce the 18C fatty acid to the alcohol, which agrees with the functional assay in yeast.

Discussion

The first FAR involved in pheromone biosynthesis was identified in B. mori (Moto et al.,

2003). Since then several FARs have been identified in the PGs of Lepidoptera at the molecular level. In addition, PGs transcriptome analysis found many FARs (Ding and Löfstedt, 2015; Li et al., 2015; Zhang et al., 2017). Most of the FARs are not PG specific and are expressed in different tissues (Xia et al., 2015; Zhang et al., 2014). Choi et al (2016) found large amounts of 16:Ald and smaller amounts of 18:Ald in male and female tarsi. We compared the transcriptome of female

PGs, tarsi and male tarsi, and found 20 FARs (Dou et al., 2019). Based on the FPKM values,

HzeaFAR1 was highly expressed in PGs but had lower expression in female and male tarsi. While

HzeaFAR4 and HzeaFAR5 were expressed at higher levels in tarsi than in PGs. According to the sex pheromone biosynthetic pathway described previously (Choi et al., 2002), we assumed that the aldehyde in tarsi is also produced in a similar pathway, but without the involvement of a desaturase.

In this study, we functional characterized three FARs. There are a few studies on the structure of reductase, but it is shown that this enzyme is located in the endoplasmic reticulum in yeast cells

(Hagström et al., 2013). Also there is a general hypothesis that FARs are membrane-bound proteins

(Kolattukudy and Rogers, 1986; Metz et al., 2000). As expected, the protein sequences of three

FARs all predicted one or two transmembrane regions through the analysis of TMPRED and 64

TMHMM software. They all have the conserved region of NAD(P)(+)-binding like all FARs involved in the sex pheromone biosynthesis. Interesting, HzeaFAR1 is clustered together with the other four FARs from heliothine moths as a PGs specific gene, but through RT-PCR, we found its presence in female and male tarsi. Most studies working on identification of FARs did not consider the distribution in tarsi (Antony et al., 2016; Moto et al., 2003), or did not find the pgFAR expressed in legs (Carot-Sans et al., 2015; Liénard et al., 2010). The possible reason is that compounds requiring an alcohol precursor were not found in the legs of these moths. In our study, we found HzeaFAR1 is 96% identical to the FARs from four heliothine moths (Hagström et al.,

2012), 43.76% identical to pgFAR from Y. evonymellus (Liénard et al., 2010) and 32.98% identical to O. scapulalis (Antony et al., 2009), indicating its function should be the same as the other heliothine moths. Using the yeast functional assay, we found HzeaFAR1 could reduce 16:Me, Z9-

14:Me, Z9-16:Me, Z11-16:Me, but not Z9-18:Me, indicating HzeaFAR1, like other pgFARs from heliothines, could not reduce the 18C fatty acid. RNAi also confirmed this conclusion. The knockdown of HzeaFAR1 reduced the amount of 16:Ald in male tarsi and Z11-16:Ald in female

PGs, but 18:Ald did not show significant difference in male tarsi, indicating there is possibly another FAR reducing the 18C fatty acid in male tarsi. We did not check the substrate preference, but based on the amounts of each fatty alcohol product, it seems HzeaFAR1 has the highest activity for Z9-14:acid, which is in agreement with the FARs from heliothine moths (Hagström et al., 2012).

This is the first study to demonstrate that the same FAR is involved in sex pheromone biosynthesis but also in the 16: Ald production in tarsi. 65

Acknowledgments

This work was supported in part by the United States-Israel Binational Agricultural Research and Development Fund (BARD#IS-4722-14) and State of Iowa Funds to R.J., and a grant from the China Scholarship Council to X.D.

References

Albre, J., Liénard, M.A., Sirey, T.M., Schmidt, S., Tooman, L.K., Carraher, C., et al (2012) Sex pheromone evolution is associated with differential regulation of the same desaturase gene in two genera of leafroller moths. PLoS Genet 8: E1002489.

Antony, B., Ding, B-J., Moto, K., Aldosari, S.A., Aldawood, A.S. (2016) Two fatty acyl reductases involved in moth pheromone biosynthesis. Sci Rep 6: 29927.

Antony, B., Fujii, T., Moto, K., Matsumoto, S., Fukuzawa, M., Nakano, R., et al. (2009) Pheromone-gland-specific fatty-acyl reductase in the adzuki bean borer, Ostrinia scapulalis (Lepidoptera: Crambidae). Insect Biochem Mol Biol 39: 90-95.

Bucek, A., Matouskovam, P., Vogel, H., Sebesta, P., Jahn U., Weißflog, J., et al. (2015) Evolution of moth sex pheromone composition by a single amino acid substitution in a fatty acid desaturase. Proc Nat Acad Sci USA 112: 12586-12591.

Carot-Sans, G., Munoz, L., Piulachs, M.D., Guerrero, A., Rosell, G. (2015) Identification and characterization of a fatty acyl reductase from a Spodoptera littoralis female gland involved in pheromone biosynthesis. Insect Mol Biol 24: 82-92.

Choi, M.Y., Han, K.S., Boo, K.S., Jurenka, R.A. (2002) Pheromone biosynthetic pathways in the moth Helicoverpa zea and Helicoverpa assulta. Insect Biochem Mol Biol 32: 1353-1359.

Choi, M.Y., Ahn, S.J., Park, K.C., Meer, R.V., Carde, R.T., Jurenka, R. (2016) Tarsi of male heliothine moths contain aldehydes and butyrate esters as potential pheromone componens. J Chem Ecol 42: 425-432.

Ding, B.J., Lofstedt, C. (2015) Analysis of the Agrotis segetum pheromone gland transcriptome in the light of sex pheromone biosynthesis. BMC Genomics 16: 711.

Dou, X., Liu, S., Ahn, S.J., Choi, M.Y., Jurenka, R. (2019) Transcriptional comparison between pheromone gland-ovipositor and tarsi in the corn earworm moth Helicoverpa zea. Comp Biochem Physiol PartD Genomics Proteomics 31: 100604. 66

Foster, S.P., Roelofs, W.L. (1996) Sex pheromone biosynthesis in the tortricid moth, Ctenopseustis herana (Felder & Rogenhofer). Arch Insect Biochem Physiol 33: 135–147.

Frerot, B., Delle-Vedove, R., Beaudoin-Ollivier, L., Zagatti, P., Ducrot, P.H., Grison, C., et al. (2013) Fragrant legs in Paysandisia archon males (Lepidoptera, Castniidae). Chemoecology 23: 137–142.

Jurenka, R. (2004) Insect pheromone biosynthesis. Topics Curr Chem 239: 97-132.

Hagström, A.K., Walther, A., Wendland, J., Löfstedt, C. (2013) Subcellular localization of the fatty acyl reductase involved in pheromone biosynthesis in the tobacco budworm, Heliothis virescens (Noctuidae: Lepidoptera). Insect Biochem Mol Biol 43: 510-521.

Hagström, A.K., Lienard, M.A., Groot, A.T., Hedenstrom, E., Lofstedt, C. (2012) Semi-selective fatty acyl reductases from four heliothine moths influence the specific pheromone composition. PLoS One 7: e37230.

Hao, G., Liu, W., O’Connor, M., Roelofs, W.L. (2002) Acyl-CoA Z9 and Z10-desaturase genes from a New Zealand leafroller moth species, Planotortrix octo. Insect Biochem Mol Biol 32: 961–966.

Hofmann, K., Stoffel, W. (1993) TMbase: a database of membrane spanning proteins. Biol Chem Hoppe-Seyler 347: 166.

Kolattukudy, P.E., Rogers, L. (1986) Acyl-CoA reductase and acyl-CoA: fatty alcohol acyl transferase in the microsomal preparation from the bovine meibomian gland. J Lipid Res 27: 404–411.

Krogh, A., Larsson, B., von Heijne, G., Sonnhammer, E.L.L. (2001) Predicting transmembrane protein topology with a hidden markov model: application to complete genomes. J Mol Biol 305: 567-580.

Kumar, S., Stecher, G., Tamura, K. (2016) MEGA7: molecular evolutionary genetics analysis version 7 for bigger datasets. Mol Biol Evol 33: 1870-1874.

Liénard, M.A., Strandh, M., Hedenstrom, E., Johansson, T., Löfstedt, C. (2008) Key biosynthetic gene subfamily recruited for pheromone production prior to the extensive radiation of Lepidoptera. BMC Evol Biol 8: 270.

Liénard, M.A., Hagstrom, A.K., Lassance, J.M., Lofstedt, C. (2010) Evolution of multicomponent pheromone signals in small ermine moths involves a single fatty-acyl reductase gene. Proc Natl Acad Sci USA 107: 10955-10960.

Livak, K.J., Schmittgen, T.D. (2001) Analysis of Relative gene expression data using real-time quantitative PCR and the 2-ΔΔCt method. Methods 25: 402-408.

Metz, J.G., Pollard, M.R., Anderson, L., Hayes, T.R., Lassner, M.W. (2000) Purification of a jojoba embryo fatty acylcoenzyme A reductase and expression of its cDNA in high erucic acid rapeseed. Plant Physiol 122: 635–644.

Ozakim K., Ryuda, M., Yamada, A., Utoguchi, A., Ishimoto, H., Calas, D., et al. (2011) A gustatory receptor involved in host plant recongnition for oviposition of a swallowtail butterfly. Nat Commun 2: 542.

Ramaswamy, S.B., Ma, W.K., Baker, G.T. (1987) Sensory cues and receptors for oviposition by Heliothis virescens. Entomol Exp Appl 43: 159–168.

Xia, Y-H., Zhang, Y-N., Hou, X-Q., Li, F., Dong, S-L. (2015) Large number of putative chemoreception and pheromone biosynthesis genes revealed by analyzing transcriptome from ovipositor-pheromone glands of Chilo suppressalis. Sci Rep 5: 7888.

Zhang, Q., Denlinger, D.L. (2012) Dynamics of diapause hormone and prothoracicotropic hormone transcript expression at diapause termination in pupae of the corn earworm, Helicoverpa zea. Peptides 34: 120-126.

Zhang, Y-N., Xia, Y-H, Zhu, J-Y., Li, S-Y., Dong, S-L. (2014) Putative pathway of sex pheromone biosynthesis and degradation by expression patterns of genes identified from female pheromone gland and adult antenna of Sesamia inferens (Walker). J Chem Ecol 40: 439- 451.

Zhang, Y-N., Zhang, L-W., Chen, D-S., Sun, L., Li, Z-Q., Ye, Z-F., et al. (2017) Molecular identification of differential expression genes associated with sex pheromone biosynthesis in Spodoptera exigua. Mol Genet Genomics 292: 795-809.

Table 3. 1. Primers used in this study

Gene Method 5' 3'

F CAAAGCCAGCATCATCCTAGTTCCCTCC RACE R ATC CGG TAC CCA G CG CCC GCT GA

F ATAGGATCC ATGGTCGTTTTAACTTCGAA Full length/RT-PCR R CGCGAATTC TTATTTATTCGTAGATTTCTT FAR1 TAATACGACTCACTATAGGGAGA F GGGCTATCTGTTAGCGAGAG RNAi TAATACGACTCACTATAGGGAGA R CCAGTTGCCTAGCCAACC

F GAATTC ATGTCATCCCCTTCGATTCG FAR4 Full length/RT-PCR R CTCGAG TTAGTCCTGCTTTTCAATGAC

F AAGCTT ATGGCTTCGGAGACCATGT FAR5 Full length/RT-PCR R GAATTC CTACGCGTCCATAGCGATGA

F GACAAGAACCAGCAGACCACTA S7 RT-PCR/qPCR R TTACAAGTAGGGTTCGGGGAAT

TAATACGACTCACTATAGGGAGA F CCCTGAAGTTCATCTGCACC GFP RNAi TAATACGACTCACTATAGGGAGA R GTGCTCAGGTAGTGGTTGTC

Table 3. 2. Percent ratio of alcohols produced in the yeast assay with and without added

substrates

Substrate 10: OH 12:O 14:OH 9- 16:OH 9- 11-16:OH 9-18:OH H 14: 16:OH OH FAR1 None 6±1.1 8±0.6 20±3.0 0 52±5. 14±4.3 0 0 4 16:ME 5±1.0 8±0.3 23±0.8 0 45±1. 19±1.9 0 0 1 Z9-16:ME 1±1.3 4±1.3 23%±0. 0 43±1. 28±0.8 0 0 1 9 Z11-16:ME 2±0.2 5±0.1 13±0.2 0 28±0. 12±0.3 40±3.9 0 6 Z9-14:ME 0.2±0.3 2±0.5 11±0.3 50± 25±0. 12±0.7 0 0 1.3 7 Z9-18:ME 4±1.9 11±0. 25±0.9 0 42±1. 18±3.9 0 0 5 5 Empty none 0 0 0 0 0 0 0 0 vector

Figure 3. 1. Multiple amino acid sequence alignments of HzeaFAR1, HzeaFAR4,

HzeaFAR5 with other functional FAR amino acids from different moths: HarmFAR,

Heliocverpa armigera (GenBank: AFD04728); BmoriFAR, Bombyx mori (GenBank:

NP_001036967); SlitFARI, Spodoptera littoralis (GenBank: CDG50833); YrorFARII,

Yponomeuta rorellus (GenBank: ADD62441); OscaFARXIII, Ostrinia scaplulalis (GenBank:

ACJ06520). The alignment was conducted with ClustalW. The black or grey shaded amino acids are identical residues or conserved regions. Red box: NADPH-binding motif; Blue line: N- terminal Rossmann fold (NAD(P)(+)-binding domain); Green line: Sterile protein domain. Grey box: Transmembrane domains 71

Figure 3. 2. RT-PCR of three HzeaFARs in female PGs, female tarsi and male tarsi. The ribosomal protein S7 was used as a control.

Figure 3. 3. Phylogenetic tree of the fatty acyl reductase from different Lepidoptera species.

The tree was conducted using the amino acid sequences with the neighbor-joining algorithm with the Jones-Taylor-Thornton (JTT) model and 1,000 bootstrap replicates using MEGA 7.0. The

FAR sequences were retrieved from Genbank and EST database with BLASTP searches and

HzeaFAR1, HzeaFAR4 and HzeaFAR5 were used as query. The amino acid sequences were aligned using ClustalW. The GenBank numbers were shown in parenthesis. FAR specific to PGs were highlighted in green. 73

Figure 3. 4. GC-MS analysis of HzeaFAR1 expressed in yeast cells without additional substrates. A: Empty vector only in yeast cells, negative control. B: HzeaFAR1 expressed in yeast cells. E4-13:OAc was used as internal standard (IS). 74

Figure 3. 5. GC-MS result of HzeaFAR1 expressed in yeast cells with various substrates. A:

Z9-14:Me, B: Z11-16:Me, C: Z9-16:Me, D: 16: Me. The top chromatograms are from the yeast cells with empty vector, and below are chromatograms from the yeast cells expressing FAR1.

Figure 3. 6. RNAi knockdown of FAR1 in pheromone gland. A: qPCR determined the dsRNA knockdown efficiency in PGs. B: Z11-16: Ald amount in PG after dsRNA injection.

Pheromone amounts were quantified by GC-MS. dsEGFP treatment was used as control. Two tailed student t-test was used to check significance. *: p< 0.05, **: p< 0.01.

Figure 3. 7. RNAi knockdown of FAR1 in male tarsi. A: qPCR determined the dsRNA knockdown efficiency in male tarsi. B: 16: Ald and 18: Ald amount in male tarsi after dsRNA injection. Aldehyde amounts were quantified by GC-MS. dsEGFP treatment was used as control.

Two tailed student t-test was used to check significance. **: p< 0.01. NS: non-significant.

CHAPTER 4. COMPARISON OF PINK BOLLWORM PHEROMONE GLAND

TRANSCRIPTOME IN TWO POPULATIONS: LABORATORY AND FIELD

Modified from a manuscript published in PLOS ONE 14(7): e0220187

Xiaoyi Dou, Sijun Liu, Victoria Soroker, Ally Harari, Russell A. Jurenka

Abstract

The pink bollworm, Pectinophora gossypiella, is a world-wide pest of cotton and in some parts of the cotton growing region is controlled by the mating disruption technique using synthetic sex pheromone. The sex pheromone consists of two compounds, (Z,Z)- and (Z,E)-7,11- hexadecadienyl acetates, in about a 50:50 ratio. However, recently, a population with sex pheromone compound ratios of about 62:38 were found in cotton fields that use mating disruption in Israel. To investigate how the change developed, we compared the pheromone gland transcriptomes between a reference laboratory population and a population obtained from an

Israeli cotton field utilizing mating disruption. We analyzed four biological replicates from each population and found transcripts encoding 17 desaturases, 8 reductases, and 17 candidate acetyltransferases in both populations, which could be involved in sex pheromone biosynthesis.

The expression abundance of some genes between the two populations was different. Some desaturases and candidate acetyltransferases were found to have mutated in one of the populations.

The differentially expressed genes play potential roles in sex pheromone biosynthesis and could be involved in causing altered female sex pheromone ratios in the field population. 78

Introduction

The pink bollworm (PBW), Pectinophora gossypiella (Lepidoptera: Gelechiidae), is a key pest of cotton in the old and new world (Henneberry et al., 2007). The sex pheromone of PBW females consists of two compounds, (Z,Z)- and (Z,E)-7,11-hexadecadienyl acetates (Z,Z-7,11-

16:OAc and Z,E-7,11-16:OAc) in about a 50:50 ratio (Hummel et al., 1973). The biosynthesis starts with the production of the saturated fatty-acid, stearic acid (Foster and Roelofs, 1988)

(Figure 4.1), from the catalysis of acetyl-CoA by acetyl-CoA carboxylase (ACC) and fatty acid synthase (FAS). Then a double bond is introduced into the fatty acid chain at the ∆9 position by a

Z9-desaturase to form oleic acid. After peroxisomal chain shortening by 2-carbons, another double bond is introduced by a Z11-desaturase producing both the Z and E isomers. Then the carbonyl group is modified to form a primary alcohol by fatty acyl reductase (FAR) and subsequently modified to an acetate ester by a fatty alcohol acetyltransferase (FAT). The pheromone biosynthetic pathway has been investigated using stable isotope precursors (Foster and Roelofs,

1988), but none of the genes encoding these enzymes have been identified in the PBW.

Mating disruption (MD) is an environmentally safe pest control method that has allowed growers to significantly reduce insecticide use and is now widely applied for the control of various moth pests (Evenden, 2016; Harari et al., 2007). Typically, mating disruption is achieved when the pheromone is released at a high dose in the active space of the pest, which negatively affects the ability of males to locate females. This technique has been applied in cotton fields all over the world including Israel and the USA as an effective control measure (Henneberry et al., 1981). In

Israel, practically all cotton fields are treated with sex pheromone in mating disruption, supported by occasional use of insecticides when the pest population levels rise. 79

In Israel, recent repeated outbreaks in the pink bollworm population have suggested a change in population characteristics. Comparison of pheromone gland extracts of females from a recent field population outbreak to laboratory-reared females supports the possibility of a shift in sex pheromone characteristics [Harrari et al., unpublished]. Based on the sex pheromone biosynthetic pathway, we hypothesized that changes in key enzymes involved in the sex pheromone biosynthetic pathway (Foster and Roelofs, 1988) could result in changes in pheromone ratios, particularly focusing on the desaturase, reductase, and acetyltransferase enzymes (Figure 4.1). In this study, using the Illumina HiSeq 3000 platform, we compared the pheromone gland transcriptome between a reference laboratory population and a population obtained from cotton fields in which mating disruption has failed. We analyzed four biological replicates from each population and found transcripts encoding 17 desaturases, 8 reductases, and 17 candidate acetyltransferases. The expression abundance of some genes between the two populations was significantly different, which could be involved in causing altered female sex pheromone ratios.

Materials and Methods

Insect collection and pheromone gland extraction and analysis

Two populations were maintained in the laboratory in Bet-Dagan, Israel as described (Bartlett and Wolf, 1985). The laboratory (Lab) population has been reared in the laboratory without any prior exposure to synthetic pheromone. The field (Field) population originated from a PBW infested field near Ein Shemer, Israel in which mating disruption was utilized. Males and females were sexed in the last larval stage by a black line on the 6th abdominal segment representing the developing testicles. Males and females were housed separately and newly-emerged moths were removed daily and placed into age cohort single sex cages. Adult moths were fed on ~10% sucrose 80

solution provided ad libitum. Pheromone glands from 3-day-old virgin females were removed 2 hours after the start of the scotophase when pheromone titers were the highest (Rafaeli and Klein,

1994). The glands were placed in hexane containing 25 ng tridecyl acetate as an internal standard and removed after twenty minutes. Gland extracts were sent to the Jurenka lab in the United States by express courier. Pheromone amounts and ratios were determined using a Hewlett Packard 5890

GC coupled to a 5972 mass selective detector. The column used to separate the extracts was a DB

Wax (J&W Scientific, 30mx0.25mm). The GC was programmed at 60 ºC for one minute, subsequently ramping at 5 ºC/min to the final temperature of 230 ºC which was held for 15 minutes.

The mass spectrometer was set in single ion monitor mode for ions 43, 55, 67 and 81 (the 4 most abundant ions of Z,Z- and Z,E- 7,11-16:OAc) and 43, 55, 61, and 69 (the 4 most abundant ions of tridecyl acetate). A 2-tailed students t-test (Microsoft Excel) was used to determine statistical significance.

RNA isolation, cDNA library construction and Illumina sequencing

Ten pheromone glands from each population were removed from 3-day-old females during the third hour of scotophase, which is the peak period of pheromone production. Pheromone glands were immediately placed in RNAlater and frozen to -80°C and then shipped to the Jurenka lab in the United States by express courier. Total RNA was isolated using TRIzol regent (Invitrogen,

Carlsbad, CA, USA) according to the manufacturer’s protocol. The quantity of RNA was determined using the 2100 Bioanalyzer (Agilent Technologies). One mRNA library and three stranded total RNA-Seq libraries were prepared in each population by the DNA facility of Iowa

State University, Ames, Iowa, USA. The library preparations were sequenced on an Illumina

HiSeq 3000 platform. The stranded total RNA-Seq libraries were sequenced with 150 pair-end and 81

the mRNA library with 100 single reads. All sequencing reads were submitted to the SRA of NCBI under the accession number “SRP140160”.

De novo assembly of short reads and gene annotation

The quality of all raw reads was checked using FastQC (Babraham Bioinformatics, UK). Low quality sequences and adaptors were removed using the Fastx-toolkit (Hannon Lab, CSHL, USA) and Trim Galore! (Felix Krueger, Babraham Bioinformatics). The de novo assembly was carried out using the merged reads and reads from each library respectively with the short reads assembling program Trinity (Grabherr et al., 2011). After Trinity assembly, the resulting sequences were then processed using CAP3 with default parameters (Huang et al., 1999) in order to decrease the redundance of BLAST searches. The resulting clusters and singletons of more than

200 bases were locally searched against the NCBI non-redundant protein database using the

BLASTx program, to obtain protein annotations of the assembled contigs.

Gene Ontology terms were performed by the Blast2GO program (Conesa et al., 2005) and the GO functional classification was obtained using WEGO software (Ye et al., 2006).

Expression abundance analysis

The expression abundance of the transcripts was calculated using the method of RNA-Seq by

Expectation-Maximization (RSEM) with the Trinity model. We use the RPKM (Reads Per

Kilobase per Million mapped reads) value as the abundance level. The differential expression between the two populations was measured by using the multiple test false discovery rate (FDR) calculation in the R program package ‘edgeR’ with the contigs derived from the merged reads.

Identification of candidate genes involved in sex pheromone biosynthesis

We focused on several important genes, including acetyl-CoA carboxylase, limited β- oxidation enzymes, fatty acid synthases, desaturases, reductases and acetyltransferases. First we started by selecting the transcripts that encode these genes from the BLASTx results. Then we translated these transcripts to their corresponding proteins using the TransDecoder program in

Trinity. The encoding proteins were used in BLASTp to identity the genes based on the homology to the NCBI non-redundant proteins database. The amino acid sequence alignment was conducted using BioEdit (http://www.mbio.ncsu.edu/bioedit/bioedit.html).

Relative expression of several candidate genes by qPCR

One µg of total RNA was used for first-strand cDNA synthesis using ProtoScript® II first strand cDNA synthesis Kit (NEW ENGLAND BioLabs inc.) according to the protocol. The cDNAs from replicates of each population were used as templates for qPCR. Primers were shown in Supplementary Table 4.1. qPCR was conducted using SYBR Green Supermix on the

Applied Biosystems QuantStuidio 3 (Thermo Fisher Scientific) according to the manufacturer’s protocol. The conditions of thermal cycles were: 95 ℃ for 3 min, 40 cycles of 95 ℃ for 15 s,

60 ℃ for 20 s. The elongation factor 1 delta was used as reference gene. The data were analyzed using the 2-ΔΔCt method.

Phylogenetic analysis

Phylogenetic analysis was conducted with two genes involved in the sex pheromone biosynthetic pathway, desaturase and reductase. Here we imported 66 desaturases sequences and

70 identified reductases from other species and the genes we found in the PBW transcriptome. The 83

amino acid sequences were aligned by BioEdit program. The phylogenetic trees were constructed using the neighbor-joining method implemented in MEGA7 with default setting and 1000 bootstrap replicates (Kumar et al., 2016).

Statistics

The significance of differential expressed genes was calculated in R program through the FDR value of multiple comparison (https://www.R-project.org). FDRs less than 0.05 were considered as significantly differential expression between populations. Other statistical comparisons were calculated in Microsoft Office Excel using a two-tailed Student’s t-Test. Fold changes were calculated based on the RPKM value in two populations after converting to the log2 scale.

Results

Analysis results of pheromone gland extracts from the Lab and Field population is shown in

Table 4.1. There was not a statistical difference in the amount of the ZZ and ZE isomers but there was a statistical difference in the ratio of the ZZ and ZE isomers. The Field population had a higher

ZZ ratio than the Lab population.

Illumina sequencing and sequence assembly

Illumina sequencing of cDNA libraries prepared from the mRNA or total RNA of the pheromone glands of two PBW populations was conducted. To assess the completeness of the assembled data, the transcripts were analyzed using the BUSCO program (Benchmarking

Universal Single-Copy Orthologs) using a database of arthropod genes (Simao et al., 2015) (Table 84

4.2). These results suggest that the quality of the sequencing assembly was acceptable for both populations.

We selected the first replicate from both populations to conduct a BLASTx search with the cut-off E-value of 1.0E-5 against non-redundant protein databases in NCBI. BLASTx hits of

50,275 and 61,844 transcripts (44% and 40%) from the Lab and Field populations respectively were found. Both populations had the same top hit species. The highest hits were to Bombyx mori

(Lab: 12,424 transcripts (25%), Field: 12,863 transcripts (21%)), followed by Danaus plexippus

(9,628 hits (19%) and 10,009 hits (16%)).

All the transcripts from the two populations were annotated into different functional groups according to Gene Ontology analysis (Supplemental Figure 4.1). 70,204 (61%) Lab and 82,604

(53%) Field transcripts were annotated into one or more GO categories (Supplemental Figure 4.1).

