Regulation of Ecdysone 20-Monooxygenase Activity in the Tobacco

REGULATION OF ECDYSONE 20-MONOOXYGENASE ACTIVITY IN THE TOBACCO

HORNWORM, MANDUCA SEXTA AND THE APPARENT OCCURRENCE OF THIS

ACTIVITY IN ASCARIS SUUM (NEMATODA)

Christopher A. Drummond

A Dissertation

Submitted to the Graduate College of Bowling Green State University in partial fulfillment of the requirements for the degree of

DOCTOR OF PHILOSOPHY

May 2011

Committee:

Carmen F. Fioravanti, Advisor

Hanfeng Chen Graduate Faculty Representative

Daniel M. Pavuk

Lee A. Meserve

Scott O. Rogers

Christopher A. Drummond

ABSTRACT

Carmen Fioravanti, Advisor

At specific intervals, increased concentrations of two steroid hormones, i.e., ecdysone (E) and 20-hydroxyecdysone (20E), elicit developmental changes in arthropods.

Conversion of E to the active molting hormone, 20E, in the tobacco hornworm Manduca sexta is catalyzed by the cytochrome P450-containing ecdysone 20-monooxygenase system (E20M). During embryogenesis, M. sexta E20M activity increased for the first 72 hours at which time it peaked and subsequently significantly declined. The increased activity coincided with the increase of free ecdysteroids and the progression of two embryonic molts. In midgut tissue of fifth instar M. sexta, decreases in second messenger

3’,5’cyclic guanosine monophosphate (cGMP) concentration inhibited day five E20M activity, but increases in cGMP concentration restored E20M activity. Midgut cGMP content peaked on day five of the instar in concert with the highest level of E20M activity observed. Molecular studies with midgut tissue demonstrated that the E agonist RH-5849 elicited increases in E20M (the shade gene) expression. In the presence of two guanylate cyclase inhibitors, E20M expression significantly increased. Inhibition remediation by pharmacological means resulted in significantly decreased shade expression. While it is unclear as to where cGMP exerts its effects on E20M activity, the data indicated that the second messenger affected the level of transcription, translation, or enzyme activity either individually or in some combination. Interestingly, E20M activity also was found to be affected by six synthesized anthraquinones suggesting that these compounds can serve to iv

disrupt M. sexta development. Lastly, E20M-like activity was observed in female Ascaris suum (Nematoda) both in muscle and reproductive tissue. Although E20M localization in muscle was unclear, in reproductive tissue E20M activity resided mainly with microsomes. This work provides a number of important insights into the regulation of M. sexta E20M during development, a role of cGMP in these events, and the possible occurrence of E20M in the parasitic nematode Ascaris suum. v

This dissertation is dedicated to the fond memory of Stan L. Smith. His encouragement,

keen intellect, care, witty sense of humor and scientific acumen are sorely missed. “So

what did I do…I went to the literature…”

“Hope Lives. No matter the mistakes we make, no matter our blunders and misunderstandings, no matter the grief and sorrow and loss, no matter how deep the darkness, hope lives.”

-M. Weis and T. Hickman vi

ACKNOWLEDGMENTS

I wish to honor the memory of Dr. Stan L. Smith, for without his initial guidance and turning me on to insect endocrinology, I may not have found my true calling. In the same manner I wish to thank my advisor Dr. Carmen Fioravanti for all of his patience and guidance, and stepping up in the face of a devastating loss. My sincerest thanks are offered to my committee members Dr. Lee A. Meserve, Dr. Daniel M. Pavuk, Dr. Scott

O. Rogers, and Dr. Hanfeng Chen, for their help in preparing this manuscript, being excellent teachers, and wonderful advisors. My thanks also go to Drs. Martin A. Mitchell and Lawrence I. Gilbert for their expert consultation; I could not have completed these projects without them.

I also would like to acknowledge my lab mates Kurt P. Vandock and Brandon T.

Welsh as well as the numerous undergraduate students over the years that spent time in the lab. Your companionship and help with the Manduca sexta colony proved to be a valuable asset.

Lastly and most importantly I wish to thank my wife Amanda for all her patience and everlasting support. Also, I am forever indebted to my family for their support, kind words, and inspiration throughout this process.

vii

TABLE OF CONTENTS

Page

CHAPTER I. A REVIEW OF THE LITERATURE

Introduction ...... 1

Insects as models...... 1

Insect molting and metamorphosis ...... 3

Endocrine control of molting and metamorphosis ...... 5

Ecdysteroids: structures, occurrence, and titers ...... 7

Ecdysteroidogenesis ...... 14

Aims of this dissertation research ...... 33

CHAPTER II. ECDSYONE 20-MONOOXYGENASE ACTIVITY DURING

EMBRYOGENESIS OF THE TOBACCO HORNWORM, MANDUCA SEXTA

Introduction ...... 34

Materials and Methods ...... 37

Results and Discussion ...... 40

CHAPTER III. CYCLIC 3’,5’ GUANOSINE MONOPHOSPHATE AND LARVAL MIDGUT

ECYSONE 20-MONOOXYGENASE ACTIVITY OF THE TOBACCO HORNWORM,

MANDUCA SEXTA

Introduction ...... 46

Materials and Methods ...... 48

Results ...... 53

Discussion ...... 63

viii

CHAPTER IV. EXAMINING THE EFFECTS OF CYCLIC 3’,5’ GUANOSINE

MONOPHOSPHATE ON SHADE EXPRESSION IN MIDGUT TISSUE OF THE TOBACCO

HORNWORM, MANDUCA SEXTA

Introduction ...... 66

Materials and Methods ...... 68

Results ...... 72

Discussion ...... 79

CHAPTER V. EFFECTS OF SIX ANTHRAQUINONES ON MANDUCA SEXTA MIDGUT

ECDYSONE 20-MONOOXYGENASE ACTIVITY

Introduction ...... 84

Materials and Methods ...... 86

Results ...... 89

Discussion ...... 98

CHAPTER VI. APPARENT ECDYSONE 20-MONOOXYGENASE ACTIVITY IN ADULT

ASCARIS SUUM (NEMATODA)

Introduction ...... 101

Materials and Methods ...... 103

Results ...... 106

Discussion ...... 110

REFERENCES ...... 114

APPENDIX A. TABLES COMPARING BIOCHEMICAL AND MOLECULAR EFFECTS OF cGMP ...... 140 ix

APPENDIX B. LIST OF ABBREVIATIONS FOR FACTORS USED TO AFFECT cGMP

CONCENTRATION ...... 143

APPENDIX C. A TABLE OF ANTHRAQUINONE EFFECTIVNESS ...... 144

APPENDIX D. E20M ACTIVITY IN HYMENOLEPIS DIMINUTA ...... 146

LIST OF FIGURES/TABLES

Figure/Table Page

CHAPTER I.

1 Endocrine control of postembryonic development in holometabolous insects ...... 8

2 Ecdysteroid titer during larval-pupal-adult development of Manduca sexta...... 15

3 Enzymes involved in Drosophila melanogaster ecdysteroid biosynthesis...... 17

4 Stoichiometry of a cytochrome P450 mixed function oxidase type reaction ...... 23

5 Km and Vmax of fat body through fifth instar ...... 25

6 Km and Vmax of midgut through fifth instar ...... 26

7 Changes in M. sexta midgut and fat body shade ...... 27

8 Midgut E20M activity during the ten-day stadium of the last larval instar of Manduca

sexta ...... 29

9 Head critical period for day five midgut E20M activity ...... 31

10 Effects of day three injections of E or 20E on day five midgut E20M activity ...... 32

CHAPTER II.

1 Ecdysteroid concentrations during M. sexta embryogenesis ...... 41

2 E20M enzyme activity throughout M. sexta embryogenesis ...... 42

CHAPTER III.

1 Effects of pre-assay incubation of M. sexta midgets with dibutyryl cyclic nucleotides plus

T on day five midgut E20M activity ...... 54

2 Effects of injected G or pharmacological factors on whole animal M. sexta day five

midgut E20M activity ...... 55 xi

3 Effects of injected G or pharmacological factors on head-ligated M. sexta day five midgut

E20M activity ...... 57

4 Effects of injected G or pharmacological factors on thorax-ligated M. sexta day five

midgut E20M activity ...... 59

5 M. sexta midgut homogenate cGMP titers assessed by RIA ...... 61

CHAPTER IV.

1 Effects of injected G or pharmacological factors on whole animal M. sexta day five

midgut shade expression ...... 73

2 Effects of injected G or pharmacological factors on head-ligated M. sexta day five midgut

shade expression ...... 75

3 Effects of injected G or pharmacological factors on thorax-ligated M. sexta day five

midgut shade expression ...... 77

4 A model of the biochemical and molecular regulation of E20M...... 82

CHAPTER V.

1 The effects of decreasing concentrations of quinizarin on E20M activity in day five

midgut homogenates of gate II fifth instar Manduca sexta larvae ...... 90

2 The effects of decreasing concentrations of 1,4-anthraquinone on E20M activity in day

five midgut homogenates of gate II fifth instar Manduca sexta larvae ...... 91

3 The effects of decreasing concentrations of 2-hydroxy-1,4-anthraquinone on E20M

activity in day five midgut homogenates of gate II fifth instar Manduca sexta larvae

...... 93 xii

4 The effects of decreasing concentrations of 2-methoxy-1,4-anthraquinone on E20M

activity in day five midgut homogenates of gate II fifth instar Manduca sexta larvae

...... 94

5 The effects of decreasing concentrations of 9-hydroxy-1,4-anthraquinone on E20M

activity in day five midgut homogenates of gate II fifth instar Manduca sexta larvae

...... 96

6 The effects of decreasing concentrations of 9-methoxy-1,4-anthraquinone on E20M

activity in day five midgut homogenates of gate II fifth instar Manduca sexta larvae

...... 97

CHAPTER VI.

1 A. suum muscle E20M activity ...... 107

2 A. suum reproductive tissue E20M activity ...... 109

CHAPTER I: A REVIEW OF THE LITERATURE

INTRODUCTION

Insects represent some of the most diverse groups in the animal kingdom in terms of both number of species and species diversity. Scientific estimates indicate that there are around

826,000 currently described species of insects and likely several million more unknown species.

Several factors have enabled the robust diversification and expansion observed in this group.

These factors include: small body size, short life cycle, high reproductive potential, the presence of an exoskeleton, and development of the processes of molting and metamorphosis (Nijhout,

1994).

INSECTS AS MODELS

Despite the diversity observed among insects there are underling threads which link all insects. One main facet of this is the use of steroid-like and lipophilic hormones, the production and secretion of which are regulated in large part by neuropeptides, to regulate development and physiological transitions. This commonality among insects allows the use of model systems to study insect development and behavior. One such model is the Lepidopteran (Sphingidae) tobacco hornworm, Manduca sexta. Over the past 50 years, M. sexta is perhaps the most widely used non-genomic insect for studies of insect biochemistry, physiology, endocrinology, and molecular biology (Gilbert and Rewitz, 2009). At this juncture, a published annotated genome for M. sexta is lacking. However, the usefulness of this insect system in physiological and biochemical studies is clearly established (Smith, 1985; Gilbert et al., 2002; Gilbert and Rewitz,

2009).

M. sexta is widely employed in physiological and biochemical studies for several reasons

(Smith, 1985; Gilbert and Rewitz, 2009). Firstly, it is large in size. M. sexta, in its fifth larval instar, attains a weight of more than 12 grams, thereby making isolation and experimentation with individual tissues quite simple. Secondly, it has a relatively short life cycle i.e., from start to finish, the life cycle lasts about 45-50 days. Throughout this life cycle, the insect undergoes a

120-hr period of embryogenesis, five vermiform larval instars, and a cleidoic pupal stage in which the insect undergoes metamorphosis and eclosure after 20 days as a reproductively active adult moth. This short generation time makes several generations of animals available throughout the year for use in experimentation. Thirdly, M. sexta is widely used because it is easy to stage. There are clear morphological markers (i.e., head-capsule slippage, browning of the mandibles, wandering stage behaviors, etc.) that occur at key intervals e.g., when the animal is passing through a developmental transition or when specific biochemical events are occurring in the animal’s life cycle (Smith, 1985; Mitchell et al., 1999). Lastly, M. sexta can be raised easily on an artificial agar-based diet which is supplemented with wheat-germ, cholesterol, sorbic acid, ascorbic acid, and a mixture of vitamins and salts (Bell and Joachim, 1976).

