Trehalulose and Multiple Day Flight in the Physiology and Ecology of Bemisia Tabaci (Gennadius)

Trehalulose and Multiple Day Flight in the Physiology and Ecology of Bemisia tabaci (Gennadius)

Item Type text; Electronic Dissertation

Authors Hardin, Jesse Andrew

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 05/10/2021 22:56:14

Link to Item http://hdl.handle.net/10150/195980

TREHALULOSE AND MULTIPLE DAY FLIGHT IN THE PHYSIOLOGY AND ECOLOGY OF BEMISIA TABACI (GENNADIUS)

Jesse Andrew Hardin

______

A Dissertation Submitted to the Faculty of the

DEPARTMENT OF ENTOMOLOGY

In Partial Fulfillment of the Requirements For the Degree of

DOCTOR OF PHILOSOPHY

In the Graduate College

THE UNIVERSITY OF ARIZONA

2009

THE UNIVERSITY OF ARIZONA GRADUATE COLLEGE

As members of the Dissertation Committee, we certify that we have read the dissertation prepared by Jesse Andrew Hardin entitled “Trehalulose and Multiple Day Flight in the Physiology and Ecology of Bemisia tabaci (Gennadius)” and recommend that it be accepted as fulfilling the dissertation requirement for the

Degree of Doctor of Philosophy

______Date: 4/17/09 David N. Byrne

______Date: 4/17/09 Martha (Molly) S. Hunter

______Date: 4/17/09 Daniel R. Papaj

______Date: 4/17/09 Michael E. Salvucci

______Date:

Final approval and acceptance of this dissertation is contingent upon the candidate’s submission of the final copies of the dissertation to the Graduate College.

I hereby certify that I have read this dissertation prepared under my direction and recommend that it be accepted as fulfilling the dissertation requirement.

______Date: 4/17/09 Dissertation Director: David N. Byrne 3

STATEMENT BY AUTHOR

This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at the University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations from this dissertation are allowable without special permission, provided that accurate acknowledgment of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his or her judgment the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author.

SIGNED: Jesse Andrew Hardin

ACKNOWLEDGEMENTS I must first give thanks and acknowledgement for the support of my wife, Kari, whose hard work and sacrifice have contributed as much to my successes in this endeavor as my own. Her patience and confidence have been my inspiration, and I hope to return her contributions as we share a life of learning together. Our son Jaxon has been the delight of my life and the greatest gift I could ask for. The steady and loving guidance I received from my parents has been a legacy I hope to bestow on my children in return. I could not have achieved this without my parents’ support. I must gratefully appreciate all the contributions of David Byrne, who has been a mentor in every sense of the word: an academic advisor, coach, and cheerleader. He has been a role model for commitment, perseverance, and passion for science and knowledge. I especially want to acknowledge his sense of humor and appreciation for the esoteric. In the Byrne laboratory, I met two of the best friends I could ask for in life: Mark Asplen and David Bellamy. A merry band of musketeers were we, and what magic we did discover together shall not be forgotten. Chris Schmidt and Alex Swanson have also been true friends and companions during my time here, and I am deeply indebted to all these fellows for the intellectual stimulation and wisdom they have shared with me over the years. There are many other friends I have made here and each has touched me personally, but I have not the space to thank them all and so my sincere gratitude must suffice. Many others have helped me along the path through discovery, especially my committee of Mike Salvucci, Molly Hunter and Dan Papaj, and I greatly appreciate their contributions and service. Others have helped with my research directly: Joshua Garcia, Suzanne Kelley, Heather Ferguson, Erich Draeger, Thann Grundy, Catherine Bartlett and Vanessa Jacobs-Lorena; I am deeply grateful for their assistance. I also want to thank Bruce Patterson for his devotion to teaching and his efforts to improve mine. The Entomology Department staff and faculty have been instrumental in their support and investments toward my graduate career here. In particular the office staff and the administration have been very kind in lending assistance and support whenever called upon for help. I have also supported by the EEB and MCB departments with many teaching assistantships, and also with funding from the College of Agriculture and Life Sciences, the Graduate College, Bio5, NSF, and USDA-NRI. 5

DEDICATION Dedicated in memoriam to the strong women in my family and life: my grandmother, Corrine Hein; my aunt, Deborah Champagne (née Hein); and committee member Dr. Jacqueline L. Blackmer. Each lived with dignity and grace, touched us with warmth, inspired us with courage in the face of cancer and left us too soon. 6

TABLE OF CONTENTS ABSTRACT ...... 7 CHAPTER 1. INTRODUCTION...... 9 1.1.Whitefly as pests ...... 9 1.2. Methods of whitefly management ...... 10 1.3. Feeding by parasitoids on whitefly honeydew ...... 11 1.4. Nutritional physiology of whitefly feeding ...... 11 1.5. Whitefly migration...... 17 1.6. Current study summary ...... 21 1.7. Explanation of dissertation format...... 22 CHAPTER 2. PRESENT STUDY ...... 23 2.1. Parasitoid feeding on trehalulose, a component of B. tabaci honeydew ...... 23 2.2. Trehalulose production under thermal stress ...... 24 2.3. Flight studies ...... 25 2.4. Conclusions ...... 27 REFERENCES...... 29 APPENDIX A: THE LONGEVITY EFFECTS OF TREHALULOSE FEEDING BY WHITEFLY PARASITOIDS...... 35 APPENDIX B: EXAMINING THE ROLE OF TREHALULOSE AND SORBITOL IN THERMAL TOLERANCE BY BEMISIA TABACI ...... 43 APPENDIX C: MULTIPLE DAY FLIGHTS BY BEMISIA TABACI (HEMIPTERA: ALEYRODIDAE) ...... 77

ABSTRACT

Physiological factors that might influence ecological dynamics were investigated to better explain the biology of Bemisia tabaci in the desert southwest. Trehalulose, a unique disaccharide only found in unusually large quantities in B. tabaci honeydew, was shown not to be different from sucrose in promoting longevity in honeydew-consuming parasitoids, indicating this insect-modified sugar does not affect the nutritional quality of aleyrodid excreta. Trehalulose is not believed to function as a feeding deterrent to natural enemies. Experiments were designed to examine the effect of temperature on sorbitol and trehalulose production by the whitefly. High performance liquid chromatography analysis of honeydew and whole body extractions revealed a negative relationship between amounts of trehalulose in the honeydew and sorbitol accumulation in the whitefly body, linking these two molecules as important to the nutritional ecology of whiteflies. In another experiment to better understand the dispersal of whiteflies across the landscape, studies of flight over multiple days were conducted to describe the role of prior flight experience on dispersal and migratory flight. Flight performance traits were measured over multiple days of flight to compare two groups of B. tabaci, those trapped moving outside of planted fields with those collected within fields. Trap-caught individuals exhibited flights of significantly shorter duration in a vertical flight chamber. Flights determined to have characteristics of migratory behaviors were initially of longer duration for trap-captured whiteflies than their field-collected counterparts. Over the context of multiple days, however, their longer flights were followed by much shorter flights on subsequent days. Although many insects from both groups were capable of 8 movement on multiple days, almost all of these flights were of a foraging nature.

Foraging flights of short duration would likely not add to dispersal distances, thereby limiting whiteflies to their originating patch. 9

CHAPTER 1.

INTRODUCTION

1.1. Whiteflies as pests

The sweet potato whitefly, Bemisia tabaci (Gennadius), has been recognized as one of the world’s most destructive agricultural pests (Byrne and Bellows, 1991; Oliviera et al., 2001; Ellsworth and Martinez-Carrillo, 2001). Removal of phloem sap though feeding by these insects impacts many crops in a variety of ways, but the most destructive effect is reduction in plant productivity (Lin et al., 1999, Inbar and Gerling, 2008).

Whiteflies extract a surplus of phloem sap and excrete a significant quantity of their intake in the form of honeydew. These liquid wastes contain many of the primary carbohydrates and amino acids found in plants (Byrne and Miller, 1990). This indicates that some of the nutrients required by whiteflies are probably in short supply in the plant phloem and their feeding method allows them to compensate by processing higher quantities of sap. In addition to passing excess nutrients through as honeydew, modification of certain carbohydrates into new components was confirmed, in particular the presence of a sucrose isomer trehalulose (Bates et al. 1990).

There has not been a clear understanding of the role for such modifications to these extracted nutrients, despite the presence of high levels of trehalulose in the honeydew in very few of the species that have been examined (Byrne et al., 2003). Other than the plant feeding hemipterans examined by Byrne et al. (2003), trehalulose, as far as is known, is otherwise limited to bacterial production, e.g., Erwinia, via different enzymes (Salvucci, 2003). The potential for this honeydew product to have some 10 adaptive value in the ecology and physiology of the whitefly is one of the key issues examined.

In addition to plant damage through sap extraction, B. tabaci serves as a vector of numerous plant pathogens. The viruses, primarily begomoviruses and closteroviruses, are transmitted exclusively by whiteflies, and are estimated to be responsible for losses in the billions of dollars (U.S.) around the world (Varma and Malathi, 2003). Damaged crops include essential food staples such as cassava (in Africa), legumes and commodities like cotton (India and Pakistan) (Brown et al. 1995). Understanding the movement of infected whiteflies has helped in characterizing the dynamics of the resulting disease outbreaks in these crops (Colvin et al., 2004). A better understanding of the flight capabilities of whiteflies could inform epidemiology and other research on the spread of these blights.

1.2. Methods of whitefly management

The threats to human agriculture posed by B. tabaci have been recently abated largely through the application of the insect growth regulators bupofezin and pyriproxyfen (Ellsworth and Martinez-Carrillo, 2001; Naranjo, 2004) and the cholornicotinyl imidachloprid (Ellsworth and Martinez-Carrillo, 2001). Concerns have been raised due to observations in the field of reductions in effectiveness to these compounds, (Li et al., 2003). In situations where pesticide use is not always possible, such as greenhouse production, control of B. tabaci and other whiteflies has often been achieved through use of parasitoid wasps (Stansly et al., 2005). Examples of successful 11 control in open fields have not been as evident, however (Hoelmer, 1996, Bellamy et al.,

2004).

1.3. Feeding by parasitoids on whitefly honeydew

Parasitoids of B. tabaci are known to utilize both the insect’s host fluids and honeydew they produce as dietary resources. Both were shown to be important for egg production and longevity in Encarsia formosa Gahan when utilizing honeydew from

Trialeurodes vaporariorum (Westwood) (Burger et al, 2004). The structures of the ovipositor and consequent egg laying behaviors have been shown to be different between two parasitoid genera, Encarsia and Eretmocerus. These differences may reflect a type of phylogenetic constraint (Price, 1997) that limits Eretmocerus from directly piercing the cuticle of the nymphal whitefly and feeding on body liquids.