Of the annotated transcripts, the most abundantly represented categories were “binding”, “cellular process”, “cell”, “cell part”, “metabolic process” and “catalytic activity” (more than 6,000 transcripts). In total, in Lab and Field, 27,879 and 31,534 annotated transcripts aligned to cellular component, 23,659 and 28,723 to the biological process, 18,666 and 22,347 to molecular function, respectively.

Differential expressed gene analysis

We used the edgeR program to do the differential expression analysis at the gene level using the full length of the contigs derived from all reads. After filtering with an absolute value of log2

(fold change) larger than 2 parameter and an FDR less than 0.05, there were 88 differentially expressed genes between the Lab and Field populations (Supplemental Figure 4.2). Forty genes were upregulated in the Field population, while 48 were downregulated. In the upregulated genes, 85

only 21 (52.5%) were annotated, while the remaining upregulated genes (47.5%) were unknown.

Number of differentially expressed genes that were classified into biological regulation, cellular process, transporter activity, etc. are similar between populations (Figure 4.2). In the downregulated genes, 20 genes (41.7%) were annotated, while the rest (58.3%) were unknown.

The 88 differentially expressed genes are shown in Supplemental Table 4.2, and none of the genes are directly involved in pheromone biosynthesis.

Putative genes related to sex pheromone biosynthesis

The biosynthetic pathway (Figure 4.1) includes the actions of acetyl CoA carboxylase (ACC), fatty acid synthase (FAS), fatty acid desaturase (DES), β-oxidation enzymes, fatty acyl reductase

(FAR), and fatty alcohol acetyltransferase (FAT). Based on BLASTx search annotation, we found members of candidate genes involved in the production of PBW sex pheromone, and then compared those genes between the two populations. In the transcriptomes, we found 1 ACC, 4

FASs, 17 DESs, 8 FARs and 17 FATs in both populations (Table 4.3). In addition, there were 17 transcripts in each population encoding putative β-oxidation enzymes, including 6 acyl-CoA oxidases (ACO), 2 enoyl CoA hydratases (ECH), 4 acyl-CoA dehydrogenase (ACD), 1 3-ketoacyl-

CoA thiolase (3-KCT) and 4 3-hydroxyacyl CoA dehydrogenase (3-HCD) (Table 4.3). We also found several G-protein coupled receptors that could be involved in regulations of sex pheromone production including 2 Pheromone biosynthesis activating neuropeptide receptors (PBANrs), 1 diapasue hormone receptor (DHr), 1ecdysis triggering hormone receptor (ETHr), 4 octopamine receptors (Octor), 4 (Lab) or 5 (Field) sex peptide receptors (SPr) (Supplemental Table 4.3). In addition, relative to pheromone transport, we found 9 chemosensory proteins (CSPs) and 7 odorant binding proteins (OBPs) (Supplemental Table 4.3). The comparison of amino acid sequences and 86

abundance levels of these transcripts based on RPKM values are shown in Table 4.4 and

Supplemental Table 4.4. However, none of them showed significant difference between populations (FDR >0.05).

Acetyl-CoA carboxylase (ACC)

ACC catalyzes the carboxylation of acetyl-CoA to form malonyl-CoA, which is one of the first reactions in the biosynthetic pathway. In the PBW pheromone gland, we found one transcript encoding ACC with a complete ORF (amino acid length 2394), in both the Lab and Field populations. The ACC shows 89% identity to Papilio xuthus ACC (GenBank: XP_013176189)

(Table 4.3). Based on RPKM value, the log2 fold change is 0.4, which shows no difference (Table

4.4).

Fatty acid synthase (FAS)

FAS catalyzes the synthesis of saturated fatty acid (stearic acid) from acetyl-CoA and malonyl-CoA. We found 4 putative FAS-like partial transcripts (Table 4.3). They all showed 100% identity between the two populations. FAS2 and FAS4 showed high similarity (78%) to

Helicoverpa species FAS (GenBank: AKD01761, XP_021186732). FAS4 had the highest RPKM value than the other three FAS, and it is lower expressed in Field population (Log2 FC = 1.14)

(Table 4.4).

β-oxidation enzymes

Once the double bonds are introduced into the fatty acyl chain, the chain will be shortened by β-oxidation enzymes to form the proper length of fatty acid precursors. This step is the action of a series enzymes, working sequentially and forming a reaction spiral. The first step is converting the acyl-CoA into E2-enoyl-CoA by acyl-CoA oxidases (ACO) and acyl-CoA 87

dehydrogenases (ACD). In PBW transcripts, we found 6 candidate ACOs in both populations

(Table 4.3). All of them show 100% identity based on amino acid comparison, and the abundance levels were similar except ACO4 (log2 FC= -1.46) and ACO5 (log2 FC= -1.17) (Table 4.4). Three of them are full sequences (ACO2, ACO3, ACO6) and three are partial sequences with different length in the two populations. The amino acid sequence of Lab_ACO5 is 87% identical to the predicted ACO of A. transitella (Protein ID: XP_013188649), while Field_ACO5 shows 78% similarity to the putative ACO of P. xuthus (Protein ID: KPJ00249). Four ACDs were found from both populations. The amino acid alignment shows 100% identity, and their RPKM values had no significant differences between the two populations. ACD1 and ACD3 show 75% and 92% amino acid identity to the short-chain ACD from S. litura and H. virescens, respectively, and

ACD4 has 80% amino acid identity to the S. litura very long chain ACD.

The next step is that enoyl-CoA hydratase hydrated E2-enoyl-CoA to L-3-hydroxyacyl-CoA.

There are two categories of enoyl-CoA hydratases identified in mitochondria. One is specific to crotonyl-CoA (4C), the other is hydrating medium-chain and long-chain substrates. In this study, we found two ECHs with the full length of 278 and 332, homologous to the ECH of S. litura and

P. xuthus with 81% and 74% identity, respectively (Table 4.3). Between these two populations, the amino acids were 100% identical and the abundance level were similar as well (Table 4.4).

The third step of β-oxidation is the production of 3-ketoacyl-CoA from L-3-hydroxyacyl-

CoA by L-3-hydroxyacyl-CoA dehydrogenase. There are three kinds of L-3-hydroxyacyl-CoA identified in mitochondria that are specific to fatty acyl chain-length. Four 3-HCDs were found from the transcripts of PBW with full sequences (Table 4.3). The amino acids show 100% identity in Lab and Field populations. The 3-HCD3 has the highest abundant level (RPKM ~ 30) among the 4 candidates. It is homologous to the predicted 3-HCD of A. transitella (Protein ID: 88

XP_013190290) with 90%, respectively. The expression levels based on the RPKM values are very low (around 10) (Table 4.4).

The last enzyme in the limited chain shortening sequence is a thiolase that cleaves the 3- ketoacyl-CoA between its α- and β-carbon atoms, making the chain two carbons shorter. We found one full sequence of 3-ketoacyl-CoA thiolase with 400 amino acids in the two populations, and it is homologous to B. mori 3-ketoacyl-CoA thiolase with 86% identity (Table 4.3). The transcript abundance level is similar between populations (Table 4.4).

Desaturase (DES)

Desaturases introduce double bonds into the fatty acid chain at specific positions. These desaturases were identified based on homology to other insect desaturases that have three histidine boxes with eight histidine residues that are involved in creating essential metal complexes. In PBW, at least two desaturases are involved in sex pheromone biosynthesis. A ∆9-desaturase introduces a double bond into the 18C saturated fatty acid and after limited chain shortening, a ∆11-desaturase introduces another double bond into the chain, generating both Z,Z-7,11-16:acid and Z,E-7,11-

16:acid precursors.

We found 17 transcripts encoding desaturases in Lab and Field populations. Through amino acid comparison, most of the candidate desaturases show 99% to 100% identity between the two populations. One desaturase with variations between the two populations is Lab_DES16 which had 94% identity with Field_DES16. Lab_DES16 encodes a protein with 66% similarity with an unknown desaturase of Danaus plexippus. It was slightly higher expressed in the Field population

(log2 FC= -0.62). The rest of the desaturases had 100% amino acid identity between the two populations and their abundances were at a similar levels except DES14 (log2 FC= -1.46) (Table

4.4) which was more abundant in the Field population. DES9 transcript was also more abundant 89

in the Field population (log2 FC= -2.02). However, DES5 was up regulated in the lab population with log2 FC= 1.47.

The most abundant desaturase transcript is DES2 (RPKM ~ 2300) that has a 79% identity with the ∆9-desaturase of Epiphyas postvittana. This desaturase is probably the ubiquitous ∆9- desaturase found in other tissues and could produce oleic acid. Sequence alignment with other

Lepidoptera ∆9 desaturases showed the conserved histidine rich motifs, four transmembrane domains, and the NPAE signature motif (Figure 4.3). The next most abundant transcripts encoding desaturases include DES6 (RPKM ~300), DES8 (RPKM ~ 400), and DES12 (RPKM ~ 100). DES6 and DES8 had 62% and 94% amino acid identity with ∆11-desaturases from the moths

Choristoneura parallela and Amyelois transitella, respectively. These transcripts could encode the

∆11-desaturase in PBW to form the Z,Z-7,11-16:acid and Z,E-7,11-16:acid. The three histidine- rich motifs and four transmembtrane domains were found in the sequences but specific signature motif varied in the two amino acid sequences (Figure 4.3). DES8 aligns with several desaturases that were found to be nonfunctional in a heterologous expression assay using yeast cells (Liu et al., 2004). In DES8 the first histidine rich motif contains a cysteine that could potentially interfere with the di-iron binding at the catalytic site (Ding et al., 2016). Relative expression level between

DES6 and DES8 showed there is no significant difference (Figure 4.4), which was in agreement with the RPKM values. Further studies on the proteins encoded in these transcripts are required to confirm their enzymatic roles.

The phylogenetic analysis of desaturases with other moth desaturases is shown in Figure 4.5.

DES5/DES17, DES2/DES10/DES11/DES16, DES12, DES13, DES3/DES6/DES8 and DES1/

DES7/DES9 are clustered in the clades of ∆6, ∆9 (C18>C16), ∆9 (C16>C18), ∆9 (C14-26), ∆11 and ∆14, respectively. DES4, DES14 and DES15 were not similar to any class of desaturases. 90

Fatty Acyl-CoA Reductase (FAR)

FARs catalyze the reduction of fatty acyl-CoA to the corresponding fatty alcohol (Moto et al.,

2003). In the PBW pheromone gland transcriptome, we found 8 transcripts homologous to known insect FAR genes (Table 4.3). Comparison of the amino acid sequence between the two populations shows 100% identity for all 8 FARs (Table 4.4). We did not find the full ORF of

Field_FAR3, but interestingly, the amino acid length is longer than the amino acid from the full

ORF of Lab_FAR3, which indicates that Field_FAR3 covers Lab_FAR3. Among these transcripts, four encode proteins that have 77% - 91% identity with fatty acyl reductases of Spodoptera exigua.

The rest have various similarities with Bombyx mori, Helicoverpa armigera and Ostrinia nubilalis.

FAR5 and FAR8 had highest abundance in PBW PG (RPKM ~ 30), while FAR5 was higher expressed in the Field population (log2 FC= -1.06). Relative expressions were checked by qPCR for these two transcripts, none of them showed significantly different (Figure 4.4). The RPKM values were also not significantly different. Other FARs had a low abundance with RPKM values of less than 10 and with 100% AA identity (Table 4.4).

Based on the phylogenetic analysis of moth FARs (Figure 4.6), two candidate FARs were identified to be likely involved in pheromone biosynthesis, FAR7 and FAR8. FAR7 forms a clade with the identified FARs found in pheromone glands of Helicoverpa (Hagstrom et al., 2012),

Spodoptera (Carot-Sans et al., 2015), Agrotis, Yponomeuta (Lienard et al., 2010) and Bombyx

(Moto et al., 2003). FAR8 is in the clade of Ostrinia (Lassance et al., 2013) reductases.

Fatty Acetyltransferases (FAT)

Fatty acetyltransferases convert fatty alcohols to acetate esters. This gene has not been identified in any insects at the molecular level (Jurenka, 2004). In the PBW PG transcriptome, 17 candidate FATs were found (Table 4.3) based on BLASTp search results, and further filtering to 91

include only potential transferases that have transmembrane domains using the prediction programs TMHMM Server and TMpred Server. Some of the transcripts hit to N- or S- acetyltransferase, but the biochemical function for these acetyltransferases has not been confirmed so we included them here. We found 14 and 15 full sequences in the Lab and Field populations, respectively. All genes were 100% identical between the two populations, except for FAT15 and

FAT17 (Table 4.4). The highest identity of Lab_FAT15 to Field transcripts is Field_FAT15, which shows 88% similarity. Lab_FAT15 is homologous to the acyltransferase AGPAT3 of H. subflexa with 93% identity, while Field_FAT15 showed 91% similarity to the AGPAT alpha-like of S. litura (Gene Bank: PCG64070). Lab_FAT17, as the partial sequence, showed 86% similarity with

Field_FAT17. Lab_FAT17 was homologous to the hypothetical protein of H. virescens (Gene bank: PCG80023) with 91% identity, while Field_FAT17 showed 87% identity to the same entry.

These two genes had low abundance levels (RPKM <1). The most abundant genes were FAT3 and FAT7 with the log2 Fold changes larger than 1.5, indicating lower expression in the Field population. Also, the abundance levels of FAT8, FAT10, FAT11, FAT17 in the Field population were lower (Table 4.4), while FAT6 and FAT16 had higher abundance in the Field population (log2

FC= -1.57, -3.73).

Identification of putative genes related to receptors

The production of sex pheromone is regulated by PBAN, which is released from the subesophageal ganglion into the hemolymph and binds to the PBAN receptors in the membrane of pheromone producing cells. We found two transcripts, PBANr1 and PBANr2 (Supplemental

Table 4.3), encoding proteins highly homologous to PBAN receptors from other moths. Amino acids sequence comparison shows they are 100% identical between the two populations but with 92

low abundance (RPKM < 3) (Supplemental Table 4.4). .PBANr2 shows higher expression in Field population (log2 FC = -2.2). In addition, we also found some G-protein-coupled receptors that show high homology to DHr of H. armigera (Protein ID: ANB78221), ETHr of D. plexippus

(Protein ID: OWR50706), Octor of P. xuthus (Protein ID: XP_013165403, XP_022827336) and

SPr of P. xuthus A. transitella S. litura (Protein ID: XP_013178706, XP_013187671,

XP_013195850, XP_022817535) (Supplemental Table 4.3). These transcripts all have low abundance level in both populations. The amino acid sequence comparison between these two populations shows 100% identity except for the Field_SPr5 which is 83% identity to Lab_SPr4

(Supplemental Table 4.4).

Pheromone biosynthesis is initiated by release of PBAN from the subesophageal ganglion that will then bind to its GPCR receptor on pheromone gland cells. In moths 3 PBAN variants have been identified that vary in the C-terminal length and sequence (Lee et al., 2012). We found two

PBAN receptors which correspond to the B and C variants. There was no difference between the populations in these PBAN receptors except for a statistical difference between Lab and Field transcripts abundance. In addition, we found transcripts encoding for four other GPCRs: DHr,

ETHr, sex peptide receptor, and octopamine receptor. It is interesting to note that the DHr and

ETHr belong to the same family of receptors as does PBAN (Jurenka, 2015). The sex peptide has been implicated in termination of pheromone production (Hanin et al., 2012), while octopamine could modulate pheromone production (Rafaeli and Gileadi, 1995).

Identification of putative carrier proteins

The odorant binding proteins (OBP) and the chemosensory proteins (CSP) are involved in olfaction and contact chemosensation (Tegoni et al. 2004). OBPs or CSPs capture chemical 93

volatiles and transport them through the aqueous lymph to the olfactory receptors (Wojtasek and

Leal, 1999). In PBW PG transcripts, we found 9 CSPs and 7 OBPs (Supplemental Table 4.3). All of them are 100% identity between the two populations. And their expression level shows the same level except OBP4 with log2 FC = 2.08 (Supplemental Table 4.4).

Discussion

The pink bollworm as a world-wide pest, has been well controlled by the mating disruption technique using artificial sex pheromone in Israel until recently when repeated outbreaks have been documented (Harari et al., unpublished). We have found that the females from the Field populations are producing a higher ratio of the ZZ isomer (Table 4.1) and that males can find these females when exposed to mating disruption pheromone (Harari et al. unpublished). Therefore to understand how females could shift the ratio of sex pheromone components we undertook a transcriptome study to identify the genes that are involved in the sex pheromone biosynthetic pathway and transport in the pheromone gland, similar to studies in other moths (Antony et al.,

2015; Ding et al., 2015; Grapputo et al., 2018; Gu et al., 2013; Li et al., 2015; Lin et al., 2018;

Zhang et al., 2017).

By sequencing the PBW pheromone gland transcriptome of Lab and Field populations, we found 64 transcripts encoding enzymes that are putatively involved in pheromone biosynthesis.

This is the first study reporting the key enzymes involved in PBW sex pheromone biosynthesis along with a comparison between Lab and Field populations. Although most transcripts were identical between populations, there were some differences in transcripts encoding desaturases and acetyltransferases. For desaturases, Lab_DES16 had a 94% identity to Field_DES16 in all four 94

replications. DES16 is likely encoding a ∆9 desaturase and their abundance level based on RPKM values is very low indicating that this desaturase is probably not involved in pheromone biosynthesis. The most abundant desaturase transcripts were DES2, DES6 and DES8 which may encode ∆9 and ∆11 desaturases, respectively, because their sequences are similar to functionally identified desaturase sequences from other moth pheromone glands (Knipple and Roelofs, 2003).

However DES8 aligns with other sequences that were found to be nonfunctional as a desaturase in the yeast expression system (Albre et al., 2012). DES6 is expressed at a lower level in the Lab population, indicating the possible role of this ∆11 in changing sex pheromone ratio, although a functional analysis needs to be conducted to confirm the type of desaturase encoded by each transcript.

The Z and E ∆11 isomers could be produced by one desaturase or by separate desaturases.

DES6 was the only desaturase that had relatively high transcript abundance and aligned with other functionally described ∆11 desaturases, indicating that this desaturase could produce both the E and Z ∆11 isomers. Some ∆11 desaturases have been shown to produce only a Z or E isomer

(Knipple and Roelofs, 2003), and some desaturases are bifunctional and could produce both isomers (Hao et al., 2002; Liu et al., 2004). It has been demonstrated that a single amino acid substitution could switch a ∆11-desaturase that produces primarily Z isomer to one that produces mostly the E isomer (Ding et al., 2016). We did not find single amino acid substitutions in the comparison of the highly expressed desaturases so other mechanisms must account for the change in the pheromone ratios we have observed in the Field population.

We found two FATs that differed in amino acid sequence between the two populations.

FAT15 and FAT17 had 88% and 86% identity between the two populations. However, the RPKM values were very low in both populations indicating that they may not be involved in pheromone 95

biosynthesis. The most abundant FATs were FAT3 and FAT7 in the Lab population but were statistically lower in the Field population. In the Field population there were 4 FATs (FAT3, FAT5,

FAT7, FAT13) that were the most abundant transcripts with RPKM between 30 and 50. These are all putative FATs because none have been identified at the molecular level in moth pheromone glands as the actual transferase that produces acetate esters.

The other major enzyme involved in pheromone production is the FAR that produces the alcohol required for the FAT to make acetate esters. We found 8 FARs in PBW pheromone glands with no sequence differences found between the two populations. The two most abundant transcripts were FAR5 and FAR8 for both populations with a significant difference in FAR5 that was higher in the Field population. Either of these could be involved in producing the alcohols required for acetate ester production since they are related to FARs identified from other moth pheromone glands.

Moreover, we found 9 CSPs and 7 OBPs from both populations in all 4 replications that had

100% identity and their expression level was also similar. These proteins could be involved in the transport of pheromone components to the surface of the pheromone gland for release. However we did not find any differences in the transcripts encoding these proteins indicating that the increased ratio of ZZ isomers in the Field populations may not be caused by the differential transportation of sex pheromone. In addition, in the analysis of differentially expressed genes, we found many genes with an unknown annotation. These unknown genes with differential expression may play roles in changing the sex pheromone ratio.

The increase in the ZZ isomer that we have found in the field population could be due to variation in abundance of key enzyme(s) found in the biosynthetic pathway. Desaturase abundance with an increase in the ∆11 desaturase that produces the Z isomer is possible. Variation in putative 96

FATs were found between the two populations, and one or more of these could be involved in preferentially acetylating the ZZ isomer. Tortricid moths utilize a FAT that prefers the Z isomer of ∆11 fatty alcohols (Jurenka and Roelofs, 1989) and a similar FAT could be involved in producing the increased ZZ isomer in the field PBW. An increase in the abundance of one of the

FARs that forms the ZZ alcohol could also be involved in increasing the ZZ isomer. A combination of enzymes that favor the ZZ isomer could also account for the increase in this isomer in the field population.

Another interesting finding is that the relative expression level of enzymes involved in pheromone biosynthesis are not similar with all enzymes in the biosynthetic pathway. The relative expression levels apparently decline in enzymes as they occur later in the biosynthetic pathway.

The desaturases are expressed at the highest level whereas the next enzyme in the pathway, the fatty-acid reductase is expressed at a lower level. This phenomenon has also been found in transcriptome studies of other moth pheromone glands (Cha et al., 2017; Li et al., 2017; Xia et al.,

2015; Zhang et al., 2014; 2015). The next enzymes in the pathway are either an alcohol dehydrogenase or an acetyltransferase, neither of which has been identified at the molecular level.

If these enzymes are expressed at relatively low levels, it becomes difficult to predict which transcript encodes these enzymes (Ding and Lofstedt, 2015).

This comparative transcriptome study has provided sequence and transcript abundance information that can be used to identify key enzymes involved in pheromone biosynthesis in the

PBW moth. Identification of the enzyme will require expressing each enzyme in a heterologous expression system followed by a functional assay. Identifying factors in the field population that are involved in the increased ZZ isomer ratios will require further research. Other factors in 97

addition to enzyme changes could be selective degradation of pheromone, the regulation of releasing pheromone, or some abiotic environmental factors.

Acknowledgements

This research was funded by the United States-Israel Binational Agricultural Research

Development fund (BARD) research grant award IS-4722-14, the Hatch Act and State of Iowa funds. Dou was supported in part by a Chinese Academic Scholarship.

References

Albre J, Liénard M.A., Sirey, T.M., Schmidt, S., Tooman, L.K., Carraher, C., et al. (2012) Sex pheromone evolution is associated with differential regulation of the same desaturase gene in two genera of leafroller moths. PLoS Genet 8: e1002489.

Antony, B., Soffan, A., Jakse, J., Alfaifi, S., Sutanto, K.D., Aldosari, S.A., et al. (2015) Genes involved in sex pheromone biosynthesis of Ephestia cautella, an important food storage pest, are determined by transcriptome sequencing. BMC Genomics 16: 532.

Bartlett, A.C., Wolf, W.W. (1985) Pectinophora gossypiella. In: Moore, R.F., Singh, P., eds. Handbook of Insect Rearing Vol 2. Elsevier, Amsterdam. pp. 415-430.

Cha, W.H., Jung, C.R., Hwang, Y-J., Lee, D-W. (2017) Comparative transcriptome analysis of pheromone biosynthesis-related gene expressions in Plutella xylostella (L.). J Asia Pac Entomol 20: 1260–1266.

Ding, B.J., Lofstedt, C. (2015) Analysis of the Agrotis segetum pheromone gland transcriptome in the light of sex pheromone biosynthesis. BMC Genomics 16: 711.

Ding, B.J., Carraher, C., Lofstedt, C. (2016) Sequence variation determining stereochemistry of a Delta11 desaturase active in moth sex pheromone biosynthesis. Insect Biochem Mol Biol 74: 68-75.

Evenden, M. (2016) Mating disruption of moth pests in integrated pest management: a mechanistic approach. In: Allison, J.D., Carde, R.T., eds. . Pheromone communication in moths: evolution, behavior, and application.. Univ California Press. pp. 365-393.

Foster, S.P., Roelofs, W.L. (1988). Pink bollworm sex pheromone biosynthesis from oleic acid. Insect Biochem 18: 281-286.

Fujii, T., Yasukochi, Y., Rong, Y., Matsuo, T., Ishikawa, Y. (2011) Multiple ∆11-desaturase genes selectively used for sex pheromone biosynthesis are conserved in Ostrinia moth genomes. Insect Biochem Mol Biol 61: 62-68.

Grabherr, M.G., Haas, B.J., Yassour, M., Levin, J.Z., Thompson, D.A., Amit, I., et al. (2011) Full- length transcriptome assembly from RNA-Seq data without a reference genome. Nat Biotechnol 29: 644-652.

Grapputo, A., Thrimawithana, A.H., Steinwender, B., Newcomb, R.D. (2018) Differential gene expression in the evolution of sex pheromone communication in New Zealand’s endemic leafroller moths of the genera Ctenopseustis and Planotortrix. BMC Genomics 19: 94.

Gu, S.H., Wu, K.M., Guo, Y.Y., Pickett, J.A., Field, L.M., Zhou, J.J., et al. (2013) Identification of genes expressed in the sex pheromone gland of the black cutworm Agrotis ipsilon with putative roles in sex pheromone biosynthesis and transport. BMC Genomics 14: 636 .

Harari, A.R., Zahavi, T., Gordon, D., Anshelevich, L., Harel, M., Ovadia, S., et al. (2007) Pest management programmes in vineyards using male mating disruption. Pest Manag Sci 63: 769-775.

Henneberry, T.J., Gillespie, J.M., Bariola, L.A., Flint, H.M., Lingren, P.D., Kydonieus, A.F. (1981) Gossyplure in laminated plastic formulations for mating disruption and pink-bollworm (Lepidoptera, Gelechiidae) control. J Econ Entomol 74: 376-381.

Henneberry, T.J. (2007) Integrated systems for control of the pink bollworm Pectinophora gossypiella in cotton. In: Vreysen, M.J.B., Robinson, A.S., Hendrichs, J., eds. Area-wide control of insect pests. pp. 567-579.

Hao, G., O'Connor, M., Liu, W., Roelofs, W.L. (2002) Characterization of Z/E11- and Z9- desaturases from the obliquebanded leafroller moth, Choristoneura rosaceana. J Insect Sci 2: 26. 99

Hanin, O., Azrielli, A., Applebaum, S.W., Rafaeli, A. (2012) Functional impact of silencing the Helicoverpa armigera sex-peptide receptor on female reproductive behaviour. Insect Mol Biol 21: 161-7.

Huang, X. and Madan, A. (1999) CAP3: A DNA sequence assembly program. Genome Res 9: 868-877 .

Hummel, H.E., Gaston, L.K., Shorey, H.H., Word, R.S., Byrni, K.J., Silverstein, RM. (1973) Clarification of chemical status of the pink bollworm sex pheromone. Science 181: 873-875.

Jurenka, R.A. (2004) Insect pheromone biosynthesis. Top Curr Chem 239: 97-132.

Jurenka, R.A., Roelofs, W.L. (1989) Characterization of the acetyltransferase used in pheromone biosynthesis in moths: Specificity for the Z isomer in Tortricidae. Insect Biochem 19: 639- 644.