Not only are insects good models for the above-mentioned reasons, but they also serve as a way to gain insights to different levels of hormonal regulation in development. Several species of insects can live without whole portions of their bodies which make regulatory inputs from differing regions discernable. Additionally, they are much simpler in design than many vertebrate systems (Brown et al., 2009). In the case of steroid hormones, insects are thought to produce only ecdysteroids whereas vertebrate species tend to synthesize several different classes of steroids which elicit differing functions. The simplicity of the M. sexta system and its ability

to be utilized in a plethora of biochemical and physiological investigations led to the utilization of this insect in the studies of insect development and steroid metabolism presented in this dissertation.

INSECT MOLTING AND METAMORPHOSIS

While providing protection against both predation and desiccation, the insect exoskeleton serves as a hindrance to growth. Insects have developed physiological processes to circumvent this problem. At highly regulated times during both embryonic and post-embryonic development the exoskeleton, in physiological response to changes in circulating levels of certain hormones and other factors, is partially digested and shed while a new larger exoskeleton is produced. This process is called molting (Smith, 1985; Lafont et al., 2005).

The molting process starts with apolysis, an event during which the insect epidermis retracts from the old exoskeleton. In this pharate (transitional) state, the epidermis begins to synthesize a new exoskeleton and molting fluid partially digests and initiates the reabsorption of the inner portion of the old exoskeleton. Eventually, the remnant of the old exoskeleton is fractured at predetermined lines of cleavage allowing the insect to wiggle its way out of the old cuticle in a process called ecdysis. The stereotypical movements that occur at this point are referred to as eclosion behaviors. The process of eclosion is under the control of several hormones including the peptides eclosion hormone, ecdysis triggering hormone, and crustacean cardio active peptide (Truman, 2005). Once the old exoskeleton has been completely shed, the animal enlarges its body mass by taking in air or water and then hormones are released (e.g., bursicon), that cause the cross-linking of proteins in the newly synthesized cuticle. At the same

time the cuticle is tanned or has insoluble calcium salts incorporated into it in order to harden

(Evans, 1984; Gullan and Cranston, 2005).

There are two types of molts: non-metamorphic and metamorphic molts. In non- metamorphic molts, the insect sheds its existing exoskeleton, which has grown too small for its current body size, to make way for a newly synthesized cuticle which is larger and accommodates its increasing body size. Non-metamorphic molts occur at regular intervals throughout the life cycle of insects. However, during the less frequent metamorphic molts the insect not only sheds its old exoskeleton, but the insect undergoes a physiological, morphological, and maturational transition to the next developmental stage of its life cycle (Smith, 1985; Gilbert et al., 2002).

In the case of the holometabolus insect model, M. sexta, there are two metamorphic molts: larval to pupal and pupal to adult. Additionally, the insect’s life cycle is divided into four stages.

Initially there is the egg or embryonic stage. During this stage, the insect undergoes embryogenesis and the larval tissues differentiate. Small nests of cells called imaginal discs may be formed at this time; these discs will eventually form several adult structures (e.g., the wings).

Following the egg stage, the insect passes through several larval stages (in M. sexta there are five larval instars). The number of larval instars through which insects pass can be variable based on the species of insect, their diet, and the environment. During the larval instars of holometabolus insects, they exponentially increase their size and store nutrients in-order to allow them to pass through the non-feeding, sessile, cleidoic pupal stage. While in the pupal stage, the insect’s larval tissues (e.g., midgut, epidermis, sensory structures, etc.) are degraded and the adult tissues begin to form. At the end of the pupal period the fully formed, reproductively active adult emerges and

travels about carrying out its main purpose, to reproduce (Evans, 1984; Gullan and Cranston,

2005). This dramatic rearrangement of tissues, which often involves changes in mode of consumption of food sources and locomotion, does not occur in all insects. Hemimetabolus insects undergo a gradual change in body form (e.g., at each molt wing buds may enlarge) during larval molts, with the final molt resulting in the generation of fully formed adult features and reproductively active adults (Gullan and Cranston, 2005). The last form of development is ametaboly. In insects displaying this form of development the only difference between nymphal

(juvenile stages) and adult stages of the insect is fully formed and functional genitalia (Gullan and Cranston, 2005).

Both non-metamorphic and metamorphic molts are regulated by complex physiological and environmental stimuli (e.g., changes in levels of certain hormones, stretch of body wall, deil cycle, nutritional state, etc.) (Nijhout, 1994; Gilbert et al., 2002; Lafont et al., 2005; Gilbert,

2008).

ENDOCRINE CONTROL OF MOLTING AND METAMORPHOSIS

Work on insect hormones started 90 years ago and many of our present concepts are attributable to the work of people like Wigglesworth, Bounhiol, Karlson, etc., but this manuscript will introduce the more current work (Gilbert, 2008). The processes of molting and metamorphosis are under hormonal control (Gilbert et al., 1980, 2002; Lafont et al., 2005;

Gilbert and Rewitz, 2009). Three principal classes of hormones are involved in this regulation: neuropeptides which include prothoracicotropic hormone (PTTH), bursicon, and eclosion hormone (Truman, 1985; Watson et al., 1989; Truman, 2005); molting hormones, now called ecdysteroids, which are polyhydroxylated ketosteroids produced from cholesterol or phytosterols,

and secreted mainly by the prothoracic glands (Smith, 1985; Gilbert and Rewitz, 2009); and the sesquiterpene derived juvenile hormones (JH), which are produced in the brain and secreted from the retrocerebral complex (Meyer et al., 1968; Judy et al., 1973; Truman, 2005).

The molting process is initiated when physiological and environmental stimuli (e.g., diet, temperature, photoperiod, and circadian rhythms) cause PTTH synthesis and release by the brain.

PTTHs are tropic neuropeptides produced in lateral neurosecretory neurons in the brain and released by the cells in the corpora allata. PTTHs then act on the prothoracic glands, or their glandular equivalents (e.g., ring glands or ventral glands), causing the glands to take in cholesterol or phytosterols obtained from the insect’s diet.

Following entry into the glands, the cholesterol or phytosterols are converted by no less than six biochemical steps to synthesize and secrete an ecdysone (E; 2β, 3β, 14α, 22R, 25- pentahydroxy-5β-choles-7-en-6-one) precursor, 3-dehydroecdysone (3dE), into the hemolymph

(Warren et al., 1988; Sakurai et al., 1989; Kirishi et al., 1990; Gilbert et al., 2002). Subsequent to release from the prothoracic glands, on the way to the peripheral tissues which are the targets of ecdysteroid action, 3dE is converted by a keto-reductase in the insect hemolymph (blood) to the precursor of the arthropod molting hormone, E (Gilbert et al., 1996). E, in turn, is hydroxylated to the more physiologically active molting hormone 20-hydroxyecdysone (20E), by a cytochrome P450 dependent steroid hydroxylase called ecdysone 20-monooxygenase (E20M;

E.C. 1.14.99.22) (Smith et al., 1979; Smith, 1985; Mitchell et al., 1999; Gilbert et al., 2002;

Lafont et al., 2005; Gilbert and Rewitz, 2009).

Under the influence primarily of a high titer of 20E, the insect undergoes the early events associated with a molt. The decline of the hemolymph titer of 20E serves as the signal for the release of the 8 kDa neuropeptide eclosion hormone (Truman, 1985). This neuropeptide elicits, via the second messenger cGMP, the stereotypical muscle movements necessary for ecdysis and, in turn, serves as the signal for the release of bursicon (Truman, 1985). This latter 40 kDa abdominal ventral nerve cord neuropeptide serves to plasticize the new exoskeleton and facilitate the uptake of tanning agent precursors.

Previous studies demonstrated that when the hemolymph 20E titer is sufficiently high, the insect undergoes a molt (Smith, 1985; Mitchell et al., 1999; Gilbert et al., 2002; Lafont et al.,

2005; Rewitz et al., 2006a, b). The character of the molt (i.e., metamorphic or non-metamorphic) is determined by the hemolymph JH titer at the time 20E triggers a molt (Riddiford, 1970, 1972;

Riddiford and Truman, 1978; Goodman and Granger, 2005). If the hemolymph JH titer is high when 20E elicits a molt, the molt will be larval to larval; if the hemolymph JH titer is lower, 20E will elicit the first metamorphic molt, i.e., a larval to pupal molt; and if the JH titer is very low or absent, 20E will elicit the second metamorphic molt, i.e., the pupal to adult (Gilbert et al., 1980,

2002; Mitchell et al., 1999; Lafont et al., 2005; Gilbert, 2008). The events associated with JH,

PTTH, and E are summarized in Figure 1.

ECDYSTEROIDS: STRUCTURES, OCCURRENCE, AND TITERS

Ecdysteroids comprise a specific family of sterol derivatives with more than 441 members bearing similar features. All members of this class of sterol derived molecule have the following features in common: a cis (5β-H) junction of rings A and B, a 7-ene-6-one

Figure 1: Endocrine control of postembryonic development in holometabolous insects. JH, juvenile hormone; P-450 monooxygenase, ecdysone 20-monooxygenase (from Gilbert et al.,

1980).

chromophore, a trans (14α-OH or –H) junction of rings C and D, and most contain a 3β-OH already present in the sterol precursor (Lafont et al., 2005). The chemical and functional diversity of ecdysteroids derives from the position, number, and presence of –OH groups (Lafont et al., 2005).

441 known ecdysteroids have been identified and shown to occur in arthropods, plants, and parasites (Abubakirov, 1980, 1982, 1984; Adler and Grebenok, 1995; Lafont et al., 2005; http://ecdybase.org). These compounds are known to be active regulators of molting activities as well as other processes in insects, algae, vascular plants, microbes, protozoans, nematodes, mollusks, annelids and crustaceans (Smith, 1985; Davey, 1988; Ahmad et al., 1996; Shirshova et al., 1999; Chitwood, 1999; Subeki et al., 2005; Rewitz et al., 2006c; http://ecdybase.org). Their presence in embryos and adults, e.g., eggs of insects in the orders Orthoptera and Lepidoptera contain complex mixtures in large amounts (up to 40µg/g fresh weight; 10 to 100 times concentrations of larvae and pupae), indicates that ecdysteroids may not only regulate molting in embryogenesis, but may have some other functions as well (Smith, 1985; Lafont et al., 2005).

Many ecdysteroid-like compounds, called phytoecdysteroids, have been isolated from several families of plants, most notably the ferns and gymnosperms (Lafont et al., 2005; http://ecdybase.org). While the definitive function(s) of ecdysteroids and/or phytoecdysteroids in non-arthropod species have yet to be determined, evidence suggests that in plants ecdysteroids serve as antifeedants for both phytophagous insects and parasitic nematodes (Soriano et al., 2004;

Gallo et al., 2006; Mitchell et al., 2007).

Interestingly, ecdysteroids have been found in a number of parasitic helminths such as the cestodes Moniezia expansa (Mendis et al., 1984), Echinococcus granulosus (Mercer et al.,

1987a), and Hymenolepis diminuta (Mercer et al., 1987b). The ability of cestodes to synthesize and be affected by ecdysteroids is the focus of some debate in the field for two reasons. First, it is unknown what function they would serve, as cestodes do not molt. One hypothesis forwarded recently is that these compounds mediate the degradation of cellular connections that connect gravid proglottids to the remainder of the animal thus facilitating release of gravid proglottids

(C.F. Fioravanti, personal communication). The basis for this hypothesis is in light of the fact that in insects at times of molts ecdysteroids mediate capsase activity in the degradation of larval tissues in preparation for synthesis of adult tissues (Iga and Sakurai, 2009). Additionally, the trematodes Schistisoma mansoni and Fasciola hepatica have been found to contain ecdysteroids

(Nirde et al., 1986; Romano et al., 2003). Debate continues over whether these animals can actually synthesize these compounds or if they obtain these compounds from the host diet. This is an area of research which should be pursued to determine biochemically and molecularly if and how these animals synthesize and utilize ecdysteroids.