Searching for separate hosts in order to lay eggs in and acquire carbohydrates for longevity may represent time limitations costs for these small wasps (Eretmocerus).

Using the same host for both resources (Encarsia) was predicted to ease these search costs (Burger et al., 2004). Alternatively, it has been proposed that honeydew-producing insects might evolve mechanisms to modify their excreta in a manner that reduces its advantage to parasitoids (Wackers, 2000). The presence of a modified sugar, represented by trehalulose, presented an opportunity to test for the presence of such an effect. The results of this study comprise Appendix A.

1.4. Nutritional physiology of whitefly feeding 12

As phloem feeders, whiteflies are supplied with an excess of liquid food because of plant turgor pressure. Carbohydrates and amino acids can then be absorbed in the gut and excess nutrients excreted in the honeydew (Byrne and Miller, 1990). Carbohydrates are usually absorbed as monosaccharides, requiring the enzymatic disassociation of polysaccharides (Chapman, 1998).

The involved enzymes, α-glucosidases, or sucrases, usually facilitate the digestion of carbohydrates in the gut. Unfortunately, the common name sucrase is not specifically descriptive to the functional nature of the enzyme as other classes of enzymes are capable of cleaving the glycosidic linkage in sucrose, including β-fructosidases (Terra and Ferriera, 1994). Additionally, the specificity of α-glucosidases are also not limited to sucrose, as many of the enzymes are capable of hydrolyzing another terminal, α-1,4- linked glucose residues from other polysaccharides. The enzyme preference for other disaccharides such as maltose has been shown in mosquitoes (Silva and Terra, 1995). The degree of preference shown by α-glucosidases for sucrose tends to correlate with animals having a high-sucrose diet such as seen in Apis (Terra and Ferriera, 1994). The broader term glycoside hydrolases are the broad group of enzymes that hydrolyze the bonds between carbohydrates as well as among carbohydrates and non-carbohydrates. The common reaction mechanism for glycosidases such as α-glucosidase involves the protonation of the substrate at the anomeric carbon, following the leaving of the non- reactive sugar and the subsequent attack of the carbonium ion by water (Silva and Terra,

1995). 13

Complementary to enzymes that hydrolyze sucrose are enzymes that synthesize or translocate the glycosidic link between two monosaccharides. An example known from both microbes and insects is sucrose isomerase, variously described as isomaltulose synthase, trehalulose or palatinose synthase. The production of trehalulose (α-1,1-) and palatinose (α-1,6-) as isomeric forms of sucrose has been described from several bacterial species including Erwinia rhapontici (Cheetham, 1984), and later the whitefly B. tabaci

(Bates et al. 1990). These sugars have attracted interest in commercial applications due to the non-cariogenic activity as sugar substitutes (see Veronese et al., 1999).

The mechanism of transglycosylation involves the glucose moity as the more strongly bound component of the sucrose constituents, as the reaction proceeds with a double displacement preserving the α-glycoside configuration (Veronese and Perlot,

1999). Additionally, glucose can serve as an inhibitor of the reaction, presumably by binding in the active site and blocking sucrose binding (Veronese and Perlot, 1998). The substrate preference has been tested through incubating sucrose and other monosaccharides with purified isomalutulose (palatinose) synthase from the bacterium

Serratia plymuthica (Veronese and Perlot, 1999). It has been proposed that the fructose moity is bound to the active site through ionic bond interactions of low energy allowing the tautomerization between fructopyranose and fructofuranose necessary for the production of trehalulose (α-D-glucopyranosyl-1,1-D-fructofuranose) (Veronese and

Perlot, 1999)

Although not characterized in three-dimensional form, the size and activity of the enzyme from B. tabaci indicates significant structural and specificity differences despite 14 the conserved sucrose isomerase functionality. The microbial palI gene product has been described as having a 69.9kD molecular mass (Zhang et al., 2002), while the purified whitefly product was 116 kD. The microbial protein is also a soluble enzyme, while the insolubility of whitefly enzyme probably indicates a transmembrane domain (Salvucci,

2003). Other functional differences include an inability of isomaltulose to bind as an inhibitor or substrate, in contrast to the enzymes of microbes known to produce large amounts of palatinose (Veronese and Perlot, 1998). These functional differences indicate the divergence between bacterial and animal enzymes despite the common sucrose isomerase functionality, indicating that more biochemical characterization is needed concerning the structure and function of whitefly sucrose isomerase. To date, the genes responsible have not been described, a detail that would help greatly in elucidating the active site differences between microbial and eukaryotic enzymes that share a common function.

The role of trehalulose in B. tabaci metabolism remains unknown. Speculation has been made about its contribution to reducing osmotic stress in B. tabaci by slowing the rate of disaccharide hydrolysis , but this relies on assumptions of differences in glucosidase activity on trehalulose and sucrose (Salvucci, 1997; Wolfe et al., 1998). This underlying assumption is untested.

Equally interesting are the pathways concerning the monosaccharides glucose and fructose and the biological significance of their subsequent modification. In mammals, the polyol pathway has been well-studied for its role in diabetic peripheral neuropathy

(Oates and David, 2002). The mammalian model pathway involves the reduction of 15 glucose to sorbitol via an NADPH-dependent aldose reductase, followed by an NAD+- dependent oxidation of sorbitol to fructose by sorbitol dehydrogenase (Banfield et al.,

2001). The buildup of sorbitol, along with the consequences of associated co-enzyme buildup, has been implicated as important in the complications from hyperglysolia in diabetes patients (Oates and David, 2002). Although sorbitol accumulation has been observed in response to elevated temperature in whiteflies (Hendrix and Salvucci, 1998) its role in mammals has been postulated to be related to osmoregulatory as well as metabolic functions similar to fructose. The ubiquity of sorbitol has been related to anoxic and reducing atmosphere of early life on earth, indicating a common use for polyol dehydrogenases among primitive enzymes (Jeffery and Jornvall, 1988).

The first step in the mammalian polyol pathway concerns the reduction of glucose to sorbitol with an NADPH-dependent aldose reductase. The important biological impacts of the accumulation of sorbitol have been known for some time from cold tolerant and diapausing insects (Chefurka, 1965; Chino, 1960; reviewed in Harvey,

1962). Sorbitol dehydrogenases are also known to catalyze the “reverse reaction” generating sorbitol from fructose (Oates and David, 2002). While the consequences of this reverse reaction are unknown on diabetes symptomology, it is noteworthy in insects.

One of the first sorbitol dehydrogenases characterized by x-ray crystallography is the so- called ketose reductase from whiteflies (Banfield et al., 2001). In B. tabaci there was shown to be an absence of aldose reductase, but a presence of high rates of ketose reductase activity suggests that fructose is the substrate for sorbitol synthesis (Wolfe et al., 1998). Although both pathways result in sorbitol, the differences between the aldose 16 reductase and ketose reductase enzymes are apparent in their structural distinctions.

Interestingly, the coenzyme used by ketose reductases is the same NADPH (Salvucci et al., 1998) used by aldose reductases, while the structurally similar sorbitol dehydrogenases are NAD(H)-dependent.

Structural models of ketose reductase enzymes such as the one from B. tabaci allow a comparison of the sorbitol dehydrogenase with the entirety of the superfamily of medium-chain dehydrogenase/reductases as well as providing an alternative model organism to study the polyol pathway. Ecologically, the consequences of this “reverse reaction” are significant for the whitefly in that it allows the production of a substance known to be beneficial in thermal and osmotic tolerance from an abundant source that does not compete with dietary resources. Sorbitol is known to function as a cryoprotectant in cold-tolerance and diapause (Storey et al., 1981; Storey and Storey,

1981), but the role in whiteflies has been shown to be related to heat tolerance (Salvucci,

2000; Wolfe et al., 1998). As B. tabaci likely contain both glycoside hydrolases such as

α-glucosidase and sucrose isomerase, as well monosaccharide reducers such as ketose reductase, an interrelationship between the oligosaccharide pathway of isomerization and the polyol pathway is conceivable. Although the location of these enzymes is likely separated within the B. tabaci body, it is possible that the ability to perform transglycosylation on the glycosidic link between glucose and fructose is related to the ability of whiteflies to produce a thermoprotectant directly from fructose. This would prevent competition between the dietary needs of glucose, which would be expected to increase at higher temperatures, and the physiological needs for stability of proteins 17 under high heat exposure provided by sorbitol (Salvucci, 2000). In order to better understand the role temperature plays on the production of trehalulose, an experiment was conducted to allow simultaneously quantification of the amounts of trehalulose and sorbitol produced by B. tabaci experiencing thermal variation. The study was designed to examine the impacts of both short-term increases in temperature over the course of 4 hr as well as longer-term acclimation to temperature after several days. The results of this study are found in Appendix B.

1.5. Whitefly migration

Studies of migration allow biologists to examine the intersection of ecological and physiological forces and how they impact an organism. A partial understanding can be gained of how individual mechanisms can affect population structure through studying migratory behaviors. In a complementary approach, an understanding of how external forces alter the physiological state of an organism to affect movement may also provide important clues to predicting population changes (Drake and Gatehouse, 1995; Hopkins and Dixon, 1997). Migration provides a specific mechanism for gene flow across geographic barriers, establishing adaptive clines and structuring the genetic variation necessary for evolutionary events such as host shifts and speciation (Feder et al., 2003).

While both internal and extrinsic forces can conspire to produce migration, the outcomes can shape populations and explain evolutionary patterns. In the narrowest sense, migration studies can measure two different aspects: individual behavior and ecological 18 outcomes. Studies that integrate these aspects together allow a more realistic understanding of the organism.

Externally, B. tabaci faces an environment with fluctuating temperature extremes and variable habitat resources. Small animals (i.e., less than 1 mg), such as a B. tabaci, face thermal stress when leaving their flight boundary layer. The body temperatures of small insects have been shown to conform to ambient air temperatures due to high surface-to-volume ratios (Harrison and Roberts, 2003). There can be additional heat generated from the muscular activity of flight itself, as well as water loss from ventilation. In larger insects (i.e., greater than 100 mg), such as the bee Centris pallida, studies have shown that flight attributes like wing-beat frequency were negatively correlated with increased air temperatures (Roberts, 2005). In general, the effects of temperature stress in small, flying insects are not well known, but whiteflies have distinctive chemical means to tolerate these thermal changes (see above).

The fitness effects for migratory individuals could be quite large, as leaving a field in the patchy, arid agroecosystem carries a risk of failure to locate quality resources

(Johnson, 1969). Although it could be considered surprising that the existence of migratory creatures is maintained in any population given those risks of failure, the potential risks could also be balanced by access to new habitats. In general, evidence exists indicating the genetic potential for migration is often variable in populations due to the trade-off between these short-term risks and long-term benefits (Roff and Fairbairn,

2007). Although there is some information about the morphometrics of migratory 19 whiteflies (Byrne and Houck, 1990), little is known about the heritability of flight and correlated genetic traits in B. tabaci.