Jurenka R. (2015) The PRXamide neuropeptide signalling system: Conserved in animales. Adv In Insect Phys 49: 123-70.

Knipple, D.C., Roelofs, W.L. (2003) Molecular biological investigations of pheromone desaturases, In: Blomquist G, Vogt R, eds. Insect pheromone biochemistry and molecular biology, San Diego: Academic Press. pp. 81-106.

Kumar, S., Stecher, G., Tamura, K. (2016) MEGA7: molecular evolutionary genetics analysis version 7 for bigger datasets. Mol Biol Evol 33: 1870-1874.

Lee, J.M., Hull, J.J., Kawai, T., Goto, C., Kurihara, M., Tanokura, M., et al. (2012) Re-Evaluation of the PBAN receptor molecule: characterization of PBANR variants expressed in the pheromone glands of moths. Front Endocrinol (Lausanne) 3: 6.

Li, R-T., Ning, C., Huang, L-Q., Dong, J-F., Li, X., Wang, C-Z. (2017) Expressional divergences of two desaturase genes determine the opposite ratios of two sex pheromone components in Helicoverpa armigera and Helicoverpa assulta. Insect Biochem Mol Biol 90: 90–100.

Lienard, M.A., Hagstrom, A.K., Lassance, J.M., Lofstedt, C. (2010) Evolution of multicomponent pheromone signals in small ermine moths involves a single fatty-acyl reductase gene. Proc Natl Acad Sci USA 107: 10955-10960. 100

Lin, X., Wang, B., Du, Y. (2018) Key genes of the sex pheromone biosynthesis pathway in female moths are required for pheromone quality and possibly mediate olfactory plasticity in conspecific male moths in Spodoptera litura. Insect Mol Biol 27: 8–21.

Liu, W., Rooney, A.P., Xue, B., Roelofs, W.L. (2004) Desaturases from the spotted fireworm moth (Choristoneura parallela) shed light on the evolutionary origins of novel moth sex pheromone desaturases. Gene 342: 303-311.

Rafaeli A, Klein Z. (1994) Regulation of pheromone production by female pink bollworm moths Pectinophora gossypiella (Sanders) (Lepidoptera: Gelechiidae). Physiol. Entomol 19: 159–164.

Rafaeli, A., Gileadi, C. (1995) Modulation of the PBAN-stimulated of pheromonotropic activity in Helicoverpa armigera. Insect Biochem Mol Biol 25: 827-34.

Simao, F.A., Waterhouse, R.M., Ioannidis, P., Kriventseva, E.V., Zdobnov, E.M. (2015) BUSCO: assessing genome assembly and annotation completeness with single-copy orthologs. Bioinformatics 31: 3210-3212.

Tegoni, M., Campanacci, V., Cambillau, C. (2004) Structural aspects of sexual attraction and chemical communication in insects. Trends Biochem Sci 29: 257-64.

Wojtasek, H., Leal, W.S. (1999) Conformational change in the pheromone-binding protein from Bombyx mori induced by pH and by interaction with membranes. J Biol Chem 274: 30950- 6

Ye, J., Fang, L., Zheng, H., Zhang, Y., Chen, J., Zhang, Z., et al. (2006) WEGO: a web tool for plotting GO annotations. Nucleic Acids Res 34: W293-297.

Zhang, Y.N., Zhang, L.W., Chen, D.S., Sun, L., Li, Z.Q., Ye, Z.F., et al. (2017) Molecular identification of differential expression genes associated with sex pheromone biosynthesis in Spodoptera exigua. Mol Genet Genomics 292: 795-809.

Zhang, Y-N., Zhu, X-Y., Fang, L-P., He, P., Wang, Z-Q., Chen, G., et al. (2015) Identification and expression profiles of sex pheromone biosynthesis and transport related genes in Spodoptera litura. PLoS One 10: e0140019. 101

Zhang, Y-N., Xia, Y-H., Zhu, J-Y., Li, S-Y., Dong, S-L. (2014) Putative pathway of sex pheromone biosynthesis and degradation by expression patterns of genes identified from female pheromone gland and adult antenna of Sesamia inferens (Walker). J Chem Ecol 40: 439–451.

102

Table 4. 1. Ratios and amounts of pheromone found in glands collected from the two populations. Ratio ZZ # glands Population ± SEM Amount per gland, ng

Lab 52.5 ± 3.6 13.8 ± 6.7 17

Field 61.8 ± 3.5 * 9.6 ± 5.3 16

*Statistically significant difference P<0.001 2-tailed t-test, compared to Lab strain for each population.

Table 4. 2. Analysis of sequencing results. BUSCO result (%) Population Rep. Library # of raw # of clean %GC # of Ave. Complete Frag Mis Preparation reads reads Contigs length (S+D)

Lab 1 mRNA-Seq 3.38*108 2.26*108 44 114125 768 97.9 1.3 0.8 2 Stranded Total 8.14*107 8.07*107 45 133684 707 98 1.4 0.6 RNA-Seq 3 Stranded Total 5.55*107 5.48*107 44 90007 800 96.8 2.1 1.1 RNA-Seq 4 Stranded Total 4.88*107 4.83*107 44 81495 867 97.6 1.7 0.7 RNA-Seq Field 1 mRNA-Seq 3.35*108 2.27*107 44 155114 671 98.1 1.2 0.7 2 Stranded Total 9.91*107 9.80*107 44 100851 817 98.5 0.8 0.7 RNA-Seq 3 Stranded Total 6.38*107 6.26*107 44 86767 805 95.5 3.5 1 RNA-Seq 4 Stranded Total 4.86*107 4.78*107 44 82709 842 95.9 3.3 0.8 RNA-Seq All 282,599 850 98 1.7 0.3

103

Table 4. 3. Putative biosynthesis related genes in PBW pheromone glands and the first BLASTp hit in GenBank. Transcripts GenBank homologue Description Accession no.* Species E value‡ % Identity Acetyl-CoA Carboxylase ACC Acetyl-CoA Carboxylase XP_013176189 Papilio xuthus 0 89 Fatty acid synthase FAS1 fatty acid synthase-like isoform X2 XP_022831709 Spodoptera litura 8E-150 37 FAS2 fatty acid synthase 2 AKD01761 Helicoverpa assulta 0 78 FAS3 PREDICTED: fatty acid synthase XP_013167810 Papilio xuthus 0 51 FAS4 fatty acid synthase-like XP_021186732 Helicoverpa armigera 0 78 Desaturase DES1 acyl-CoA Delta(11) desaturase-like XP_022825758 Spodoptera litura 7E-163 71 DES2 acyl-CoA delta-9 desaturase AAK94070 Epiphyas postvittana 0 79 DES3 Z11-fatty acid desaturase ALA65425 Manduca sexta 0 67 DES4 acyl-CoA Delta(11) desaturase-like XP_022125992 Pieris rapae 1E-114 51 DES5 PREDICTED: acyl-CoA Delta(11) XP_013183656 Amyelois transitella 0 78 desaturase-like DES6 E11-desaturase SFWGE11 AAQ12891 Choristoneura parallela 4E-151 62 DES7 sphingolipid delta(4)-desaturase DES1 XP_004930794 Bombyx mori 0 91 DES8 PREDICTED: acyl-CoA Delta(11) XP_013195132 Amyelois transitella 0 94 desaturase-like DES9 PREDICTED: acyl-CoA Delta(11) XP_011559976 Plutella xylostella 7E-179 70 desaturase-like DES10 acyl-CoA Delta(11) desaturase-like isoform XP_021183601 Helicoverpa armigera 0 77 X2 DES11 delta9-desaturase AGD98721.1 Bicyclus anynana 2E-162 71 DES12 acyl-CoA delta-9 desaturase CAJ27975.1 Manduca sexta 5E-169 91 DES13 stearoyl-CoA desaturase 5 isoform X1 XP_013192760 Amyelois transitella 0 78 DES14 Desaturase AIM40223 Cydia pomonella 0 74 DES15 acyl-CoA Delta(11) desaturase-like XP_022125992 Pieris rapae 3E-119 51 DES16 acyl-CoA desaturase HassNPVE OWR40684 Danaus plexippus 2E-142 66 DES 17 PREDICTED: acyl-CoA Delta(11) XP_013193663 Amyelois transitella 1E-179 71 desaturase β-oxidation enzymes Acyl-CoA oxidase ACO1 PREDICTED: probable peroxisomal acyl- XP_013177324 Papilio xuthus 1E-124 69 coenzyme A oxidase 1 isoform X2 ACO2 peroxisomal acyl-CoA oxidase 3 AID66678 Agrotis segetum 0 77 ACO3 PREDICTED: probable peroxisomal acyl- XP_013188704 Amyelois transitella 0 84 coenzyme A oxidase 1 ACO4 PREDICTED: probable peroxisomal acyl- XP_014366074 Papilio machaon 0 69 coenzyme A oxidase 1 ACO5 PREDICTED: probable peroxisomal acyl- XP_013188649 Amyelois transitella 3E-120 87 coenzyme A oxidase 1 ACO6 peroxisomal acyl-coenzyme A oxidase 3 XP_022819471 Spodoptera litura 0 75 Acyl-CoA dehydrogenase ACD1 short-chain specific acyl-CoA XP_022830593 Spodoptera litura 0 75 dehydrogenase, mitochondrial-like isoform X1 ACD2 acyl-CoA dehydrogenase family member 9 AID66671 Agrotis segetum 0 67 ACD3 hypothetical protein B5V51_7750 PCG80426 Heliothis virescens 0 92 ACD4 very long-chain specific acyl-CoA XP_022822499 Spodoptera litura 0 80 dehydrogenase, mitochondrial isoform X1 3 hydroxyacyl CoA dehydrogenase 3-HCD1 3-hydroxyacyl-CoA dehydrogenase type-2- XP_021183236 Helicoverpa armigera 4E-159 87 like 3-HCD2 3-hydroxyacyl-CoA dehydrogenase type-2 XP_021186997 Helicoverpa armigera 2E-160 85 3-HCD3 PREDICTED: probable 3-hydroxyacyl-CoA XP_013190290 Amyelois transitella 0 90 dehydrogenase B0272.3 3-HCD4 PREDICTED: probable 3-hydroxyacyl-CoA XP_013140866 Papilio polytes 0 87 dehydrogenase B0272.3 isoform X1 104

Table 4. 3. (Continued) 3 ketoacyl-coa thiolase 3-KCT 3-ketoacyl-CoA thiolase, mitochondrial XP_012546519 Bombyx mori 0 86 Enoyl CoA hydratase ECH1 enoyl-CoA hydratase domain-containing XP_022822615 Spodoptera litura 1E-170 81 protein 3, mitochondrial isoform X1 ECH2 PREDICTED: probable enoyl-CoA XP_013171890 Papilio xuthus 3E-177 74 hydratase Fatty-acyl reductase FAR1 fatty acyl reductase ARD71196 Spodoptera exigua 0 83 FAR2 putative fatty acyl-CoA reductase CG5065 XP_004925992 Bombyx mori 0 69 FAR3 putative fatty acyl-CoA reductase CG5065 XP_004925992 Bombyx mori 0 81 FAR4 putative fatty acyl-CoA reductase CG8306 XP_022824194 Spodoptera litura 0 80 FAR5 fatty acyl-CoA reductase wat-like XP_021197953 Helicoverpa armigera 0 48 FAR6 putative fatty acyl-CoA reductase CG5065 XP_022824237 Spodoptera litura 0 77 FAR7 putative fatty acyl-CoA reductase CG5065 XP_022823985 Spodoptera litura 0 91 FAR8 fatty-acyl CoA reductase ADI82791 Ostrinia nubilalis 40 Acetyltransferase FAT1 heparan-alpha-glucosaminide N- XP_004928101 Bombyx mori 0 79 acetyltransferase FAT2 PREDICTED: heparan-alpha-glucosaminide XP_013191695 Amyelois transitella 0 78 N-acetyltransferase-like FAT3 PREDICTED: 1-acyl-sn-glycerol-3- XP_013195392 Amyelois transitella 4E-168 83 phosphate acyltransferase alpha-like FAT4 lysophospholipid acyltransferase 7-like XP_021192342 Helicoverpa armigera 0 83 FAT5 PREDICTED: glycerol-3-phosphate XP_013165576 Papilio xuthus 0 77 acyltransferase 1, mitochondrial-like isoform X1 FAT6 lysophospholipid acyltransferase 5 XP_004933932 Bombyx mori 0 74 FAT7 PREDICTED: heparan-alpha-glucosaminide XP_013194974 Amyelois transitella 5E-132 40 N-acetyltransferase-like FAT8 probable protein S-acyltransferase 23 XP_021190174 Helicoverpa armigera 0 98 isoform X1 FAT9 PREDICTED: 2-acylglycerol O- XP_013172217 Papilio xuthus 0 74 acyltransferase 2-A-like FAT10 acyl-CoA:lysophosphatidylglycerol XP_021181103 Helicoverpa armigera 0 83 acyltransferase 1 isoform X2 FAT11 PREDICTED: glycerol-3-phosphate XP_013197894 Amyelois transitella 0 82 acyltransferase 3-like FAT12 PREDICTED: sterol O-acyltransferase 2 XP_013182360 Papilio xuthus 0 71 FAT13 hypothetical protein B5V51_10571 PCG75995 Heliothis virescens 3E-141 69 FAT14 lysophospholipid acyltransferase 1 isoform XP_004927037 Bombyx mori 0 77 X1 FAT15 acyltransferase AGPAT3 AGG55011 Heliothis subflexa 3E-145 93 FAT16 PREDICTED: glycerol-3-phosphate XP_013174439 Papilio xuthus 2E-172 82 acyltransferase 4 isoform X2 FAT17 hypothetical protein B5V51_12259 PCG80023 Heliothis virescens 2E-115 91 *Accession number of the GenBank homologue.

‡E-value for the comparison of the PBW transcript AA sequence and the GenBank homologue.

105

Table 4. 4. Comparison of candidate transcripts involved in the sex pheromone biosynthetic pathway. Lab population Field population

AA Complete AA Complete Log2 % AA Gene RPKM RPKM Length ORF Length ORF FC* Identity‡

Acetyl-CoA Carboxylase ACC 2394 Y 13.2±4.9 2394 Y 10.0±4.4 0.40 100 Fatty acid synthase FAS1 717 N 1.8±1.2 1531 N 1.6±0.5 1.12 100 FAS2 1348 N 28.9±15.6 1368 N 14.0±2.0 2.05 100 FAS3 985 N 10.6±7.8 1611 N 7.0±2.3 1.50 100 FAS4 1050 N 84.4±11.4 489 N 38.3±23.4 2.20 100 β-oxidation enzymes Acyl-CoA oxidase ACO1 251 N 13.6±1.2 244 N 13.5±2.0 0.02 100 ACO2 695 Y 3.8±0.9 695 Y 3.6±0.1 0.10 100 ACO3 668 Y 25.1±7.2 668 Y 22.4±2.0 0.16 100 ACO4 444 N 3.7±0.8 447 N 10.1±1.9 -1.46 100 ACO5 191 N 0.3 575 N 0.6±0.2 -1.17 100 ACO6 686 Y 0.6±0.2 686 Y 0.6±0.1 -0.04 100 Acyl-CoA dehydrogenase ACD1 409 Y 4.8±3.2 409 Y 4.0±1.8 0.26 100 ACD2 608 Y 7.7±1.9 608 Y 7.0±0.9 0.13 100 ACD3 407 Y 11.1±2.4 407 Y 14.5±2.1 0.38 100 ACD4 577 Y 29.5±9.0 577 Y 31.2±7.7 -0.08 100 3_hydroxyacyl_CoA_dehydrogenase 3-HCD1 256 Y 4.1±1.9 256 Y 4.0±1.5 0. 04 100 3-HCD2 255 Y 11.1±2.6 255 Y 6.0±1.1 0.89 100 3-HCD3 312 Y 35.2±18.8 309 Y 26.2±11.5 0.43 100 3-HCD4 307 Y 8.6±1.5 307 Y 6.5±1.8 0.41 100 3_ketoacyl-CoA_thiolase 3-KCT 400 Y 4.0±1.8 400 Y 2.9±0.3 0.44 100 enoyl_CoA_hydratase ECH1 278 Y 2.3±0.5 278 Y 1.7±0.3 0.44 100 ECH2 332 Y 12.7±4.8 332 Y 10.1±2.2 0.33 100 Desaturase DES1 321 Y 0.5±0.1 321 Y 0.8±0.4 -0.74 100 DES2 383 Y 2362.9±488.4 351 Y 2286.3±659.2 -0.05 100 DES3 339 Y 4.9±2.8 339 Y 6.5±1.1 -0.40 100 DES4 329 Y 5.7±3.4 329 Y 8.3±2.1 -0.55 100 DES5 331 Y 0.9±0.5 331 Y 0.3±0.03 1.47 100 DES6 318 Y 300.6±54.4 318 Y 420.0±91.5 -0.48 100 DES7 321 Y 6.1±1.8 321 Y 4.7±0.9 0.37 100 DES8 327 Y 450.3±342.4 327 Y 392.8±143.7 0.20 100 DES9 331 Y 8.0±2.9 385 Y 32.6±16.4 -2.02 100 DES10 351 Y 7.6±2.6 351 Y 7.8±2.0 -0.04 100 DES11 300 Y 0.8±0.3 159 N 0.6±0.3 0.40 100 DES12 243 N 99.1±33.6 147 N 133.7±30.2 -0.43 100 DES13 372 Y 19.3±3.3 372 Y 26.2±8.5 -0.44 100 DES14 370 Y 3.9±0.9 370 Y 10.8±4.1 -1.47 100 DES15 325 Y 0.9±0.6 362 Y 0.7±0.1 0.37 100 DES16 294 Y 1.3±0.7 303 Y 2.0±1.1 -0.62 94 DES17 367 Y 5.4±1.2 367 Y 4.0±0.5 0.42 100 Fatty-acyl reductase FAR1 519 Y 2.4±0.6 519 Y 2.8±0.7 -0.23 100 FAR2 521 Y 5.3±1.8 521 Y 9.2±3.1 -0.79 100 106

Table 4. 4. (Continued) FAR3 526 Y 9.7±2.9 552 N 9.7±0.3 0 100 FAR4 510 Y 7.2±1.8 510 Y 7.3±1.0 -0.02 100 FAR5 523 Y 19.3±10.5 523 Y 40.1±12.7 -1.06 100 FAR6 525 Y 0.9±0.3 525 Y 1.0±0.2 -0.07 100 FAR7 524 Y 0.4±0.1 524 Y 0.4±0.04 0.07 100 FAR8 451 Y 33.5±5.1 451 Y 31.6±8.8 0.09 100 Acetyltransferase FAT1 596 Y 2.1±0.5 596 Y 2.7±0.7 -0.38 100 FAT2 572 Y 11.3±2.7 586 Y 1.0±1.7 0.18 100 FAT3 270 Y 104.0±34.1 270 Y 36.5±4.8 1.51 100 FAT4 485 Y 18.5±3.7 485 Y 17.3±3.7 0.09 100 FAT5 864 Y 44.0±7.7 863 Y 45.2±5.6 -0.04 100 FAT6 480 Y 1.5±0.2 480 Y 4.6±7.0 -1.57 100 FAT7 552 Y 121.8±38.6 552 Y 42.5±9.5 1.52 100 FAT8 504 N 1.0±0.3 486 N 0.6±0.1 0.82 100 FAT9 359 Y 12.3±3.1 359 Y 9.8±0.9 0.33 100 FAT10 374 Y 2.2±0.5 378 Y 1.2±0.5 0.82 100 FAT11 497 Y 4.8±1.0 481 N 2.6±0.7 0.87 100 FAT12 469 Y 42.5±14.7 360 Y 18.7±2.5 1.12 100 FAT13 282 Y 32.3±8.6 282 Y 49.9±8.0 -0.63 100 FAT14 499 Y 24.8±1.9 499 Y 26.2±1.9 0.08 100 FAT15 245 Y 1.3±0.3 283 Y 1.8±0.3 0.51 88 FAT16 337 N 2.2±0.6 398 N 29±39.4 -3.73 100 FAT17 416 N 0.54±0 500 Y 0.2±0.1 1.17 86 * Log2 Fold Change (FC) - >0: up regulated in Lab population Log2(FC) <0: up regulated in field population ‡ % AA identity between the Lab and field populations.

107

Figure 4. 1. Pink bollworm sex pheromone biosynthetic pathway.

108

Figure 4. 2. Number of differentially regulated genes in each population, grouped by gene ontology. All genes had a full length ORF with FDR less than 0.05 and logFC larger than 2 (UP) or less than -2 (DOWN). Up: Field population upregulated genes; Down: Field population downregulated genes.

109

. 110

Figure 4. 3. Sequence alignments of DES2, DES6 and DES8 with other desaturases from

Lepidoptera. Epo: Epiphyas postvittana, Cpa: Choristoneura parallela, Pex: Planotortrix excessana, Cob: Ctenopseustis obliquana, Aip: Agrotis ipsilon, Cro: Choristoneura rosaceana,

Ave: Argyrotaenia velutinana, Hzea: Helicoverpa zea, Poc: Planotortrix octo, Pno: Planotortrix notophaea, Msex: Manduca sexta. The protein ID were shown in parenthesis. Histidine-rich motifs shown in red box; Signature motif in green box; Predicted transmembrane domain under grey line.

111

Figure 4. 4. Relative expression of selected genes between Lab and Field population as determined by qPCR. ns: Non-significant, P > 0.05 (two-tailed Student’s t-Test).

112

Figure 4. 5. Phylogenetic relationship of DESs from Lepidoptera constructed using amino acid sequences as described in Experimental Methods. DESs from Lab population were marked with brown cycle, and from field population were green triangle. Each family was classified to different color. Red: ∆11 desaturase; Maroon: ∆9 desaturase (C18>C16); Green: ∆9 desaturase (C16>C18); Lime: Z/E 6 desaturase; Olive: ∆9 desaturase (C14-26); Navy: Z/E 14 desaturase; Purple: Z/E 14+ E13 desaturase; Teal: ∆6 desaturase 113

Figure 4. 6. Phylogenetic relationship of FARs from Lepidoptera constructed using amino acid sequences as described in Materials and Methods. FARs from Lab population were marked with brown circle, and from field population with green triangle.

114

CHAPTER 5. IDENTIFICATION OF RNA VIRUSES FROM THE TRANSCRIPTOME OF PHEROMONE GLAND IN THE PINK BOLLWORM MOTH

Modified from a manuscript submitted to PLOS ONE

Xiaoyi Dou, Sijun Liu, Russell Jurenka

Abstract

Transcriptomic analysis of insects and their separate tissues have provided the identification of unique viral genomes. In this study, we utilized the transcriptome obtained from the pheromone gland of the pink bollworm, Pectinophora gossypiella, in two populations in Israel to identify several viruses. We found the composition of viruses were the same between populations. Based on sequence alignment and phylogenetic analysis we identified two positive-sense single-stranded

RNA viruses belonging to the Picornaviridae, one to iflavirus and one genome with the reversed type, and two negative-sense single-stranded RNA viruses, one belonging to Bunyaviridae and one is unknown with only one segment. In another study reported to GenBank, the same iflavirus was found in the transcriptome from the midgut of pink bollworm larvae in the USA, indicating that this iflavirus exists in other pink bollworm populations. The other viruses were currently found only in Israel populations and it remains to be determined if these viruses are found in other populations in this pest species found world-wide.

115

Introduction

An RNA virus utilizes its RNA as genetic material, and the International Committee on

Taxonomy of Viruses (ICTV) classifies these as RNA viruses without considering the DNA intermediates. RNA viruses could be classified as double-stranded RNA and single-stranded RNA.

Single-stranded RNA is further classified as negative-sense and positive-sense or antisense RNA viruses according to the sense or polarity of the RNA.

Positive-strand RNA viruses are assigned to Group IV in the Baltimore classification system

(Baltimore, 1971). Three orders were classified as positive-sense single-stranded RNA viruses, the

Nidovirales, Picornavirales and Tymovirales. The order Picornavirales comprises several families including Marnaviridae, Dicistroviridae, Secoviridae, Picornaviridae and Iflaviridae, and the genera Bacillarnavirus and Labyrnavirus. The picornavirus-like viruses have two distinct genomic structures: one is similar to the mammalian picornaviruses with the structural region at the 5’ terminus and non-structural region at the 3’ terminus. Sequences of the non-structural region consist of the conserved domains of helicase, protease and RNA-dependent RNA polymerase (Le

Gall et al., 2008). Many viruses belonging to this type have been identified in insects. The sacbrood virus (SBV) that causes a fatal infection in honey bee larvae has been identified and was found to be longer than mammalian picornaviruses (Ghosh et al., 1999). Solenopsis invicta virus 2 that infects the red imported fire ant, Solenopsis invicta, belongs to a new family of Picornavirales

(Olendraite et al., 2017); the other one is a calicivirus-like with the order reversed (Koonin &

Gorbalenya, 1992). The genome structure has been determined for Acyrthosiphon pisum virus

(APV) (van der Wilk et al., 1997), Drosophila C virus (DCV) (Johnson & Christian, 1998), and

Rhopalosiphum padi virus (RhPV) (Moon et al., 1998). 116

The structure of iflavirus resembles the mammalian picornaviruses. Iflaviridae is a family of small non-enveloped viruses and have a positive-sense, single-stranded (+ss)-RNA genome with approximately 9-11 kilobases. Iflaviruses infect invertebrates, primarily insects. The genome contains a single open reading frame with virus-encoded protein (VPg) at the 5’ end and a poly A tail at the 3’ end (Lee et al., 1977). Also, there is an internal ribosome entry site located at the 5’ untranslated region that controls translation (Ongus et al., 2006; Wu et al., 2007). Currently, more than 30 iflaviruses have been found in different species in the orders Lepidoptera, Hymenoptera,

Diptera and Orthoptera. In this study, we found two novel (+ss)-RNA viruses that meet all of the genomic features of iflaviruses, and one of them resembles the iflavirus found in Helicoverpa armigera (Yuan et al., 2017) and the other one resembles the caliciviruses.

Negative-sense single-stranded RNA viruses need to be converted to a positive sense and then replicated. They belong to Group V in the Baltimore classification (Baltimore, 1971). Some (-ss)-

RNA viruses are important human pathogens that include viruses that cause hemorrhagic fever, rabies, and influenza. The (-ss)-RNA viruses are classified based on the number of distinct segments of the genome structure: one segment or unsegmented (Mononegavirales), two

(Arenaviridae), three (Bunyaviridae), three to four (Ophioviridae), and six to eight segments

(Orthomyxoviridae) (King et al., 2012). Many (-ss)-RNA viruses have been described in insects, including Diptera (Ballinger et al., 2014), Hymenoptera (Wang et al., 2017), and Lepidoptera (Li et al., 2015). Bunyaviridae is the largest and most diverse family of RNA viruses. It is divided into five genera: Orthobunyavirus, Phlebovirus, Nairovirus, Hantavirus and Tospovirus. These viruses are a unique group that infect a wide range of hosts including vertebrates, invertebrates and plants

(Webster et al., 2011). The genome consists of three to eight segments of (-ss)-RNA, such as small

(S), medium (M) and large (L), which encode the RNA-dependent RNA polymerase (RdRp), two 117

glycoproteins and a nucleocapsid protein, respectively (Elliott and Blakqori, 2011; Léger and

Lozach, 2015). In this study, we found two viruses that are closely related to the Bunyaviridae.