The most complete studies of ecdysteroid action and synthesis in parasitic helminths have been with nematodes. To date the following parasitic nematodes were shown to have ecdysteroids (including E and 20E) present in their bodies: Dirofilaria immitis, Brugia pahangi,

Ascaris suum, Anasakis simplex, and Onchocerca vovulus (Mendis et al., 1983; Chitwood, 1999;

Gagliardo et al., 2002). In the equine nematode, Parascaris equorum, polar ecdysteroid conjugates are produced when injected with radiolabelled E (O’Hanlon et al., 1991). Similar results were found in Caehnorhabditis elegans (Mercer et al., 1988). There has been some debate

if the free living nematode C. elegans contains these compounds since early studies using less accurate methods of detection lead to a positive indication for ecdysteroids (Antebi et al., 2000), but more detailed studies indicated that the compound was actually a structurally similar product called dafachronic acid (Antebi et al., 2000; Jia et al., 2002; Beckstead and Thummel, 2006).

Evaluations from several laboratories indicate that nematodes can metabolically alter ecdysteroids (Koolman et al., 1984; Mercer et al., 1989; Chitwood and Feldlaufer, 1990).

However, other work has shown that incubating some species of nematodes in the presence of radiolabelled sterols failed to generate any ecdysteroid-like material (Chitwood, 1999).

While there is controversy over the ability of nematodes to synthesize ecdysteroids, it is clear that ecdysteroids have certain effects on nematodes (Chitwood, 1999), e.g., concentrations of 20E in the range of 10-5 to 10-8 M promoted molting in Trichanella spiralis (Hitcho and

Thorson, 1971), A. suum (Flemming, 1985), and D. immitis (Barker et al., 1991). Exogenous application of E also stimulated microfilarial release in B. pahangi, meiotic reinitiation in oocytes of D. immitis (Barker et al., 1991), and egg laying in Nippostrongylus brasilensis

(Goudey-Perriere et al., 1992). Together these studies indicate that ecdysteroids play a key role in helminth development. Clearly, more studies should be devoted to understanding whether these organisms can synthesize ecdysteroids, what biochemical and molecular mechanisms regulate ecdysteroid actions and ecdysteroidogenesis, and what lead to the evolution of sensitivity to these compounds.

A recent emergence of literature indicates that there are functional signaling components for ecdysteroids in parasites. In Brugia malayi, the cloning and molecular characterization of an ecdysone receptor (EcR) and ultraspiracle (USP) homologue have been reported (Tzertzinis et al.,

2010). EcR and USP are the major nuclear receptors which mediate the responses elicited by 20E.

Additionally, EcR and USP (RXR) orthologs have been discovered in D. immitis (Shea et al.,

2004). Meanwhile, in the trematode S. mansoni, an RXR ortholog has been identified (de

Mendonca et al., 2000). This line of evidence is exciting in that it demonstrates that the machinery to respond to ecdysteroids is present and presumably has a function. As molecular methods develop and more parasite genomes are sequenced, there is little doubt that not only will the presence of the ecdysteroid signaling machinery be discovered more frequently, but also the synthetic machinery for ecdysteroids will be discovered in these animals as well. This is an exciting emerging avenue of study in the biology of parasites that may have chemotherapeutic advantage.

Both the precursor of the main arthropod molting hormone E and its more active metabolite 20E have been detected in varying amounts in all developmental stages of insects

(Smith, 1985; Lafont et al., 2005). The major source of E throughout much of the insect life cycle appears to be the prothoracic glands (PG; or their glandular counter parts) (Smith, 1985).

However, in stages where prothoracic glands are missing or non-functional, other tissues may play a role (Grieneisen et al., 1991). In several species of adult holometabolous insects, where the PG are missing, the ovaries and/or testes have been shown to synthesize ecdysteroids

(Hagedorn, 1985). In addition the ecdysteroids synthesized by these adult tissues are commonly packaged into eggs, which are not themselves capable of complex ecdysteroid biosynthesis, in the form of polar conjugates, or they are packaged with vitellogenins during the initial stages of embryogenesis (Brown et al., 2009). Once packaged into the eggs, they serve as the source of

ecdysteroids regulating development (which includes molting) during embryogenesis while the

PG or their equivalents are forming (Truman, 1985).

While the functionality of these hormones in embryogenesis remains to be determined, during post-embryonic development there are clear peaks in the titers of E and 20E in each larval and pupal stage (Smith, 1985). Just prior to each larval molt the hemolymph ecdysteroid titer increases significantly. However, in the last larval stadium of M. sexta, a holometabolous insect, there are two peaks in ecdysteroids, a smaller peak of ecdysteroids near the middle of the instar and a larger peak closer to the end of the instar (Bollenbacher et al., 1981). The first peak typically occurs in the middle of the last larval instar and tends to reprogram the larval tissues and structures. This so-called “reprogramming peak” initiates molecular events in the larval tissues that target these structures at the next molt to be broken-down. This “reprogramming peak” also causes the insect’s tissues to allow for proliferation of the adult tissues and structures in place of the larval tissues during the pupal period. The second peak, which occurs towards the end of the instar, is the molt triggering peak (Bollenbacher et al., 1981) causing pharate pupal development and the initiation of puparium formation. During the pupal stage there is a prolonged increase in the titer of ecdysteroids that is thought to be necessary for the dramatic change in body composition, structurally and physiologically from the vermiform larvae to the winged, reproductively-active adult. Subsequently, toward the end of the pupal stage, there is a decrease in ecdysteroids that allows eclosion of the fully formed adult (Bollenbacher et al., 1981).

Not only are the timing and magnitude of the ecdysteroid peaks important in guiding the development of these organisms, but the composition of the peaks is critical as well. Typically, molt-triggering peaks are composed mostly of 20E (i.e., ratios of 20E to E that are higher than 10

to 1). Unlike molt-triggering peaks or periods, reprogramming periods are composed of almost equal amounts of E to 20E (Bollenbacher et al., 1981). During the pupal period (i.e., during the metamorphic period, in which the body form, reproductive competence, and physiology of the animal are being dramatically changed) the hemolymph titer of E to 20E is vastly skewed towards E (3 parts E to 1 part 20E). Interestingly, at least in the case of M. sexta, this composition of ecdysteroids directs the physiological remodeling events that are occurring

(Bollenbacher et al., 1981). These qualitative, quantitative, and temporal relationships are critical to ensuring proper development of arthropods. Figure 2 depicts changes in hemolymph ecdysteroid content throughout the last portion of M. sexta larval development to adult development (Bollenbacher et al., 1981).

ECDYSTEROIDOGENESIS

Much of ecdysteroidogenesis occurs in the PG or its glandular equivalents i.e., in crustaceans the Y-organ (Gilbert et al., 2002). Unlike vertebrates, insects and most arthropods cannot synthesize cholesterol from simple precursors such as acetate, but must obtain it from dietary sources i.e., cholesterol or phytoecdysteroids obtained by phytophagy or feeding on vertebrate hosts (Gilbert et al., 2002; Lafont et al., 2005). Interestingly, ecdysteroids can be made by other tissues and organs when the PG is not present or not active. These organs and tissues include the ovaries (Hoffman and Lageux, 1985), the epidermis (Delbecque et al., 1990), and the amnioserosa (Kozlova and Thummel, 2002).

In insects, at least six discrete biosynthetic steps have been proposed to convert cholesterol to 20E based on biochemical, pharmacological, and inhibitor studies. Until recently,

Figure 2: Ecdysteroid titer during larval-pupal-adult development of Manduca sexta. Peaks in ecdysteroid titers are observed at developmentally important periods. Stages represented are fourth larval instar (abbreviations as used for x axis) (IV L), fifth larval instar (V L), pharate pupae (PP), pupae (P), pharate adult (PA), and adult (A). The developmental events during these stages are: (abbreviations as used above data) ecdysis (E), spiracle apolysis (SA), wandering (W), and apolysis (A). During adult life, female (▲) and male (Δ) data have been plotted separately.

Each datum point is the mean ± SEM of four to eight separate determinations (from

Bollenbacher et al., 1981).

all steps were thought to be catalyzed by cytochrome P450-type enzymes (Lafont et al., 2005;

Ono et al., 2006; Gilbert and Rewitz, 2009; Rewitz et al., 2009). These P450-type enzymes reside in either the endoplasmic reticulum or the mitochondria, and intermediates are likely shuttled between these subcellular compartments by unknown mechanisms (Gilbert et al., 2002).

The uncertainty in the actual number of enzymatic steps comes from a lack of understanding of the chemistry that takes place at the second step. This so called “black-box” reaction involves the conversion of 7-dehydrocholesterol (7-dC) into a Δ4 diketol. The Δ4 diketol is then converted to the first identified downstream substrate, 2, 22, 25-trideoxyecdysone (ketodiol) as the result of

5β and 3β hydroxysteroid oxidoreductase activity (Fig. 3). It is not clear, however, whether this reaction requires the activity of one or more P450s and/or additional regulatory proteins (Gilbert and Rewitz, 2009).

As a necessary prerequisite to determining how pulses of E are generated in vivo, much of the work on ecdysteroidogenesis in the last several years has been aimed at identifying the enzymes involved in E biosynthesis and determining where they act in the pathway. Since E is necessary to produce new cuticle during larval to larval and larval to pupal molts, it was postulated that it might also be required to produce embryonic cuticle. Therefore a number of groups sought to characterize several mutants identified in the original Drosophila melanogaster embryonic lethal screens of Nüsslein-Volhard and Wieschaus (Jürgens et al., 1984; Nüsslein-

Volhard et al., 1984; Wieschaus et al., 1984). These mutants exhibited a poorly differentiated cuticle phenotype and the loci for these include disembodied (dib), phantom (phm), shadow (sad), shade (shd), spook (spo), and shroud (sro) (Fig. 3). Collectively, this set of genes is referred to as the Halloween group (Chavez et al., 2000).

Figure 3: Enzymes involved in Drosophila melanogaster ecdysteroid biosynthesis (from Gilbert and Rewitz, 2009).

Interestingly, it was discovered that in addition to cuticle defects, mutations in these genes also disrupt many aspects of D. melanogaster late (after stage 15) embryonic morphogenesis including dorsal closure, midgut morphogenesis, and head involution (Warren et al., 2002). These genes have now been cloned and characterized for all six of these loci. Five of these (spo, phm, dib, sad, and shd) code for cytochrome P450-type proteins similar to vertebrate steroidogenic enzymes, while the sixth, sro, is allelic to kay, a gene that encodes the D. melanogaster Fos transcription factor (Chavez et al., 2000; Warren et al., 2002, 2004; Giesen et al., 2003; Petryk et al., 2003; Niwa et al., 2004; Namiki et al., 2005; Ono et al., 2006; Yoshiyama et al., 2006). Of the P450 genes, dib, phm, and sad are all expressed in the larval PG and follicle cells of the ovary, two known steroidogenic tissues. The shd and spo genes are unusual in that they are not expressed in the PG. Instead, spo is expressed only in embryonic yolk nuclei and the amnioserosa, a potential embryonic endocrine tissue (Kozlova and Thummel, 2002), while shd is transcribed only in mesodermal derivatives such as the gut, fat body and malpighian tubules. The exact substrate-product relationships for phm, dib, sad, and shd were determined using biochemical methods, including RP-HPLC/RIA, mass spectroscopy, and NMR, and these enzymes were found to catalyze the last four steps in the E biosynthetic pathway (Fig. 3).

The exact biochemical activity of spo remains unclear, although rescue experiments in which various compounds were perfused into spo mutant embryos demonstrated that spo functions at the “black-box” step (Ono et al., 2006). However, D. melanogaster S2 cells transfected with a spo-expressing transgene are not able to convert 7-dC into the Δ4 diketol suggesting that other components are required at this step (Ono et al., 2006). The embryo- specific expression of spo is also enigmatic since the “black-box” step is required for production

of E at all developmental stages. This suggests that there is another gene product with spo-like activity, and a spo paralog spookier (spok) was found in heterochromatin (Ono et al., 2006). The expression of spok complements spo in that it is expressed in the PG of larvae and not in the early embryo when spo is transcribed. Additionally, RNAi-mediated loss of spok as well as spok mutants arrest at the first instar stage suggesting that spok provides the spo-like activity after embryogenesis (Ono et al., 2006).