We do not know all the important factors that affect the individual migratory behavior of whiteflies. Some of the characteristics of migratory B. tabaci have been identified, and Blackmer et al. (1995a) estimated that 6% of the population from a greenhouse colony displayed such indicators. B. tabaci has not displayed an oogenesis- flight tradeoff (Johnson, 1969) in field studies examining the link between migration and reproduction (Isaacs and Byrne, 1998; Veenstra and Byrne, 1999). In the laboratory, however, prior experience with egg laying was found to be an important physiological factor explaining oviposition activity (Veenstra and Byrne, 1998). Although host plant experience was found to be a factor that could influence eggload and female weight

(Veenstra and Byrne, 1998), flight ability was not examined under more controlled circumstances in order to explicitly test the influence of those variables on dispersal.

Consequently, the inclusion of egg production was considered in the studies on multiple day movement. Age has also been shown to be important with respect to B. tabaci migration and dispersal, with histological studies of flight musculature indicating that degeneration of the flight muscles and depletion of glycogen began on the fifth day and was markedly changed by the seventh day (Blackmer et al, 1995b). These studies suggested that age would be an important factor in B. tabaci flight, limiting significant movement beyond day 5. All these studies utilized untested individuals in flight measurements over consecutive days. Recent findings (D.N. Byrne, unpublished) showed that whiteflies were capable of flight on multiple days in the laboratory. Over the process 20 of investigating host plant effects on flight initiation, we began testing whiteflies over several days. After observing some individuals engaging in flight over a longer period that initially thought possible, it seemed important to look outside our laboratory populations. In order to better understand the implications for multiple day flight events on migratory movement by B. tabaci, the present study examines field populations in the flight chamber in Appendix C. 21

1.6. Current study summary

B. tabaci presents a unique study system with many opportunities to examine insect biology at the level of the organism within a changing environment. Examining the causes and consequences of honeydew production provides a way to approach multiple trophic levels, since whiteflies using different plant hosts obtain different nutritional resources that can be reflected in their honeydew (Byrne and Miller, 1990). This honeydew represents an important nutritional resource to the parasitoids that attack whiteflies, since it is shown here that the modifications to their honeydew do not negatively impact their natural enemies. Instead, since B. tabaci honeydew was not shown to adversely affect natural enemies, other physiological needs of whiteflies may explain the production of trehalulose. An examination of the relationship between temperature and sorbitol production indicates an alternative consideration for the role of trehalulose. Finally, investigating B. tabaci flight performance over multiple days brings about another method to examine how age and prior flight experience might explain the ecologically relevant phenomenon of migration. Future studies examining the tolerance of heat accumulated during movements between fields could incorporate the negative relationship between trehalulose and sorbitol to help explain the ability of whiteflies to move across the desert Southwest. The subjects of this study express successful strategies for survival in a patchy, arid, agroecosystem, and can be used to link basic and applied research.

1.7. Explanation of dissertation format

The chapters of this dissertation are designed to provide a summary to the main conclusions of my research. The appendices (A-C) consist of either published manuscripts or manuscripts being prepared for submission to peer-reviewed journals. I was responsible for the design, collection and conduct of the trehalulose feeding studies.

The conception of the multiple flight and sorbitol studies was more collaborative, but I was responsible for much of the design and all data analysis for these studies. I am the primary author of each manuscript attached. Appendix A is a study of the effects of feeding on trehalulose for parasitoids, including several B. tabaci parasitoids and the fly parasitoid Nasonia vitripennis. This was published in 2008 Physiological Entomology

(vol. 33, pp. 95-100). Appendix B contains the results of an experiment quantifying the production of trehalulose and sorbitol under thermal variation and has been prepared for submission to Entomologia Experimentalis et Applicata. The final appendix (C) contains a combined field and laboratory study to measure the importance of multiple day flights in contributing to migratory movement by B. tabaci, prepared for submission to Journal of Animal Ecology. 23

CHAPTER 2.

PRESENT STUDY

The methods, results and conclusions for this study are presented in the papers appended to this dissertation. The important findings of these manuscripts are summarized below.

2.1 Parasitoid feeding on trehalulose, a component of B. tabaci honeydew

The paper presented in Appendix A was designed to test the hypothesis that certain carbohydrate components of B. tabaci honeydew, particularly those modified by the insect itself, might be detrimental when eaten by parasitoid natural enemies.

Whiteflies are known to produce trehalulose by isomerization of sucrose, with the product comprising a large proportion of their honeydew, depending on the plant utilized for feeding. Parasitoids of whiteflies were fed diets of either; sucrose, trehalulose or a water solution. While it was shown that certain levels of carbohydrates were essential in prolonging the longevity of these parasitoids, there was no difference detected between the longevities of wasps consuming sucrose or trehalulose. In order to reduce the likelihood that these parasitoids were capable of digesting all sugar components of B. tabaci honeydew due to their parasitoid-host relationship, a parasitoid was examined that would not be expected to have contact with B. tabaci. Nasonia vitripennis is a pteromalid reared commercially for control of the flesh fly (Sarcophaga sp.). None of the parasitoids tested showed any significant difference in longevity on either carbohydrate diet when compared at the same concentration. This indicates that it is highly unlikely that 24 trehalulose functions in the B. tabaci honeydew as a deterrent to parasitoid feeding.

These studies did not exclude the possibility that other compounds in B. tabaci honeydew might have effects on parasitoid longevity or other important variables. The scope of this study only examines trehalulose and there are known to be other insect-modified carbohydrates present in B. tabaci honeydew, e.g. bemisiose (Hendrix & Wei, 1994).

This experiment assayed the digestibility of trehalulose using parasitoids, and found that even at low concentrations there was no difference in survival. This contradicted the suggested role of trehalulose inside the B. tabaci digestive system to prevent osmotic stress from hydrolysis of sucrose through isomerization to a form that could be less digestible.

2.2. Trehalulose production under thermal stress

Another aspect of carbohydrate chemistry that is notable in B. tabaci is production of sorbitol under high temperatures. This polyhydric alcohol has been shown to accumulate in B. tabaci exposed to high temperatures. It appears that sorbitol is rapidly synthesized in whiteflies through a ketose reductase enzyme that uses fructose as a substrate for creation of sorbitol. It seemed plausible to connect sorbitol production with trehalulose production using a common component of both, fructose. Biochemical models of the isomerization of trehalulose in bacteria have modeled fructose as the less tightly bound of the carbohydrates components, fructose and glucose (Veronese & Perlot,

1999). I wanted to measure trehalulose production under varying temperatures, specifically looking at conditions known to produce high levels of sorbitol predicting that 25 sorbitol and trehalulose would be negatively related. The experiment was designed to examine how whiteflies conditioned at high temperatures would compare with those switched from low to high temperatures and vice versa. The results indicate that increased levels of sorbitol are found under conditions that correspond with production of significantly lower levels of trehalulose. The data indicate that sorbitol and trehalulose are negatively related with changing temperature and change over the course of a few hours. An unknown peak was identified at approximately the same retention time as trehalulose under high temperatures. This peak has not yet been identified, but this was the first experiment to examine trehalulose specifically at high temperatures. While the particular mechanisms relating sorbitol and trehalulose are not clear, the relationship is striking and indicates that further research into these pathways could yield valuable information.

2.3. Flight studies

In order to understand how the physiological state of B. tabaci can impact population dynamics, another key biological characteristic must be investigated. The consequence of understanding B. tabaci movement is important scientifically andpractically, particularly in light of their ability immigrate into field crops. The constraints faced by whiteflies while dispersing are therefore important to understand when predicting their spread and impact in agroecosystems. Models of the evolution of resistance by whiteflies to pesticides have recently been developed that utilize estimates about migration by incorporating single movement events by 5 day old whiteflies 26

(Crowder et al., 2008). After identifying the occurrence of multiple day flights, a concern about the extent to which laboratory-generated data reflected behavior in the field was apparent. Flight chamber experiments used animals from the field that represented two groups, those that were trapped moving outside the field and those collected directly from crop plants. Flight performance by field-collected whiteflies could be analyzed by measuring their flight in the vertical flight chamber while taking into account several limitations. Although prior history was not known for animals caught inside fields, by flying them on multiple days within the laboratory I was able to analyze the effects of previous flight experience and age. B. tabaci was found to be capable of movement on successive days, however, I found migratory flights were always limited to a single event.

Total distance and duration decreased as the animals aged. This was found after adjusting for the effects of sex, host, and collection methods. The findings all demonstrate that migratory flight likely does not continue over multiple days for the same individual, and that movement over multiple days is unlikely to dramatically increase current estimates of migration. Additionally, egg production was not different between migratory and non- migratory or foraging fliers, and also was not affected by flight on multiple days. 27

2.4 Conclusions

Applying knowledge concerning B. tabaci biology generated in the laboratory to the field relies on extending what is known about their survival in the arid landscape of the southwestern United States. These studies indicate that parasitoids are probably not affected by honeydew constituents made by the B. tabaci, even though the chemistry differs from typical nectar resources. I confirmed that some carbohydrates are essential for the survival of natural enemies. It is beneficial to know that supplemental carbohydrates are probably not necessary for these agents of biological control, as they can gain longevity through feeding on host-provided resources and extend their usefulness in the field. The role of trehalulose is more likely related to the ability to synthesize sorbitol, as both utilize a common element of fructose. Since trehalulose and sorbitol were shown to be negatively correlated in the presence of high temperature, the pathways for trehalulose and sorbitol synthesis might be affected by similar approaches through novel technology that targets either substance. In particular, the upregulation of trehalulose isomerization could prevent adequate sorbitol synthesis under high temperatures and potentially affect B. tabaci survival. Additionally, the effect of high temperatures on the production of trehalulose should be better characterized, as there may be some implications for the stability of the compound at high temperatures. Finally, flight characteristics over multiple days do not reveal the need for a reconsideration of dispersal distances, especially migratory flight. Given that multiple day flights are possible, migratory flight may be important on an expanded scale across the lifetime of an adult B. tabaci. Movement within the field may be of greater importance, as an animal 28 bringing pathogenic agents into the field could still be capable of spreading disease rapidly within the field during subsequent foraging. The relationship between migration and reproduction are less clear and should be investigated in an experiment that controls for flight experience and reproduction independently. Understanding the heritability of flight performance traits may more accessible with better techniques for handling whiteflies that have arisen in the process of these experiments. The results indicate clearly that migratory flights by whiteflies are not expected on multiple days, reducing the likelihood that whiteflies could be spreading viruses across multiple fields in a single generation. This has important considerations for the understanding of disease spread and the ability to estimate the distances involved in vector movement, as well as B. tabaci dispersal across the landscape in general. This dissertation investigates the links between physiological and ecological dynamics using the whitefly B. tabaci as a subject, demonstrating that this organism has several unique characteristics that allow investigations linking these two approaches. 29

REFERENCES

Banfield, M.J., Salvucci, M.E., Baker, E.N., Smith, C.A. 2001. Crystal structure of the NADP(H)-dependent ketose reductase from Bemisia argentifolii at 2.3 angstrom resolution. Journal Of Molecular Biology, 306, 239-250.