The pink bollworm, Pectinophora gossypiella, (Lepidoptera: Gelechiidae) is a worldwide pest that causes significant yield loss in cotton fields. Control measures include the nuclear polyhedrosis virus in a bait formulation that could significantly decrease the number of larvae and thus boll damage (Bell & Kanavel, 1977). In addition, transgenic cotton producing Bacillus thuringiensis (Bt) toxin has been used effectively against this pest, but resistance to Bt toxin

Cry1Ac has been found in the field (Tabashnik et al., 2002). Also, it has been reported in Israel that the mating disruption technique against pink bollworm has been failing indicating resistance has developed (Harari et al., unpublished). So the finding of novel pathogens active against this pest is desirable.

In this study, we used the transcriptome of pheromone glands and ovipositor in PBW from

Israel to identify the viruses. Through blast searches, sequence alignment and phylogeny analysis, we found one iflavirus, one picorna-like virus, one bunyaviruses and one unknown (-ss)-RNA virus. These viruses are novel and we herein describe their unique features.

Materials and Methods

Insect collection and Pheromone gland extraction

Two different populations of pink bollworm were collected in Israel. One is from a laboratory colony (Lab) maintained at the Volcani Institute, Bet-Dagan, Israel and the other one was collected from a cotton field (Field) near Ein Shemer, Israel where mating disruption has failed. Pheromone 118

glands along with ovipositors from the two populations were removed and immediately placed in

RNAlater (Invitrogen) and frozen to -80 °C and then shipped to the Jurenka lab in the United States.

RNA isolation and Illumina sequencing

Total RNA was isolated using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol. A 2100 Bioanalyzer (Agilent Technologies) was used to check the quantity of RNA. Library preparation and sequencing on an Illumina HiSeq 3000 platform was conducted by the DNA facility at Iowa State University, Ames, Iowa, USA. The stranded total

RNA-Seq libraries were sequenced with 150 pair-end and the mRNA library with 100 single reads.

All sequencing reads were submitted to the SRA of NCBI under the accession number

“SRP140160”.

De novo Assembly of Short Reads and Gene Annotation

The quality of all raw reads was checked using FastQC (Babraham Bioinformatics, UK). The de novo assembly was carried out with the short reads assembling program Trinity (Grabherr et al., 2011). After Trinity assembly, the resulting sequences were then locally searched against the virus genomes and NCBI non-redundant (nr) protein database using the BLASTx program, to obtain protein annotations of the assembled contigs.

Sequence and Phylogenetic analysis

The conserved and functional domain of the predicted viral proteins were identified by the

Conserved Domain Database (CDD) in NCBI

(https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi) and InterProtScan 119

(https://www.ebi.ac.uk/interpro/search/sequence-search). Multiple sequence alignments of the core motifs were conducted using by CLC genomics Workbench 9.5. For the phylogenetic analysis the sequences were aligned using the ClustalW method. MEGA 7.0 (Kumar et al., 2016) was used for tree construction using the neighbor-joining method with 1,000 bootstrap replicates, the Poisson model and pairwise gap deletion options.

Results

Viruses found in the transcriptome

After a blast search of the NCBI non-redundant (nr) protein, Swiss-prot and virus genomes databases, the transcript hits to viruses were filtered according to the nucleotide length longer than

500 and an amino acid length longer than 100. In Total, there were 16 transcript hits to viruses in the Lab population and 18 hits to viruses in the Field population (Table 5.1, Supplementary Table

5.1&Table 5.2). Out of all the viruses found, there are 9 and 11 transcripts hit to dsDNA in Lab and Field populations, respectively, 7 are homologous to ssRNA viruses in both populations with

2 of them positive-sense and 5 negative-sense. Two (+ss)-RNA viruses and 4 (–ss)-RNA viruses were selected to do the downstream analysis since they all had long nucleotide and amino acid lengths, as well as a full open reading frame (ORF). The sequences of these six ssRNA were submitted to GenBank “MN164617 to MN164623”. The full-length ORF was not found in the other viral transcript encoding a (–ss)-RNA virus (Contig 1 in Lab population and Contig 18 in

Field population). According to the length of nucleotide and amino acid sequences, full ORF, and blast description, we predict the source of viruses. The source of all of the candidate ssRNA viruses were predicted from the virus genome, while the transcripts with a shorter length and partial ORF may have come from the host genome. 120

Amino acid alignment of the six ssRNA viruses between the two populations showed that they were 99% to 100% identical while the gene expressions were not significantly different (q value > 0.05) (Table 5.2). So we assumed that in the two populations, the PBW had the same six ssRNA viruses. Blastp analysis of the amino acid sequences found PBWV1 and PBWV4 hit to the

Iflaviridae, PBWV2 (with one segment) and PBWV3 (with three segments) hit to the family of

Bunyaviridae.

Positive-sense single-strand viruses

The PBWV1 was homologous to a virus from Helicoverpa armigera with 78% identity

(Supplementary Table 5.1 & Table 5.2), which represented a new species in Iflaviridae. The genome of 9,677 nt comprised a 593 nt 5’ untranslated region (UTR) with a putative internal ribosome entry site (IRES), a single ORF polyprotein of 2,948 aa with a predicted molecular weight of 337.77 kDa and a 429 nt 3’ UTR followed by a poly A tail (Figure 5.1A). Sequences are available in supplementary 3. The PBWV4 had 37% identity to a hypothetical protein from

Wuhan insect virus 13 and Bee iflavirus. PBWV4 is 9,782 nt in length, encoding a polyprotein of

2,846 aa with a predicted molecular weight of 320.55 kDa. Based on the conserved domain prediction, the structural protein domains in PBWV1 are at the N-terminus, and the nonstructural protein domains are at C-terminus including the helicase, protease and RdRp domains. All these domains were found in other iflaviruses (Figure 5.1A). Interestingly, the location of structural and nonstructural protein domains in PBWV4 are different from PBWV1. They are in an opposite direction compared to the PBWV1 domains (Figure 5.1A). RdRp from PBWV1 and PBWV4 contained eight picornaviral conserved motifs as recognized by Koonin and Dolja (1993) (Figure

5.1B). The three conserved helicase domains: Motif-A (GxxGxGKS), Motif-B (QxxxxxDD) and 121

Motif-C (KKxxxxPxxxxxNTN) were recognized by Koonin and Dolja (1993). However, the

Motif-A of PBWV1 was changed to GxxExGKS, which is the same in Lab and Field populations and also in some identified Lepidoptera iflaviruses found in Helicoverpa armigera, Bombyx mori,

Lymantria dispar, Antheraea pernyi and Thaumetopoea pityocampa. While motif-A of PBWV4 was GxxGxMKS which is unique (Figure 5.1C). Motif-C also contained unique sequences. The

KK sequences were changed to KG in PBWV1 in both PBW populations and Lepidoptera iflaviruses, and changed to KD in PBWV4 (Figure 5.1C). It appears that the helicase conserved domain is unique in Lepidoptera compared to other insects. Similarly, the protease conserved sequence GxCG and GxHxxG were also found in PBWV1 and PBWV4 in both Lab and Field populations (Figure 5.1D).

The PBWV1 was from the PBW in Israel. We named the virus in Lab as PBWV1-Lab and in

Field as PBWV1-Field. Through the NCBI/SRA transcriptomic database, we found another almost identical sequence from larval PBW originating in the USA (SRA: SRP062867) (Tassone et al.,

2016). Through the gene annotation, we found one transcript encoding a single ORF of 2,948 aa which was the same as PBWV1 from Israel. We proposed ‘Pink bollworm virus in USA’

(PBWV1-USA) as a tentative name for the virus found in USA. Blastp showed PBWV1-Lab,

PBWV1-Field and PBWV1-USA hit to the Helicoverpa armigera iflavirus with 78% identity.

Amino acid comparison of these PBWV1 viruses found in different populations and areas showed they were 99% identical to each other. The conserved domains of helicase, RdRp and protease were all 100% similar (Figure 5.1). Since the virus is found in USA and Israel, in midgut and pheromone gland, in larvae and adult, respectively, we could assume that it may exist in all pink bollworm populations. 122

The ORF sequence of the polyprotein were aligned to other members of the iflaviruses

(Supplementary 1) and a phylogenetic tree wasconstructed (Figure 5.2). The tree revealed that the three PBWV1 were clustered together, and they were closely related to the iflavirus from H. armigera. Several Lepidoptera iflaviruses were clustered together. However, PBWV4 is different from all iflaviruses from Lepidoptera, but it is close to the Dinocampus coccinellae wasp paralysis virus, Bee iflavirus and some unclassified (-ss)-RNA viruses.

Negative-sense single-strand RNA viruses

Blast results showed one transcript was 62% identical to the Seattle Prectang virus which is in the family Peribunyaviridae, we named it ‘PBWV2’, while three transcripts hit to Hubei lepidoptera virus 1 of the unclassified viruses, we named it “PBWV3”. For PBWV2, we only found one large segment containing 6,419 nt, encoding a putative RdRp of 2,094 aa with a predicted molecular weight of 241.89 kDa (Figure 5.3A). The genome of PBWV3 is a (-ss)-RNA virus with three segments, small (S), medium (M) and large (L). The large segment had 7,827 nt and encodes a putative RdRp of 2,503 aa with a predicted molecular weight of 287.09 kDa (Figure

5.3A). CDD analysis recognized the conserved domain Bunya_RdRp (interval: 618-1231, 809-

1485, E-value: 8.09e-07, 3.68e-74) in the PBWV2-L and PBWV3-L protein, respectively.

PBWV2-L and PBWV3-L only showed 12% identity, indicating they are two different viruses.

The alignment of the RdRp from PBWV2-L and PBWV3-L, and other (–ss)-RNA viruses indicates that the amino acid sequences of their RdRps consisting of the pre-motif A and motifs A through

E, which are highly conserved in negative sense RNA viral polymerases and common to all other

Bunyavirales characterized to date (Reguera et al., 2010) (Figure 5.3B). We found three basic residues (K, R, and R/K) in promotive A and a glutamic acid (E) downstream of premotif A, which 123

are conserved in bunyavirus RdRps (Bruenn, 2003; Elbeaino et al., 2009). SDD sequences in motif

C were also found in both viruses. The other motifs were slightly different between these two viruses. The medium segment of PBWV3 (PBWV3-M) was 5,042 nt in length and only encodes one surface glyprotein of 469 aa with a mass of 43.17kDa. Blastp result showed the M segment had 27% identify to the putative glycoprotein from Hubei lepidoptera virus 1. The conserved domain predicted by CDD showed a single transmembrane region close to the C terminus of the glycoprotein (interval: 1052-1520, E-value: 1.14e-15), however, we did not find a glycoprotein domain at the N terminus or a non-structural domain in the center. The small segment is 2,362 nt in length and encodes for a unique protein of 276 aa with a predicted mass of 31.08 kDa. Blastp result found it was closely associated to the N protein of members of the genus Phasivirus; CDD analysis showed there is a Tenuivirus/Phlebovirus nucleocapsid protein domain (interval: 23-245,

E-value: 3.98e-06).

Polygenetic trees were constructed based on the amino acid sequences of different segments

(Figure 5.4). The accession number of the other viruses is shown in Supplementary 1. PBWV2

RdRp did not cluster together with the known families in the order Bunyavirales. In all three trees

PBWV3 was clustered together with Hubei lepidoptera virus1 which is an unclassified RNA virus but closely related to other viruses in the Phenuiviridae. All of the data showed that PBWV2 and

PBWV3 are different viruses, and PBWV3 belongs to the Order of Bunyavirales.

Discussion

New viruses are difficult to discover with the traditional viral detection methods because they require whole genome sequences. However, the high-throughput next-generation sequencing technologies and bioinformatics have been applied to virus discovery in various organisms, 124

including humans (Chiu, 2013; Lipkin and Firth, 2013), arthropods (Kobayashi et al., 2017; Smith et al., 2014; Sparks, et al., 2013), and plants (Barba et al., 2014; Hany et al., 2014). Transcriptome sequencing and deep sequencing of viral small RNA analysis have also been used for the discovery of viruses in plant and insect hosts (Li et al., 2012; Zografidis et al., 2015).

In the present study, through the transcriptome analysis, we found some novel viruses.

According to the genomic characteristics, structure and phylogenetic proximity to other viruses, two viruses are (+ss)-RNA viruses, one is an iflavirus (PBWV1) and one is picorna-like with a calicivirus structure order (PBWV4), while the other two RNA viruses are (-ss)-RNA viruses, one is unclassified (PBWV2) and one belongs to Bunyaviridae Phenuiviridae (PBWV3). PBWV1 and

PBWV4 shared similar properties with Picornavirales, such as the three domains helicases, RdRp, proteinase, genome bound VPg at 5’ terminus and poly A tail at 3’ terminus. However, the three conserved domains in the genome structure are in a different order, indicating that they may be classified into a different viral group. PBWV2 and PBWV3 share the common features as (-ss)-

RNA viruses, including the homologous RdRp, inverted complementary genome ends (data not shown) and as encapsidated genome (King et al., 2012). These four viruses had the same expression level and 99% to 100% identity in the comparison of lab and field populations, indicating that they are consistently expressed in the pink bollworm.

The pink bollworm is a pest species that quickly spread around the world as it was first reported on infested cotton from Egypt in 1913 and in 1917 in the USA. In the transcriptome of

Israel and USA populations, we found one common iflavirus (PBWV1). They share many common features, including the same motifs in RdRp, helicase and protease. The transcriptome of Israel population was from the pheromone gland with ovipositors while the USA population was from the midgut of larvae. We assume that PBWV1 is continuously infecting PBW in different 125

development stages, at least in adult and larvae, and it is passed to the next generation vertically.

The other three viral transcripts were only found in the pheromone gland-ovipositor and were not reported from the larval midgut transcriptome.

PBWV2 and PBWV3 were defined as (-ss)-RNA viruses. However, we just found one segment of PBWV2 that encodes RdRp. We found all three segments of PBWV3. Through the amino acid alignment and phylogenetic analysis, it is clear that PBWV2 and PBWV3 were different viruses. The PBWV2 large segment was related to other viruses in the families of

Fimoviridae and Tospoviridae which commonly used plants as hosts, while PBWV3 was related to viruses in the family of Phenuiviridae that can use both vertebrates and invertebrates as their hosts.

Acknowledgements

This research was funded by the United States-Israel Binational Agricultural Research

Development fund (BARD) research grant award IS-4722-14, the Hatch Act and State of Iowa funds. Dou was supported in part by a China Scholarship Council.

References

Ballinger, M.J. Bruenn, J.A., Hay, J., Czechowski, D., Taylor, D.J. (2014) Discovery and evolution of bunyavirids in Arctic phantom midges and ancient bunyavirid-like sequences in insect genomes. J Virol 88: 8783-8794.

Baltimore, D. (1971) Expression of animal virus genomes. Bacteriol Rev 35: 235-241.

Barba, M., Czosnek, H., Hadidi, A. (2014) Historical perspective, development and application of next-generation sequencing in plant virology. Viruses 6: 106-136.

126

Bell, M.R., and Kanavel, R.F. (1997) Field tests of a nuclear polyhedrosis virus in a bait formulation for control of pink bollworms and Heliothis spp. in cotton in Arizona. J Econ Entomol 70: 625-629.

Bruenn, J.A. (2003) A structural and primary sequence comparison of the viral RNA dependent RNA polymerases. Nucleic Acids Res 31: 1821-1829.

Chiu, C.Y. (2013) Viral pathogen discovery. Curr Opin Microbiol 16: 468-478.

Elbeaino, T., Diagiaro, M., Alabdullah, A., De Stradis, A., Minafra, A., Mielke, N., et al. (2009) A multipartite single-stranded negative-sense RNA virus is the putative agent of fig mosaic disease. J Gen Virol 90: 1281-1288.

Elliott, R.M., and Blakqori, G. (2011) Molecular biology of orthobunyaviruses. Bunyaviridae: Molecular and cellular Biology (eds A. Plyusnin & R.M. Elliot), pp. 1-39. Caister Academic Press, Norfolk, UK.

Lee, Y.F., Nomoto, A., Detjen, B.M., Wimmer, E. (1977) The genome linked protein of picornaviruses: I. A protein covalently linked to poliovirus genome RNA. Proc Natl Acad Sci USA 74: 59-63.

Le Gall, O., Christian, P., Fauquet, C.M., King, A.M., Knowles, N.J., et al. (2008) Picornavirales, a proposed order of positive-sense single-stranded RNA viruses with a pseudo-T=3 virion architecture. Arch Virol 153: 715-727.

Li, C.X., Shi, M., Tian, J.H., Lin, X.D., Kang, Y.J., Chen, L.J., et al. (2015) Unprecedented genomic diversity of RNA viruses in arthropods reveals the ancestry of negative-sense RNA viruses. Elife 4: e05378.

Li, R., Gao, S., Hernandez, A.G., Wechter, W.P., Fei, Z. and Ling, K.S. (2012) Deep sequencing of small RNAs in tomato for virus and viroid identification and strain differentiation. PLoS One 7: e37127.

Lipkin, W.I. and Firth, C. (2013) Viral surveillance and discovery. Curr Opin Virol 3: 199-204

Ghosh, R.C., Ball, B.V., Willcocks, M.M., Carter, M.J. (1999) The nucleotide sequence of sacbrood virus of the honey bee: an insect picorna-like virus. J Gen Virol 80: 1541-1549.

Hany, U., Adams, I.P., Glover, R., Bhat, A.I., Boonham, N. (2014) The complete genome sequence of Piper yellow mottle virus (PYMoV). Arch Virol, 159, 385-388.

127

Johnson, K.N. and Christian, P.D. (1998) The novel genome organization of the insect picorna- like virus Drosophila C virus suggets this virus belongs to a previously undescribed virus family. J Gen Virol 79: 191-203.

King, A.M.Q., Adams, M.J., Carstens, E.B., Lefkowitz, E.J. (2012). Virus Taxonomy: Ninth Report of the International Committee on Taxonomy of Viruses. Elsevier Academic Press. pp. 651-782.

Kobayashi, D., Isawa, H., Fujita, R., Murota, K., Itokawa, K., Higa, Y., et al. (2017) Isolation and characterization of a new iflavirus from Armigeres spp. mosquitoes in the Philippines. J Gen Virol 98: 2876-2881.

Koonin, E.V., Dolja, V.V., Morris, T.J. (1993) Evolution and taxonomy of positive-strand RNA viruses: implications of comparative analysis of amino acid sequences. Crit Rev Biochem Mol Biol 28: 375-430.

Koonin, E.V., Gorbalenya, A.E. (1992) An insect picornavirus may have genome organization similar to that of caliciviruses. FEBS Letters 297: 81-86

Kumar, S., Strcher, G., Tamura, K. (2016) MEGA7: Molecular evolutionary genetics analysis version 7.0 for bigger datasets. Mol Biol Evol 33: 1870-1874.

Léger, P., Lozach, P.Y. (2015) Bunyaviruses: from transmission by arthropods to virus entry into the mammalian host first-target cells. Future Virol 10: 859-881.

Miranda, J.R.D., Hedman, H., Piero, O., Jörg, S., Olof, K., Helena, B., et al. (2017) Characterization of a novel RNA virus discovered in the autumnal moth Epirrita autumnata in Sweden. Viruses 9: 214.

Moon, J.S., Domier, L.L., McCoppin, N.K., D’ Arcy, C.J., Jin, H. (1998) Nucleotide sequence analysis shows that Rhopalosiphum padi virus is a member of a nover group of insect- infecting RNA viruses. Virology 243: 54-65.

Olendraite, I., Lukhovitskaya, N.I., Porter, S.D., Valles, S.M., Firth, A.E. (2017) Polycipiviridae: a proposed new family of polycistronic picorna-lie RNA viruses. J Gen Virol 98: 2368- 2378.

Ongus, J.R., Roode, E.C., Pleij, C.W.A., Valk, J.M., van Oers, M.M. (2006) The 5’ non-translated region of Varroa destructor virus 1 (genus Iflavirus): structure prediction and IRES activity in Lymantria dispar cells. J Gen Virol 87: 3397-3407.

Reguera, J., Weber, F., Cusack, S. (2010) Bunyaviridae RNA polymerases (L-protein) have an N- terminal, influenza-like endonuclease domain, essential for viral cap-dependent transcription. PLoS Pathog 6: e1001101. 128

Smith, G., Macias-Munoz, A., Briscoe, A.D. (2014) Genome sequence of a novel iflavirus from mRNA sequencing of the butterfly Heliconius erato. Genome Announc 2: e00398-14.

Sparks, M.E., Gundersen-Rindal, D.E., Harrison, R.L. (2013) Complete genome sequence of a novel iflavirus from the transcriptome of Halyomorpha halys, the brown marmorated stink bug. Genome Announc 1: e00910-13.

Tabashnik, B.E., Liu, Y.B., Dennehy, T.J., Sims, M.A., Sisterson, M.S., Biggs, R.W., et al. (2002) Inheritance of resistance to Bt toxin Cry1Ac in a field-derived strain of pink bollworm (Lepidoptera: Gelechiidae). J Econ Entomol 95: 1018-1026.

Tassone, E.E., Zastrow-Hayes, G., Mathis, J., Nelson, M.E., Wu, G., et al. (2016) Sequencing, de novo assembly and annotation of a pink bollworm larval midgut transcriptome. Gigascience 5: 28.

Van der Wilk, F., Dullemans, A.M., Verbeek, M., van den Heuvel, J.F.J.M. (1997) Nucleotide sequence and genomic organization of Acyrthosiphon pisum virus. Virology 238: 353-362.

Wang, F., Fang, Q., Wang, B., Yan, Z., Hong, J., Bao, Y., et al (2017) A novel negative-stranded RNA virus mediates sex ratio in its parasitoid host. PLoS Pathog 13: e1006201.

Webster, C.G., Reitz, S.R., Perry, K.L., Adkins, S. (2011) Anatural M RNA reassortant arising from two species of plant-and insect-infecting bunyaviruses and comparison of its sequence and biological properties to parental species. Virology, 413, 216-225.

Wu, T.Y., Wu, C.Y., Chen, Y.J., Chen, C.Y., Wang, C.H. (2007) The 5’ untranslated region of Perina nuda virus (PnV) possesses a strong internal translation activity in baculovirus- infected insect cells. FEBS Letters 581: 3120-3126.

Yuan, H., Xu, P., Yang, X., Graham, R.I., Wilson, K., Wu, K. (2017) Characterization of a novel member of genus iflavirus in Helicoverpa armigera. J Invertebr Pathol 144: 65-73.

Zografidis, A., Nieuwerburgh, F.V., Kolliopoulou, A., Apostolou-Karampelis, K., Head, S.R., Defore, D., et al. (2015) Viral small-RNA analysis of Bombyx mori larval midgut during persistent and pathogenic cytoplasmic polyhedrosis virus infection. J Virol 89: 11473- 11486.

129

Table 5. 1. Number of viruses found in the transcriptome of pheromone gland in two populations Population Virus dsDNA virus ssRNA virus Positive-sense Negative-sense Lab 16 9 2 5 Field 18 11 2 5

Table 5. 2. Comparison of ssRNA viruses between two populations of the PBW ssRNA Viruses AA length Identity (Field vs Lab) log2 FC q value

PBWV1 2948 99% 0.44 0.219

PBWV2 2093 99% -1.1 0.321

PBWV3-S 277 100% -1.64 0.219

PBWV3-M 1558 99% -1.53 0.219

PBWV3-L 2502 99% -1.42 0.219

PBWV4 2846 99% 0.09 0.744

130

131

Figure 5. 1. Genome organization and amino acid alignment of PBWV1 and PBWV4. A:

CDD and InterProScan analysis of genome structure of PBWV1 (top) and PBWV4 (below).

Rhv: Picornavirus capsid protein domain_like; CRPV_c: CRPV capsid protein like; Hel: RNA helicase; Pro: protease; RdRp: RNA-dependent RNA polymerase; UTR: untranslated region;

ORF: open reading frame. B: RdRp amino acid alignment. C: Helicase amino acid alignment; D:

Protease amino acid alignment. The accession number for sequences were shown in

Supplementary Table 5.3.

132

Figure 5. 2. Phylogenetic tree of PBWV1 and PBWV4 with other iflavirus. Full length polyprotein sequences were used. The amino acid sequences were aligned using ClastalW and then the tree constructed tree using Mega 7.0 with neighbor-joining method and evaluated with

1,000 bootstrap replicates. The accession number for sequences were shown in Supplementary

Table 5.3. 133

Figure 5. 3. Genome organization and amino acid alignment of PBWV2 and PBWV3. A:

CDD and InterProScan analysis of genome structure of PBWV2 (top) and PBWV3 (below). B:

Amino acid alignment between conserved RdRp premotif A and motifs A-E of PBWV2 and

PBWV3, and selected (-ss)-RNA viruses. The accession number for sequences are listed in

Supplementary Table 5.4. 134

135

136

137

Figure 5. 4. Phylogenetic tree of PBWV2 and PBWV3 with other (-ss)-RNA. The RdRP (A), glycoprotein (B), and nucleoprotein (C) were used. The accession number for sequences were shown in Supplementary Table 5.5. The amino acid sequences were aligned using ClastalW and then the tree constructed tree using Mega 7.0 with neighbor-joining method and evaluated with

1,000 bootstrap replicates.

138

CHAPTER 6. IDENTIFICATION OF AN ACETYLTRANSFERASE IN CABBAGE LOOPER

Abstract

Fatty alcohol acetyltransferases (FATs) play important roles in metabolism in various organisms. They are involved in the biosynthesis of fatty acids, cholesterol esters, waxes and gene regulation among other functions. Many moths utilize acetate esters derived from the acetylation of fatty alcohols as the sex pheromone for mating communication. This is the last step in the biosynthetic pathway of some sex pheromones. However, it is hard to predict the function of FATs only based on the sequence information since they have multiple roles. FAT has been biochemically characterized in sex pheromone biosynthesis in some moths, but at the molecular level, the FAT gene has not been identified. The cabbage looper, Trichoplusia ni, utilizes the Z-7- dodecenyl acetate (Z7-12:OAc) as the major sex pheromone compound with some other acetate esters as minor compounds. In this study, we selected one candidate FAT from T ni which is homologous to mouse wax synthase. We utilized qPCR to check the expression level, RNAi to knockdown the gene expression, and a functional assay in yeast and insect cells to confirm the role in vitro. Unfortunately, despite considerable research the candidate acetyltransferase has not been conclusively demonstrated to be involved in pheromone biosynthesis. However, this study indicates that one candidate FAT may not be the enzyme making pheromone acetate esters and confirmed that RNAi did not work well in the cabbage looper.