The final biosynthetic enzyme, neverland (nvd) was initially identified in Bombyx mori as a gene that is strongly up-regulated in the PG during ecdysteroidogenesis (Yoshiyama et al.,

2006). Unlike the majority of the other Halloween genes, nvd does not code for a P450, but instead produces a Reiske-type oxygenase. A Reiske-type oxygenase is an iron-sulfur protein that has one of its iron atoms coordinated to two histidine residues rather than cystein residues

(Reiske et al., 1964). Loss-of-function and substrate feeding experiments in D. melanogaster suggest that it acts at the first step of the biosynthetic pathway where cholesterol is converted to

7-dC. A similar conclusion was reached for the free-living nematode C. elegans, where a homolog coded for by the daf-36 locus was observed (Rottiers et al., 2006). In a similar manner to the molecular examination of these gene products in D. melanogaster, the laboratory of Dr.

Larry Gilbert, has characterized the Halloween genes in M. sexta (Gilbert and Rewitz, 2009).

Ecdysteroidogenesis in the PG is regulated by a complex of factors including PTTH, a fat body stimulatory factor (presumed to be a sterol carrier protein), and by interendocrine interactions involving several hormones including juvenile hormones (Smith, 1985). In addition, the ecdysteroids themselves may affect ecdysteroidogenesis either indirectly, by actions on the brain or corpora allata, or directly by actions at the level of the PG (Smith, 1985).

The best understood model for ecdysteroidogenesis involves the actions of PTTH on the

PG. These interactions are quite complex and PTTHs are known to stimulate E biosynthesis by cyclic 3’, 5’ adenosine monophosphate (cAMP)-dependent and –independent mechanisms. Both free calcium ions and those bound to calmodulin have also been shown to play a role in the

PTTH stimulated events (Smith et al., 1984, 1986, Smith and Combest, 1985; Gilbert et al., 1988;

Rybczynski, 2005; Rewitz et al., 2009).

Ecdysone 20-Monooxygenase

In contrast to the extensive molecular and biochemical studies devoted over the last several years to the occurrence and regulation of the enzymes regulating ecdysteroidogenesis in the PGs, far less work has gone into studying the regulation of the enzyme mediating the final step in the biosynthetic sequence. This step results in the formation of the active molting hormone 20E, by mediating its conversion from E. The enzyme responsible for this conversion is ecdysone 20-monooxygenase (E20M; E.C. 1.14.99.22; the gene product shade). It is a cytochrome P450 enzyme which was initially the focus of extensive biochemical studies regarding its regulation (Smith et al., 1979, 1980, 1983; Smith, 1985; Keogh and Smith, 1991;

Gilbert and Rewitz, 2009).

E20M activity has been localized to several tissues peripheral to the PGs including several mesodermal derivatives, e.g., the epidermis, midgut, fat body, and malpighian tubules.

Other tissues such as imaginal discs, nerve and muscle tissue, hemolymph, and salivary glands appear to be devoid of E20M activity or expression (Smith, 1985; Lafont et al., 2005; Rewitz et al., 2006a, b). Further reports demonstrated that E20M activity fluctuates in a stage and tissue

specific fashion during the life cycle of an insect, as demonstrated in D. melanogaster and M. sexta (Smith et al., 1980, 1983; Mitchell and Smith, 1986, 1988a; Mitchell et al., 1999; Gilbert et al., 2002; Lafont et al., 2005).

Depending on the insect species and/or tissue, E20M activity has been localized either to the mitochondria and/or endoplasmic reticulum (Smith et al., 1979, 1980; Gilbert et al., 2002). In the Orthopteran Schistocerca gregaria, E20M activity has been found to be associated only with the mitochondria (Johnson and Rees, 1977; Greenwood and Rees, 1984) whereas in another

Orthopteran Locusta migratoria, E20M appears to be localized exclusively in the microsomes

(Feyereisen and Durst, 1978). In the Lepidoptera such as M. sexta, E20M activity appears to be mostly associated in the mitochondria of fat body (Smith et al., 1979, 1980). However, in

Spodoptera littoralis fat body (Hoggard and Rees, 1988; Hoggard et al., 1989) and M. sexta midgut (Weirich et al., 1985), E20M activity is both microsomal and mitochondrial. In the dipterans, Aedes aegypti, Musca domestica, and D. melanogaster (Smith and Mitchell, 1986;

Mitchell, 1987; Agosin et al., 1988; Mitchell and Smith, 1988a), E20M activity has been localized in both the mitochondria and microsomes. The molecular studies on M. sexta shade reveal that at the amino terminus targeting sequences for both the endoplasmic reticulum and mitochondria are present, and thus the location of the enzyme depends on alternative splicing mechanisms (Rewitz et al., 2006a, b).

The localization of this invertebrate steroid hydroxylase varies as compared to some well studied vertebrate steroid hydroxylases. For example, the vertebrate cholesterol side chain cleavage and 11β-hydroxylase activities are localized in the mitochondria, whereas the 17α- hydroxylase is exclusively in the endoplasmic reticulum (Waterman and Simpson, 1990).

However, even in vertebrates some steroid hydroxylase activity may be localized to both mitochondria and endoplasmic reticula e.g., cholecalciferol 25-hydroxylase (Ichikawa and

Hiwatashi, 1983).

Biochemical studies of E20M activity in numerous insects, e.g., A. aegypti (Smith and

Mitchell, 1986), Diploperta punctata (Halliday et al., 1986), D. melanogaster (Mitchell and

Smith, 1986), Locusta migratoria (Feyereisen and Durst, 1978), M. sexta (Nigg et al., 1976;

Bollenbacher et al., 1977; Smith et al., 1977, 1979, 1980; Mayer et al., 1978; Kaplanis et al.,

1980; Weirich et al., 1985; Mitchell et al., 1999), M. domestica (Agosin et al., 1988), Pieris brassicae (Blais and Lafont, 1986), S. gregaria (Johnson and Rees, 1977; Greenwood and Rees,

1984), and Gryllus bimaculatus (Liebrich and Hoffman, 1991) revealed a number of characteristics. E20M has been identified and isolated biochemically and/or molecularly in several additional species: Spinacia oleracea (Grebenok et al., 1996); Spodoptera littoralis

(Shergill et al., 1995); Neobellieria bullata and Parasarcophaga argyrostoma (Darvas et al.,

1993); A. aegypti (Mitchell and Smith, 1986); B. mori (Maeda et al., 2008); Ajuga reptans

(Alekseeva, 2004); Drosophila virlillis (Chentsova et al., 2007); and the crayfish Orconectes limosus (Boecking et al., 1995). This enzyme is much like those of vertebrate steroid hydroxylases in that E20M requires NADPH as a source of reducing equivalents (Fig. 4), and can also function with cofactors such as NADH or Krebs cycle intermediates presumably by intramitochondrial generation of NADPH via transhydrogenation of NADP+ (Greenwood and

Rees, 1984; Smith, 1985; Vandock et al., 2008, 2010). E20M is also inhibited by carbon monoxide in a dose dependent fashion and such inhibition is photoreversed maximally at

Figure 4: Stoichiometry of a cytochrome P450 mixed function oxidase type reaction and the C20 hydroxylation of α- to β- ecdysone (from Bollenbacher et al., 1977).

450nm wavelength (Smith et al., 1979; Smith, 1985). These properties are consistent with a cytochrome P450-dependent nature for this insect steroid hydroxylase.

Further reports demonstrated that E20M activity and expression fluctuate dramatically in a stage and tissue specific fashion during the life cycle of an insect (Smith et al., 1980, 1983;

Mitchell and Smith, 1986, 1988a; Mitchell et al., 1999; Gilbert et al., 2002; Lafont et al., 2005;

Rewitz et al., 2006a, b; Gilbert and Rewitz, 2009). Indeed, in vitro experiments in M. sexta fat body and midgut (Mitchell et al., 1999) (Figs. 5 and 6) have shown that during larval-larval

(Beckage and Templeton, 1986), larval-pupal (Smith et al., 1980), and pupal-adult (Warren and

Gilbert, 1986) development the E20M activity varies based on tissue type. Further studies demonstrated that these changes in activity reflect changes in abundance and expression of the enzyme (Smith et al., 1983; Gilbert and Rewitz, 2009). Moreover, studies have also shown that

E20M activity varies throughout the development of other insect species such as D. melanogaster (Mitchell and Smith, 1988), adult female A. aegypti during early adult life and the gonadotropic cycle (Smith and Mitchell, 1986), and in the Orthopterans L. migratoria, D. punctata, and S. gregaria (fat body, Malpighian tubules and/or the midgut) during-larval adult development (Johnson and Rees, 1977; Feyereisen and Durst, 1978; Halliday et al., 1986). The stage and tissue specific fashion in which the expression and biochemical activity of E20M changes suggest that this steroid hydroxylase is probably regulated as noted for corresponding enzymes of vertebrates and are depicted in Figure 7.

Figure 5: Km and Vmax of fat body through fifth instar. The profiles of the apparent Km (inset) and Vmax of the fat body E20M activity during the last larval stadium of Manduca sexta. The dotted line depicts the E20M activity of fat body when incubated with substrate levels (3.4 X 10-

7 M) which approximate Km (from Smith et al., 1983).

Figure 6: Km and Vmax of midgut through fifth instar. The profiles of the apparent Km (inset) and Vmax of the midgut E20M activity during the last larval stadium of Manduca sexta. The dotted line depicts the E20M activity of midgut when incubated with substrate levels (3.4 X 10-7

M) which approximate Km (from Smith et al., 1983).

Figure 7: Changes in M. sexta midgut and fat body shade. Developmental changes in Manduca shade expression in the fat body and midgut during the fifth instar and through day 6 of pupal- adult development. Expression was analyzed by qPCR and values are means +/−S.E.M. (Data from Rewitz et al., 2006b). Ecdysteroid titer data and enzyme activity are from the Gilbert and

Smith laboratories with references given in the above cited publication (from Gilbert and Rewitz,

2009). W, wandering; E, ecdysis

Regulation of Ecdysone 20-Monooxygenase Activity

E20M has received considerable attention with respect to its biochemical properties, its developmental stage expression, and its tissue and subcellular distribution. This attention reflects the status of E20M as a model enzyme for insect steroid hydroxylase activity leading to the formation of E from cholesterol. While the intensity of study has fallen off in the past two decades, molecular and biochemical characterization of the earlier enzymes involved in E synthesis have been undertaken, however there is still much to be learned from E20M activity regulation. The stage and tissue specific fluctuations in E20M activity suggest that this enzyme is regulated in a complex fashion. Additionally, the clear and decisive increase in enzyme activity observed in midgut tissue between days four and five of the M. sexta fifth larval instar (Fig. 6), tied to morphological (i.e., dorsal vessel exposure, lateral pink pigmentation, etc.), behavioral

(i.e., cessation of feeding and wandering stage development) and developmental (i.e., reprogramming of the larval genome to pupal synthesis) markers makes this enzyme an excellent model for the study of the regulation of steroid hydroxylases generally (Smith et al., 1983; Smith,

1985; Gilbert and Rewitz, 2009). Therefore, the regulation of this enzyme has been the focus of much study.