Bates, R.B., Byrne, D.N., Kane, V.V., Miller, W.B., Taylor, S.R. 1990. Nmr Characterization Of Trehalulose From The Excrement Of The Sweet-Potato Whitefly, Bemisia-Tabaci. Carbohydrate Research, 201, 342-345.

Bellamy, D.E., Aplen, M.K. Byrne, D.N. 2004. Impact of Eretmocerus eremicus (Hymenoptera: Aphelinidae) on open-field Bemisia tabaci (Hemiptera: Aleyrodidae) populations. Biological Control, 29, 227-234.

Bellamy, D.E., Byrne, D.N. 2001. Effects of gender and mating status on self-directed dispersal by the whitefly parasitoid Eretmocerus eremicus. Ecological Entomology, 26, 571-577.

Blackmer, J., Byrne, D.N., Tu, Z. 1995. Behavioural, morphological, and physiological traits associated with migratory Bemisia-tabaci (Homoptera, Aleyrodidae). Journal of Insect Behaviour, 8, 251-267.

Brown, J. K., D. R. Frohlich, and R. C. Rosell. 1995. The sweetpotato or silverleaf whiteflies: biotypes of Bemisia tabaci or a species complex? Annual Reviews of Entomology, 40, 511-534.

Burger, J.M.S., Hemerik, L., Lenteren, J.C.V., Vet, L.E.M. 2004. Reproduction now or later: optimal host-handling strategies in the whitefly parasitoid Encarsia formosa. Oikos, 106, 117-130.

Byrne, D.N., Bellows, T.S., 1991. Whitefly biology. Annual Reviews of Entomology, 36, 431-457.

Byrne, D.N., Hendrix, D.L., Williams, L.H. 2003. Presence of trehalulose and other oligosaccharides in hemipteran honeydew, particularly Aleyrodidae. Physiological Entomology, 28, 1-6.

Byrne, D.N., Houck, M., 1990. Morphometric identification of wing polymorphism in Bemisia-tabaci (Homoptera, Aleyrodidae). Annals of the Entomological Society of America, 83, 487-493.

Byrne, D.N., Miller, W.B. 1990. Carbohydrate and amino acid composition of phloem sap and honeydew produced by Bemisia tabaci. Journal of Insect Physiology, 36, 433- 439.

Chapman, R.F. 1998 The Insect: Structure and Function, 4th ed. Cambridge University Press.

Cheetham, P.S.J. 1984. The extraction and mechanism of a novel isomaltulose- synthesizing enzyme from Erwinia-rhapontici. Biochemical Journal, 220, 213- 220.

Chefurka, W. 1965. Some comparative aspects of the metabolism of carbohydrates in insects. Annual Review of Entomology, 10, 345-382.

Chino, H. 1960. Enzymatic pathways in the formation of sorbitol and glycerol in the diapausing egg of the silkworm, Bombyx-mori.1. On the polyol dehydrogenases. Journal of Insect Physiology, 5, 1-15.

Colvin, J., Omongo, C.A., Maruthi, M.N., Otim-Nape, G.W., Thresh, J.M. 2004. Dual begomovirus infections and high Bemisia tabaci populations: two factors driving the spread of a cassava mosaic disease pandemic. Plant Pathology, 53, 577-584.

Crowder, D.W., Ellsworth, P.C., Tabashnik, B.E., Carriére, Y. 2009. Effects of operational and environmental factors on evolution of resistance to pyriproxyfen in the sweetpotato whitefly (Hemiptera: Aleyrodidae). Environmental Entomology, 37, 1514-1524.

Drake, V.A., Gatehouse, A.G. (eds) 1995. Insect migration: tracking resources through space and time. Cambridge University Press.

Ellers, J., Alphen, J.J.M.V., Sevenster, J.G. 1998. A field study of size-fitnes relationships in the parasitoid Asobara tabida. Journal of Animal Ecology, 67, 318-324.

Elliott, C. & Evenden, M., 2009. Factors influencing flight potential of Choristoneura conflictana. Physiological Entomology, 34, 71-78.

Ellsworth, P.C., Martinez-Carrillo, J.L., 2001. IPM for Bemisia tabaci: a case study from North America. Crop Protection, 20(9), 853-869.

Feder, J.L., Berlocher, S.H., Roethele, J.B., Dambroski, H., Smith, J.J., Perry, W.L., Gavrilovic, V., Filchak, K.E., Rull, J., Aluja, M. 2003. Allopatric genetic origins for sympatric host-plant shifts and race formation in Rhagoletis. Proceedings of the National Academy of Sciences of the United States of America, 100, 10314- 10319. 31

Harrison, J.F., Roberts, S.P. 2003. Flight respiration and energetics. Annual Review of Entomology, 62, 179-205.

Harvey, W.R. 1962. Metabolic aspects of insect diapause. Annual Review of Entomology, 7, 57-80.

Hendrix, D.L., Salvucci, M.E. 1998. Polyol metabolism in homopterans at high temperatures: accumulation of mannitol in aphids (Aphididae: Homoptera) and sorbitol in whiteflies (Aleyrodidae: Homoptera). Comparative Biochemistry And Physiology A-Molecular And Integrative Physiology, 120, 487-494.

Hendrix, D.L. Wei, Y. 1994. Bemisiose: an unusual trisaccharide in Bemisia honeydew. Carbohydrate Research, 253, 329-334.

Hoelmer, K.A. 1996. Whitefly parasitoids: can they control field populations of Bemisia? In D Gerling & RT Mayer (eds.) Bemisia: 1995 Taxonomy, Biology, Control and Management, pp. 451-476. Intercept Ltd. Andover, Hants, UK.

Hopkins, G.W., Dixon, A.F.G. 1997. Enemy-free space and the feeding niche of an aphid. Ecological Entomology, 22, 271-174.

Hopkins, G.W., Dixon, A.F.G. 1997. Enemy-free space and the feeding niche of an aphid. Ecological Entomology, 21, 150-154.

Inbar, M., Gerling, D. 2007. Plant-mediated interactions between whiteflies, herbivores, and natural enemies. Annual Review of Entomology, 53, 431-448.

Isaacs, R., Byrne, D.N. 1998. Arial distribution, flight behaviour and eggload: their interrelationship during dispersal but the sweetpotato whitefly. The Journal of Animal Ecology, 67, 741-750.

Jeffery, J., Jornvall, H. 1988. Sorbitol Dehydrogenase. Advances in enzymology and related areas of molecular biology, 61, 47-106.

Johnson, C.G. 1969. Migration and Dispersal of Insects by Flight. London: Methuen and Company, Ltdd.

Johnson, S.J. 1995. Insect migration in North America: synoptic-scale transport in a highly seasonal environment. In: Insect Migration: tracking resources through space and time (ed. V.A. Drake and A.G. Gatehouse), pp 31-66. Cambridge University Press.

Kennedy J.S., Booth, C.O. 1963. Free flight of aphids in the laboratory. Journal of Experimental Biology, 40, 67-85.

Li, A., Dennehy, T., Nichols, R., 2003. Baseline susceptibility and development of resistance to pyriproxyfen in Bemisia argentifolii (Homoptera : Aleyrodidae) in Arizona. Journal of Economic Entomology, 96, 1307-1314.

Lin, T., Schwartz, A., Saranga, Y., 1999. Photosynthesis and productivity of cotton under silverleaf whitefly stress. Crop Science, 39, 174-184.

Loxdale , H.D., Lushai G. 1999. Slaves of the environment: the movement of herbivorous insects in relation to their ecology and genotype. Philosophical Transactions of the Royal Society of London B, 354, 1479-1495.

Naranjo, S., Ellsworth, P. Hagler, J., 2004. Conservation of natural enemies in cotton: role of insect growth regulators in management of Bemisia tabaci. Biological Control, 30, 52-72.

Oates, P.J., David, T. 2002. Polyol pathway and diabetic peripheral neuropathy. International Review of Neurobiology: Academic Press. p 325.

Oliviera, M.R.V., Henneberry, T.J., Anderson, P. 2001. History, current status, and collaborative research projects for Bemisia tabaci. Crop Protection, 20, 709-723.

Price, P.W. 1997. Insect Ecology, 3rd ed. Wiley, New York.

Roberts, S.P., 2005. Effects of flight behaviour on body temperature and kinematics during inter-male mate competition in the solitary desert bee Centris pallida. Physiological Entomology, 30, 151-157.

Roff, D., Fairbairn, D., 2007. The evolution and genetics of migration in insects. Bioscience, 57, 155-164.

Salvucci, M.E. 2000. Sorbitol accumulation in whiteflies: evidence for a role in protecting proteins during heat stress. Journal Of Thermal Biology, 25, 353-361.

Salvucci, M.E. 2003. Distinct sucrose isomerases catalyze trehalulose synthesis in whiteflies, Bemisia argentifolii, and Erwinia rhapontici. Comparative Biochemistry And Physiology B-Biochemistry & Molecular Biology, 135, 385- 395.

Salvucci, M.E., Wolfe, G.R., Hendrix, D.L. 1998. Purification and properties of an unusual NADPH-dependent ketose reductase from the silverleaf whitefly. Insect Biochemistry And Molecular Biology, 28,357-363. 33

Silva, C.P., Terra, W.R. 1995. An alpha-glucosidase from perimicrovillar membranes of Dysdercus-peruvianus (Hemiptera, Pyrrhocoridae) midgut cells - purification and properties. Insect Biochemistry And Molecular Biology, 25, 487-494.

Stansly, P., Calvo, F., Urbaneja, A., 2005. Augmentative biological control of Bemisia tabaci biotype "Q" in Spanish greenhouse pepper production using Eretmocerus spp. Crop Protection, 24, 829-835.

Storey, K.B., Baust, J.G., Storey, J.M. 1981. Intermediary metabolism during low- temperature acclimation in the overwintering gall fly larva, Eurosta-Solidaginis. Journal Of Comparative Physiology, 144, 183-190.

Storey, K.B., Storey, J.M. 1981. Biochemical strategies of overwintering in the gall fly larva, Eurosta-Solidaginis - effect of low-temperature acclimation on the activities of enzymes of intermediary metabolism. Journal Of Comparative Physiology, 144, 191-199.