Introduction

Acyltransferase enzymes transfer acyl groups to thiol, amino, or hydroxyl group of an acceptor to form an acyl ester derivative. Acyltransferases could be classified into several groups 139

based on the substrate they utilize, including 1-O- acylglucosides, acylated-acyl carrier protein, quinic acid ester, and acyl coenzyme A. In all, coenzyme A esters are the major acyl donor in most groups of acyltransferases. In plants, acyl CoA-utilizing acetyltransferases involved in the production of cysteine, fatty acid, phospholipids, polyamines, acetyl-CoA and others (St-Pierre and De luca, 2000), wax (Kalscheuer et al., 2006; Li et al., 2008), and plant volatile esters

(Beekwilder et al., 2004; Günther et al., 2011). In animals, some acetyltransferases play important roles in producing cholesterol esters (Vaziri et al., 2001), and acetylcholine neurotransmitters (Itoh et al., 1986), in the arylamine carcinogens detoxification and bioactivation (Hanna et al., 1994), and in the gene transcription regulation by histone acetylation (Kuo and Allis, 1998).

FATs that catalyze the fatty alcohol to acetate ester has been described in many organisms.

A moth-pollinated flower, Clarkia breweri, an enzyme acetyl-CoA: benzylalcohol acetyltransferase (BEAT) was identified that is involved in the production of benzylacetate from benzyalcohol (Dudareva et al., 1998). An apple alcohol acyltransferase (MPAAT1) was cloned and characterized that could utilize a range of alcohol substrates with different chain lengths (C3-

C10), branched chain, terpene or aromatic alcohols (Souleyre et al., 2005). An alcohol acetyltransferase that catalyzes short chain alcohol conversion to butyl acetate has been characterized in kiwifruit (Günther et al., 2011). A novel acyl-CoA acyltransferase also has been studied in Acinetobacter calcoaceticus where it is involved in the biosynthesis of both wax esters and triacylglycerols (Kalscheuer and Steinbüchel, 2002).

Some moths utilize an acetate ester as the sex pheromone components. The biosynthesis has been widely described, including the fatty acid precursor production from fatty acid synthesis, double bond introduction by fatty acyl desaturase (DES), limited chain shortening by β-oxidation, fatty alcohol production by fatty acyl reductase (FAR) and fatty acetate synthesis by FAT. Various 140

DESs and FARs have been identified in moths, but the gene involved in the last step of acetylation has not been identified in any insect at the molecular level. It is very difficult to predict the function of FAT only according to the sequence information, because FATs belong to a large enzyme family and have various functions.

In Choristoneura fumiferana, it has been shown that the pheromone components Z11-14:OAc is catalyzed by a FAT which utilizes C12 to C15 chain length alcohols and acetyl-CoA (Morse and Meighen, 1987). In three noctuid species, the activity of acetylation has very low substrate specificity (Teal and Tumlinson, 1987; Bestmann et al., 1987: Dunkelblum et al., 1989). Jurenka and Roelofs (1989) demonstrated that the FAT, specifically found in pheromone glands, was located in microsomes and exhibited specificity for the Z isomer of 11-tetradecenol (Z11-14:OH) in Argyrotaenia velutinana and three other species of tortricid moths, Choristoneura rosaceana,

C. fumiferana and Platynota idaeusalis. However in three strains (E, Z and hybrid) of Ostrinia. nubilalis there was no preference for Z or E isomers. In the adzuki bean borer moth, Ostrinia scapulalis, one acetyl-CoA acetyltransferase (OsAT1) was expressed in all tissues, and the molecular and functional role characterized in the production of acetoacetyl-CoA (Fujii et al.,

2010). Interestingly, a plant-derived diacyltransferase acetyltransferase (EaDAcT) (EC 2.3.1.20) from the burning bush, Euonymus alatus, was found to catalyze the production of some moth sex pheromone acetate esters from fatty alcohols but with low efficiency (Ding et al., 2014). Similarly, a yeast acetyltransferase (ATF1) also acetylates a broad range of sex pheromone alcohols to acetate esters (Ding et al., 2016).

The cabbage looper, Trichoplusia ni, is difficult to control due to its broad distribution and resistance to insecticides, utilizes cis-7-dodecenyl acetate (Z7-12:OAc) as the major sex pheromone component, and cis-5-dodecenyl acetate (Z5-12:OAc), cis-7-tetradecenyl acetate (Z7- 141

14:OAc), cis-9-tetradecenyl acetate (Z9-14:OAc), and 11-dodecenyl acetate (11-12:OAc) as minor components. The biosynthesis of these compounds utilizes the enzymatic activities of desaturases, β-oxidation, FAR, and FAT. One desaturase in T. ni has been identified (Knipple et al., 1998) but FAR and FAT have not been characterized at the molecular level.

In this study, we selected a candidate FAT which is homologous to the wax synthase in mice.

Through the tissues distribution, RNAi, insect cell and yeast cell expression, functional assay and

GC-MS analysis, the candidate FAT is likely not involved in the last step of sex pheromone biosynthesis in this moth. However, this study showed RNAi did not work well in T ni, especially in adult and excluded one candidate FAT. Further work on FAT identification needs to be conducted.

Materials and Methods

Insect

The T. ni population was maintained at 25°C±1°C under a photoperiod of L15: D9 in the laboratory. Larvae were reared using an artificial diet (Stonefly heliothis diet). Females and males were separated in their pupal stage and placed into two cages. Adult moths were fed a 10% sucrose sollution. Pheromone glands were extracted from virgin females.

Cloning of the FAT

Ten pheromone glands were extracted from 2-3 day old females. The total RNA was isolated from PGs with TRIzol reagent (Gibco, Paisley, UK) according to the manufacturer’s instructions and quantity measured with a Nanodrop 2000/2000c (Thermo Scientific). The 5’ RACE and 3’ 142

RACE first-strand cDNA were synthesized by SMARTer RACE cDNA amplification kit

(Clontech) according to the user manual with the gene-specific primers 5’-

GGGGAAGTAATCGCGGAGGCAATACCACCACGTCCAGT-3’ and 5’-

GGAGGGGCGGCTGAAGCATTGGATGCTCATCCC -3’ The touchdown PCR conditions for

RACE were 5 cycles of 95℃ for 30s, 72℃ for 3 min, 5 cycles of 95℃ for 30s, 70℃ for 30s, 72℃ for 3 min, 25 cycles of 95℃ for 30s, 68℃ for 30s, 72℃ for 3 min. The PCR products were gel- purified and cloned into the pGEM-T easy vector (Promega). Plasmid DNA was collected and sequenced to obtain the 5’ and 3’ end of FAT sequences. Using the 5’ RACE or 3’ RACE first- strand cDNA as templet to conduct PCR with a sense primer 5’-ATGGAACACGTCCTTTGT -3’ and an antisense primer 5’-CTATGTGACGACTAGCTTGG-3’ to get the full length ORF. PCR conditions were 95℃ for 3 min, 35 cycles of 95℃ for 30s, 55 ℃ for 90s, 72℃ for 30s, and a final extension 72℃ for 10 min. PCR product was cloned into pGEM-T easy vector and sequenced to obtain the ORF of FAT.

Tissue specificity studies by qPCR

Tissues distribution of FAT were investigated by qPCR. Total RNAs from PG, fat body, ovary, and remaining abdominal segments of 10 females and testis, fat body, aedeagus/hair pencil complex and remaining abdominal segments of 10 males were isolated from 2-day-old-adults. 1µg of total RNA was used for first-strand cDNA synthesis using the same procedures as described in

Chapter 3. The cDNA from each tissues was used as template for qPCR. The primers were: 5-

CGCCTGTTACTACCGTTGTT-3’ and 5’-TCACTTTCTCCGTCTCGAAG-3. qPCR was conducted using SYBR Green Supermix (Invitrogen) on the Applied Biosystems QuantStudio 3

(Thermo Fisher Scientific) according to the manufacturer’s protocol. The conditions of the thermal 143

cycles were: 95℃ for 3 min, 40 cycles of 95℃ for 15 s, 60℃ for 20 s. Three technical replicates were used for each sample. The ribosomal protein S5 was used as a reference gene. The data were analyzed using the 2-ΔΔCt method (Livak and Schmittgen, 2001).

Functional Assay in insect and yeast cells

Two expression systems were selected to conduct the gene expression and functional assay.

The first one was the Insect SelectTM BSD system (Invitrogen) with Sf9 cells following manufacturer’s protocols. Briefly, the ORF was cloned into the pIB-V5 vector containing the

OpIE1 promoter and His tag at the C terminus. PCR was used to amplify from the cDNA we got previously using the primers: 5’-CGCGCGAAGCTTATGGAACACGTCCTT TGT-3’ (forward primer with HindIII site) and 5’- CGCGCGGGATCCTGTGACGACTAGCTT-3’ (reverse primer with BamHI site). The PCR conditions are the same as in the full length clone. The product was double digested and purified with the gel purification kit (QIAquick gel extraction kit, Qiagen).

Then the product was ligated into a pIB-V5 vector which was predigested by the same double digestion enzymes, and transformed into JM109 competent cells. Single clones were selected and plasmid DNA was purified with a QIAprep kit (Qiagen). The plasmid was sequenced in the DNA facility at Iowa State University. FAT-PIB and EGFP-PIB recombinant DNA was used for gene expression. In order to confirm the expression of FAT, EGFP gene (without start codon) was cloned into the prepared FAR-PIB plasmid to from a new FAT-GFP-PIB plasmid. Sf9 cells (2.5 ×

106 cells/ml, >96% viability) were plated into a 75 cm2 tissue culture treated flasks (FALCON®) supplemented with Sf900 SFM medium (Invitrogen). After 1h sediment, the medium was removed and then transfected the recombinant DNA (5 µg) into Sf9 cells with Cellfectin® reagent

(Invitrogen) according the manufacturers protocol. After 72 h incubation at 27℃, the medium was 144

removed and the precursor 0.5 mM (Z)-7- dodecenol (Z7-12: OH) in 0.7% DMSO in fresh medium was add. Sf9 cells were washed with PBS after another 48 h incubation and then cells removed by scraping the flask and cells were extracted with 2 ml n-hexane for 30 min. The extra hexane was evaporated under N2 gas and 2 µl to 5 µl was analyzed by GC-MS system (A Hewlett Packard

5890 GC coupled to a 5972 mass selective detector, the column used to separate the extracts was a DB Wax (J&W Scientific, 30mx0.25mm). The oven temperature was set as follows: 60℃ for 1 min, increase to 203℃ by 5℃/min, final 15 min at 230 ℃. EGFP was used as a positive control for gene expression and empty vector was a negative control. 100 ng of E4-13: OAc was used as the internal standard.

The second system was the yeast expression system. The full length FAT was cloned to pYES

2.1 expression vector (Invitrogen). The plasmids were transformed into INVSc1 yeast host strain using the S.C. EasyCompTM kit (Invitrogen) and incubated for 72 h on SC-U plate. Individual colonies were selected and inoculated in 6 mL selective medium ((SC-U)+ 2% Glucose) and then incubated for 48 h at 30 ℃ and 200 rpm. The cells were diluted to OD600nm/mL = 0.4 in a 20 ml

SC-U medium with 2% galactose and 0.1% glucose. After incubation for 24 h at 30℃ and 200 rpm, the yeast cell culture was diluted to 1: 10 in 2 mL fresh induction medium containing 0.5mM fatty alcohol (Z7-12: OH, Z9-14:OH and Z11-14:OH) as the substrate. Cells were incubated at 30℃ and 200 rpm, after 24 h the cells were pelleted at 2,000 ×g and washed with sterile water twice. 1 mL n-hexane was used to extract the cell pellets and stored at -20 ℃ until analyzed by GC-MS.

RNAi

The double-stranded RNA for FAT was synthesized using MEGAscript RNAi kit (Ambion).

The templates contained T7 polymerase at both 5’ and 3’ end. The PCR cycle conditions were set 145

as follows: 95℃ for 3 min, 35 cycles of 95℃ for 1 min, 60℃ for 1 min, 72℃ for 1min, and a final elongation at 72℃ for 10 min. The purified PCR product was used as templates for dsRNA synthesis. The dsRNA were treated with DNase I and RNase to remove DNA and single strand

RNA contaminations, followed by purification to remove proteins, free nucleotides and nucleic acid degradation products. The dsRNA was dissolved in elution solution and the concentrations were measured using the Nanodrop 2000/2000c. The enhanced green fluorescent protein (EGFP) dsRNA was used as a control and the ∆11 desaturase was used as positive control.The final dsRNA had 470 nt, 485 nt and 680 nt for EGFP, FAT and ∆11 desaturase, respectively.

20 µg of dsRNA (4 µg/ul) for EGFP, FAT and DES were injected between the 7th and 8th abdominal segments which is close to pheromone glands in 12 h and 24 h old adult females. The injected females were incubated for 48 h at 25°C±1°C. The PGs were then removed and extracted with n-hexane containing 150 ng of E4-13:OAc as an internal standard. GC-MS analysis was conducted as described above. RNAi knockdown effects were measured with semi-quantitative

PCR. 1 µg of total RNA from treated females was isolated for the synthesis of first-strand cDNA as described above. The cDNA was diluted ten times and used as template for semi-quantitative

RT-PCR as described above.

Results

Cloning of biosynthetic FAT homolog candidates

The candidate FAT was selected based on the wax synthase in the mouse (Gene bank accession: AY611031) (Cheng and Russell, 2004) and the transcriptome of the cabbage looper

(Chen et al., 2014). The wax synthase catalyzes the formation of wax esters using the substrate 146

palmitoyl-CoA and various fatty alcohols. BLASTn search of the wax synthase resulted in a hit that was 39% identical to a transcript fromT ni. The full-length T. ni FAT cDNA transcript contains an open reading frame (ORF) of 1,074 base pair (bp), encoding a protein of 368 AA with predicted molecular weight of 40.51 kDa. Submit to GenBank and list accession “MN.164616”. TniFAT shares 32.78% identify to the wax synthase in mouse, 32.03% to human wax synthase, 12.32% identity to the plant Euonymus alatus diacylglycerol acetyltransferase (EaDAcT) (Ding et al.,

2014), and only 9.67% identity to Saccharomyces cerevisiae acetyltransferase I (AFT1) (Ding et al., 2016), respectively. Through the conserved domain search in NCBI and InterProScan, we found that there is a lysophospholipid acyltransferase (LPLAT) of glycerophospholipid biosynthesis domain (Interval: 63-346, Evalue: 2.93e-108), or a 2-acylglycerol O-acyltransferase

(MGAT), which catalyzes a N-acylglycerol to diacylglycerol. Since the acetyltransferase has been characterized as transmembrane protein (Jurenka and Roelofs, 1989), the amino acid sequence was analyzed using the transmembrane region prediction software TMPRED (Hofmann and Stoffel,

1993) and TMHMM (Krogh et al., 2001). One putative membrane bound region was predict between residues 42 to 65, with both algorithms.

Tissue distribution of FAT

The results from qPCR showed that FAT was not specific expressed in PGs but the ∆11 desaturase and fatty acyl reductase were expressed only in PGs. FAT transcripts were found all tissues including the fat body, ovaries, pheromone gland, and abdomen in adult females and fat body, aedeagus/hair pencil complex, testis, and abdomen in adult males (Figure 6.1). The candidate FAT transcripts were also found at higher levels in males than females with the highest levels found in aedeagus/hair pencil complex. 147

Functional expression in insect and yeast cells

When expressed FAT using the InsectSelectTM BSD system with pIB/V5-His, we indeed found the protein expressed based on fluorescence microscopy (Figure 6.2). EGFP had higher expression while FAT-EFGP coexpressed at a lower level. FAT-PIB gene expressed in SF9 cells were conducted for downstream analysis. However, adding the substrate Z7-12:OH did not result in the production of Z7-12:OAc. When we expressed FAT in yeast cells and added Z7-12:OH, Z9-

14:OH, or Z11-14:OH we did not find an increased amount of the corresponding acetate ester compared to control. Also to determine if FAT can produce wax esters, 16:Me and 18:OH, Z9-

14:OH, or Z11-14:OH were incubated with yeast cells expressing FAT. No corresponding wax ester products were found.

RNAi

EGFP was used as a negative control and ∆11-desaturase as a positive control. Semi- quantitative RT-PCR result did not show any knockdown after the treatment of dsRNA for ∆11- desaturase and FAT. The amount of Z7-12:OAc were checked by GC-MS, but there was no significant decrease in the amount between treatments.

Discussion

Acyltransferases belong to a huge family of enzymes with various functions making a functional prediction very hard using just sequence information. The CoA-dependent acyltransferases are also diversified in molecular weights, oligomeric structures and sequences.

Some enzymes classes have been described in prokaryotes, animals, yeast, and plants, but many of them do not have conserved sequences between the groups of organisms. Even within one class, 148

two families of evolutionarily unrelated genes have been found, such as chloramphenicol acetyltransferases (Murray and Shaw, 1997). The FAT we selected in this study was based on homology to wax synthase in the mouse. The mouse wax synthase catalyzes the production of wax esters using palmitoyl-CoA and different fatty alcohols as substrates, including the fatty alcohols:

C10-C18 saturated, C16-C29 monounsaturated, and C18 with two or three double bonds (Cheng and Russell, 2004). We assumed the homologous gene in T. ni could be the candidate FAT that produces Z7-12:OAc from Z7-12:OH and acetyl-CoA. Through the tblastn in NCBI searched for the database of Transcriptome Shotgun Assembly within the organism Lepidoptera (taxid: 7088), we found one transcript (GenBank: GEEM01018130) in T. ni that was 39% identical to the mouse wax synthase. It is important to note that the yeast ATF1 can produce acetate esters from fatty alcohols; however, it is not homologous to the FAT sequence in T. ni. As our expectation, we found this candidate FAT has membrane bound region as determined by the transmembrane prediction software. Since DESs and FARs in various moths are uniquely expressed in PGs, and

FATs has been characterized in PGs of A. velutinana (Jurenka and Roelofs, 1989), we speculated that this candidate FAT should be highly expressed in PGs. However, we found the candidate FAT was present in various tissues in both females and males. It seems this gene may encode a protein involved in wax ester rather than acetate ester biosynthesis. Recently, FATs have been predicted in many moths in transcriptomic studies, including Agrotis segetum (Ding and Löfstedt, 2015),

Agrotis ipsilon (Gu et al. 2013), Ephestia cautella (Antony et al., 2015), Grapholita molesta and

Grapholita dimorpha (Jung and Kim, 2014), Spodoptera litura (Zhang et al., 2015), Spodoptera exigua (Zhang et al., 2017), and Ctenopseustis and Planotortrix (Grapputo et al., 2018). The 34 candidate FATs of A. segetum were expressed in yeast cells but none of them showed the function of producing acetate esters, including the FAT homolog from T.ni (Ding and Löfstedt, 2015). 149

In our study, we used the InsectSelectTM BSD expression system using a pIB/V5-His vector with OpIE promoter. I could find only one study that utilized a similar expression system for the identification of FAR in Ostrinia scapulalis (Antony et al., 2009). The corresponding Z11-14:OH was produced from Z11-14:acid in the FAR expressed cells. The main problem that I encountered with this expression system was the apparent low level of expression which was indicated by the comparison of fluorescent microscopy images. Even with low level of expression we added the precursor Z7-12:OH and incubated for a long period of time, I cound not detect the corresponding acetate ester. Several reasons could have caused these results: First, since the candidate FAT distributes in the several tissues and is not specific to the PGs like DESs or FARs, it maybe not be involved in the biosynthesis of sex pheromone. Some other functions of this gene need to be considered. Secondly, the candidate FAT is a membrane bound protein and it could be hard to express the protein at a high level. A functional assay with low amounts of enzyme may not work as exprected. Some secretory proteins were identified in this system with high expression level, but for the transmembrane proteins, as far as I know only the FAR from O. scapulalis pheromone glands has been described using SF9 cells. Third, technique problems. This was the first time I attempted to use insect cells and there was not a good positive control to do the trouble shooting.

After using insect cells, we also tried to conduct the functional assay in yeast cells. Since

ATF1 in yeast cells has the ability to produce pheromone acetate esters, and I did not have the mutate yeast strain without ATF1, we compared the amount of acetate ester to the control.

However, there was no significant difference. Taken all together, we made the conclusion that the

FAT we choose is not the acetyltransferase that is involved in sex pheromone biosynthesis. We also tried to determine if this enzyme was involved in the production of wax esters since it is 150

homologous to the mouse wax synthase. 16:Me was used as one precursor, plus 18:OH, 9-14:OH,

11-14:OH. However, we did not find any wax ester produced.

The first reported RNA interference in Lepidoptera was the gene silencing in pupae of

Cecropia (Bettencourt et al., 2002). It was considered as a method to investigate gene function in

Lepidoptera. However, it has been shown that RNAi was not as straight-forward as we thought in

Lepidoptera, even though there are some studies showing successful knockdown(Fabrick et al.,

2004; Wang et al., 2013; Wang et al., 2015; Choi and Vander Meer, 2018). Only a few studies utilized RNAi to knockdown genes involved in the sex pheromone biosynthesis in moths (Ohnishi et al., 2006; Lin et al., 2017). In our study, even using a very high amount of dsRNA (100

µg/insect), I did not find knockdown of the target gene. Some factors may lead to the inefficiency of RNAi in Lepidoptera, such as various species, tissue delivery methods, gene targets, amounts of dsRNA and so on (Terenius et al., 2011). Recently, a study showed a nuclease, termed REase

(GeneBank: AYE20402) which is specific to Lepidoptera, results in the suppression of RNAi

(Guan et al., 2018). Some other methods maybe good candidates to knockdown target genes, such as CRISPR-Cas9 that has been used in multiple lepidopteran insects (Zhang and Reed, 2017;

Zhang et al., 2019). This study is an incomplete work with many negative results. Further work could include the gene expression in a baculovirus expression system, other candidate gene selection based on the transcriptome and proteomics, and using the CRISPR-Cas9 to show the function in vivo.

151

References

Antony, B., Fujii, T., Moto, K., Matsumoto, S., Fukuzawa, M., Nakano, R., et al. (2009). Pheromone-gland-specific fatty-acyl reductase in the adzuki bean borer, Ostrinia scapulalis (Lepidoptera: Crambidae). Insect Biochem Mol Biol 39: 90-95.

Antony, B., Soffan, A., Jakše, J., Alfaifi, S., Sutanto, K.D., Aldosari, S.A., et al. (2015) Genes involved in sex pheromone biosynthesis of Ephestia cautella, an important food storage pest, are determined by transcriptome sequencing. BMC Genomics 16:532.

Beekwilder, J., Alvarez-Huerta, M., Neef, E., Verstappen, F.W.A., Bouwmeester, H.J., Aharoni, A. (2004) Functional characterization of enzymes forming volatile esters from strawberry and banana. Plant Physiol 135:1865–78.

Chen, Y.R., Zhong, S., Fei, Z., Gao, S., Zhang, S., Li, Z., et al. (2014) Transcriptome responses of the host Trichoplusia ni to infection by the baculovirus Autographa californica multiple nucleopolyhedrovirus. J Virol 88: 13781-13797.

Cheng, J.B., Russell, D.W. (2004). Mammalian wax biosynthesis: II. Expression cloning of wax synthase cDNAs encoding a member of the acyltransferase enzyme family. J Biol Chem 279: 37798-37807.

Choi, M-Y., Vander Meer, R.K. (2018) Phenotypic effects of PBAN RNAi using oral delivery of dsRNA to corn earworm (Lepidoptera: Noctuidae) and tobacco budworm larvae. J Econ Entomol 112: 434-439.

Ding, B.J., Hofvander, P., Wang, H.L., Durrett, T.P., Stymne, S., Löfstedt, C. (2014) A plant factory for moth pheromone production. Nat Commun 5: 3353.

Ding, B.J. and Löfstedt C. (2015) Analysis of the Agrotis segetum pheromone gland transcriptome in the light of sex pheromone biosynthesis. BMC Genomic 16:711.

Ding, B.J., Lager, I., Bansal, S., Durrett, T.P. Stymne, S., Löfstedt, C. (2016). The yeast ATF1 acetyltransferase efficiently acetylates insect pheromone alcohols: Implications for the biological production of moth pheromones. Lipids 51: 469-475.

Dudareva, N., D’Auria, J.C., Nam, K.H., Raguso, R.A., Pichersky, E. (1998) Acetyl-CoA: benzylalcohol acetyltransferase – an enzyme involved in floral scent production in Clarkia breweri. Plant J 14: 297-304.

Fabrick, J.A., Kanost, M.R., Baker, J.E. (2004) RNAi-induced silencing of embryonic tryptophan oxygenase in the Pyralid moth, Plodia interpunctella. J Insect Sci 4: 15. 152

Fujii, T., Ito, K., Katsuma, S., Nakano, R., Shimada, T., Ishikawa, Y. (2010) Molecular and functional characterization of an acetyl-CoA acetyltransferase from the adzuki bean borer moth Ostrinia scapulalis (Lepidoptera: Crambidae). Insect Biochem Mol Biol 40: 74-78.

Gu, S.H., Wu, K.M., Guo, Y.Y., Pickett, J.A., Field, L.M., Zhou, J.J, Zhang, Y.J. (2013) Identification of genes expressed in the sex pheromone gland of the black cutworm I with putative roles in sex pheromone biosynthesis and transport. BMC Genomic 14:v636. Guan, R-B., Li, H-C., Fan, Y-J., Hu, S-R., Christiaens, O., Smagghe, G. (2018) A nuclease specific to lepidopteran insects suppresses RNAi. J Biol Chem 293: 6011-6021.

Günther, C.S., Chervin, C., Marsh, K.B., Newcomb, R.D., Souleyre, E.J. (2011) Characterisation of two alcohol acyltransferases from kiwifruit (Actinidia spp.) reveals distinct substrate preferences. Phytochem 72:700–710.

Hanna, P.E. (1994) N-Acetylransferases, O-Acetyltransferases, and N, O-Acetyltransferases: Enzymology and Bioactivation. Adv Pharmacol Sci 27: 401-430.

Itoh, N., Slemmon, J., Hawke, D. (1986) Cloning of Drosophila choline acetyltransferase cDNA. Proc Natl Acad Sci USA 83:4081–4085.

Kalscheuer, R., Steinbüchel, A. (2002) A novel bifunctional wax ester synthase/Acyl-CoA: diacylglycerol acyltransferase mediates wax ester and triacylglycerol biosynthesis in Acinetobacter calcoaceticus ADP1. J Biol Chem 278: 8075-8082.

Kalscheuer, R., Stoveken, T., Luftmann, H., Malkus, U., Reichelt, R., Steinbüchel, A. (2006) Neutral lipid biosynthesis in engineered Escherichia coli: jojoba oil-like wax esters and fatty acid butyl esters. Appl Environ Microbiol 72:1373–1379.

Kuo, M-H. and Allis, D.C. (1998) Roles of histone acetyltransferases and deacetylases in gene regulation. Bioessays 20: 615-626.

Knipple, D.C., Rosenfield, C-L., Miller, S.J., Liu, W., Tang, J., Ma, P.W.K., et al. (1998) Cloning and functional expression of a cDNA encoding a pheromone gland-specific acyl-CoA ∆11- desaturase of the cabbage looper moth, Trichoplusia ni. Proc Natl Acad Sci USA 95: 15287-15292.

Jung, C.R. and Kim, Y. (2014) Comparative transcriptome analysis of sex pheromone glands of two sympatric lepidopteran congener species. Genomics 103: 308-315. 153

Jurenka, R.A. and Roelofs, W.L. (1989) Characterization of the acetyltransferase used in pheromone biosynthesis in moths: Specificity for the Z isomer in tortricidae. Insect Biochem 19: 639-644.