Utilizing the ten day stadium of the M. sexta fifth larval instar, it was discovered that

E20M activity and expression exhibit a single significant increase between days four and five of the instar. This increase in midgut enzyme activity is reflected in a temporally coincident increase in the Vmax of the enzyme system, although no significant changes in the apparent Km were noted and the increase was not regulated by the loss of an endogenous inhibitor(s) (Fig.8 as

Figure 8: Midgut E20M activity during the ten-day stadium of the last larval instar of Manduca sexta. Abbreviations as used in this figure: E, ecdysis; W, onset of wandering stage development;

A, apolysis and onset of pharate-pupal development; open circle depicts enzyme activity in day three plus day five tissues mixed. Inset depicts effects of day three injected cycloheximide on day five midgut E20M activity (control activity, 387 pg 20E formed/min/mg tissue) (from Keogh et al., 1989).

indicated by the open circle; Smith et al., 1983; Keogh et al., 1989). Further studies demonstrated that the increase in midgut E20M activity is predictable based on mRNA and protein synthesis

(Figs. 7, 8). Increased activity exhibits a head-critical period that is temporally coincident with the release of PTTH (Keogh et al., 1989). However, injections of M. sexta brain sonicates containing high levels of PTTH into competent pre-wandering head- or thorax-ligated larvae were found to elicit the fifty fold increase in enzyme activity only in head-ligated animals (Fig. 9;

Keogh et al., 1989). By contrast, injections of E, 20E or the E agonist RH-5849 (Wing, 1988;

Wing et al., 1988) into competent pre-wandering head- or thorax-ligated larvae elicited the significant increase in midgut E20M activity (Fig. 10; Keogh et al., 1989; Keogh and Smith,

1991). These data are consistent with the view that PTTH released from the brain activates the

PGs to release E (Fig. 3) and this, in turn, serves as the physiological signal for the significant increase in midgut E20M activity.

Several studies have implicated cyclic nucleotides and calcium ions in the regulation of the ecdysteroidogenic enzymes in the PGs in D. melanogaster and M. sexta (Rybczynski, 2005;

Rewitz et al., 2009). Also studies have shown that E20M in the fat body of the lepidopteran S. littoralis, can be reversibly activated/deactivated via cAMP-mediated protein kinases (Hoggard and Rees, 1988; Hoggard et al., 1989). These studies suggest that the regulation of E20M is not as clear as would be expected of a steroid-mediated enzyme. Perhaps this enzyme is regulated by second messengers which would be novel in terms of a steroid-mediated event.

Figure 9: Head critical period for day five midgut E20M activity. Control activity, 374 pg 20E formed/min/mg tissue. Inset depicts day five midgut E20M activity in animals Head ligated (HL) or thorax ligated (IA, isolated abdomen) on day three and injected with 5 sonicated day one pupal brains (from Keogh et al., 1989).

Figure 10: Effects of day three injections of E or 20E (inset) on day five midgut E20M activity.

Prior to the ecdysteroid injections on day three the larvae were either head ligated (closed circles) or thorax ligated (open circles). Day five control monooxygenase activity for the E experiments was 444 pg 20E formed/min/mg tissue; day five control activity for 20E experiments was 401 pg

20E formed/min/mg tissue (from Keogh et al., 1989).

AIMS OF THIS DISSERTATION RESEARCH

The aims of the research presented in this dissertation are to expand the knowledge of

E20M regulation as it relates to insect development. From the 1970’s to the 1990’s, studies of

E20M activity were at the forefront of investigations dealing with insect endocrinology and development. However, with interests in molecular studies and the discovery of the other enzymes in the prothoracic glands that mediate the initial steps of ecdysteroidogenesis, little attention has been paid to the regulation of E20M activity. Therefore data relating to the regulation and activity of E20M, derived from biochemical and molecular studies using the M. sexta model are presented here. The first of these studies examined of the role of E20M in ecdysteroidogenesis during embryogenesis as reflected in changes in E20M enzyme activity throughout the embryonic period. A second study pursued the role of the second messenger cyclic 3’, 5’ guanosine monophosphate (cGMP) in the regulation of midgut E20M activity in the fifth larval instar of M. sexta. Thirdly, an examination of the effects of several anthraquinone compounds derived from a commercially available precursor on midgut E20M activity were carried out to determine if these compounds served as inhibitors of enzyme activity as has been shown for other quinone derived compounds. Lastly, we undertook an initial examination of ecdysteroidogenesis and E20M activity in the helminth parasites Ascaris suum and Hymenolepis diminuta, utilizing biochemical analyses.

CHAPTER II: ECDYSONE 20-MONOOXYGENASE ACTIVITY DURING

EMBRYOGENESIS OF THE TOBACCO HORNWORM, MANDUCA SEXTA

INTRODUCTION

During insect postembryonic development the conversion of the arthropod molting pre- hormone, ecdysone (E), to its physiologically more active metabolite, 20-hydroxyecdysone

(20E), is carried out by a cytochrome P450-dependent steroid hydroxylase, viz., ecdysone 20- monooxygenase (E20M; E.C. 1.14.99.22) (Smith et al., 1979, 1980, 1983; Smith, 1985; Mitchell et al., 1999; Gilbert and Rewitz, 2009). In both Drosophila melanogaster and Manduca sexta, this enzyme is transcribed by the mRNA shade (Feyereisen, 2005; Rewitz et al., 2006a, b;

Gilbert and Rewitz, 2009). The activity and gene expression of this enzyme has been shown to vary in a tissue and stage specific fashion throughout the postembryonic development of several insects including M. sexta (Mitchell et al., 1999; Feyereisen, 2005; Rewitz et al., 2006a,b;

Gilbert and Rewitz, 2009). These variations in enzyme activity inherently control the proportion of circulating E and 20E during postembryonic development. This proportional control is known to be vital to the regulation of the biochemical and physiological events that are elicited during both the metamorphic and non-metamorphic molts in insects (Smith et al., 1983; Adams et al.,

1985; Gilbert et al., 1996).

Several lines of evidence indicate that ecdysteroids play key roles during insect embryogenesis, i.e., control of cuticle deposition, epidermal line elongation, vitellogenesis, etc.

(Gilbert et al., 1980, 1996; Adams et al., 1985; Lafont et al., 2005). A number of studies also revealed that the ecdysteroids available during embryogenesis are maternally derived and that

these ecdysteroids are packaged into embryos mostly in the form of polar conjugates (i.e., as phosphate, lipid, and amide conjugates) (Gilbert et al., 1980; Isaac et al., 1983, 1984; Isaac and

Rees, 1985; Tawfik et al., 1999; Makka et al., 2002; Lafont et al., 2005). Once embryonic development begins, the conjugated ecdysteroids are hydrolyzed, thereby causing the level of free ecdysteroids to increase. The resulting accumulated ecdysteroids are then further metabolized to a 20-hydroxylated “active” form, as in the E to 20E sequence observed during postembryonic development (Warren et al., 1986; Dorn et al., 1987; Smith, 1985). The 20- hydroxylated product, which in the case of M. sexta is 20,26-dihydroxyecdysone (20,26E), is then used to control the development of the embryo (Isaac et al., 1983; Isaac and Rees, 1985;

Warren et al., 1986; Thompson et al., 1990).

In several insect species such as Bombyx mori (Mizuno et al., 1981; Horike and Sonobe,

1999; Makka and Sonobe, 2000; Sonobe and Yamada, 2004), Aedes aegypi (Gilbert et al., 1980),

Oncopeltus fasciatus (Dorn and Romer, 1976), Blaberus craniifer (Bulliere et al., 1979),

Nauphoeta cinerea (Imboden and Lanzrein, 1982), clear correlations between molting events in the embryo and peaks in molting hormone bioactivity or ecdysteroid immunoreactivity have been noted. These studies suggest that the activation of ecdysteroids is necessary for the progression of insect embryonic development to proceed properly.

In M. sexta, changes in E20M enzyme activity and expression have been extensively studied in most developmentally important periods i.e., during larval to larval, larval to pupal, and pupal to adult transitions (Mitchell et al., 1999; Gilbert and Rewitz, 2009). Yet the changes in enzyme activity during embryogenesis have been largely unexamined. Therefore, the present study evaluated the changes in E20M enzyme activity and related these variations in enzyme

activity to previously established changes in ecdysteroid content as well as physiological events that occur during embryogenesis of M. sexta.

MATERIALS AND METHODS:

Animals:

Eggs (containing embryos) of the tobacco hornworm, Manduca sexta, were collected from a breeding stock of adults maintained under non-diapausing conditions (16L:8D, 25°C, 60% relative humidity). Eggs were obtained by placing a tomato or pepper plant free of eggs into a breeding chamber with the breeding stock of male and female adult M. sexta 1 hour prior to the beginning of the light phase. The plant remained in the breeding chamber for 2 hours, and then was immediately removed from the breeding chamber and the eggs collected manually from the plant. After collection, eggs were stored under the non-diapausing conditions noted above. Eggs collected over the 2 hour time period (0-2 hour) were considered to have been laid at 1 ± 1 hour.

Ecdysteroids and Chemicals:

[23, 24 3H]-E (stocks of 45 and 70 Ci/mmol), which served as the substrate for the radioassays, was purchased from New England Nuclear, Boston, MA. Ecdysteroid standards were purchased from Fluka Chemical Corp., Ronkonkoma, NY. NADPH and scintillation fluid

(Ultima Gold) were purchased from Sigma Chemical Co., St. Louis, MO. Salts and solvents were purchased from Fischer Scientific Co., Cleveland, Ohio.

Tissue Preparation:

Whole M. sexta embryos (in the form of intact eggs) were homogenized in a sodium phosphate buffer (50 mM, pH 7.5, containing 250 mM sucrose) at a concentration of 1 egg per

10 µl of homogenization buffer. Homogenizations were carried out using a Potter-Elvejhem tissue grinder with a motor driven Teflon pestle (275 rpm, 25 strokes, 0-4◦C).

E20M assay:

An established radioassay was used to measure E20M activity in M. sexta embryos

(Smith et al., 1979, 1980; Keogh et al., 1989; Keogh and Smith, 1991; Mitchell et al., 1999).

E20M activity was assayed at 12 hour intervals starting from 1 ± 1 hour past egg deposition

(considered the start of embryogenesis), and ending with 120 ± 1 hours past egg deposition. For the assays, aliquots of 0.05 ml (containing 5 egg equivalents) of homogenate were added to 0.05 ml aliquots of 0.05 M sodium phosphate buffer, pH 7.5, containing 1000 pg [23,24-3H] E, and

NADPH (1.6 mM assay concentration). Assays were incubated for 2 hours at 30◦C with constant agitation. All assays were run in duplicate with zero-time controls. E20M activity for each time point in the 120 hour period was the mean of at least 6 and as many as 20 determinations in duplicate. Following incubation, assays were terminated via addition of 1.5 ml of ethanol. The assay tubes were centrifuged at 8,000xg for 10 minutes to remove precipitated protein. After centrifugation, 0.15 ml aliquots of assay supernatant were added to 200 µg each of cold carrier E and 20E, and the mixture evaporated to dryness. Residual assay supernatant and cold carrier were then resuspended in 50 µl of methanol, and streaked onto thin layer chromatography (TLC) plates (0.25 mm silica gel 60, F-254; E. Merck Darmstadt, Germany). The plates were developed in a solvent system of chloroform: 95% ethanol (4:1, v/v), and the E and 20E bands visualized under short wavelength UV light. The visualized bands were scraped into scintillation vials, resuspended in scintillation fluid, and radioactivity counted using a Beckman model 3801

scintillation counter (3H counting efficiency, 60%; Beckman Instruments, Irvine, CA). E20M activity was expressed as pg of 20E formed per minute per gram of eggs (± SEM).

Statistical Analyses:

Significant differences between mean E20M activities for each of the time points of the developmental profile were found by ANOVA analysis of variance, followed by pairwise comparisons of mean activity using Tukey’s-HSD post-hoc analysis. All significant differences in mean values have a P value ≤0.05.

RESULTS AND DISCUSSION

Throughout embryogenesis of M. sexta, the titers of E and 20E are low and are consequently, not considered to be the active hormones which elicit molting of the first embryonic cuticle to make way for the second embryonic cuticle (Warren et al., 1986; Dorn et al.,

1987). Previous studies have shown that during M. sexta embryogenesis, 26-hydroxyecdysone

(26E), released via hydrolysis of maternally derived 26E-phosphate conjugates, is hydroxylated at carbon 20 to form 20,26E. 20,26E then appears to direct molting during embryogenesis

(Warren et al., 1986; Dorn et al., 1987). 20,26E titers peak 72 hours into embryogenesis, a time which lies between secretion of the first and second embryonic cuticle (Fig. 1) (Warren et al.,

1986; Dorn et al., 1987). In turn, bioassays have demonstrated that 20,26E is capable of directing molting and metamorphosis. Thus, it has been hypothesized that 26E is the precursor molting hormone that is hydroxylated to 20,26E to form the active molting hormone in M. sexta embryonic development (Bergmasco and Horn, 1980; Warren et al., 1986). Despite this hypothesis the process by which 20,26E titers increase in embryogenesis of M. sexta remains largely uninvestigated. The data presented here show that E20M activity changes during M. sexta embryogenesis and that these changes correlate with changes in 20,26E titers.