Tenhumberg, B., G. Siekmann, M. A. Keller. 2006. Optimal time allocation in parasitic wasps searching for hosts and food. Oikos, 113, 121-131.

Terra, W.R., Ferreira, C. 1994. Insect digestive enzymes - properties, compartmentalization and function. Comparative Biochemistry And Physiology B- Biochemistry & Molecular Biology, 109, 1-62.

Varma, A. & Malathi, V., 2003. Emerging geminivirus problems: A serious threat to crop production. Annals of Applied Biology, 142(2), 145-164.

Veenstra, K.H., Bryne, D.N. 1998. Effect of starvation and oviposition activity on the reproductive physiology of the sweet potato whitefly, Bemisia tabaci. Physiological Entomology, 23, 62-68.

Veenstra, K.H., Byrne, D.N. 1999. Does dispersal affect the reproductive physiology of the sweet potato whitefly, Bemisia tabaci? Physiological Entomology, 24, 72-75.

Veronese, T, Perlot, P. 1998. Proposition for the biochemical mechanism occurring in the sucrose isomerase active site. Febs Letters, 441, 348-352.

Veronese, T, Perlot, P. 1999. Mechanism of sucrose conversion by the sucrose isomerase of Serratia plymuthica ATCC 15928. Enzyme And Microbial Technology, 24, 263-269.

Wäckers, F.L. 2000. Do oligosaccharides reduce the suitability of honeydew for predators and parasitoids? A further facet to the function of insect-synthesized honeydew sugars. Oikos, 90, 197-201.

Wolfe, G.R., Hendrix, D.L., Salvucci, M.E. 1998. A thermoprotective role for sorbitol in the silverleaf whitefly, Bemisia argentifolii. Journal Of Insect Physiology, 44, 597- 603.

Zhang, D.H., Li, X.Z., Zhang, L.H. 2002. Isomaltulose synthase from Klebsiella sp strain LX3: Gene cloning and characterization and engineering of thermostability. Applied And Environmental Microbiology, 68, 2676-2682. 35

APPENDIX A THE LONGEVITY EFFECTS OF TREHALULOSE FEEDING BY PARASITOIDS

Reprint from Physiological Entomology (2008) 33, 95-100

Permissions to reprint from publisher:

APPENDIX B

EXAMINING THE ROLE OF TREHALULOSE AND SORBITOL IN

THERMAL TOLERANCE BY BEMISIA TABACI 44

Examining the role of trehalulose and sorbitol in

thermal tolerance by Bemisia tabaci

Jesse A. Hardin*1, Michael E. Salvucci2, and David N. Byrne1

1Department of Entomology, 410 Forbes Bldg., University of Arizona, Tucson, AZ,

85721; USA; 2USDA-ARS Arid-Lands Agricultural Research Center, 21881 North

Cardon Lane, Maricopa, AZ, 85239, USA

*Author for correspondence: Tel: +1-520-360-3094; Fax +1-520-621-1150; E-mail: [email protected]

Running headline: Trehalulose, sorbitol, and Bemisia thermal tolerance

For Submission to: Entomologia Experimentalis et Applicata 45

Abstract

Studies have shown that Bemisia tabaci produces the polyhydric alcohol sorbitol when under heat stress, for them > 30ºC. Contrastingly, this polyhydric alcohol is known to function as a cryoprotectant in many insects. The presence of high rates of ketose reductase activity has indicated that fructose is the substrate for sorbitol synthesis.

Fructose is also a component of the sucrose isomer trehalulose [α-D-glucopyranosl-(1,1)-

β-D-fructofuranoside], which constitutes a major component of B. tabaci honeydew. It was predicted that trehalulose levels in honeydew should be reduced under high temperature conditions when fructose is shifted to sorbitol assembly. Experiments were designed to examine the effect of temperature on sorbitol and trehalulose production by this insect. HPLC analysis of honeydew and B. tabaci whole body extractions revealed a negative relationship between trehalulose production in the honeydew and sorbitol accumulation in the whitefly body.

Keywords: trehalulose; sorbitol; fructose; thermal; honeydew

Introduction

Bemisia tabaci (Gennadius), the sweet potato whitefly, is a pest in agricultural systems worldwide for several well-documented reasons (Gullan and Martin, 2003). For these concerns, and because of general biological interest, B. tabaci is worthy of our attention.

Allthough well-studied as a pest insect, it is worth noting that B. tabaci is unusual among members of Aleyrodidae because it is one of only two species in this family that is truly 46 polyphagous while being associated with primarily herbaceous plant hosts. This is precautionary because results for this insect may not be representative of other aleyrodid species.

One of the challenges for all poikilotherms is the threat of the appearance of high temperatures that exceed survivable levels. This is particularly true of B. tabaci, a very small insect (approximately 35 µg [Byrne et al., 1996]) with a relatively large surface/volume ratio. Consequently, B. tabaci must defend itself against extremely high environmental temperatures such as those found in the southwestern United States where

B. tabaci is common. Ambient air temperatures often approach 48º C during summer months. Damage at these temperatures can be immediate and often occurs at the cellular level where compounds such as DNA, lipids, and other macromolecules are destabilized.

Damage may in turn affect life histories (Denlinger and Yocum, 1999).

One response by B. tabaci under heat stress is to produce sorbitol (Wolfe et al.,

1998; Salvucci, 2000). The presence of this polyhydric alcohol can reduce protein aggregation (Salvucci, 2000) and may provide stability, similar the to effects known for sorbitol in other insects at low temperatures (Chino, 1957; Storey and Storey, 1983; Lee,

1991

The presence of high levels of ketose reductase in insects indicates that fructose is the substrate for sorbitol synthesis. Fructose is also a component of the sucrose isomer trehalulose [α-D-glucopyranosl-(1,1)-β-D-fructofuranoside], which is as a major component of B. tabaci honeydew (Byrne et al., 2003). Since these two compounds 47 compete for the same substrate, it is predicted that trehalulose should be reduced under high temperature conditions because fructose is being made available for sorbitol production. Experiments described here were designed to examine the effect of temperature on sorbitol and trehalulose levels.

Depending on plant host (Byrne and Miller, 1990), feeding by B. tabaci on phloem sap supplies these whiteflies with a carbohydrate diet consisting, at least in cotton, primarily of sucrose [α-D-glucopyranosl-(1,2)-β-D-fructofuranoside] (Tarczynski et al., 1992), which is either hydrolyzed to glucose and fructose or can be isomerized to trehalulose [α-D-glucopyranosl-(1,1)-β-D-fructofuranoside] (Salvucci et al., 1997).

Trehalulose was thought by Wolfe et al. (1998) to aid in the excretion of excess carbohydrates. Another study contradicted the notion of a difference in hydrolysis between these two sugars as the digestibility of trehalulose was shown to be similar to sucrose in parasitoids provided diets of either disaccharide (Hardin et al., 2008).

It is important to consider the role of fructose as a substrate in the synthesis of both sorbitol and trehalulose. If both molecules use the same substrate, a negative relationship is likely to exist between the two compounds. Having knowledge of the role of sorbitol in heat stressed whiteflies, it was predicted that trehalulose production in the honeydew should be reduced under such conditions. To better explore this relationship, experiments were designed to examine the effect of temperature on the carbohydrate constituents of honeydew, thought to represent gut contents, and whole body extracts, thought to represent hemocoel contents. In addition, a population was acclimated to 48 higher temperatures to examine the impacts of rapidly changing thermal regimes on the carbohydrate composition of these animals.

Materials and Methods

Chemicals

Reagents were purchased from Sigma Chemical Co.® (St. Louis, MO) unless otherwise indicated.

Insect and plant material

B. tabaci were reared on cotton, Gossypium hirsutum L., in laboratory growth chamber colonies at the University of Arizona, Tucson, AZ, under a 16:8 h light regimen.

For plants and B. tabaci in the high temperature preconditioned experiment, environmental chamber temperatures were programmed to increase from 27 to 40º C within 6 h of light exposure. Temperatures were maintained at 40º C for 4 h and decreased to 27º C for the duration of the daily cycle.

Thermal variation experiments

B. tabaci were transferred via gentle aspiration to cotton plants in environmental chambers for a preconditioning period at either 26 or 40º C for 4 h. Animals were contained within modified clip cages constructed from 35x10mm Petri dishes using non- medicated callus cushions (Walgreen Co.®, Deerfield, IL) as a barrier to minimize escape. Approximately 1.75cm2 of leaf area was exposed to the feeding adults. Petri dish lids served as collection surfaces for honeydew. Approximately 10-20 B. tabaci adults were gently aspirated from colony plants and briefly chilled over ice to allow transfer to 49 clip cages. Four temperature treatments were utilized: (1) a low temperature (26-26ºC for preconditioning and experimental treatment), (2) an increasing temperature (26-40ºC), (3) a high temperature (40-40ºC) , and (4) a decreasing temperature (from 40-26ºC) treatment that was preconditioned at 40ºC . Following the thermal exposure periods of 4 h, clip cages were carefully opened and B. tabaci were collected by aspiration into glass tubes. Petri dish lids containing honeydew deposits were sealed with parafilm and B. tabaci in tubes were immediately stored at -80ºC until whole-body extractions were made.

Extraction and analysis of carbohydrates

Honeydew was resuspended in Petri dish lids with ultra-pure water and placed under vacuum centrifugation to allow evaporation. Samples were stored at -80º C and resuspended in H2O for HPLC analysis. Frozen adult B. tabaci were extracted into 80%

(v/v) ethanol at 80º C. The liquid phase was removed after 30min, and the solid phase re- extracted for another 30 min. These extractions were pooled, filtered through activated charcoal and lyophilized (Wolfe et al., 1998). The extracts were dissolved in ultra-pure water and carbohydrates were separated by anion-exchange HPLC and detected by pulsed amperometry (Wolfe et al., 1998). Peak retention times of carbohydrates were detected using Chromeleon® 7 software package (Dionex Corp.®, Sunnyvale, CA, USA) comparing samples with standards. Quantities of identified carbohydrate peaks were estimated through Chromeleon® by comparison of samples with best-fit regression lines against standards injected in 2, 5, 10, 20 and 25µl volumes. Any missing values were 50 back-calculated from regression lines. Results represent means (± SE) of replicate clip cages for each temperature treatment, which contained 5-19 adult B. tabaci per cage

(Table 1).

Results

Honeydew high performance liquid chromatograph (HPLC) analysis

Analysis of HPLC trace graphs from honeydew is based on identified peak retention times corresponding to standardized samples. Unless indicated, the following comparisons between compounds within a treatment group are based solely on data from

HPLC trace graphs.