Li, F., Wu, X., Lam, P., Bird, D., Zheng, H., Samuels, L., et al. (2008) Identification of the wax ester synthase/acyl-coenzyme A: diacylglycerol acyltransferase WSD1 required for stem wax ester biosynthesis in Arabidopsis. Plant Physiol 148: 97-107.

Lin, X., Wang, B., Du, Y. (2017) Key genes of the sex pheromone biosynthesis pathway in female moths are required for pheromone quality and possibly mediate olfactory plasticity in conspecific male moths in Spodoptera litura. Insect Mol Biol 27: 8-21.

Livak, K.J., Schmittgen, T.D. (2001) Analysis of Relative gene expression data using real-time quantitative PCR and the 2-ΔΔCt method. Methods 25: 402-408.

Morse, D. and Meighen, E. (1987) Biosynthesis of the acetate ester precursor of the spruce budworm sex pheromone by an acetyl CoA: Fatty alcohol acetyltransferase. Insect Biochem 17: 53-59.

Murray, I.A. and Shaw, W.V. (1997) O-Acetyltransferases for chloramphenicol and other natural products. Antimicrob Agents Chemother 41: 1-6.

Ohnishi, A., Hull, J., Matsumoto, S. (2006) Targeted disruption of genes in the Bombyx mori sex pheromone biosynthetic pathway. Proc Natl Acad Sci USA 103: 4398-4403.

Souleyre, E.J.F., Greenwood, D.R., Friel, E.N., Karunairetnam, S., Newcomb, R.D. (2005) An alcohol acyl transferase from apple (cv. Royal Gala), MpAAT1, produces esters involved in apple fruit flavor. FEBS J 272: 3132-44.

St-Pierre, B. and De Luca, V. (2000) Evolution of acyltransferase genes: origin and diversification of the BAHD superfamily of acyltransferases involved in secondary metabolism. In: Ibrahim, R., Varin, L., De Luca, V., Romeo, J.T., eds. Recent Advances in Phytochemistry Evolution of Metabolic Pathways. 34, pp. 285–315.

Terenius, O., Papanicolaou, A., Garbutt, J.S. Eleftherianos, I., Huvenne, H., Kanginakudru, S., 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.

Vaziri, N.D., Liang, K., Parks, J.S. (2001) Acquired lecithin-cholesterol acyltransferase deficiency in nephrotic syndrome. Am J Physiol Renal Physiol 280: 823-828.

Wang, Z., Dong, Y., Desneux, N., Niu, C. (2013) RNAi silencing of the HaHMG-CoA reductase gene inhibits oviposition in the Helicoverpa armigera cotton bollworm. PLoS One 8: e67732. 154

Wang, J., Gu, L., Ireland, S., Garczynski, S.F., Knipple, D.C. (2015) Phenotypic screen for RNAi effects in the codling moth Cydia pomonella. Gene 572: 184-190.

Zhang, L. and Reed, R.D. (2017) A practical guide to CRISPR/Cas9 genome editing in Lepidoptera. In: Sekimura, T., Nijhout, H.F., eds. Diversity and evolution of butterfly wing patterns, DOI 10.1007/978-981-10-4956-9_8.

Zhang, Y-N., Zhang, L-W., Chen, D-S., Sun, L., Li, Z-Q., Ye, Z-F., Zheng, M-Y. et al. (2017). Molecular identification of differential expression genes associated with sex pheromone biosynthesis in Spodoptera exigua. Mol Genet Genomics 292: 795-809. Zhang, Y-N., Zhu, X-Y., Fang, L-P., He, P., Wang, Z-Q., Chen, G. et al. (2015). Identification and expression profiles of sex pheromone biosynthesis and transport related genes in Spodoptera litura. PLoS One 10: e0140019. Zhang, Y-N., Zhang, X-Q., Zhu, G-H., Zheng, M-Y., Yan, Q., Zhu, X-Y. (2019) A ∆9 desaturase (SlitDes11) is associated with the biosynthesis of ester sex pheromone components in Spodoptera litura. Pest Biochem Physiol 156: 152-159.

155

Figure 6. 1. Transcript expression of TniFAT relative to S5 reference genes in tissues from

T. ni. Tissues analyzed were: pheromone gland(PG), abdomen(AB), fatbody(FB), aedeagus/hair pencil complex(ADE), testis(TSE). F is Female and M is male.

156

Figure 6. 2. Florencence comparison of the EGFP and FAT expression in insect Sf9cells.

Cells Bright: Cells under bright light; EmptyVector (PIB): Negative control with empty vector pIB/V5-His expresed in insect cells; EGFP-PIB: Positive control with EGFP expression; FAT-

EGFP-PIB: the candidate FAT expression under fluorescent light, coexpressed with EGFP in pIB/V5-His vector. 157

Figure 6. 3. RNAi treatment of pheromone glands. A. Semi-quantitative PCR showed there was no knockdown of target genes after the injection of dsRNA. dsEGFP was used as negative control and ds∆11-desaturase (ds11DES) was used as a positive control. B. GC/MS analyzed the amounts of Z7-12: OAc after different treatments. Left: 12h post-emergence female; Right: 24h post-emergence female. ES: Elusion solution; EGFP: Treated with dsEGFP; FAT: Treated with the candidate dsFAT; 11DES: Treated with ∆11 desaturase. ES and EGFP as negative control;

11DES as positive control. NS: non-significant, P > 0.05, two-tailed t-test.

158

CHAPTER 7. CONCLUSIONS

Reproduction is one of the most important features for all life forms. Most insects can reproduce quickly within a relatively short amount of time. This fast reproductive capability means that they can evolve and adjust quickly to a changing environment. Female moths release sex pheromones in a calling behavior, and males capture these components with their antennae to locate females. Female and male moths attach together at their abdomens then the male passes a spermatophore to the female. Female will lay eggs shortly after fertilization. Some species could lay more than 100 eggs at one time, usually near a food source. My dissertation research was to understand one aspect of the mating behavior of moths, particularly sex pheromone biosynthesis in female moths. I utilized multiple approaches in this research, including transcriptomic analysis,

RNA interference, GC/MS, gene expression in yeast and insect cells, qPCR, etc. The overall goal of this dissertation was to identify genes involved in sex pheromone biosynthesis in several species of moths using the techniques just listed.

This dissertation supplies the first transcriptomic difference between female pheromone gland

(PG), female tarsi and male tarsi (Chapter 2), and also the first identification of the genes involved in producing aldehydes in the tarsi (Chapter 3). All of the genes that are related to sex pheromone biosynthesis, regulation, transportation, single transduction found in PGs were also found in tarsi, indicating that there is some overlap of metabolic activity. Fatty acid derivatives are also de novo biosynthesized in tarsi. However, since PBAN is not the factor regulating biosynthesis of aldehyde in tarsi, some other unknown mechanism maybe responsible for inducing aldehyde production, similar to ecdysone regulating sex pheromone biosynthesis in T ni.

Tarsi and ovipositors have been recognized as gustatory tissues since many gustatory receptors are present. In our transcriptome, we indeed found gustatory receptors of different 159

families, but also some odorant receptors, indicating their roles in the reception of odor. However, the expression level of genes encoding odorant receptors is at a low level, indicating ovipositors and tarsi are not the major tissues inovled in odor reception. Trehalose is necessary for sex pheromone production, but the relationship of the expression of the trehalose transpoter and sex pheromone amount is unknown. In our data, we found some genes encoding trehalose transpoter were highly expressed in either PGs or tarsi, indicating they could be very important in aldehyde production in the tarsi.

As for sex pheromone biosynthesis, the PGs had more genes expressed at higher levels that encode DESs, FARs, and AOs, indicating PGs are the major source for sex pheromone production.

However, even with levels of lower expression, DESs, FARs, and AOs were still found in female and male tarsi. We assumed that female and male tarsi utilize an independent system to form the fatty acid derivatives. 16:Ald and 18:Ald were found in female tarsi and male tarsi, and in larger amounts than PGs, so some different genes may be involved in the production of aldehyde in tarsi.

The function for aldehydes in tarsi is unknown but it seems they could be involved in female acceptance and male mating competition. Three candidate genes for producing alcohols were selected based on the expression level in different tissues. Only FAR1 showed the function in production of Z11-16:Ald in PGs and 16:Ald in tarsi. None of them seemed to be involved in converting 18:Me to 18:OH, no matter the functional assay in yeast cells or RNA knockdown. This result is in agreement with experiments on FARs in other heliothine moths. There should be another FAR in tarsi that converts 18C acid to the corresponding 18:OH. There could also be some other mechanism such as α-oxidation of longer chain fatty acids to form the aldehyde. Utilizing stable isotope precursors could help to understand the mechanism of aldehyde production in tarsi. 160

An interesting finding is that FAR1 clusters as a pheromone gland specific FAR, same as other functional FARs in different moths through the phylogeny analysis, but we still found it was present in tarsi. The possible reason is that so far only heliothine moths were found to have aldehydes in the tarsi. The last step of sex pheromone biosynthetic pathway is the production of aldehyde from alcohol by AOs, but no AOs have been identified at the molecular level. We found one AO (AO7) expressed in PG and one AO (AO8) expressed at high levels in both tissues. We assumed these two might be the gene involved in the pathway. However, injection of dsRNA for

AO7 into female moths did not show any decrease in aldehyde levels (data not shown). Future research is required to identify the FAR to make 18:OH and the AO to make aldehyde.

Another significant finding of this dissertation is the study on pink bollworm from fields where mating disruption has failed (Chapter 4). Mating disruption is considered a powerful and effective method to control moth pests. Where mating disruption has been used extensively to control moth pests, only one occurrence of resistance was reported so far. That was for the smaller tea tortrix, and mating disruption was restored when an additional component was added to the disruption blend. In the pink bollworm a change in the ratio of the two component sex pheromone blend apparently created the failure of mating disruption in Israel cotton fields. Some possible reasons that could account for the ratio change of sex pheromone components include environmental effects, food resoures, evolution and gene mutations. In this study, we checked the mRNA level of some genes. In order to understand how the sex pheromone ratio could change, we compared the transcriptomes between the lab population which has never been exposed to mating disruption, and a field population which is mating disruption resistant. Many key genes possibly involved in sex pheromone biosynthesis were found in the transcriptome, such as the ACCs, FASs,

β-oxidation enzymes, DESs, FARs, and FATs. Base on the expression level in both populations, 161

some DESs and FARs were recognized as the genes involved in the ∆11 and ∆9 double insertion and fatty alcohol production. The enzymes including DESs, FARs and FATs could be responsible for the production of different isomers. Since one amino acid variation leads to the functional change of ∆11 desaturase from producing Z isomer to E isomer in Manduca sexta, and some moth desaturases are bifunctional and could produce both isomers, one of the desaturases in PBW could account for the different isomers ratio changes. Similar to Ostrinia nubilalis strains different FARs produce Z or E isomers. Different FARs in PBW may produce the ZZ or ZE isomers.

Through the abundance comparison, we found some differentially expressed genes. The FAT has not been identified at the molecular level, but we found some putative FATs that were differentially expressed between the lab and field population. Unfortunately, we did not find any statistically differentially expressed DESs or FARs. Since the ratio change (50:50 to 60:40) of the two components was not high enough, we assumed the slightly different expression of key enzymes may cause the ratio change. Taken altogether, some changes of candidate DES, FARs and

FATs may lead to the change of pheromone ratios. However, around 50% of differential expressed genes are unknown. So some other mechanisms may account for the change in the pheromone ratios. Further research needs to be conducted to identify their functions.

In addition, I also used the Next Generation Sequencing (NGS) technologies, combining with the bioinformatics to identify some novel viruses, which is different from the conventional approaches. In the transcriptomes, we also found some negative-sense single stranded and positive-sense single stranded RNA viruses. The PBWV1 which was an iflavirus was thought to be present in all PBW since it was found in PBW from Israel and USA. It seems the iflavirus did not cause any symptom in PBW, and it is inherited vertically from parents and distributes to various tissues and development stages, such as larval midgut and adult pheromone gland. Another 162

interesting virus is a (+ss)-RNA virus, which has the same conserved domains such as RdRp, helicase, protease, but the genome structure is in a different order from the common iflavirus.

Three segments of a bunyavirus, which can also infect humans, was found in the PBW PG transcriptome. However, the PBW populations did not have any apparent symptoms of viral infection. Further research is required to determine how wide spread these viruses are and if they can form lethal infections. Since field documented cases of resistance to BT and mating disruption are known in PBW populations, other methods such as biocontrol using a viral pathogen would be a potential approach. The identification of viruses supplies novel pathogens against pest insects.

In the Chapter 6 in this dissertation, I was unable to identify the gene encoding fatty alcohol acetyltransferase involved in sex pheromone biosynthesis. Many methods were attempted including qPCR, RT-PCR, gene expression in insect cells/yeast cells, functional assay by supplying different substrates, and GC/MS analysis. None of them showed the expected results, but it supplied some data that RNA interference did not work well in the cabbage looper, showed the transmembrane protein cannot be heavily expressed in Insect SelectTM BSD system, and potentially exclude one candidate gene. Moreover, since the gene was homologous to wax synthase,

I tried the functional assay by adding the fatty acid plus fatty alcohol. Still, no corresponding wax ester was detected by GC/MS. More work combining transcriptomes with the proteomics to screen all the candidate FATs, CRISPR-Cas9 to analyze the function in vivo, and gene expression with baculovirus expression system, need to be conducted in the future. Also, since RNAi is very inefficient in Lepidoptera, identify FATs in transgenetic Drosophila strains would be a better option.

This dissertation supplies some significant results for the sex pheromone communication in moths with the combination of molecular work and bioinformatics. The gene identification from 163

tarsi and PGs gives new sight on the gene function in different tissues, which is different from the gene that is specific expressed in PGs if it is involved in sex pheromone biosynthesis. Moreover, the screening of genes from the transcriptome helps identify the function of these tissues. A significant finding is that odorant receptors were also present in the ovipositor and tarsi which are known gustatory tissues. In addition, this dissertation provides important evidence for the resistance of PBW to mating disruption and analyzes the potential mechanism of this resistance.

Also, it gives an up to date approach to identify RNA viruses which could provide candidate novel pathogens for PBW control. This research will inform future investigation on FATs and AOs identification in moths.

164

APPENDIX A. ADDITIONAL FIGURES AND TABLES FOR CHAPTER 2

Supplemental Table 2.1. Common up regulated genes in female PG and male tarsi compared to female tarsi

Female PG vs Male tarsi vs female Gene description female tarsi tarsi

logFC FDR logFC FDR

Elongation of very long chain fatty acids protein 2.19493 9.65E-07 2.56384 1.86E-13

Acetylcholinesterase 2.12580 0.00011 2.27659 1.95E-08

Glutathione hydrolase-like YwrD proenzyme 3.59602 2.98E-08 2.57919 0.00029

Larval cuticle protein 3.06059 0.00073 2.35464 0.001734

165

Supplemental Table 2.2. Common down regulated genes in female PG and male tarsi compared to female tarsi

Female PG vs

Gene description female tarsi male_tarsi vs female tarsi

logFC FDR logFC FDR

Alpha-tocopherol transfer protein-like -4.57671 8.95E-26 -7.278563 8.097

Nose resistant to fluoxetine protein -6.06688 4.08E-25 -6.464636 4.12E-34

Lipase -2.06582 7.14E-06 -4.365309 7.82E-32

Alpha-tocopherol transfer protein-like -3.66422 1.89E-19 -3.296347 3.22E-24

Unknown -3.93746 1.59E-17 -3.839762 8.29E-22

Presenilin homolog -3.66087 6.92E-21 -2.925512 1.01E-21

Integrator complex -4.48124 2.46E-22 -2.585715 5.48E-18

Unknown -5.9991 2.51E-14 -5.76572 5.66E-17

Probable RNA-directed DNA polymerase from transposon X-element -3.88341 4.25E-20 -2.772272 8.91E-17

Cytochrome P450 -5.5583 9.85E-28 -3.855442 4.12E-16

Multiple C2 and transmembrane domain- containing protein -4.31153 1.69E-27 -2.339409 2.25E-14

Cytochrome P450 -6.21514 4.82E-38 -2.244592 1.37E-13

Facilitated trehalose transporter -6.86152 1.57E-28 -3.158977 4.96E-13

Unknown -5.0064 2.87E-28 -2.185825 2.13E-12

Iron-sulfur cluster co-chaperone protein -4.85516 3.40E-22 -2.26096 5.57E-12

ATP-binding cassette sub-family G member 1 -5.13522 1.21E-31 -2.141378 2.60E-11

Cytochrome P450 -4.63469 9.03E-21 -2.747327 1.46E-10

Alpha-tocopherol transfer protein-like -2.37468 1.19E-08 -2.147295 2.94E-10 166

Supplemental Table 2.2. (Continued)

Ras-related and estrogen-regulated growth inhibitor -2.81733 4.14E-08 -3.049069 3.05E-10

RNA-directed DNA polymerase from mobile element jockey -5.63636 2.32E-20 -2.426548 4.13E-08

Nose resistant to fluoxetine protein -2.17833 3.07E-05 -2.576758 4.21E-08

Unknown -6.30093 4.69E-27 -2.342703 5.24E-07

Viral cathepsin -3.20437 0.012085 -6.046498 1.72E-06

Viral cathepsin -3.80477 0.00454 -6.332263 6.56E-06

Unknown -5.14321 3.79E-09 -3.141028 7.21E-06

Cathepsin B -4.11765 0.002136 -6.004942 2.86E-05

Putative serine protease -3.70828 0.014605 -7.443515 9.33E-05

Probable RNA-directed DNA polymerase -3.44598 4.36E-10 -2.090642 0.000164

Viral cathepsin -4.64818 0.008122 -7.333303 0.000273

N-acetylgalactosaminidase -3.80484 0.016426 -6.924415 0.000321

Venom serine carboxypeptidase -2.16394 0.044877 -3.657403 0.000789

Unknown -3.30531 5.89E-05 -3.342193 0.00626

Unknown -4.58154 1.01E-06 -2.119653 0.007932

Unknown -4.48105 0.007223 -5.335475 0.010555

167

Supplementary Table 2.3. Blastp results of all selected genes from H.zea PG and tarsi

Transcripts GenBank homologue description e- identity(%) Accession Heliothinae_PG* value ACC acetyl-CoA carboxylase [Helicoverpa 0 99 ALS92678 ar, v, as armigera] FAS1 fatty acid synthase-like [Helicoverpa 1E-33 99 XP_021186749 ar, v, as armigera] FAS2 fatty acid synthase-like [Helicoverpa 0 99 XP_021182516 v, as armigera] FAS3 atty acid synthase-like [Helicoverpa 0 93 XP_021181026 armigera] LPAQ acyl-CoA delta-11 desaturase [Helicoverpa 0 100 AAF81787 ar, v, as zea] PDSN acyl-CoA Delta(11) desaturase [Helicoverpa 0 99 XP_021183629 armigera] NPAE acyl-CoA Delta(11) desaturase-like 0 99 XP_021200693 [Helicoverpa armigera] GATD acyl-CoA desaturase 4 [Helicoverpa assulta] 0 96 AKU76410 v QPVE acyl-CoA desaturase-like [Helicoverpa 0 96 XP_021195370 armigera] NPVE acyl-CoA delta-9 desaturase [Helicoverpa 0 100 AAF81790 ar, v, as zea] KPSE acyl-CoA Delta(11) desaturase-like 0 100 XP_021190176 v, as [Helicoverpa armigera] KSVE acyl-CoA desaturase 6 [Helicoverpa 0 97 AKU76405 v armigera] DES9 acyl-CoA desaturase 8 [Helicoverpa assulta] 0 100 AKU76414 ar, as FAR1 fatty acyl reductase 1 [Helicoverpa assulta] 0 98 ATJ44516 ar, v, as FAR2 putative fatty acyl-CoA reductase CG5065 0 99 XP_021181288 ar, as [Helicoverpa armigera] FAR3 putative fatty acyl-CoA reductase CG5065 0 99 XP_021197383 ar [Helicoverpa armigera] FAR4 fatty acyl-CoA reductase 9 [Helicoverpa 0 99 AKD01770 ar, as armigera] FAR5 fatty acyl-CoA reductase 1 [Helicoverpa 0 99 XP_021197389 ar, as armigera] FAR6 putative fatty acyl-CoA reductase CG5065 0 93 XP_021194316 [Helicoverpa armigera] FAR7 fatty acyl-CoA reductase 10 [Helicoverpa 2E-167 99 AKD01771 ar, v armigera] FAR8 fatty acyl reductase 12 [Helicoverpa 0 99 ATJ44469 ar, v armigera] FAR9 fatty acyl-CoA reductase wat-like 0 98 XP_021197953 [Helicoverpa armigera] FAR10 fatty acyl reductase 5 [Helicoverpa armigera] 0 99 ATJ44463 ar, as FAR11 putative fatty acyl-CoA reductase CG8306 0 100 XP_021198391 ar, v isoform X1 [Helicoverpa armigera] FAR12 putative fatty acyl-CoA reductase CG5065 0 100 XP_021199436 ar, as [Helicoverpa armigera] FAR13 putative fatty acyl-CoA reductase CG5065 0 99 XP_021197384 ar isoform X1 [Helicoverpa armigera] FAR14 fatty acyl-CoA reductase 13 [Helicoverpa 0 99 AKD01791 ar assulta] FAR15 fatty acyl-CoA reductase 1 [Helicoverpa 0 100 AKD01779 ar, as assulta] FAR16 fatty acyl-CoA reductase wat-like 0 98 XP_021192389 v [Helicoverpa armigera]

168

Supplementary Table 2.3. (Continued) FAR17 putative fatty acyl-CoA reductase CG5065 2E-104 81 XP_021199763 [Helicoverpa armigera] FAR18 putative fatty acyl-CoA reductase CG5065 0 98 XP_021192560 ar, as [Helicoverpa armigera] FAR19 fatty acyl reductase 14 [Helicoverpa 0 100 ATJ44467 ar armigera] FAR20 fatty acyl-CoA reductase 1-like [Helicoverpa 0 95 XP_021199138 ar armigera] AO1 PREDICTED: alcohol dehydrogenase 2E-164 99 XP_016937524 [Drosophila suzukii] AO2 alcohol dehydrogenase-like [Helicoverpa 2E-102 80 XP_021201213 armigera] AO3 alcohol dehydrogenase class-3 [Helicoverpa 9E-179 99 XP_021189392 ar, as armigera] AO4 alcohol dehydrogenase [NADP(+)] isoform 0 99 XP_021193613 ar, v, as X2 [Helicoverpa armigera] AO5 aldehyde reductase 8 [Helicoverpa armigera] 0 98 ATJ44503 v AO6 zinc-type alcohol dehydrogenase-like protein 0 100 XP_021192204 ar, as C1773.06c [Helicoverpa armigera] AO7 alcohol oxidase 1 [Helicoverpa armigera] 0 91 ATJ44473 v, as AO8 hypothetical protein B5X24_HaOG209534 0 98 PZC73438 ar, v, as [Helicoverpa armigera] ADC1 adenylate cyclase type 9 isoform X2 0 99 XP_021193347 as [Helicoverpa armigera] ADC2 adenylate cyclase type 2-like [Helicoverpa 0 97 XP_021186709 ar, as armigera] ADC3 adenylate cyclase type 9 isoform X1 0 99 XP_021193346 ar [Helicoverpa armigera] ADC4 adenylate cyclase type 2-like isoform X1 9E-150 99 XP_026733556 as [Trichoplusia ni] ADC5 LOW QUALITY PROTEIN: 0 99 XP_021181195 as Ca(2+)/calmodulin-responsive adenylate cyclase-like [Helicoverpa armigera] ADC6 adenylate cyclase type 8-like [Helicoverpa 0 99 XP_021190791 ar armigera] ADC7 adenylate cyclase type 6-like [Helicoverpa 0 99 XP_021194809 armigera] ADC8 adenylate cyclase type 8-like [Helicoverpa 0 98 XP_021190757 ar, as armigera] PKA1 PREDICTED: cAMP-dependent protein 0 100 XP_013134221 ar, as kinase catalytic subunit [Papilio polytes] PKA2 cAMP-dependent protein kinase type II 0 100 XP_021183092 regulatory subunit isoform X1 [Helicoverpa armigera] PKA3 cAMP-dependent protein kinase type I 0 100 XP_021181426 ar, as regulatory subunit isoform X1 [Helicoverpa armigera] IP3R1 inositol 1,4,5-trisphosphate receptor isoform 0 99 XP_021188707 ar X1 [Helicoverpa armigera] IP3R2 inositol 1,4,5-trisphosphate receptor isoform 0 99 XP_021188709 ar, as X3 [Helicoverpa armigera] IP3R3 inositol 1,4,5-trisphosphate receptor isoform 0 97 XP_021188708 ar, as X2 [Helicoverpa armigera] PLC1 1-phosphatidylinositol 4,5-bisphosphate 0 99 XP_022826645 ar phosphodiesterase isoform X1 [Spodoptera litura] PLC2 1-phosphatidylinositol 4,5-bisphosphate 0 99 XP_021201241 ar, as phosphodiesterase classes I and II isoform X1 [Helicoverpa armigera]

169

Supplementary Table 2.3. (Continued) PLC3 1-phosphatidylinositol 4,5-bisphosphate 0 100 XP_021187906 ar, as phosphodiesterase gamma-1 [Helicoverpa armigera] Orai calcium release-activated calcium channel 4E-161 100 XP_021180862 ar protein 1 isoform X3 [Helicoverpa armigera] AMPK1 5'-AMP-activated protein kinase subunit 3E-110 84 XP_022834520 ar, as gamma-2-like isoform X2 [Spodoptera litura] AMPK2 5'-AMP-activated protein kinase subunit 0 100 XP_022820753 ar, as gamma-2-like [Spodoptera litura] AMPK3 5'-AMP-activated protein kinase catalytic 0 100 XP_021182530 ar, as subunit alpha-2 isoform X3 [Helicoverpa armigera] AMPK4 5'-AMP-activated protein kinase subunit 0 100 XP_021190181 ar, as beta-1 isoform X2 [Helicoverpa armigera] CaMKII1 Ca2+/calmodulin-dependent protein kinase 0 99 AID54518 ar, as II [Helicoverpa armigera] CaMKII2 calcium/calmodulin-dependent protein 0 100 XP_021182298 ar, as kinase type 1-like [Helicoverpa armigera] CaMKII3 calcium/calmodulin-dependent protein 2E-157 100 XP_021193715 ar kinase kinase 2 isoform X1 [Helicoverpa armigera] PP2B1 calcineurin B [Bombyx mori] 7E-120 100 NP_001037026 ar, as PP2B2 serine/threonine-protein phosphatase 2B 0 100 XP_021181213 ar, as catalytic subunit 3-like [Helicoverpa armigera] Stim1 Stromal interaction molecule-like [Papilio 0 100 KPI98201 ar, as xuthus] Stim2 stromal interaction molecule homolog 0 100 XP_021185698 ar, as isoform X2 [Helicoverpa armigera] Calmodulin1 calmodulin-A-like isoform X5 [Helicoverpa 2E-127 100 XP_021198995 ar, as armigera] Calmodulin2 Calmodulin [Trichinella pseudospiralis] 4E-105 100 KRX92830 ar, v, as Calmodulin3 calmodulin-like protein 4 [Helicoverpa 6E-110 100 XP_021199968 armigera] OBP1 odorant-binding protein 31 [Helicoverpa 2E-100 99 ASA40067 as armigera] OBP2 general odorant binding protein 2 7E-116 100 AAG54078 [Helicoverpa zea] OBP3 odorant-binding protein [Helicoverpa 8E-133 98 AEX07273 v, as assulta] OBP4 odorant binding protein 10 [Ostrinia 8E-100 63 BAV56797 furnacalis] OBP5 odorant-binding protein 19 [Helicoverpa 1E-142 98 AGC92793 ar, v assulta] OBP6 odorant binding protein 9 [Spodoptera litura] 0 81 ALD65883 OBP7 general odorant-binding protein 28a-like 9E-68 63 XP_022826775 isoform X4 [Spodoptera litura] OBP8 general odorant-binding protein 70 8E-131 99 XP_021188671 [Helicoverpa armigera] OBP9 general odorant-binding protein 69a-like 1E-90 82 XP_021194655 [Helicoverpa armigera] OBP10 odorant-binding protein [Helicoverpa 8E-87 98 AEX07270 v, as assulta] OBP11 general odorant-binding protein 1-like 8E-121 98 XP_021192650 [Helicoverpa armigera] OBP12 general odorant-binding protein 1 2E-110 100 XP_021192665 [Helicoverpa armigera]