Under the rearing conditions used, the time period from egg deposition (considered the beginning of embryogenesis i.e., 0 hours) to eclosion of first instar larvae was 120 hours (±4 hours). Measurements of E20M activity were carried out at 12 hour intervals throughout embryogenesis. During the initial stages of embryogenesis, E20M activity was 4.36 ± 0.49 pg

20E/min/g 1 ± 1 hour after egg laying (Fig. 2). By 12 ± 1 hours after the beginning of

Figure 1: Ecdysteroid concentrations during M. sexta embryogenesis. -♦- indicates titer of 26E; -

■- indicates titer of 20,26E during the 120 hour period of embryonic development. Reproduced with permission from (Warren et al., 1986).

Figure 2: E20M enzyme activity throughout M. sexta embryogenesis. Significant changes in enzyme activity are noted between 1 and 12 hours; 60 and 72; 72 and 84 hours; and finally between 108 and 120 hours of embryogenesis. Error bars indicate ±SEM. N= 6 to 20 determinations in duplicate with zero time controls.

embryogenesis, E20M activity increased significantly to 8.03 ± 1.07 pg 20E/min/g (Fig. 2). Over the next 48 hours (i.e., 12 ± 1 hours to 60 ± 1 hours), E20M enzyme activity did not show any significant changes. This 48 hour period is one in which secretion of the serosa and other early embryonic events are occurring (Dorn et al., 1987). Additionally, this is a period when the polar form of 26E is being hydrolyzed, free 26E is peaking, and the gradual accumulation of 20,26E is initiated (Fig. 1). 72 hours into embryogenesis, E20M enzyme activity reaches its peak at, 12.20

± 1.81 pg 20E/min/g (Fig. 2). This peak in enzyme activity corresponds to the time which lies between the secretion of the first and second embryonic cuticle and is coincident with the peak in

20,26E titer (Fig. 1) (Warren et al., 1986, Dorn et al., 1987). This is of interest because studies on the regulation of E20M enzyme activity in M. sexta fifth larval instar show that increases in enzyme activity can be elicited by its substrate E and to a lesser extent 20E (Keogh et al., 1989).

It would be interesting to determine if the same holds true for E20M during embryogenesis.

Figure 1 shows that during the first 12 hours of embryogenesis, the 26E titer increases nearly 30- fold. Meanwhile, the current study found that during the first 12 hours of embryogenesis a significant increase in E20M activity occurs. The increase in 26E in conjunction with the increase in E20M activity suggests that as in postembryonic development, increases in substrate for E20M may cause increased enzyme activity. Conversely, one might expect a pulse of 26E to be necessary for the significant increase in E20M activity observed between 60 to 72 hours, yet there does not appear to be such a peak (Fig. 1). Clearly, more investigation into the factors which elicit E20M activity in embryogenesis is necessary.

Interestingly, during the period directly following 72 hours into embryogenesis, the titer of 20,26E begins to decline and E20M enzyme activity drops significantly to 0.92 ± 0.20 pg

20E/min/g. E20M activity then remains at this low level until just before eclosion of the first instar larvae (i.e., 120 ± 1 hours) when activity increases significantly to 3.07 ± 0.58 pg

20E/min/g (Fig. 2). The correlation between decreasing titers of 20,26E (Fig. 1) and the significant drop in E20M enzyme activity (Fig. 2) suggest that the enzyme is the only source of

20-hydroxylation for 26E during embryogenesis. The rapidity of the decrease in E20M activity suggests that the rapid regulation of the enzyme is necessary for accurate developmental events to occur. Finally, the additional significant peak in E20M activity that is observed just prior to eclosion of the first instar larvae suggests a possible role in eliciting egg hatching. However, the latter assumption remains to be validated.

Prior to the current study, few groups sought to determine the role and level of E20M activity during embryogenesis of M. sexta, with only one group (Mitchell et al., 1999) suggesting that E20M activity was present in eggs. To our knowledge, only one other group, utilizing the silk moth, Bombyx mori, biochemically characterized E20M activity during embryogenesis

(Mizuno et al., 1981; Horike and Sonobe, 1999; Makka and Sonobe, 2000; Sonobe and Yamada,

2004). Clearly, detailed studies on M. sexta E20M enzyme activity during embryogenesis should be carried out utilizing both biochemical (e.g., using selective inhibitors, determining binding affinities and rates of conversion, etc.) and molecular techniques (i.e., measuring changes in levels of shade expression) to fully characterize E20M and evaluate its role in embryogenesis.

These studies would allow definitive evidence that E20M is indeed catalyzing the hormonal conversion of 26E to 20,26E and will give a better understanding of the regulation of this developmentally important enzyme. This is necessary in order to determine if this enzyme works in a similar manner to that known for postembryonic regulation (Smith, 1985; Mitchell et al.,

1999; Feyereisen, 2005). This initial report of E20M enzyme activity throughout M. sexta embryogenesis provides valuable initial data on which future studies can be based.

CHAPTER III: CYCLIC 3’, 5’ GUANOSINE MONOPHOSPHATE AND LARVAL MIDGUT

ECDYSONE 20-MONOOXYGENASE ACTIVITY OF THE TOBACCO HORNWORM,

MANDUCA SEXTA

INTRODUCTION

Ecdysone 20-monooxygenase (E20M; E.C. 1.14.99.22; mRNA transcript shade) is the insect cytochrome P450 (CYP)-dependent steroid hydroxylase responsible for conversion of E to the principal arthropod molting hormone 20E (Smith, 1985; Gilbert et al., 2002; Rewitz et al.,

2006a,b; Gilbert and Rewitz, 2009). Studies of the nature and regulation of E20M made apparent the CYP-dependency of this enzyme (Feyereisen, 2005; Rewitz et al., 2006a,b) and the similarities of this enzyme to vertebrate steroid hydroxylases (Gilbert et al., 2002; Feyereisen,

2005). Significantly, it has been demonstrated that in the tobacco hornworm, Manduca sexta

E20M activity and expression fluctuates dramatically in a stage- and tissue-specific fashion during the insect’s life cycle (Smith et al., 1983; Mitchell et al., 1999; Gilbert and Rewitz, 2009).

In point of fact, studies with the model system M. sexta, have demonstrated that during the fifth larval instar, midgut tissue E20M activity increases 50-fold between days four and five of the stadium. The magnitude and developmentally significant periods of E20M activity fluctuations suggest regulation as noted for vertebrate steroid hydroxylases (Smith et al., 1983; Waterman et al., 1988; Pikuleva, 2006).

Studies relative to the earlier steps of ecdysteroidogenesis that occur in the prothoracic glands of insects have demonstrated the involvement of second messengers (e.g., calcium ions, cyclic 3’,5’-adenosinemonophosphate [cAMP], inositol triphosphate) in regulating the enzymes

and processes responsible for synthesis and release of E (Rybczynski, 2005). However, to our knowledge no examination of the role(s) of second messengers in the steroid-mediated regulation of E20M activity at the target tissue level has been reported. The studies presented here make evident both in vitro and in vivo effects of cyclic 3’,5’ guanosine monophosphate (cGMP) on M. sexta E20M enzyme activity. Additionally, changes in midgut tissue cGMP content throughout the ten day fifth (final) larval stadium of M. sexta were measured via radioimmunoassay. These changes were found to be coincident with developmentally significant periods. Collectively, the findings herein represent the first demonstration of cGMP involvement in E20M regulation occurring in E target tissues.

MATERIALS AND METHODS

Animal Cultivation

Animals used for assessment of second messenger effects on E20M activity were days three and five, gate II, fifth instar larvae of the tobacco hornworm M. sexta. Larvae were reared on an artificial diet under a non-diapausing photoperiod (L:D, 16:8) at 26º C and ~60% relative humidity (Bell and Joachim, 1976). Gate II animals were staged from fourth larval instar head capsule slippage-brown mandibles (pharate fifth instar larvae-day zero) and also were selected on the basis of empirically generated weight curves (Goodman et al., 1985). Such animals displayed the ability to respond to increasing E titers, but had not yet experienced an E increase and thus were deemed competent to respond (Keogh et al., 1989; Keogh and Smith, 1991).

Wandering stage animals were defined on the basis of morphological (dorsal vessel exposure with lateral pink pigmentation) and behavioral (wandering) markers.

Animal Dissection and Tissue Preparation

Determination of E20M activity

Midguts from day five fifth instar cold-anesthetized (20 min at 4º C) larvae were dissected in Lepidopteran Ringers (Weevers, 1966). The excised tissues were rinsed in homogenization medium (50 mM sodium phosphate, 250 mM sucrose, pH 7.5), blotted, weighed, and homogenized (10 to 100 mg tissue/ml medium) using a Potter Elvehjem tissue grinder with a motor driven Teflon pestle (275 rpm, 20 strokes, 0-4º C).

In addition to the 30 min E20M assay incubation, performed prior to homogenization and assessment of E20M enzyme activity, some midguts were incubated for up to 2 h in homogenization medium. The medium contained one or more concentrations of cAMP, dibutyryl cAMP (dbcAMP), cGMP, dibutyryl cGMP (G), and/or theophylline (T) as indicated. The latter are referred to as “pre-assay” incubations.

cGMP titer measurements by radioimmunoassay

Midguts from larvae representing the last day of the fourth larval stadium and each of the ten days in the fifth larval stadium were cold-anesthetized and dissected in radioimmunoassay

(RIA) buffer (50 mM Tris-HCl, 4 mM EDTA, pH 7.5). The excised tissues were rinsed in RIA buffer, blotted, weighed, and homogenized (0.3 to 0.5g per ml in RIA buffer).

Ligations and Injections

Animals to be ligated and/or injected were cold-anesthetized. Head ligations were performed between the head and first prothoracic segment. Decapitation was accomplished using a razor blade to make a cut immediately anterior to the ligature. Thorax ligations were performed between the final metathoracic and first abdominal segments and the head and thorax removed following ligation using a razor blade. Ligations were done with waxed dental floss.

Intrahemocoel injections, via the abdominal prolegs, were performed using 10 μl

Hamilton syringes on day three of the last larval stadium. After injection, the prolegs were ligated with waxed dental floss to prevent leakage and/or infection. Sham injected whole animals served as controls for all injection experiments. The following were injected either alone or in combination: cAMP, dbcAMP, cGMP, G, T, 6-anilinoquinoline-5,6-quinone (LY-

83583, LY), methylene blue (MB), 1H-[1,2,4]oxadizolo[4,3-a]-quinoxalin-1-one (ODQ) and S- nitroso-N-acetylpenicillamine (SNAP, S) as indicated. In addition to the compounds listed above, all head-ligated and thorax-ligated animals received injections of 1,2-dibenzoyl-1-tert- butylhydrazine (RH-5849, RH) as described. The injection vehicle for all compounds and combinations thereof was 95% ethanol

E20M Assay

E20M activity of midgut tissue was assessed by radioassay as previously described

(Smith et al., 1979; Mitchell et al., 1999) with the following exceptions: the direct effects of cAMP, dbcAMP, cGMP, G, either alone or in combination with T, and other additives as indicated, were assessed for 30 min in 50 mM sodium phosphate buffer (pH 7.5).

Midgut cGMP titer determinations

The cGMP contents of M. sexta fifth larval instar midguts were assessed using a TRK-

500 cGMP RIA kit (GE Healthcare, UK). Assessments were performed according to manufacturer’s protocol as summarized here. Midguts, representing each day of the M. sexta fifth larval instar, were dissected from precisely staged animals, rinsed in RIA buffer, blotted and weighed. Midguts were homogenized (0.3 to 0.5 g per ml in RIA buffer). Homogenates were boiled for 3 min and proteins removed by centrifugation at 3800xg for 10 min. 100 µl aliquots of supernatant were added to tubes containing 50 µl 3H-cGMP (each assay tube contained 8 X 10-3

µCi) and 50 µl of stock anti-cGMP antibody solution prepared per manufacturer’s protocol.