For the low temperature treatments, the major peaks observed corresponded with trehalulose standards, which indicated greater amounts of this substance in the honeydew than levels of free glucose and fructose combined. There was a small amount of sucrose detected in some samples (Fig. 1).

The increasing temperature treatments indicated the presence of trehalulose and another unknown peak very closely aligned in retention time. This was an unexpected finding and is discussed below. Again, glucose and fructose were present, as were small amounts of sucrose (trace graph not shown).

The high temperature treatment samples presented major peaks corresponding to glucose and fructose in the honeydew, with both of these monosaccharides appearing in greater quantity than trehalulose. Also found in these samples was a double peak at the approximate retention time of our standard trehalulose (Fig. 2,). 51

Major peaks corresponding with the retention times of glucose and fructose standards were present in the decreasing temperature treatments. As observed in the high temperature treatments, both of these monosaccharides were estimated to be greater in quantity than trehalulose. Similarly, there was a double peak near the trehalulose standard retention time (graph not shown).

Bemisia tabaci whole body extract HPLC analysis

The whole body extraction samples for insects at a low temperature also displayed major peaks of glucose and fructose. Trehalulose was present in small amounts (Fig. 3).

A relatively high amount of sucrose (6.43 nMol) was found in one sample, but this compound was undetected in the other samples at this temperature. Small amounts of sorbitol were also found, but in a lower relative quantity than fructose.

The increasing temperature samples indicated the presences of large quantities of glucose and sorbitol. Quantities of fructose and trehalulose were estimated to be present in relatively equivalent amounts.

The major peaks of sorbitol and glucose were prevalent in the higher temperature samples. Sorbitol was present at higher levels than glucose. Fructose and trehalulose were both present in these samples, but at lower relative quantities compared to either sorbitol or glucose (Fig. 4).

The decreasing temperature samples were similar in composition to the other samples from groups experiencing high temperature exposure. These were observed to 52 have the major peaks of glucose and sorbitol, but with slightly lower amounts of sorbitol than glucose in all samples. Peaks corresponding to fructose and trehalulose were both present, but in lower relative quantities.

Temperature exposure comparisons

Comparisons between temperature groups from both the honeydew and the whole insect samples were performed using analysis of variance (one-way ANOVA) models that utilize transformed data (type indicated where appropriate) to better meet the assumptions of equal variances between treatment and normal distributions of the data.

Differences between means of temperature treatments were compared using Tukey-

Kramer HSD tests with α = 0.05. The corresponding paired treatment comparisons that were not significantly different are indicated by sharing a letter in graphs found in figures.

Honeydew

There was a significant difference (F3, 10 = 45.15, p. < 0.0001) between the mean quantity (data transformed as [nMol/insect]1/3) of all putative trehalulose peaks across the four temperature treatments, in such a manner that any group experiencing high (40ºC) temperature exposure made less trehalulose than the low (26-26ºC) temperature treatment 53

(Fig. 5). These data combine the estimated quantities of trehalulose for both peaks, even though these double peaks were only found in the higher temperature treatments, thereby making this a liberal estimate of the quantity of trehalulose production. If both peaks can be interpreted to represent two different constituent molecules, then the estimated quantities of trehalulose in the high temperature treatments would be lower.

There was also a significant difference (F3, 10 = 6.05, p. = 0.0129) between the mean quantity (nMol/insect) of glucose across the four temperature treatments. In these comparisons, the groups of animals experiencing a consistently high temperature treatment were found to have a significantly greater amount of glucose in their honeydew than those insects experiencing low temperature treatment (Fig. 6).

In contrast, there was a statistically significant difference in the mean quantity

(nMol/insect) of sucrose between the temperature treatments (F3, 10 = 5.08, p. = 0.0216).

A Tukey-Kramer HSD test (α = 0.05) revealed a significantly greater quantity of sucrose produced in the honeydew of the consistently low (26-26ºC) temperature treatment when compared to the increased (26-40ºC) temperature treatment (Fig. 7).

There was not a significant difference between the treatments in terms of the mean quantities of sorbitol in the honeydew samples (F3, 6 = 2.45, p. = 0.1611). There was not enough sorbitol detected in several of the samples to estimate the quantity, thus reducing the power of this comparison. There was also no significant difference in the amount of fructose (F3, 10 = 3.22, p. = 0.0701) across all treatments (data not shown).

Whitefly while body extracts

It was determined that levels of sorbitol in B. tabaci varied significantly (F3, 10 =

9.13, p. < 0.01) in response to the temperature (data transformed as (nMol/insect) 1/2). The trend is indicative of evidence of a significant difference between the consistently low

(26-26ºC) and high (40-40º) temperature treatments (Tukey-Kramer HSD test, α = 0.05,

Fig. 8).

In contrast, across all temperature treatments for the whole body extractions, there was no evidence of a difference for either mean quantities of glucose (F3, 10 = 0.79, p. =

0.53) or fructose (F3, 10 = 0.84, p. = 0.50).

Traces of trehalulose (mean = 0.01 ± 0.01nMol/insect, n = 14) were observed in the whole body extractions, while peaks corresponding to sucrose standards were not found in enough of the samples to allow comparison.

Discussion

When B. tabaci were preconditioned at 26ºC, and maintained at that temperature, there were significantly higher levels of trehalulose in honeydew than was found when insects were exposed to 40ºC. Also, when B. tabaci that experienced 40ºC in any of the treatments were examined there were significantly higher levels of sorbitol in their bodies than in those of insects experiencing only the lower temperature. Importantly, when B. tabaci were preconditioned at the lower temperature, and then exposed to 40ºC over 4-h- 55 period, significantly lower quantities of trehalulose were found in their honeydew. There was also a trend toward intermediate quantities of sorbitol in their bodies.

Previously, quantities of sorbitol have been shown to rise in B. tabaci (at the time described as B. argentifolii) in response to increases in temperature (Wolfe et al. 1998).

Increased sorbitol resulted in a decrease in protein aggregation in the heat-stressed B. tabaci (Salvucci 2000). The presence of high activity rates of an NADPH-dependant ketose reductase enzyme has been suggested as evidence of fructose acting as the substrate for sorbitol synthesis in B. tabaci (Wolfe et al. 1998). This is also indicated by structural differences between these same whitefly enzymes and other members of the medium-chain dehydrogenase/reductase family (Wolfe et al. 1999). Assuming the molecular pathway for sorbitol synthesis proceeds via a fructose substrate, one would expect high demand for fructose inside the whitefly body to mediate elevated temperature exposure through production of sorbitol.

In microbial models of trehalulose isomerization it has been proposed that fructose is less tightly bound to the active site of sucrose isomerase than glucose

(Veronese & Perlot 1999). The importance of free fructose in both trehalulose isomerization, which takes place in the insect gut, and as a substrate for its’ reduction to sorbitol in the hemocoel provides a possible mechanism to link the results from both locations at varying temperatures. Finding significantly greater amounts of trehalulose in the honeydew of B. tabaci under the same temperature treatments that correspond to significantly lower levels of sorbitol in whole body extracts (assumed to include a greater 56 fraction of hemocoel contents) is strong evidence favoring a temperature-dependant correlation in the production of these two molecules. If fructose is in high demand inside the hemocoel as a substrate for sorbitol, one might expect to find lower than normal levels within the gut and resulting low levels in the honeydew. While not significantly different in all cases, the one treatment where B. tabaci was acclimated at 40º C trehalulose levels were lowest and sorbitol was highest. This suggests that under the same conditions where fructose is depleted via reduction to sorbitol, glucosidase activity in the gut might also be depleting available sucrose. As a consequence, the substrate (fructose) for the formation of trehalulose would be limited. As expected under this model, free glucose in the honeydew was observed in the honeydew of those animals facing high thermal exposure. Although carbon metabolism is generally increased under high temperature, the significant increase in glucose in the honeydew indicates that this substance is not metabolised in the same manner as fructose in order for sorbitol levels to increase. Therefore, higher metabolic rates at higher temperatures could account for increased quantities of glucose relative to glucose, but it was observed that sucrose levels are not higher in the honeydew of experimental animals. Increased glucosidase activity could create proportionally higher levels of glucose in those animals that have both higher feeding rates and high reduction rates of fructose, which would explain the lack of trehalulose product in the honeydew. Neither glucosidase activity, nor the levels of any other enzyme activity, were measured in this study, however this approach would be suggested as a method to better elucidate the role of carbohydrate metabolism under thermal variation. 57

There are several interesting findings that also would benefit from more detailed examinations of these pathways. Sorbitol was found in the honeydew of insects from all temperature treatments. The small amounts of sorbitol (< 0.05nMol/insect) detected in the honeydew samples seems to be likely due to hemolymph contaminants from accidentally crushed insects on the inside of the experimental cages and not to represent the presence of sorbitol in the honeydew because this component has rarely been observed in honeydew samples (Hendrix & Salvucci, 1998).

The double peak observed at the approximate retention time of trehalulose has not been described and was only observed in samples that were subjected to 40ºC during any of the temperature regimens. Interestingly, these peaks were consistently observed in all of these higher temperature treatments and not found in samples from our low (26-26ºC) treatments. It is possible that both peaks are present in all samples, but are only observed when levels of trehalulose are relatively small and resolve separately in terms of retention times. Further identification of the compounds detected at this retention time would be beneficial to a better understanding of this phenomenon, but currently no explanation for this finding exists. For the purpose of estimating total trehalulose production, these analyses combine the estimated quantities of the compounds for both peaks making these estimates more likely to indicate the maximum quantity of trehalulose possible in the samples. By combining these two peaks one can still perform analyses comparing all the temperature treatments without knowing the reason for the difference. If the similarity in retention time is assumed to indicate similar but slightly different structures, it could be postulated that the trehalulose isomerase responsible for translocation of the α-glycosidic 58 link in this molecule is unstable at higher temperatures. This would indicate an even more striking difference in honeydew profiles between insects acclimated to lower temperatures and those that experienced variation in environmental temperatures.

Acknowledgements 59

References

Banfield MJ, Salvucci ME, Baker EN, Smith CA (2001) Crystal structure of the

NADP(H)-dependent ketose reductase from Bemisia argentifolii at 2.3 angstrom

resolution. Journal of Molecular Biology 306: 239-250.

Bates RB, Byrne DN, Kane VV, Miller WB, Taylor SR (1990) Nmr characterization of

trehalulose from the excrement of the sweet potato whitefly, Bemisia tabaci.

Carbohydrate Research 201: 342-345.

Chapman RF (1998) The Insect: Structure and Function, 4th ed. Cambridge University

Press.

Cheetham PSJ (1984) The Extraction And Mechanism Of A Novel Isomaltulose-

Synthesizing Enzyme From Erwinia rhapontici. Biochemical Journal 220: 213-

220.