170

Supplementary Table 2.3. (Continued) OBP13 general odorant-binding protein 72- 2E-38 99 XP_021193810 v like[Helicoverpa armigera] CSP1 chemosensory protein [Helicoverpa 6E-84 98 AIW65099 v armigera] CSP2 putative chemosensory protein [Sesamia 1E-87 84 AGY49270 inferens] CSP3 chemosensory protein 23 [Helicoverpa 6E-85 98 ASA40079 armigera] CSP4 chemosensory protein 5 [Athetis dissimilis] 4E-70 95 AND82447 as CSP5 chemosensory protein 14 [Adelphocoris 2E-82 94 AXS78220 v lineolatus] CSP6 chemosensory protein [Helicoverpa assulta] 9E-81 94 ABB91378 v CSP7 chemosensory protein 3 [Athetis dissimilis] 2E-80 93 AND82445 CSP8 chemosensory protein [Helicoverpa 1E-60 92 AIW65097 v, as armigera] CSP9 chemosensory protein 5 [Spodoptera exigua] 3E-66 74 AKT26482 CSP10 chemosensory protein [Helicoverpa 0 99 AIW65104 ar, as armigera] CSP11 chemosensory protein 25 [Helicoverpa 6E-76 95 ASA40086 v assulta] CSP12 chemosensory protein 25 [Helicoverpa 4E-91 97 ASA40087 v armigera] OR1 odorant receptor coreceptor [Helicoverpa 0 100 XP_021195606 armigera] OR2 olfactory receptor 4 [Helicoverpa armigera] 3E-155 97 ACF32962 OR3 odorant receptor 4-like [Helicoverpa 3E-176 97 XP_021184412 armigera] OR4 odorant receptor 13a-like [Helicoverpa 0 99 XP_021200762 armigera] OR5 odorant receptor 4-like [Helicoverpa 0 99 XP_021191841 armigera] OR6 odorant receptor [Helicoverpa armigera] 0 98 AIG51879 as OR7 odorant receptor 4-like [Helicoverpa 0 97 XP_021191840 armigera] OR8 odorant receptor [Helicoverpa armigera] 0 99 AIG51888 OR9 putative odorant receptor 92a [Helicoverpa 0 97 XP_021200876 armigera] OR10 odorant receptor [Helicoverpa armigera] 1E-137 98 AIG51889 OR11 olfactory receptor 14a [Helicoverpa 1E-176 94 AGK90005 armigera] OR12 olfactory receptor 30 [Helicoverpa assulta] 9E-64 99 AJD81565 OR13 odorant receptor 23a-like [Helicoverpa 0 99 XP_021185068 armigera] OR14 hypothetical protein B5X24_HaOG208188 0 98 PZC74198 [Helicoverpa armigera] OR15 putative odorant receptor [Peridroma saucia] 4E-94 84.15 AVF19676 OR16 odorant receptor 67c-like [Helicoverpa 1E-163 98 XP_021186882 armigera] OR17 odorant receptor [Helicoverpa armigera] 2E-94 99 AIG51869 OR18 putative odorant receptor [Peridroma saucia] 9E-109 90 AVF19625 v GR1 gustatory receptor [Helicoverpa armigera] 1E-97 94 AIG51911 GR2 gustatory and odorant receptor 22 0 99 XP_021185659 [Helicoverpa armigera] GR3 gustatory receptor 1 [Helicoverpa armigera] 0 99 AGK90010 GR4 gustatory receptor 4 [Helicoverpa armigera] 0 99 ASW18693 GR5 gustatory receptor 7 [Helicoverpa armigera] 0 98 ASW18696 GR6 gustatory receptor for sugar taste 64e-like 4E-179 98 XP_021201486 [Helicoverpa armigera] 171

GR7 gustatory receptor [Helicoverpa armigera] 1E-154 98 AGA04648 Supplementary Table 2.3. (Continued) GR8 hypothetical protein B5X24_HaOG200922 0 98 PZC81404 ar, as [Helicoverpa armigera] GR9 gustatory receptor 68a isoform X1 0 99 XP_021188277 [Helicoverpa armigera] IR1 ionotropic receptor 25a isoform X1 0 100 XP_021185959 [Helicoverpa armigera] IR2 ionotropic receptor 93a [Helicoverpa 0 99 XP_021190111 armigera] IR3 uncharacterized protein LOC110383944 0 99 XP_021200603 as [Helicoverpa armigera] IR4 ionotropic receptor 21a [Helicoverpa 2E-131 99 XP_021200448 armigera] IR5 uncharacterized protein LOC110379185 0 98 XP_021194402 [Helicoverpa armigera] IR6 ionization receptor 64a [Helicoverpa 0 93 ARB05670 armigera] IR7 hypothetical protein B5X24_HaOG200842 0 97 PZC86345 [Helicoverpa armigera] SNMP sensory neuron membrane protein 2 0 99 XP_021182698 ar, v, as [Helicoverpa armigera] ODE1 acetate esterase 6 [Helicoverpa assulta] 0 98 ATJ44550 ar, as ODE2 juvenile hormone esterase-like isoform X1 0 97 XP_021192216 ar [Helicoverpa armigera] ODE3 acetate esterase 5 [Helicoverpa armigera] 0 98 ATJ44480 ar, v, as ODE4 odorant degrading enzyme CXE20 [Sesamia 0 72 AII21992 v, as inferens] ODE5 juvenile hormone esterase-like isoform X2 0 99 XP_021190399 ar, as [Helicoverpa armigera] ODE6 carboxylesterase 1E [Helicoverpa armigera] 0 99 XP_021194420 ar ODE7 carboxylesterase 5A [Helicoverpa armigera] 0 96 XP_021191546 ar, as ALDH1 aldehyde dehydrogenase, dimeric NADP- 0 99 XP_021193191 ar, v, as preferring isoform X2 [Helicoverpa armigera] ALDH2 aldehyde dehydrogenase X, mitochondrial- 0 99 XP_021182274 v, as like [Helicoverpa armigera] ALDH3 putative aldehyde dehydrogenase family 7 0 99 XP_021184994 ar, as member A1 homolog [Helicoverpa armigera] ALDH4 aldehyde dehydrogenase, mitochondrial 0 98 XP_021201072 v, as [Helicoverpa armigera] ETHR ecdysis triggering hormone receptor 0 77 AAX19163 ar subtype-A [Manduca sexta] DHR diapause hormone receptor [Helicoverpa 0 99 AGR34305 ar, v, as zea] SPR sex peptide receptor [Helicoverpa armigera] 0 99 XP_021183461 v, as OctoR1 octopamine receptor beta-2R-like 0 99 XP_021187627 ar, as [Helicoverpa armigera] OctoR2 octopamine receptor Oamb isoform X1 0 100 XP_021194395 [Helicoverpa armigera] PBANR_B pyrokinin-1 receptor-like isoform X1 0 99 XP_021183657 ar, v [Helicoverpa armigera] PBANR_C pyrokinin-1 receptor-like isoform X2 0 99 XP_021183658 ar, v [Helicoverpa armigera] Tret1 facilitated trehalose transporter Tret1-like 0 100 XP_021187379 ar, v, as [Helicoverpa armigera] Tret2 facilitated trehalose transporter Tret1-like 0 98 XP_021195229 ar, v [Helicoverpa armigera] Tret3 hypothetical protein B5V51_3992 [Heliothis 0 96 PCG69525 ar virescens] 172

Supplementary Table 2.3. (Continued) Tret4 facilitated trehalose transporter Tret1-2 0 100 XP_021187107 as homolog isoform X1 [Helicoverpa armigera] Tret5 facilitated trehalose transporter Tret1-2 0 100 XP_021191421 ar, v, as homolog [Helicoverpa armigera] Tret6 facilitated trehalose transporter Tret1-2 0 100 XP_021186692 ar, as homolog isoform X3 [Helicoverpa armigera] Tret7 facilitated trehalose transporter Tret1-2 0 99 XP_021193322 ar, as homolog [Helicoverpa armigera] Tret8 facilitated trehalose transporter Tret1-like 0 99 XP_021188615 ar, as [Helicoverpa armigera] Tret9 facilitated trehalose transporter Tret1-like 0 99 XP_021200332 ar [Helicoverpa armigera] Tret10 facilitated trehalose transporter Tret1-like 0 91 XP_021190825 ar, as [Helicoverpa armigera] Tret11 facilitated trehalose transporter Tret1-like 0 99 XP_021197355 ar, as [Helicoverpa armigera] Tret12 hypothetical protein B5X24_HaOG212390 0 96 PZC71834 [Helicoverpa armigera] Tret13 facilitated trehalose transporter Tret1-like 0 99 XP_021194570 isoform X1 [Helicoverpa armigera] Tret14 solute carrier family 2, facilitated glucose 0 99 XP_021191477 ar transporter member 6-like [Helicoverpa armigera]

173

Supplementary Table 2.4. Comparison of candidate transcripts involved in pheromone biosynthesis, chemosensation, and signal transduction in H. zea PG and tarsi. FPKM values are shown for each replication.

Pheromone gland-ovipositor Female tarsi Male tarsi Transcripts Rep1 Rep2 Rep3 Rep1 Rep2 Rep3 Rep1 Rep2 Rep3 Acetyl-CoA Carboxylase ACC 94.3 82.5 50.4 38.6 26.4 21.2 24.1 18.2 26.5 Fatty acid synthetase FAS1 1270.64 714.05 927.15 28.55 16.34 20.07 31.58 27.85 31.66 FAS2 0.7 0.1 0.3 0.3 0.5 0.3 0.4 0.0 0.1 FAS3 0.2 0.0 0.1 0.3 0.2 1.0 1.0 0.5 0.0 Desaturase LPAQ 8663.4 7949.9 7233.2 30.7 52.4 45.3 32.1 23.1 37.1 PDSN 0.4 0.0 0.4 0.0 4.7 4.1 2.7 3.9 0.0 NPAE 0.1 0.3 0.4 0.2 1.1 0.6 1.0 0.1 0.1 GATD 1.0 25.5 2.3 1.6 1.4 0.9 2.2 1.6 3.3 QPVE 0.0 0.0 0.0 0.0 0.3 0.0 0.0 0.2 0.0 NPVE 53.9 101.5 77.7 18.2 11.9 17.9 13.5 8.6 11.3 KPSE 216.4 273.3 418.8 63.1 56.2 36.7 23.2 27.5 25.3 KSVE 0.8 0.3 0.2 0.0 0.0 0.4 0.5 0.0 0.6 DES9 10.5 10.4 15.6 8.2 11.5 7.1 12.1 7.7 10.8 Fatty-acyl reductase FAR1 1197.8 1453.0 1838.2 10.4 10.4 14.2 17.6 7.2 11.1 FAR2 9.9 14.7 9.6 12.9 15.2 7.0 9.1 9.0 7.8 FAR3 0.5 0.2 0.2 0.0 0.0 0.0 0.0 0.0 0.0 FAR4 189.9 141.6 172.0 243.2 152.0 118.7 37.0 10.1 17.0 FAR5 31.0 25.1 41.3 116.2 192.8 132.9 156.6 178.1 203.9 FAR6 0.4 1.1 1.5 27.8 21.3 28.1 109.6 51.8 59.7 FAR7 63.5 49.0 61.1 215.5 158.7 150.3 255.4 180.6 183.5 FAR8 4.2 5.2 6.1 14.1 12.2 10.2 7.2 2.6 4.7 FAR9 0.0 0.0 0.0 2.4 1.4 0.3 0.0 0.0 0.2 FAR10 4.7 8.2 9.4 11.7 12.7 9.4 12.9 10.1 18.1 FAR11 91.9 66.4 62.8 170.8 103.9 94.5 221.1 113.9 129.4 FAR12 1.1 1.9 6.1 1.1 0.2 0.3 0.2 0.1 0.0 FAR13 10.6 17.5 28.9 2.6 1.5 4.5 9.2 6.1 3.9 FAR14 3.8 4.9 14.7 53.3 46.3 39.8 74.5 46.5 65.2 FAR15 39.7 45.1 16.6 78.3 70.9 90.0 98.3 98.9 103.8 FAR16 0.6 0.4 0.2 0.0 0.1 0.2 0.2 0.0 0.4 FAR17 0.4 0.1 0.0 0.0 0.6 0.2 0.0 0.0 1.0 FAR18 11.3 12.3 8.3 10.7 12.3 6.8 10.1 4.0 6.8 FAR19 1.9 2.7 2.1 2.3 2.9 1.4 1.8 1.3 2.3 FAR20 0.7 0.4 0.5 0.0 0.0 0.0 0.0 0.0 0.0 Alcohol oxidase AO1 0.1 0.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 AO2 13.3 6.6 3.6 52.3 55.0 68.1 39.7 34.7 38.5 AO3 6.4 8.1 20.9 11.4 12.2 8.5 17.7 13.7 17.8 AO4 6.9 7.0 19.0 18.3 25.9 16.5 20.8 28.4 26.6 AO5 32.7 43.6 57.4 4.1 2.7 3.2 4.9 6.5 7.0 AO6 70.3 51.2 36.7 27.7 20.1 18.3 18.3 11.7 11.6 AO7 941.1 684.0 696.7 8.0 10.8 7.2 8.2 6.2 7.2 AO8 290.7 257.5 371.2 349.2 342.6 262.0 260.7 353.4 282.0 Adenylate cyclase ADC1 2.8 2.7 2.3 0.0 0.5 0.5 2.3 0.9 1.0 ADC2 5.8 0.1 1.6 2.1 3.3 3.5 5.4 0.8 4.3 ADC3 3.8 4.5 5.5 4.9 4.9 3.5 3.0 1.7 2.8 174

Supplementary Table 2.4. (Continued) ADC4 2.1 2.0 6.4 1.4 0.4 0.6 2.6 2.5 3.1 ADC5 18.0 11.7 11.2 2.1 0.0 0.8 4.1 3.9 4.8 ADC6 1.8 0.7 0.7 0.0 0.0 0.0 0.9 0.9 0.4 ADC7 0.0 0.0 0.0 2.3 0.4 0.4 3.3 2.7 2.4 ADC8 0.0 0.0 0.0 2.9 2.5 1.2 3.6 1.6 2.0 Protein Kinase A PKA1 85.4 85.3 90.5 25.9 16.2 12.9 50.1 34.3 42.6 PKA2 12.8 12.2 22.4 10.5 1.6 4.4 11.6 12.3 9.8 PKA3 35.0 25.3 31.3 14.2 6.0 6.2 6.0 2.2 5.3 Inositol trisphosphate receptor IP3R1 4.9 5.4 6.7 12.3 10.1 9.9 14.9 12.1 14.9 IP3R2 3.0 3.1 3.6 2.0 2.5 2.2 3.5 3.2 4.3 IP3R3 2.6 2.4 4.3 2.2 1.8 1.2 4.5 1.1 2.5 Phospholipase C PLC1 16.1 11.4 8.7 23.0 18.2 22.8 26.4 18.4 24.7 PLC2 18.5 14.2 10.1 15.9 13.0 19.1 14.0 14.4 14.4 PLC3 9.3 10.5 6.5 26.5 31.8 27.4 21.0 23.2 23.5 Calcium release-activated calcium channel protein Orai 19.4 12.7 17.3 27.4 23.6 24.2 32.6 25.2 37.6 5′-AMP-activated protein kinase AMPK1 3.2 2.9 3.2 1.5 1.8 1.0 1.3 1.4 1.6 AMPK2 36.3 39.6 50.8 2.9 3.2 3.8 10.4 11.8 7.4 AMPK3 70.3 67.3 65.5 44.5 42.4 28.7 42.8 26.0 31.3 AMPK4 33.7 37.2 39.7 33.9 33.7 31.3 19.5 14.7 19.2 Calcium/calmodulin-dependent protein kinase CaMKII1 9.5 8.1 10.6 39.1 33.5 38.0 28.9 16.7 16.7 CaMKII2 5.5 9.6 2.0 12.1 9.6 12.3 22.9 8.2 13.4 CaMKII3 7.1 4.1 3.7 2.5 3.0 4.4 17.0 18.8 22.4 Protein phosphatase 2B PP2B1 37.6 40.8 142.8 33.8 22.3 24.3 59.9 32.5 32.2 PP2B2 87.5 93.7 109.6 70.3 54.9 56.8 72.6 53.5 64.1 Stromal interaction molecule Stim1 0.0 0.0 0.0 0.0 0.6 0.0 33.6 22.8 20.6 Stim2 16.8 26.9 10.1 3.7 6.1 2.6 7.9 5.9 16.7 Calmodulin Calmodulin1 2.0 3.4 8.0 0.0 0.0 0.2 1.2 1.9 1.0 Calmodulin2 537.5 595.0 692.5 817.6 1033.0 845.5 860.8 953.6 955.8 Calmodulin3 1.9 1.7 7.4 4.8 6.0 5.8 6.9 3.4 5.3 Odorant binding protein OBP1 0.2 0.0 0.3 5.2 7.0 2.1 3.3 1.8 2.0 OBP2 1.5 2.1 2.8 19.6 19.6 16.6 24.5 14.2 20.1 OBP3 3.9 6.2 13.2 2.6 3.1 2.6 3.3 2.0 2.8 OBP4 0.2 0.0 0.3 0.4 0.1 1.3 0.6 0.9 1.9 OBP5 8.1 8.7 17.8 7.7 3.4 8.4 8.8 10.0 9.6 OBP6 0.0 1.7 0.3 0.8 4.3 1.3 4.7 1.3 1.2 OBP7 2.1 0.5 2.1 25.9 21.8 12.8 8.3 6.0 10.5 OBP8 51.3 74.8 48.8 237.5 228.7 305.5 322.0 386.9 384.1 OBP9 453.2 458.6 130.3 195.1 109.9 78.5 52.9 77.0 69.2 OBP10 1.6 1.4 2.0 310.3 185.6 84.7 30.6 37.4 40.5 OBP11 10.2 8.8 5.0 48.4 30.2 30.4 24.0 32.7 28.5 OBP12 11.7 12.7 7.8 31.6 33.7 29.2 20.1 24.9 22.8 OBP13 80.8 82.3 169.3 15966.7 18626.7 14449.7 8107.4 6359.7 7864.3 Chemosensory protein CSP1 110.8 147.0 173.5 1344.9 1277.5 1078.3 2731.5 2196.2 2495.8 175

Supplementary Table 2.4. (Continued) CSP2 0.4 0.1 0.0 0.8 0.7 0.1 0.2 0.4 0.0 CSP3 0.3 0.0 0.0 5.6 9.3 3.6 5.0 4.2 3.7 CSP4 2.6 2.1 4.3 10.3 8.1 16.2 14.4 13.0 15.3 CSP5 38.9 59.8 152.6 1434.9 1682.9 914.6 1565.9 1056.4 1118.0 CSP6 65.7 81.6 220.1 2181.2 1614.1 2465.6 4964.7 3282.8 1779.2 CSP7 33.5 12.8 7.2 37.7 51.9 29.4 15.9 18.0 14.4 CSP8 4613.1 4867.9 10303.0 62.8 60.9 58.7 112.4 54.3 59.7 CSP9 6.9 2.8 4.1 59.1 134.2 111.1 176.7 195.1 188.9 CSP10 77.3 65.6 54.8 75.3 85.6 143.8 182.8 215.6 212.9 CSP11 155.2 182.8 331.8 5598.2 7422.1 3248.2 4945.3 3305.9 3925.4 CSP12 1.4 1.2 3.1 106.9 87.5 59.7 191.8 68.8 90.9 Odorant receptor OR1 0.0 0.0 0.1 0.9 1.0 0.3 0.6 0.6 0.6 OR2 0.0 0.0 0.1 0.7 0.5 0.0 0.2 0.8 0.5 OR3 0.4 0.0 0.0 1.1 0.0 0.4 0.3 0.4 0.0 OR4 3.5 3.8 5.1 4.0 4.9 5.1 7.6 4.2 4.3 OR5 0.2 0.0 0.2 1.5 0.7 1.8 1.4 3.8 1.5 OR6 68.8 74.4 38.9 51.2 41.0 51.7 49.2 34.8 32.1 OR7 0.0 0.1 0.0 1.7 2.0 1.0 2.5 3.5 2.8 OR8 0.0 0.0 0.0 0.1 0.7 0.2 0.2 0.4 0.0 OR9 0.0 0.0 0.0 0.9 0.7 1.5 9.4 4.0 6.9 OR10 0.1 0.1 0.1 0.2 0.0 0.4 0.6 0.1 0.1 OR11 0.4 0.0 0.2 0.0 0.1 0.0 0.6 0.0 0.0 OR12 0.2 0.1 0.1 0.2 0.6 0.2 0.2 0.4 0.3 OR13 0.0 0.0 0.0 0.1 0.2 0.1 0.7 0.1 0.4 OR14 0.0 0.0 0.0 0.0 0.4 0.5 0.6 0.2 0.4 OR15 31.8 30.0 41.3 42.6 37.9 30.7 40.9 37.9 37.7 OR16 4.7 2.0 6.2 5.7 7.2 4.7 6.1 4.0 10.3 OR17 1.2 1.1 0.6 0.0 0.7 0.0 0.6 0.5 0.4 OR18 0.1 0.2 0.5 0.0 0.1 0.1 0.1 0.0 0.0 Gustatory receptor GR1 0.0 0.0 0.0 1.2 1.1 3.8 1.4 0.9 1.2 GR2 3.1 7.3 4.5 2.9 1.2 2.6 4.1 7.8 1.5 GR3 12.1 17.6 14.1 16.9 24.4 20.8 18.5 13.9 21.0 GR4 1.3 0.6 0.6 4.7 4.9 1.8 3.0 2.8 3.0 GR5 0.1 0.0 0.0 4.3 2.9 2.2 0.9 2.0 2.5 GR6 0.0 0.0 0.0 1.3 2.1 1.3 2.1 2.1 2.7 GR7 0.0 0.0 0.1 0.7 0.5 0.0 0.2 0.8 0.5 GR8 1.6 1.3 3.0 2.5 1.9 2.9 3.0 3.0 3.2 GR9 0.0 0.0 0.0 1.2 0.7 0.8 0.2 0.2 0.8 Ionotropic receptor IR1 2.8 5.2 5.8 10.4 11.5 7.9 7.7 9.5 10.2 IR2 1.6 0.6 1.8 2.4 3.0 4.0 1.7 1.4 1.7 IR3 0.1 0.1 0.1 2.5 5.2 2.4 3.1 2.4 3.6 IR4 13.2 19.0 15.4 15.1 20.6 17.7 13.4 17.6 15.1 IR5 27.5 19.1 13.2 59.2 41.1 92.0 53.2 59.6 68.3 IR6 10.4 4.5 6.3 5.8 5.6 5.7 5.9 3.5 6.2 IR7 1.9 2.6 5.0 4.8 11.8 3.8 5.8 3.9 5.1 Sensory neuron membrane protein SNMP2 57.9 78.3 91.0 446.8 375.9 374.0 1537.6 1060.6 1387.8 Odorant degrading enzyme ODE1 98.5 77.4 104.4 145.8 91.0 73.3 16.1 5.6 6.6 ODE2 2.7 4.5 4.2 7.3 9.6 15.9 5.8 11.7 16.4

176

Supplementary Table 2.4. (Continued) ODE3 34.8 38.4 75.0 82.8 103.1 55.2 60.9 62.2 65.2 ODE4 20.0 12.1 26.3 6.8 5.6 7.2 27.8 29.6 32.7 ODE5 26.6 35.5 35.0 20.2 13.2 15.0 14.2 15.3 12.3 ODE6 0.5 2.1 1.6 3.9 6.2 5.7 3.8 8.0 6.5 ODE7 22.0 10.6 14.5 9.0 18.7 11.2 8.7 9.5 8.5 Aldehyde dehydrogenase ALDH1 698.4 604.3 562.9 1326.4 1266.4 1221.0 1633.0 1382.3 1486.8 ALDH2 22.5 22.2 38.4 54.4 58.6 57.6 53.6 61.9 52.4 ALDH3 0.3 0.6 1.2 2.6 0.9 0.2 0.7 0.2 0.0 ALDH4 3.4 6.0 9.7 22.3 21.7 15.2 22.2 23.5 28.8 G-protein coupled receptors ETHR 28.1 19.6 18.0 0.6 1.4 0.1 2.2 2.0 1.4 DHR 17.0 15.0 12.8 8.7 7.0 5.5 36.6 22.7 33.8 SPR 2.7 3.0 1.4 3.7 3.5 7.2 3.2 3.1 1.0 OctoR1 3.3 1.7 2.3 3.7 1.3 1.3 0.0 0.0 0.0 OctoR2 0.0 0.0 0.0 2.6 1.3 0.9 3.5 2.1 3.1 PBANR_B 2.0 2.2 1.9 0.0 0.0 0.0 0.0 0.0 0.0 PBANR_C 1.7 0.8 6.4 0.2 0.1 0.0 0.3 0.3 0.1 Facilitated trehalose transporter Tret1 99.85 128.67 74.42 25.39 39.28 40.95 32.87 37.63 36.78 Tret2 46.49 55.52 50.34 0.77 0.86 0.53 0.51 0.41 0.92 Tret3 41.66 54.96 40.22 46.93 42.73 59.31 48.35 42.08 49.99 Tret4 54.68 32.27 42.19 99.65 150.36 93.15 85.25 93.87 112.59 Tret5 147.11 222.55 311.53 79.55 93.51 75.16 81.84 61.84 92.12 Tret6 61.82 63.94 42.1 275.32 356.93 300.7 95.33 99.28 88.61 Tret7 77.75 67.94 53.69 1.7 1.98 2.82 2.83 8.69 4.84 Tret8 39.34 50.39 54.26 7.22 2.44 2.93 4.83 3.77 1.8 Tret9 50.16 54.77 35.34 207.46 179.65 168.27 134.03 99.49 152.3 Tret10 33.21 34.32 27.56 80.25 124.98 80.06 56.6 74.96 70.61 Tret11 40.11 65.83 65.33 0.86 2.08 9.04 0.52 0.46 1.39 Tret12 47 45.22 40 48.94 38.34 45.98 56.8 53.69 73.96 Tret13 3.26 4.33 4.79 10.33 11.38 19.09 12.01 11.69 11.94 Tret14 24.97 40.7 21.77 38.67 28.92 44.35 35.14 32.16 24.93