After mixing, tubes were allowed to stand for 1.5 h at 4° C. The reaction was terminated and the cGMP-bound antibody precipitated via the addition of 1 ml of 60% (NH4)2SO4 and collected by

centrifugation at 9000xg for 10 min. The pellet was resuspended in 1.1 ml of water.

Subsequently, 1 ml of the suspension was taken for scintillation counting. The amount of cGMP present in the tissue samples was calculated by comparison to a cGMP standard curve made from stock solutions per manufacturer’s protocol.

Statistical analyses

Differences between control and treated midgut E20M activity were considered significant with a p-value ≤ 0.05 employing the unpaired Student’s t test. All differences in mean day five midgut E20M enzyme activities of treated versus control animals presented in the

Results, are significant (p≤0.05). Analysis of variance (ANOVA, JMP 8, SAS) followed by posthoc Tukey’s tests were carried out to determine differences between means of cGMP titers.

Chemicals

The ecdysteroid standards, i.e., E and 20E, were purchased from Fluka Chemical

Corporation (Ronkonkoma, NY, USA). Radiolabelled [23,24-3H]-E (stocks of 45 and 70

Ci/mmol) was purchased from GE Life Sciences-Amersham (UK). Animal diet supplies were obtained from United States Biochemical Corporation, Cleveland, OH, USA. NADPH, cyclic and dibutyryl cyclic nucleotides as well as T were obtained from Sigma Chemical Company, St.

Louis, MO, USA. LY-83583 and SNAP, were purchased from Calbiochem-Novabiochem

Corporation, San Diego, CA, USA. Methylene blue was purchased from J.T. Baker Chemical

Company, Phillipsburg, NJ, USA. ODQ was purchased from Cayman Chemical Company, Ann

Arbor, MI, USA. Scintillation fluid (Ultima Gold) was purchased from PerkinElmer, Waltham,

MA, USA. Organic solvents and all other reagents were obtained from Fisher Scientific

Company, Cleveland, OH, USA. RH-5849 was a gift from Dr. Keith Wing, Rohm and Haas

Company, Spring House, PA, USA.

RESULTS

Pre-assay incubation of isolated M. sexta day five midguts with G and the phosphodiesterase inhibitor T (the combination of which is referred to as G/T), resulted in a significant increase in E20M activity after 0.5 h and reached its maximum significant activity

(~74% greater than controls) after 1 h (Fig. 1). dbcAMP did not substitute for G and incubations in excess of 1 h with either, cAMP, dbcAMP, cGMP, or G, with or without T, did not significantly alter midgut E20M activity versus controls (Fig. 1). Incubation of day five midguts with dbcAMP plus T for up to 2 h did not result in a significant change in activity (Fig. 1).

Neither cAMP nor cGMP, in the presence or absence of T, nor T alone significantly changed midgut E20M activity.

Since the only in vitro cyclic nucleotide effect on E20M activity was that of the G/T combination, in vivo assessments were made comparing G/T injected animals with corresponding controls. Fifth instar, day three pre-wandering stage larvae were injected with G/T and midgut homogenate E20M activity assessed on day five. Use of day three larvae allowed for injections of competent animals that would be expected to respond by day five (Keogh et al.,

1989).

Interestingly, injections of day three fifth instar larvae with G/T essentially yielded no change in day five midgut E20M activity as compared to whole animal sham injected controls

(Fig. 2). Injection of the guanylate cyclase inhibitors LY, MB, or ODQ (Diamond, 1987;

Deutsch et al., 1996; Friebe and Koesling, 2003) however, significantly decreased average day five midgut E20M activity to 31, 24, and 34%, respectively, that of controls. These inhibitions

Figure 1: Effects of pre-assay incubation of M. sexta midguts with dibutyryl cyclic nucleotides plus T on day five midgut E20M activity. Midguts from gate II day five last instar larvae were incubated for up to 2 h prior to assay in 50 mM sodium phosphate buffered medium (pH 7.5) containing 125 mM sucrose, and either G/T (-♦-) or dbcAMP plus T (-■-). E20M activity is expressed as percent of untreated incubated controls. Control E20M activity (100%) was 337 ±

35 pg of 20E/min/mg tissue. Values are the means of 4-9 determinations done in duplicate.

Cyclic nucleotide and T concentrations were: G and dbcAMP, 1 x 10-3 M, and T, 1 x 10-4 M.

Error bars indicate SE. Number of stars indicate level of statistical difference from zero time pre- incubation.

Figure 2: Effects of injected G or pharmacological factors on whole animal M. sexta day five midgut E20M activity. Injected materials and abbreviations of these materials were as follows:

G/T; LY (LY-83583); LY plus G/T; LY plus S (SNAP); MB (methylene blue); MB plus G/T;

MB plus S; ODQ (1H-[1,2,4]oxadizolo[4,3-a]-quinoxalin-1-one); ODQ plus G/T; and ODQ plus

S. Injections were carried out on gate II day three fifth instar larvae. Injected materials expressed as nmol per g body mass either alone or in combination as indicated were: G/T, 20 and 2, respectively; LY, MB, and S, 20; ODQ, 13.4. Midgut E20M activities were assessed on day five of the stadium and are expressed as a percent of whole animal sham injected control day five midgut E20M activity. Control E20M activity (100%) was 419 ± 54 pg 20E/min/mg tissue.

Values are the means of 10 to 17 determinations in duplicate. Error bars indicate SE.

were lessened by co-injections of G/T or the activator of soluble guanylate cyclases, S

(Matsumoto et al., 2006). With LY, MB, or ODQ, G/T co-injection restored activity to 78, 89, and 85%, of controls, respectively. Similarly, co-injections of S with LY, MB, or ODQ restored activity to 73, 92, and 75%, respectively that of controls (Fig. 2).

To determine if the G/T effects entailed changes in larval brain physiology (e.g., by endogenous synthesis of E induced by PTTH release), head-ligated day three fifth instar larvae were injected and E20M evaluated on day five. Injections of head-ligated larvae with G/T, as used for intact animals, resulted in essentially no change in the expected low levels of day five midgut E20M activity as compared to whole animal sham injected controls (Fig. 3A). Injections of the E agonist, RH, restored day five midgut E20M activity to 99% of controls, whereas injections with an amount of RH deemed to be suboptimal resulted in day five E20M activities that were only 20% of controls (Fig. 3A). However, when G/T was co-injected with suboptimal

RH, day five E20M activity increased to 81% that of controls. Moreover, the guanylate cyclase activator S elicited similar significant increases in midgut E20M activity when co-injected with suboptimal RH, yielding day five midgut E20M activities that were 92% of controls (Fig. 3A).

Co-injecting RH with the guanylate cyclase inhibitor LY into head-ligated animals reduced midgut E20M activity to 20% that of controls (Fig. 3B). In contrast, the addition of G/T to the injection mixture of RH and LY resulted in day five midgut E20M activity that was 85% of controls. Likewise, when the guanylate cyclase activator S was included with the RH and LY mixture, day five midgut E20M activity was 78% that of controls (Fig. 3B).

Figure 3: Effects of injected G or pharmacological factors on head-ligated M. sexta day five midgut E20M activity. Injected materials and abbreviations of those materials were as follows:

A. G/T, RH (RH-5849), suboptimal RH (3/4 RH), 3/4 RH plus G/T, 3/4 RH plus S (SNAP); B.

RH plus LY (LY-83583), RH plus LY plus G/T, RH plus LY plus S; C. RH plus MB (methylene blue), RH plus MB plus G/T, RH plus MB plus S; D. RH plus ODQ (1H-[1,2,4]oxadizolo[4,3- a]-quinoxalin-1-one), RH plus ODQ plus G/T, RH plus ODQ plus S. Injections were carried out on gate II day three fifth instar head-ligated larvae. Injected materials expressed as nmol per g body mass either alone or in combination as indicated were: G/T, 20 and 2, respectively; RH, 8;

3/4 RH, 6; LY, MB, and S 20; ODQ, 13.4. Midgut E20M activities were assessed on day five of the stadium and are expressed as a percent of whole animal sham injected control day five midgut E20M activity. Control E20M activity (100%) was 419 ± 54 pg 20E/min/mg tissue.

Values are the means of 10 to 17 determinations in duplicate. Error bars indicate SE.

Inclusion of the guanylate cyclase inhibitor MB with RH into head-ligated animals reduced midgut E20M activity to 14% that of controls (Fig. 3C). A marked decrease in the degree of inhibition was noted if the day three injections of the RH and MB mixture were augmented with

G/T. This resulted in E20M activities that averaged 87% of whole animal sham injected controls.

Similarly, if S was added to the injection cocktail, E20M activities were 79% of controls (Fig.

3C).

Lastly, co-injections of head-ligated larvae with the guanylate cyclase inhibitor ODQ plus

RH resulted in E20M activities that were 27% of controls (Fig. 3D). This decrease in activity was lessened considerably by co-injecting G/T with the RH and ODQ mixture yielding E20M activities that were 79% of controls. Similarly, when S was included in the RH plus ODQ mixture, E20M activity was 63% of controls (Fig. 3D).

Utilizing thorax-ligated animals, we sought to determine if G affects prothoracic gland E synthesis and release. Thus, day three fifth instar thorax-ligated M. sexta larvae were evaluated for day five midgut E20M activity changes in response to the same treatments listed for head- ligated animals. G/T injections of thorax-ligated larvae resulted in essentially no change in the expected low level of day five midgut E20M activity (Fig. 4A). Injections of RH, restored

E20M activity to levels (77%) similar to those of controls. Suboptimal RH injections resulted in

E20M activities that were 20% of controls. But, with co-injection of G/T with suboptimal RH, a significant increase in E20M activity was noted (62% of controls). S elicited similar significant increases when co-injected with suboptimal RH, resulting in E20M activities, which were 71% of controls (Fig. 4A).

Figure 4: Effects of injected G or pharmacological factors on thorax-ligated M. sexta day five midgut E20M activity. Injected materials were as follows: A. G/T, RH, suboptimal RH (3/4 RH),

3/4 RH plus G/T, 3/4 RH plus S; B. RH plus, RH plus LY plus G/T, RH plus LY plus S; C. RH plus MB, RH plus MB plus G/T, RH plus MB plus S; D. RH plus ODQ, RH plus ODQ plus G/T,

RH plus ODQ plus S. Injections were carried out on gate II day three fifth instar thorax-ligated larvae. Injected materials expressed as nmol per g body mass either alone or in combination as indicated were: G/T, 20 and 2, respectively; RH, 8; 3/4 RH, 6; LY, MB, and S, 20; ODQ, 13.4.

Midgut E20M activities were assessed on day five of the stadium and are expressed as a percent of whole animal sham injected control day five midgut E20M activity. Control E20M activity

(100%) was 419 ± 54 pg 20E/min/mg tissue. Values are the means of 10 to 17 determinations in duplicate. Error bars indicate SE.

Co-injecting RH with LY reduced E20M activity to 4% of controls (Fig. 4B). The addition of G/T to the injection mixture resulted midgut E20M activity which was 65% of controls (Fig. 4B). Likewise, when S was added to the RH and LY mixture, the resulting E20M activity was 61% of controls (Fig. 4B).

The combination of RH and MB significantly reduced E20M activity to 16% of controls, and this inhibition was relieved (72% of controls) if the injection mixture was augmented with

G/T. Similar results (70% of controls) were obtained if S was used in lieu of G/T with the RH-

MB cocktail (Fig. 4C).

When ODQ and RH were injected on day three the resulting E20M activities were 3% of controls (Fig. 4D). This was relieved by injecting G/T with the RH and ODQ mixture, yielding

67% of the control activity. When S was injected with the RH-ODQ mixture, the animals displayed E20M activity that was 57% of controls. Injections of either whole or ligated animals with cAMP, dbcAMP, or cGMP, with or without T, as well as with T alone, were without significant effect.

Changes in midgut homogenate cGMP levels throughout the M. sexta fifth larval instar were evaluated by RIA (Fig. 5). From day zero (last day of the fourth larval instar) through day three, the variations in cGMP content were not significant (day zero mean = 46.1 pmol/g tissue).