Chefurka W (1965) Some Comparative Aspects of the Metabolism of Carbohydrates in

Insects. Annual Review of Entomology 10: 345-382.

Chino H (1960) Enzymatic Pathways In The Formation Of Sorbitol And Glycerol In The

Diapausing Egg Of The Silkworm, Bombyx-Mori.1. On The Polyol

Dehydrogenases. Journal Of Insect Physiology 5: 1-15.

Denlinger DL, Yocum GD (1999) Physiology of heat sensitivity. In: Temperature

Sensitivity in Insects and Application in Integrated Pest Management (eds. G.J. 60

Hallman and D.L. Denlinger), p. 7-53. Westview Press, Boulder, Colorado.

Ellsworth, P.C., Martinez-Carrillo, J.L., 2001. IPM for Bemisia tabaci: a case

study from North America. Crop Protection, 20, 853-869.

Gullan PJ, Martin JH (2003) Sternorrhyncha (jumping plant lice, B. tabaci, aphids, and

scale insects). pp 1079-1089 In (eds. V.H. Resh, R.T. Cardé) Encyclopedia of

Insects. Academic Press, San Diego, CA 1266 pp.

Hardin JH, Asplen MK, Byrne DN (2008) The longevity effects of trehalulose feeding by

parasitoids. Physiological Ecology 33: 95-100.

Harvey WR (1962) Metabolic Aspects of Insect Diapause. Annual Review of Entomology

7: 57-80.

Heinrich B (1993) The Hot Blooded Insects. Harvard University Free Press, Cambridge,

Massachusetts.

Hendrix DL, Salvucci ME (1998) Polyol metabolism in homopterans at high

temperatures: accumulation of mannitol in aphids (Aphididae: Homoptera) and

sorbitol in whiteflies (Aleyrodidae: Homoptera). Comparative Biochemistry And

Physiology A-Molecular And Integrative Physiology 120: 487-494.

Ouedraogo RM, Cusson M, Goettel MS, Brodeur J (2003) Inhibition of fungal growth in

thermoregulating locusts, Locusta migratoria, infected by the fungus Metarhizium

anisopliae var acridum. Journal of Invertebrate Pathology 82: 103-109. 61

Jeffery J, Jornvall H (1988) Sorbitol Dehydrogenase. Advances in Enzymology and

Related Areas of Molecular Biology 61: 47-106.

Oates PJ, David T (2002) Polyol pathway and diabetic peripheral neuropathy.

International Review of Neurobiology: Academic Press. p 325.

Salvucci ME (2000) Sorbitol accumulation in whiteflies: evidence for a role in protecting

proteins during heat stress. Journal Of Thermal Biology 25: 353-361.

Salvucci ME (2003) Distinct sucrose isomerases catalyze trehalulose synthesis in

whiteflies, Bemisia argentifolii, and Erwinia rhapontici. Comparative

Biochemistry And Physiology B-Biochemistry & Molecular Biology 135: 385-

395.

Salvucci ME, Wolfe GR, Hendrix DL (1998) Purification and properties of an unusual

NADPH-dependent ketose reductase from the silverleaf whitefly. Insect

Biochemistry And Molecular Biology 28: 357-363.

Silva CP, Terra WR (1995) An Alpha-Glucosidase From Perimicrovillar Membranes Of

Dysdercus-Peruvianus (Hemiptera, Pyrrhocoridae) Midgut Cells - Purification

And Properties. Insect Biochemistry And Molecular Biology 25: 487-494.

Storey JM, Storey KB (1983) Regulation of cryoprotectant metabolism in the

overwintering gall fly larva, Eurosta solidaginis – temperature control of glycerol

and sorbitol levels. Journal of Comparative Physiology 149: 495-502. 62

Storey KB, Baust JG, Storey JM (1981) Intermediary Metabolism During Low-

Temperature Acclimation In The Overwintering Gall Fly Larva, Eurosta

solidaginis. Journal Of Comparative Physiology 144: 183-190.

Storey KB, Storey JM (1981) Biochemical Strategies Of Overwintering In The Gall Fly

Larva, Eurosta solidaginis - Effect Of Low-Temperature Acclimation on the

Activities of Enzymes of Intermediary Metabolism. Journal Of Comparative

Physiology 144: 191-199.

Storey JM, Storey KB (2004) Cold Hardiness and Freeze Tolerance. In: Functional

Metabolism: Regulation and Adaptation (ed: K.B. Storey), p. 473-503. Wiley-Liss,

Inc., Hoboken, New Jersey.

Tarczynski MC, Byrne DN, Miller WB (1992) High performance liquid chromatography

analysis of carbohydrates of cotton phloem sap and of honeydew produced by

Bemisia tabaci feeding on cotton. Plant Physiology 98: 753-756.

Terra WR, Ferreira C (1994) Insect Digestive Enzymes - Properties,

Compartmentalization And Function. Comparative Biochemistry And Physiology

B-Biochemistry & Molecular Biology 109: 1-62.

Veronese T, Perlot P (1998) Proposition for the biochemical mechanism occurring in the

sucrose isomerase active site. FEBS Letters 441: 348-352. 63

Veronese T, Perlot P (1999) Mechanism of sucrose conversion by the sucrose isomerase

of Serratia plymuthica ATCC 15928. Enzyme And Microbial Technology 24:

263-269.

Wolfe GR, Hendrix DL, Salvucci ME (1998) A thermoprotective role for sorbitol in the

silverleaf whitefly, Bemisia argentifolii. Journal of Insect Physiology 44: 597-603.

Zhang DH, Li XZ, Zhang LH (2002) Isomaltulose synthase from Klebsiella sp strain LX3:

Gene cloning and characterization and engineering of thermostability. Applied and

Environmental Microbiology 68: 2676-2682.

Table Legends

Table 1. Numbers of animals used in replicates of temperature treatments. Numbers differ between cages due to escape or mortality from handling (see methods).

Thermal treatment # of cages # of adults in each cage low (26-26ºC) 3 5, 13, 13 increasing (26-40ºC) 3 4, 7, 19 high (40-40ºC) 4 5, 6, 18, 3 decreasing (40-26ºC) 4 12, 14, 17, 10

Figure Legends

Figure 1. Identification of sugars in the honeydew samples taken from B. tabaci, collected from whiteflies in the low (26-26ºC) temperature treatment after 4 hours at

26ºC. Sugars have been identified based on comparison to retention times of injected standards and amounts estimated by comparing areas of the appropriate peaks with quantified standards. Detector response is reported in nanocoulombs. 65

Fig 1

Figure 2. Identification of sugars in the honeydew samples taken from B. tabaci, collected from whiteflies in the high (40-40ºC) temperature treatment after 4 hours at

40ºC. A trehalulose peak has been identified at 9.484 min. with a second unidentified peak at 9.584 min.

Fig 2

Figure 3. Identification of polyols and sugars in extracts of B. tabaci, collected from 5 whiteflies in the low (26-26ºC) temperature treatment after 4 hours at 26ºC. Sorbitol has been identified at a peak retention time of 4.48 min. with a calculated quantity of 0.343 nMol.

Fig 3

Figure 4. . Identification of polyols and sugars in extracts of B. tabaci, collected from 6 whiteflies in the high (40-40ºC) temperature treatment after 4 hours at 40ºC. Sorbitol has been identified at a peak retention time of 4.45 min. with a quantity of 4.057nMol.

Fig 4

Figure 5. Mean quantities of all putative trehalulose peaks in honeydew identified and calculated from anion-exchanged HPLC for all temperature treatments. Treatments are indicated in both precondition and exposure temperatures (ºC). Analysis of variance (one- way ANOVA) model utilizing transformed data (nMol/insects)1/3 results in a significant difference between treatment levels F3, 10 = 45.15, p. < 0.0001. Differences between means of temperature treatments compared using Tukey-Kramer HSD test with alpha =

0.05, and the corresponding paired treatment comparisons that were not significantly different are indicated by sharing a letter (e.g., B). 70

Fig 5

Figure 6. Mean quantities of glucose in honeydew identified and calculated from anion- exchanged HPLC for all temperature treatments. Analysis of variance (one-way

ANOVA) model (nMol/insects) results in a significant difference between treatment levels F3, 10 = 6.05, p. = 0.0129. Differences between means of temperature treatments compared using Tukey-Kramer HSD test with alpha = 0.05, and the corresponding paired treatment comparisons that were not significantly different are indicated by sharing a letter (e.g., B). 72

Fig 6

Figure 7. Mean quantities of sucrose in honeydew identified and calculated from anion- exchanged HPLC for all temperature treatments. Analysis of variance (one-way

ANOVA) model (nMol/insects) results in a significant difference between treatment levels F3,10 = 5.08, p. = 0.0216. Differences between means of temperature treatments compared using Tukey-Kramer HSD test with alpha = 0.05, and the corresponding paired treatment comparisons that were not significantly different are indicated by sharing a letter (e.g., A or B). 74

Fig 7

Figure 8. Mean quantities of sorbitol in honeydew identified and calculated from anion- exchanged HPLC for all temperature treatments. Analysis of variance (one-way

ANOVA) model (nMol/insects) results in a significant difference between treatment levels F3, 10 = 9.13, p. < 0.01. Differences between means of temperature treatments compared using Tukey-Kramer HSD test with alpha = 0.05, and the corresponding paired treatment comparisons that were not significantly different are indicated by sharing a letter (e.g., A or B). 76

Fig 8

APPENDIX C

MULTIPLE DAY FLIGHTS BY BEMISIA TABACI (HEMIPTERA: ALEYRODIDAE) 78

Multiple day flights by Bemisia tabaci (Hemiptera: Aleyrodidae)

Jesse A. Hardin* and David N. Byrne

Department of Entomology, 410 Forbes Bldg., University of Arizona, Tucson, AZ, 85721,

USA;

*Author for correspondence: Tel: +1-520-360-3094; Fax +1-520-621-1150; E-mail: [email protected]

Running headline: Multiple day flights by Bemisia tabaci

For Submission to: Journal of Animal Ecology 79

Summary

Considerations of prior experience with migration are important when examining dispersal in animals. A study designed to measure flight performance by the whitefly

Bemisia tabaci across multiple days was used to investigate two distinct groups from the same population; B. tabaci caught in traps placed at least 100 m from agricultural fields

(indicating they had already dispersed) and those collected from those same fields. Trap- caught individuals exhibited flights of significantly shorter duration in the laboratory using a vertical flight chamber. It was also possible to determine in the chamber if individuals from these two sample groups were engaging in migratory behavior by measuring responses to vegetative cues (a criterion traditionally used to identify insect migration). Migratory flights on day 1 were of longer duration for trap-captured whiteflies, however, these were followed by much shorter flights on subsequent days.