177

APPENDIX B. ADDITIONAL FIGURES AND TABLES FOR CHAPTER 4

Supplementary Table 4.1. The 88 differentially expressed genes between Lab and Field populations

AA Gene description length Log2FC P Value FDR Field Upregulated facilitated trehalose transporter Tret1-like 526 4.243 3.26E-17 1.07E-14 Unknown 104 3.379 5.03E-16 1.42E-13 Unknown 661 4.298 1.28E-13 3.01E-11 Unknown 709 3.611 3.91E-11 6.21E-09 trypsin 5G1-like 258 2.071 7.75E-11 1.12E-08 Unknown 195 7.290 1.21E-10 1.69E-08 digestive cysteine proteinase 1-like 549 2.345 3.09E-10 3.95E-08 laccase-1-like 592 3.503 4.80E-10 5.94E-08 Unknown 108 3.990 5.36E-10 6.59E-08 Unknown 337 2.056 7.46E-10 8.92E-08 Unknown 330 3.147 9.16E-10 1.06E-07 Unknown 110 2.851 8.97E-09 8.93E-07 serine protease 24 512 2.315 1.31E-08 1.27E-06 Unknown 1680 2.008 4.07E-08 3.55E-06 Insect cuticle protein 153 2.284 1.08E-07 8.80E-06 regucalcin-like 339 2.107 1.87E-07 1.48E-05 protein-L-isoaspartate(D-aspartate) O-methyltransferase 300 4.194 2.44E-07 1.86E-05 Unknown 311 2.283 4.01E-07 2.97E-05 ysozyme-like 161 4.363 1.39E-06 9.41E-05 cytochrome P450 4C1-like 507 2.593 3.35E-06 0.000204 cuticle protein 16.5, isoform B-like isoform X1 314 3.552 6.52E-06 0.000371 aldo/keto reductase 112 2.604 1.76E-05 0.000857 60S ribosomal protein L9 195 2.079 2.16E-05 0.001014 Unknown 539 2.091 2.55E-05 0.001178 Unknown 116 2.652 2.92E-05 0.001321 Unknown 126 2.346 8.56E-05 0.003283 Unknown 983 3.321 0.000106 0.003939 gamma-interferon-inducible lysosomal thiol reductase- 238 2.018 0.000165 0.005629 Unknownlike 164 2.002 0.000227 0.00737 PH, RCC1 and FYVE domains-containing protein 1 373 2.899 0.000282 0.008783 attacin-like 155 3.411 0.000389 0.011385 RNA-directed DNA polymerase from mobile element 497 2.109 0.000428 0.012351 Unknownjockey-like 128 2.684 0.000748 0.019236 PREDICTED: venom peptide BmKAPI-like 138 2.184 0.000776 0.019773 phospholipid scramblase 1-like 222 2.381 0.000805 0.020281 gloverin-like 194 3.377 0.000972 0.023594 Unknown 775 2.173 0.00108 0.02556 Unknown 170 2.674 0.001824 0.038556 Unknown 1407 2.037 0.00211 0.043083 cytochrome P450 6B2-like 356 2.077 0.002191 0.04434 Field Downregulated Unknown 838 -4.640 3.47E-29 2.54E-26 gag-pol polyprotein 952 -3.337 8.05E-22 3.89E-19 178

Supplementary Table 4.1. (Continued) Unknown 978 -3.411 5.35E-18 1.87E-15 Unknown 387 -4.476 9.07E-18 3.10E-15 Unknown 187 -2.381 2.87E-13 6.40E-11 Unknown 905 -2.964 2.92E-13 6.47E-11 Unknown 1312 -3.666 3.23E-13 7.05E-11 Unknown 115 -3.416 3.16E-12 6.01E-10 hypodermin-A-like isoform X2 277 -4.339 5.60E-12 1.03E-09 olfactory receptor 16 354 -3.484 2.14E-11 3.64E-09 glutathione S-transferase siama 2 205 -3.598 4.96E-11 7.58E-09 Unknown 898 -2.639 2.66E-10 3.43E-08 Unknown 152 -3.008 5.46E-10 6.68E-08 collagen alpha-1(XV) chain-like 128 -3.186 5.52E-10 6.73E-08 Unknown 219 -2.337 1.62E-09 1.79E-07 Unknown 117 -2.449 1.01E-08 1.00E-06 synaptic vesicle glycoprotein 2C-like 530 -2.407 1.37E-08 1.32E-06 Unknown 182 -2.480 1.65E-08 1.56E-06 Unknown 133 -4.374 1.91E-08 1.77E-06 mucin-19-like isoform X1 391 -2.494 2.38E-08 2.17E-06 Unknown 1761 -2.980 3.49E-08 3.10E-06 Unknown 410 -3.976 6.88E-08 5.77E-06 Beta-ureidopropionase 417 -2.020 1.48E-07 1.18E-05 zinc carboxypeptidase-like 218 -2.065 3.17E-07 2.39E-05 Unknown 564 -2.575 5.64E-07 4.08E-05 Unknown 291 -9.109 9.00E-07 6.35E-05 piggyBac transposable element-derived protein 4-like 274 -2.697 1.32E-06 9.04E-05 Unknown 1052 -2.255 2.44E-06 0.000156 Unknown 678 -4.467 3.85E-06 0.000231 Unknown 117 -3.368 7.04E-06 0.000395 cytochrome P450 4c3-like 499 -2.009 1.75E-05 0.000852 glutathione S-transferase epsilon 8 235 -3.199 1.78E-05 0.000862 facilitated trehalose transporter Tret1-like 488 -2.064 5.75E-05 0.002376 probable cytochrome P450 6a23 609 -3.472 6.37E-05 0.002582 Unknown 621 -2.972 8.35E-05 0.003227 trichohyalin isoform X1 297 -2.170 0.000177 0.005958 Unknown 104 -3.435 0.00024 0.007695 Unknown 738 -2.658 0.000338 0.010195 DNA ligase 408 -2.593 0.000364 0.010832 Unknown 192 -2.338 0.000462 0.013103 Unknown 965 -2.603 0.000522 0.014284 calexcitin-2-like 201 -2.351 0.001155 0.026934 synaptic vesicle glycoprotein 2B-like 512 -2.269 0.0012 0.027804 Unknown 107 -4.277 0.001307 0.029929 Unknown 152 -2.428 0.001318 0.03009 Unknown 844 -2.376 0.001532 0.033786 15-hydroxyprostaglandin dehydrogenase NAD(+)]-like 273 -2.339 0.001635 0.035712 monocarboxylate transporter 12-like 521 -2.438 0.002265 0.045546

179

Supplementary Table 4.2. Putative G-protein coupled receptors, odorant binding proteins, chemosensory proteins in PBW pheromone glands and the first BLASTp hit in GenBank.

E Identity Transcripts GenBank homologue description Accession no.* Species value‡ (%)

Diapause hormone receptor

Helicoverpa DHr diapause hormone receptor ANB78221 0 66 armigera

Ecdysis-triggering hormone receptor

ETHr Ecdysis triggering hormone OWR50706 Danaus plexippus 0 74 receptor subtype-A Octopamine receptor

Octor1 octopamine receptor beta-2R-like XP_013165403 Papilio Xuthus 0 86

octopamine receptor Oamb isoform Octor2 XP_022827336 Spodoptera litura 0 92 X2

Sex peptide receptor

SPR1 sex peptide receptor isoform X1 XP_013178706 Papilio Xuthus 0 85

PREDICTED: sex peptide SPR2 XP_013187671 Amyelois transitella 0 84 receptor-like

SPR3 sex peptide receptor-like XP_013195850 Amyelois transitella 0 94

sex peptide receptor-like isoform SPR4 XP_022817535 Spodoptera litura 0 82 X2

Field_SPR PREDICTED: sex peptide XP_013187496 Amyelois transitella 0 79 5 receptor-like

PBAN receptor

pyrokinin-1 receptor-like isoform PBANr1 XP_022814705 Spodoptera litura 0 83 X2

pyrokinin-1 receptor-like isoform PBANr2 XP_022814703 Spodoptera litura 0 83 X1

Chemosensory proteins

CSP1 chemosensory protein 16 BAV56820 Ostrinia furnacalis 1E-20 44

Dendrolimus CSP2 chemosensory protein AII01028 5E-71 81 kikuchii

CSP3 chemosensory protein 7 BAV56811 Ostrinia furnacalis 6E-65 72

CSP4 putative chemosensory protein AGY49269 Sesamia inferens 2E-60 69

CSP5 chemosensory protein 7 precursor NP_001037068 Bombyx mori 1E-46 62

CSP6 chemosensory protein 2 BAV56806 Ostrinia furnacalis 2E-61 69 180

Supplementary Table 4.2. (Continued)

Eogystia CSP7 chemosensory protein AOG12895 2E-47 56 hippophaecolus

putative chemosensory protein CSP8 ALJ302214 Spodoptera litura 1E-50 62 CSP3

CSP9 chemosensory protein CSP15 ATD12158 Cydia pomenella 8E-47 57

Odorant binding proteins

general odorant-binding protein XP_013136623 OBP1 Papilio polytes 8E-79 79 72-like 6

Eogystia OBP2 odorant-binding protein AOG12878 3E-78 89 hippophaecolus

OBP3 odorant-binding protein OBP14 ATD12155 Cydia pomonella 3E-70 54

OBP4 odorant-binding protein 7 AKI87968 Spodoptera litura 2E-71 70

OBP5 odorant-binding protein OBP13 ATD12154 Cydia pomonella 5E-55 53

Eogystia OBP6 odorant-binding protein AOG12879 6E-57 62 hippophaecolus

OBP7 odorant-binding protein AGK24577 Chilo suppressalis 6E-28 39

181

Supplementary Table 4.3. Comparison of candidate transcripts of G-protein coupled receptors, odorant binding proteins, chemosensory proteins in PBW pheromone glands.

Lab population Field population

AA Complete AA Complete Log2 Fold % Gene RPKM RPKM Length ORF Length ORF Change Identity*

Diapause hormone receptor

DHr 521 Y 2.68±0.34 521 Y 4.98±0.79 -0.9 100

Ecdysis-triggering hormone receptor

ETHr 550 Y 0.74±0.33 550 Y 0.5±0.089 0.56 100

Octopamine receptor

Octo_r1 388 Y 3.21±1.09 388 Y 3.15±0.64 0.03 100

Octo_r2 520 Y 0.3±0.06 520 Y 0.4±0.09 -0.44 100

Sex peptide receptors

SPr1 429 Y 1.3±0.64 429 Y 1.94±0.81 -0.58 100

SPr2 243 N 0.17±0 302 N 0.18±0.04 -0.07 100

SPr3 388 Y 2.85±0.45 388 Y 3.26±0.50 -0.19 100

SPr4 424 Y 0.39±0 424 Y 0.23±0.08 0.77 100

Lab_SPr4 vs 424 Y 0.39±0 423 Y 0.45±0.07 -0.23 83 Field_SPr 5

PBAN receptor

PBANr1 400 Y 2.57±0.52 400 Y 2.04±0.15 0.33 100

PBANr2 470 Y 0.76±0.28 470 Y 3.49±0.37 -2.2 100

Chemosensory proteins

CSP1 107 Y 0.22 106 Y 0.34 -0.09 100

CSP2 122 Y 32.1±7.0 122 Y 34.3±7.1 -0.50 100

2248.7 3177.9 CSP3 127 Y 127 Y 0.86 100 ±733.8 ±1046.9

CSP4 131 Y 21.5±8.3 131 Y 26.4±16.7 -0.30 100

CSP5 120 Y 28.5±9.3 120 Y 15.74±8.6 0.86 100

182

Supplementary Table 4.3. (Continued)

909.3 777.9 CSP6 126 Y 126 Y 0.23 100 ±652.0 ±418.1

CSP7 121 N 0.35 115 N 0.37 -0.08 100

283.8 332.7 CSP8 126 Y 126 Y -0.23 100 ±83.1 ±199.7

CSP9 122 Y 0.48±0.07 122 Y 0.46±0.19 0.08 100

Odorant binding proteins

OBP1 141 Y 12.8±7.9 141 Y 14.0±5.5 -0.13 100

OBP2 143 Y 13.8±9.9 143 Y 19.7±10.8 -0.52 100

OBP3 213 N 59.9±7.2 191 Y 58.7±7.8 0.03 100

OBP4 143 Y 1.8±1.3 143 Y 0.4±0.2 2.08 100

OBP5 149 Y 1.6±1.4 103 Y 1.9±1.6 -0.27 100

OBP6 136 Y 4.8±4.4 136 Y 9.0±8.8 -0.90 100

147.8 184.4 OBP7 141 Y 141 Y -0.32 100 ±98.9 ±42.6

* % identity between the Lab and Field populations.

183

Supplementary Figure 4.1. Gene ontology classification and comparison of PBW transcripts of two populations 184

Supplementary Figure 4.2. Volcano plot for differential expressed genes between Lab and Field populations. FC: Fold Change. FDR: False discovery rate. Black dots: FDR>0.05. Not significant; Red dots: FDR < 0.05, Significant.

185

APPENDIX C. ADDITIONAL FIGURES AND TABLES FOR CHAPTER 5

Supplementary Table 5.1 Viruses found in PBW Lab population

Lab Full Identity Access Potential nt aa GenBank description Type population ? (%) number sourse L polymerase RdRp Contig1 2153 345 N 32.83 AYW51538 (-ss)RNA Unknown [Formica fusca virus 1] Contig2 polyprotein [Helicoverpa 9677 2948 Y 78.25 YP_009344960 (+ss)RNA Virus (PBWV1) armigera iflavirus] RNA-dependent RNA Contig3 6419 2093 Y polymerase [Seattle 62.05 AOF41423 (-ss)RNA Virus (PBWV2) Prectang virus] RNA-dependent RNA Contig4 7827 2502 Y polymerase [Hubei 33.26 YP_009330283 (-ss)RNA Virus (PBWV3-L) lepidoptera virus 1] Contig5 putative nucleoprotein 2362 276 Y 31.49 YP_009330256 (-ss)RNA Virus (PBWV3-S) [Hubei lepidoptera virus 1] Contig6 putative glycoprotein 5042 1558 Y 27.73 YP_009330257 (-ss)RNA Virus (PBWV3-M) [Hubei lepidoptera virus 1] Contig7 hypothetical protein 9782 2846 Y 37.36 YP_009342321 (+ss)RNA Virus (PBWV4) [Wuhan insect virus 13] hypothetical protein Host Contig8 1009 259 Y B5V51_1641 [Heliothis 34.46 PCG71644 dsDNA genome virescens] orf10-like protein Host Contig9 2059 256 Y [Peridroma 30.4 YP_009049835 dsDNA genome alphabaculovirus] ORF61 [Xestia c-nigrum Host Contig10 1784 437 Y 27.89 NP_059209 dsDNA granulovirus] genome orf10-like protein Host Contig11 1439 257 Y [Peridroma 27.73 YP_009049835 dsDNA genome alphabaculovirus] unknown similar to MacoNPV-B orf57 Host Contig12 765 170 Y 31.34 YP_008004381 dsDNA [Choristoneura biennis genome entomopoxvirus] cathepsin-like protein Contig13 791 231 N [Helicoverpa armigera 50.45 AIY24949 dsDNA Virus nucleopolyhedrovirus] cathepsin-like cysteine proteinase [Spodoptera Contig14 758 130 N 46.88 NP_258322 dsDNA Virus litura nucleopolyhedrovirus] cathepsin [Helicoverpa Contig15 1056 144 N armigera 46.81 NP_075125 dsDNA Virus nucleopolyhedrovirus G4] cathepsin [Choristoneura Contig16 1249 353 N fumiferana multiple 47.57 XP_028167656 dsDNA Unknown nucleopolyhedrovirus]

186

Supplementary Table 5.2 Viruses found in PBW Field population

Field Full Identi Potential nt aa GenBank description Access Type Population ? ty(%) sourse unknown similar to MacoNPV-B orf57 YP_008 Contig1 754 189 N 30.88 dsDNA Unknown [Choristoneura biennis 004381 entomopoxvirus] cathepsin-like cysteine NP_258 Contig2 861 136 N proteinase [Spodoptera litura 46.62 dsDNA Unknown 322 nucleopolyhedrovirus] transposase mut [Lambdina YP_009 Contig3 593 191 N fiscellaria 49.73 dsDNA Unknown 133324 nucleopolyhedrovirus] zingipain-2-like [Ostrinia XP_028 Contig4 1288 333 N 46.25 dsDNA Unknown furnacalis] 167656 cathepsin-like protein AIY249 Contig5 776 226 N [Helicoverpa armigera 50.45 dsDNA Unknown 49 nucleopolyhedrovirus] hypothetical protein [Samia BBD51 Contig6 1390 376 N 33.67 dsDNA Unknown ricini nucleopolyhedrovirus] 232 Mabr_orf10 [Mamestra AFP957 Contig7 631 164 N brassicae multiple 32.7 dsDNA Unknown 29 nucleopolyhedrovirus] Contig8 polyprotein [Helicoverpa YP_009 10063 2948 N 79 (+ss)RNA Virus (PBWV1) armigera iflavirus] 344960 RNA-dependent RNA Contig9 AOF41 6423 2093 Y polymerase [Seattle Prectang 62 (-ss)RNA Virus (PBWV2) 423 virus] Contig10 putative nucleoprotein [Hubei YP_009 2352 276 Y 31.49 (-ss)RNA Virus (PBWV3-S) lepidoptera virus 1] 330256 Contig11 putative glycoprotein [Hubei YP_009 (PBWV3- 5115 1559 Y 26.86 (-ss)RNA Virus lepidoptera virus 1] 330257 M) RNA-dependent RNA Contig12 YP_009 7762 2502 Y polymerase [Hubei lepidoptera 35 (-ss)RNA Virus (PBWV3-L) 330283 virus 1] Contig13 hypothetical protein [Wuhan YP_009 9711 2846 N 37 (+ss)RNA Virus (PBWV4) insect virus 13] 342321 orf10-like protein [Peridroma YP_009 Host Contig14 948 257 Y 28 dsDNA alphabaculovirus] 049835 genome unknown [Antheraea pernyi ABQ12 Host Contig15 2164 256 Y 27 dsDNA nucleopolyhedrovirus] 247 genome ORF61 [Xestia c-nigrum NP_059 Host Contig16 1710 437 Y 27 dsDNA granulovirus] 209 genome transposase mut [Lambdina YP_009 Host Contig17 2215 204 Y fiscellaria 46 dsDNA 133324 genome nucleopolyhedrovirus] L polymerase RdRp [Formica AYW5 Contig18 4219 574 N 52 (-ss)RNA Virus fusca virus 1] 1538

187

Supplementary Table 5.3 Virus accession number in sequence alignment and phylogenetic tree (Iflavirous)

Abbreviation Virus Accession MKV Moku virus isolate Big Island YP_009305421 VDV Varroa destructor virus YP_145791 BrBV Brevicoryne brassicae picorna-like virus isolate IL AKJ70949 TpIV Thaumetopoea pityocampa iflavirus AJC98140 DWV Deformed wing virus APV34355 EoPV Ectropis obliqua picorna-like virus NP_919029 IFV Infectious flacherie virus NP_620559 PnPV Perina nuda virus NP_277061 SeIV Spodoptera exigua iflavirus YP_004935363 PaIV Psammotettix alienus iflavirus AYD38337 ArmIV Armigeres iflavirus gene for polyprotein YP_009448183 DmIV Diamondback moth iflavirus isolate Guangzhou YP_009361829 HarIV Helicoverpa armigera iflavirus strain HBLF20 YP_009344960 BmIV Bombyx mori iflavirus gene for polyprotein YP_009162630 LdIV Lymantria dispar iflavirus YP_009047245 HeIV Heliconius erato iflavirus YP_009026409 ApIV Antheraea pernyi iflavirus isolate LnApIV-02 YP_009002581 IhIV Ixodes holocyclus iflavirus isolate AQZ42314 BatIV Bat iflavirus clone Bat/CAM/IfaV-P YP_009345906 LsIV Laodelphax striatellus iflavirus 3 AYU66736 CqIV Chequa iflavirus isolate A YP_009444707 OiIV Opsiphanes invirae iflavirus YP_009165593 YgIV Yongsan iflavirus AXV43887 PrIV Pityohyphantes rubrofasciatus iflavirus isolate UW YP_009351892 SBPV Slow bee paralysis virus strain PP APZ86804 ScV Sacbrood virus strain APT42957 FeV Formica exsecta virus 2 isolate Fex2 YP_008888537 NlhV Nilaparvata lugens honeydew virus- YP_009505599 LlV Lygus lineolaris virus YP_009505598 EVV Euscelidius variegatus virus YP_009328891 GNV Graminella nigrifrons virus YP_009129265 LSPV Laodelphax striatellus picorna-like virus YP_009110667 HahV Halyomorpha halys virus isolate Beltsville YP_008719809 DVVV Diabrotica virgifera virgifera virus APF29088 WIV Wuhan insect virus 13 YP_009342321.1 BIV Bee iflavirus 1 AVH76848.1 HPV Hubei picorna-like virus 30 YP_009337722.1 WCV Wuhan coneheads virus 1 YP_009342053.1 DCPV Dinocampus coccinellae paralysis virus YP_009111311.1 188

Supplementary Table 5.3. (Continued) CIV Culex Iflavi-like virus 3 AXQ04785.1 HarmIV Helicoverpa armigera iflavirus strain HBLF20 YP_009344960

Supplementary Table 5.4 Virus accession number in sequence alignment (Bunyavirus)

Abbreviation Virus Accession SPV Seattle Prectang virus AOF41423 WCV Wuchang Cockroach Virus 1 YP_009304995 HOV Hubei odonate virus 8 YP_009329887 GBV Ganda bee virus APT68154 HDV Hubei diptera virus 6 APG79294 WMV Wuhan mosquito virus 1 YP_009305130 HLV Hubei lepidoptera virus 1 YP_009330283 UKV Uukuniemi virus AIU95041 KMV Kabuto mountain virus YP_009449450 HTV Huangpi Tick Virus 2 YP_009293590 HDV Hubei diptera virus 5 YP_009330277 RVFV Rift Valley fever virus ABD51533

189

Supplementary Table 5.5 Virus accession number in phylogenetic tree (Bunyavirus)

Viruses L segment M segment S segment Family Punta Toro virus AKF42419 YP_009512940 ABQ23558 Phenuiviridae Chandiru virus AEA30057 YP_004347992 YP_004347994 Phenuiviridae Munguba virus YP_009346010 YP_009346034 YP_009346011 Phenuiviridae Bujaru virus API68880 API68881 API68883 Phenuiviridae Capira virus AKF42424 AKF42409 ABQ23564 Phenuiviridae Anhanga virus YP_009346019 YP_009346033 YP_009346021 Phenuiviridae Cocle virus AKF42422 AKF42407 AKF42389 Phenuiviridae Chagres virus AEL29642 AEL29641 AEL29643 Phenuiviridae Echarate virus AEA30058 AEA30046 AEA30073 Phenuiviridae Rift Valley fever virus ABD51538 AUM56908 AAF00695 Phenuiviridae Adana virus AJK91618 YP_009227128 YP_009227130 Phenuiviridae Wuhan Fly Virus 1 YP_009305000 YP_009304997 YP_009304999 Phenuiviridae Phasi Charoen-like phasivirus YP_009505332 YP_009505331 YP_009505333 Phenuiviridae Badu phasivirus YP_009505327 YP_009505328 YP_009505329 Phenuiviridae Wutai mosquito phasivirus YP_009305140 YP_009305141 YP_009305142 Phenuiviridae Rice stripe tenuivirus NP_620522 NP_620521 AAM49589 Phenuiviridae Rice grassy stunt tenuivirus BAA32246 None ABU53645 Phenuiviridae Cumuto virus AHH60917 AHH60918 AHH60919 Phenuiviridae Yichang Insect virus AJG39273 YP_009305144 YP_009305145 Phenuiviridae Gouleako virus ABP68557 AEJ38174 AEJ38173 Phenuiviridae European mountain ash ringspot- associated emaravirus AAS73287 YP_003104765 SPN63240 Fimoviridae Wheat mosaic virus AML03181 AML03208 AML03175 Fimoviridae Pigeonpea sterility mosaic emaravirus 1 YP_009237282 CEJ09716 ANQ90740 Fimoviridae Raspberry leaf blotch emaravirus YP_009237274 YP_009237265 YP_009237266 Fimoviridae Tomato spotted wilt tospovirus AIA24440 APG79634 AUX16838 Topovirus Polygonum ringspot virus AOO95322 ABX65310 YP_009513002 Topovirus Groundnut bud necrosis virus ATD12162 NP_619703 ADF58354 Topovirus Watermelon bud necrosis virus AIL30564 ACN53946 ACB12338 Topovirus Iris yellow spot virus YP_009241381 AIS23797 AWK91367 Topovirus Impatiens necrotic spot tospovirus AWK77935 NP_619691 AWK77939 Topovirus Watermelon silver mottle tospovirus NP_620752 NP_620767 NP_620771 Topovirus Zegla virus AXP33562 AXP33554 AXP33568.1 Peribunyaviridae Shark River virus axp33561 AXP33551 AXP33567 Peribunyaviridae Patois virus AXP33560 AXP33555 AXP33566 Peribunyaviridae Babahoya virus AXP33559 AXP33549 AXP33565 Peribunyaviridae Abras virus AXP33558 AXP33547 AXP33564 Peribunyaviridae Lokern virus AXP32026 AXN72389 AXP32027 Peribunyaviridae Herbert herbevirus AGX32061 AGX32056 AGX32064 Peribunyaviridae 190

Supplementary Table 5.5. (Continued) Kibale virus YP_009362027 YP_009362025 YP_009362035 Peribunyaviridae Tai virus YP_009362026 YP_009362024 YP_009362028 Peribunyaviridae Shuangao Insect Virus 1 YP_009300681 AJG39311 YP_009300682 Peribunyaviridae Hantaan orthohantavirus CAA39394 APH07634 QBH69869 Hantaviridae Crimean-Congo hemorrhagic fever orthonairovirus AAZ76532 ATG31911 ABD98121 Nairoviridae Hubei bunya-like virus 1 APG79293 APG79328 None Unclassified Hubei lepidoptera virus 1 YP_009330283 YP_009330257 YP_009330256 Unclassified Wuhan Insect virus 1 AJG39261 None AJG39323 Unclassified Shayang bunya-like virus 1 APG79324 None None Unclassified Xinzhou bunya-like virus 1 APG79358 None None Unclassified Jingmen ascaridia virus 1 APG79256 None None Unclassified Rhizoctonia solani negative- stranded virus 4 ALD89133 None None Unclassified