However, in comparisons of day four larval midguts with the prior days’ assessments, cGMP content significantly declined to an average of 25 pmol cGMP/g tissue. Thereafter, the level of cGMP significantly increased by more than three-fold to 81 pmol cGMP/g tissue. On day six, the cGMP titer declined significantly (37 pmol/g tissue) as compared to day five (Fig. 5).

Figure 5: M. sexta midgut homogenate cGMP titers assessed by RIA. Homogenates equivalent to

0.3g to 0.5g of tissue per ml of homogenization buffer were prepared from the last day of the fourth larval instar and each day of the fifth larval stadium. Significant changes occurred between days three and four, four and five, five and six, and eight through ten (F=14.23, p<

0.0001). Number of stars indicates level of significance. *, not significantly different from day zero; **, significantly different from day zero; ***, significantly different from day zero and day four. Values are the means of 15 to 25 determinations; error bars indicate SE.

It is noteworthy, that on days seven through ten of the instar the amount of cGMP/g tissue increased significantly, seemingly reaching a plateau of about 83 pmol cGMP/g tissue on days nine and ten of the instar (Fig. 5).

DISCUSSION

Based on both in vitro and in vivo evaluations, the data obtained support an involvement of cGMP in the regulation of M. sexta midgut E20M activity. In terms of in vitro studies, incubation of isolated day five midgut tissue with either cAMP or cGMP, at concentrations ranging from 10-8 to 10-3 M, with or without 10-4 M T or with T alone, did not significantly alter

E20M activity. However, incubation of isolated midguts for 0.5 h with G/T, increased E20M activity significantly and this increase was even more pronounced after 1 h. Following this,

E20M activity returned to levels like those of the controls. Substitution of dbcAMP for G in the cyclic nucleotide-T combination was essentially without effect on E20M activity.

Given the in vitro effects of G/T, injections of G/T, in the absence or presence of pharmacological factors known to affect cGMP content in other systems (Friebe and Koesling,

2003), were administered to day three fifth instar larvae and midgut E20M activity assessed on day five. Although, G/T injections were without significant effect versus controls, injections of the guanylate cyclase inhibitors LY and MB (Diamond, 1987; Deutsch et al., 1996) significantly reduced E20M activity. Such reductions were equally apparent with the more specific soluble guanylate cyclase inhibitor, ODQ. Furthermore, these inhibitor-dependent reductions were relieved with co-injections of G/T or the soluble guanylate cyclase activator, S (Matsumoto et al.,

2006).

Considering the data obtained employing head- or thorax-ligated animals, it appears that neither the larval brain (PTTH) nor prothoracic glands (ecdysteroid synthesis) were targeted by

G/T and/or pharmacological agents, but rather the effects of G/T and/or these agents were those

occurring at the level of the target tissues. The need for E to elicit E20M activity, which is apparently modulated by cGMP, is noted with the data obtained using the E agonist RH (or suboptimal RH) in the presence of G/T and/or pharmacological factors.

The ability of RH to elicit increases in midgut E20M activity supports previous observations demonstrating that injections of RH into head- or thorax-ligated larvae elicit a 50- fold increase in E20M activity and the onset of wandering stage development by day five (Keogh et al., 1989; Keogh and Smith, 1991). A novel aspect of the present study is the observation that suboptimal levels of RH elicited less E20M activity than that observed for the optimal dose. In this context, it is noteworthy that G/T and S synergize with suboptimal RH to elicit levels of

E20M activity that matched that of controls.

Clearly, a difference is apparent when G/T is applied directly to M. sexta midgut tissue as compared to injection of this combination. While this difference warrants further investigation, our data suggest the potential for short-term effects of cGMP on midgut tissue (in vitro), and a longer-term effect(s) of the cyclic nucleotide when injected into intact animals. Regardless, the data obtained support an effect of cGMP on the regulation of E20M activity. This is particularly evident inasmuch as all of the pharmacological factors tested are known to affect cGMP metabolism in other systems (Diamond, 1987; Deutsch et al., 1996; Friebe and Koesling, 2003;

Matsumoto et al., 2006), and were very effective in altering E20M activity. Moreover, it is of interest that a soluble guanylate cyclase system apparently acts in the maintenance of cGMP titers in M. sexta based on the effects of ODQ and S on E20M activity. In keeping with this consideration, preliminary data suggest that the guanylate cyclase inhibitors employed in this study act to diminish cGMP content of midgut tissue.

Assessments of cGMP titers in M. sexta fifth larval instar midguts demonstrated that the level of cGMP varies throughout the stadium. While a marked decrease in titer occurs between days three and four, our data indicate that midgut cGMP titers initially peak significantly between days four and five of the fifth larval instar. This peak aligns perfectly with the 50-fold increase in E20M activity (Mitchell et al., 1999) and the maximal expression of the E20M gene, shade (Gilbert and Rewitz, 2009). Thus, there are changes in cGMP content that occur in a temporally coincident manner with both changes in E20M activity and expression.

Cyclic nucleotides have been implicated as physiological regulators in a number of insect systems, e.g., the cotton leaf worm, Spodoptera littoralis (Hoggard et al., 1989); the flesh fly,

Sarcophaga crassipalpis (Denlinger and Wingard, 1978); and the Bertha armyworm, Mamestra configurata (Bodnaryk, 1983). With respect to M. sexta larvae, cGMP acts in the priming of the central nervous system in response to eclosion hormone (Truman, 2005). To date, all data indicate that E is not only a substrate and inducer of E20M (Keogh et al., 1989), but a hormone with specific functions (Beckstead et al., 2007) that may be acting through a second messenger system. The data of the present study make evident for the first time that in the M. sexta system cGMP enhances E-dependent E20M activity at the level of the target tissues. Moreover, our data provide a framework for further studies of the effects of cGMP at the biochemical level.

CHAPTER IV: EXAMINING THE EFFECTS OF CYCLIC 3’,5’ GUANOSINE

MONOPHOSPHATE ON SHADE EXPRESSION IN MIDGUT TISSUE OF THE TOBACCO

HORNWORM, MANDUCA SEXTA

INTRODUCTION

Ecdysone 20-monooxygenase (E20M; E.C. 1.14.99.22; mRNA transcript shade) is the insect cytochrome P450 (CYP)-dependent steroid hydroxylase responsible for conversion of edysone (E) to the principal arthropod molting hormone 20-hydroxyecdysone (20E) (Smith,

1985; Gilbert et al., 2002; Rewitz et al., 2006a,b; Gilbert and Rewitz, 2009). The regulation of

E20M activity has been studied previously in the insect model system Manduca sexta (Keogh et al., 1989; Keogh and Smith, 1991; Feyereisen, 2005). These studies demonstrated that E20M activity is regulated in a fashion similar to that observed for other steroid hydroxylases (Pikuleva,

2006). Specifically these studies demonstrated that increases in E20M activity are dependent upon a head-critical period, are induced by increasing titers of the enzyme substrate E, and are dependent upon transcriptional and protein synthetic activities (Keogh et al., 1989). In a recent study by Drummond et al., (2010), E20M activity was shown to be regulated by cyclic 3’,5’ guanosine monophosphate (cGMP) and other pharmacological factors which affect intracellular cGMP concentration. This finding is of interest because this is counter to the accepted model of steroid hormone action. This model indicates that following entrance of the steroid hormone into the cell, and subsequent binding to cytoplasmic or nuclear receptors, the activated receptors directly affect transcriptional events in the nucleus. If cGMP were involved, that would suggest a different mode of action for this steroid hormone, viz. action through a second messenger cascade in addition to or in place of the accepted steroidal signaling model.

In order to determine if the effects of cGMP are related to molecular regulation or simply reflect biochemical activity, the current study examines changes in shade expression after exposure to cGMP or factors which affect intracellular cGMP concentration. Collectively, the findings herein represent the first demonstration of cGMP involvement in E20M expression occurring in E target tissues.

MATERIALS AND METHODS

Animal Cultivation

Animals used for assessment of second messenger effects on E20M expression were days three and five, gate II, fifth instar larvae of the tobacco hornworm M. sexta. Larvae were reared on an artificial diet under a non-diapausing photoperiod (L:D, 16:8) at 26º C and ~60% relative humidity (Bell and Joachim, 1976). Gate II animals were staged from fourth larval instar head capsule slippage-brown mandibles (pharate fifth instar larvae-day zero) and also were selected on the basis of empirically generated weight curves (Goodman et al., 1985). Such animals displayed the ability to respond to increasing E titers, but had not yet experienced an E increase and thus were deemed competent to respond (Keogh et al., 1989; Keogh and Smith, 1991).

Wandering stage animals were defined on the basis of morphological (dorsal vessel exposure with lateral pink pigmentation) and behavioral (wandering) markers.

Animal Dissection and Tissue Preparation

cGMP effects on shade expression

Animals were staged and dissected as in Drummond et al., 2010. Subsequent to dissection the tissues were stored in 150 µl of RNAlater (Ambion). The samples were stored at between 20º C and 4º C for use in total RNA extraction. At the time of RNA extraction tissue was subsequently homogenized by syringe in lysis buffer provided in the RNeasy kit (Qiagen).

Ligations and Injections

Intrahemocoel injections, via the abdominal prolegs, were performed using 10 μl

Hamilton syringes on day three of the last larval stadium. After injection, the prolegs were ligated with waxed dental floss to prevent leakage and/or infection. Sham injected whole animals served as controls for all injection experiments. The following were injected either alone or in combination: cyclic 3’5’ adenosine monophosphate (cAMP), dibutyryl cAMP

(dbcAMP), cGMP, dibutyryl cGMP (G), theophylline (T), 6-anilinoquinoline-5,6-quinone (LY-

Total RNA isolation and cDNA synthesis

Total RNA was isolated using an RNeasy minikit (Qiagen) and contaminating genomic

DNA was removed by on-column DNase digestion according to the manufacturer’s protocol.

The RNA was quantified spectrophotometrically at 260 nm and cDNA was synthesized from

total RNA using an oligo(dT)20 primer, 2X First-Strand Reaction mix, and SuperScript

III/RNaseOUT Enzyme mix, according to the manufacturer’s protocol (Invitrogen).

Quantitative real-time PCR (qPCR) analysis of Msshd

The expression of Manduca sexta shade (Msshd) was analyzed by qPCR performed with a LightCycler 480 (Version 1.5, Roche). Primers for qPCR analysis were designed using the

Primer3 program (Rozen and Skaletsky, 2000). Primers were: forward, 5’-

AGGCACCAGCGAAGAGACC-3’ and reverse, 5’-TGAAGCCATTTGTCACCTTGT-3’. PCR reactions were carried out with FastStart Taq DNA polymerase (Roche) in a final volume of 18

µl, containing 1 µl of each primer; 10µl of 2X master mix (containing: FastStart Taq polymerase reaction buffer, nucleoside triphosphate mix, SYBR Green dye I, and MgCl2), and 2µl of cDNA synthesized from a standardized amount of total RNA. The amplification program consisted of a pre-incubation step of 95°C for 5 min followed by 45 amplification cycles consisting of 95 °C for 15 sec, 59 °C for 15 sec, and 72 °C for 15 sec repeated. The specificity of the PCR products was verified by melting curve analysis for all samples, and checked electrophoretically for each primer pair. Samples were corrected for differences in amplification efficiencies, determined by serial 1:10 dilutions of cDNA. Transcript levels of shade were normalized to mRNA levels of the

(ribosomal) housekeeping gene, SrpL32. Primers for M. sexta SrpL32 were forward, 5’-

GGCGTAAACCGAGAGGTATTG-3’ and reverse, 5’-GAGCATGTGACGGGTCTTCT-3’.

Potential genomic DNA contamination was checked by including non-reverse transcribed total

RNA in control reactions. Facilities, supplies, equipment and expertise for completion of these molecular studies were generously provided in part by Drs. Michael O’Connor and Kim F.

Rewitz of the University of Minnesota, Department of Genetics. Michael O’Connor is supported

by funding from the Howard Hughes Medical Institute. Additional funding was provided by the

Department of Biological Sciences at Bowling Green State University.