Although many insects from both groups were capable of movement on multiple days, almost all of these flights were of a foraging nature, i.e., they consistently responded to vegetative cues. Foraging flights of short duration would likely not greatly add to dispersal distances. It was also determined that migratory status was not found to be a significant factor in the variation of egg production.

Keywords: dispersal; vertical flight chamber; oogenesis-flight syndrome; repeated measures

Introduction

Dispersal is an important mechanism for change in insect population dynamics, yet this phenomenon remains one of the least understood aspects of insect ecology

(Wiens, 2001). Prior experience has been shown to affect insect flight initiation, duration, and cessation (Senger et al., 2007), as well as stored energy resources (Elliot & Evenden,

2009). These internal factors should be considered equally important in shaping the outcomes of dispersal, as opposed to such extrinsic factors as resource distribution

(Roitberg & Mangel, 1997). Of specific interest in this study is the effect of prior flight experience in the context of subsequent dispersal.

The small size of most insects often presents a challenge to making accurate measurements of dispersal by preventing the use of technology such as radio transponders that have aided similar research in other animal groups. In some instances, specially designed equipment has been used to ameliorate many of these problems in the laboratory, e.g., tethered flight mills. In field studies, many obstacles have been avoided by observing the movement of insect populations, rather than individuals. Also, mark- release-recapture studies allow assessment of a small portion of a population, rather than individuals, by allowing estimations of dispersal components of larger units. Mark- release-recapture studies estimates are extremely useful, but are associated with statistical concerns due to low recapture rates (Johnson, 1969). Methods that combine both laboratory and field studies in the context of multiple testing days should contribute to an enhanced understanding of the ecological significance of movement. 81

Bemisia tabaci (Gennadius), the sweet potato whitefly, is an important pest insect causing extensive economic damage to agricultural crops and commodities around the world (Oliviera et al., 2001). As a result, a close examination of its movement is important from both the basic and applied points of view.

Increased mobility

To appreciate movement by B. tabaci, it is important to recognize the degree of difference in dispersal ability between this whitefly and other members of Aleyrodidae.

Many of these differences relate to degrees of polyphagy, which are assumed to affect levels of general mobility. Examining the fossil record for clues to the evolutionary history of this taxon, a few specimens indicate that the two extant subfamilies

(Aleyrodicinae and Aleyrodinae) diverged sometime after the Late Cretaceous (Campbell et al., 1996). This coincides with what is known about the diversification of the angiosperms, which provides a potential mechanism for increased speciation within

Aleyrodidae (Mound and Halsey, 1978; Martin et al., 2000). The subfamily Aleyrodinae has the highest density of species in pantropical regions (Campbell et al., 1996). The two most well studied genera within this subfamily are Bemisia and Trialeurodes. These genera are believed to have originated in tropical Africa (Campbell et al., 1996). The ancestral life history of Aleyrodidae is likely represented by a number of genera in the more primitive subfamily Aleyrodicinae, which is closely associated with the woody plants in families such as Loranthaceae, Myrtaceae, and Rubiaceae (Mound and Halsey,

1978). As aleyrodids expanded their range northward, they must have encountered host plants that we not actively growing during parts of the winter and perhaps sparser in 82 distribution. As a consequence, there could have been selective advantages to becoming more mobile in order to leave seasonally deteriorating patches and find new suitable host plants. The most speciose genera may be so because they were able to adapt to new host plants found in more northerly latitudes. These would include Bemisia and Trialeurodes, which are members of the Subfamily Aleyrodinae. These genera are unique among whiteflies because they are polyphagous and feed primarily on herbaceous host plants

(Mound and Halsey, 1978). The remaining aleyrodids are usually associated with monophagy or oligophagy on perennial host plants. It is speculated that the polyphagous nature of these few aleyrodid genera may have led to a migratory nature. In fact, it is only

B. tabaci that has empirically demonstrated migratory behavior. Attempts made to coax

T. vaporariorum to fly in the laboratory were unsuccessful [DNB, personal observation])

Blackmer & Byrne (1993a & b, 1995a) described non-migratory and migratory flight behavior by B. tabaci using a vertical flight chamber in the laboratory. Byrne, et al.

(1996) examined whitefly dispersal in the field.

Importantly for this study, an examination of B. tabaci flight musculature revealed that their flight apparatus degrades rapidly with age (Blackmer et al., 1995b), leading to the prediction that sustained flight was likely limited to single day events. This led to questions about the role of age and prior experience in the earlier estimates of whitefly dispersal, and subsequently to the current study on the importance of multiple day flights within the field.

Determining the presence or absence of an oogenesis-flight syndrome (Johnson,

1969) is important when trying to understand whitefly population dynamics. Veenstra & 83

Byrne (1998) found no evidence of an oogenesis-flight syndrome in relation to dispersal distances for B. tabaci caught in the field studies (Isaacs & Byrne, 1998; Veenstra &

Byrne, 1999). There was, however, some evidence that insects caught in higher traps, i.e.,

2 m above ground, might display a reallocation strategy different from those dispersing closer to the ground. Because of these different patterns, reproductive performance was observed here in concert with multiple day flight performance.

To measure dispersal by B. tabaci over multiple days, a study was designed to sample field populations and monitor their flight in the laboratory. It was hoped these combined field and laboratory studies would provide answers concerning the potential for long-range migration by other small biota (Westbrook & Isard, 1999; Sandstrom et al.,

2007).

Materials and Methods

Trap Collection

Crop fields located at Yuma Agricultural Center near Yuma, AZ were used as sources for whiteflies engaging in dispersal between and within crop fields. Fields were utilized until plants showed obvious signs of senescence. At this time collection locations were relocated to new fields. Muffin fan traps (after Bellamy & Byrne, 2001) were mounted on wooden poles at heights of 0.5 m, 1.5 m and 2.5 m. The traps consisted of plastic collection cups with organdy mesh bottoms fitted into 7.6 cm long sections of 7.6 cm diameter polyvinyl chloride (PVC) pipe. Air was drawn through the plastic cups by 8.0 cm, 12V DC, 0.12A muffin fans (Model AD0812MS-A70, Adda Corp., Ping Tung City,

Taiwan) mounted on one end of the PVC pipe. Fan traps were powered by 12V deep- 84 cycle marine batteries. Wooden poles with fan traps attached were mounted in the cardinal directions from planted fields. These were placed at the field edge, 100 and 200 m from the collection fields. On collection days (Table 1), batteries powered traps at

0700 h. Trap cups were replaced after 2 h during the 4-h collection period (Byrne & von

Bretzel, 1987). Collection cups containing adult whiteflies were placed in a cooler and transported to a laboratory at the Yuma Agricultural Center. Following an approximately

3-min period at 7ºC, individual adults were removed from collection cups and placed onto leaf discs (see below). They were transported that day to the University of Arizona’s

Tucson campus.

Field Collection

A hand-held vacuum was used to collect whiteflies directly from leaves in source fields.

Collection sites were randomly chosen within fields and were located not less than 3.8 m from field margins. Vials containing adult whiteflies were treated as above and transported to Tucson, AZ on leaf disks.

Leaf Disks

Plant material for individual whiteflies consisted of 2.5 cm diameter leaf punches from host plants (Tables 1 and 2). These were placed in Petri dishes half-filled with 1% agar to maintain humidity. During the laboratory flight experiments, whiteflies were transferred to new leaf discs every 2 d. Eggs laid were counted daily.

Vertical Flight Chamber

The vertical flight chamber consists of a large open-fronted box, 0.56 m3, with a central opening in the roof permitting light from a sodium vapor light (sky cue) to shine from 85 above. A circular fan mounted above the box forces air downward into the box, with wind speed controlled by altering the aperture of an air entry hole. The setting of the aperture is constantly adjusted to keep the flying insect 5 to 12 cm below the screen of the roof of the chamber (Blackmer & Byrne, 1993a). Muslin screens dampen air turbulence. Digital readings from an anemometer (Model AS-201 hotwire anemometer,

GreyWolf Sensing Solutions®, Trumbull, CT) measured wind flow from above and were encoded to a data logging program (Wolfsense, GreyWolf Sensing Solutions®,

Trumbull, CT) recording to a laptop computer. These data were used as an indirect measure of insect rate of ascent in m/sec. Airflow rates were measured once every second and recorded to the computer data log. Logged data could be summed over the total flight period to yield an estimate of flight distance. Mounted on the sidewall of the chamber was a 550 nm filter illuminated by a 50w light source that simulates a vegetative cue

(Blackmer et al., 1995a). The light source of the cue was connected to a timer providing illumination of the cue for 10s at a cycle of 3 times/min. Individuals were recorded as responding to this cue when they moved closer to the cue during the time it was illuminated than during the remainder of their flight (David & Hardie, 1988). Since

Kennedy’s definition of migration (1985) includes “…persistent and straightened-out movement…”, actual landing on the cue would be considered a sign of vegetative responses and therefore indicative of foraging flight. Therefore, it was necessary to establish a criterion for response to the cue that could be observed during continual flight.

In this study, observing movement that brought the insect within 10cm of the illuminated vegetative cue source was said to be a response. This deviation of movement from the 86 center point of the overhead light source has been documented in prior studies as significantly different from the mean horizontal displacement observed during phototactic flight by whiteflies (Blackmer & Byrne, 1993b). Whiteflies responding to fewer than 20% of presented vegetative cues while flying within the flight chamber for longer than 3 min were identified as engaging in migratory flight (after Blackmer et al.,

1995a).

For vertical flight testing, adult whiteflies were gently removed from leaf discs and placed into plastic vials. Racks containing adults for testing were placed in the chamber between 0800-1100h (Byrne & von Bretzel, 1987), to begin preconditioned to the overhead mercury vapor light and chamber temperatures at least 30 min prior to testing. Test tube racks containing preconditioned animals were removed from the chamber to begin flight testing. For individual flight measurements, adult whiteflies in plastic vials were placed in the chamber on a platform resting on the chamber floor. The vial cap was removed for 3 min to begin the flight testing period and allow the insect the opportunity to engage in flight (Blackmer & Byrne, 1993a). If flight was initiated, whiteflies were allowed to engage in free flight while observers controlled the downward air current as described above to control whitefly ascent. After an individual terminated flight and remained flightless for more than 1 min, it was captured and returned to its tube by gentle aspiration. At the conclusion of the testing period, animals were returned to leaf discs for testing on subsequent days or dispatched by freezing.

Host plant effects 87

There was interest in whether field source influenced flight duration. It was not possible to draw conclusions due, in a large part, to small sample sizes in the interaction terms for host*day and host*day*source model. This was not considered in the flight duration analyses unless otherwise indicated by the results.

Results

Flight characteristics between individuals

Whiteflies that engaged in migratory flight, i.e., non-responsive to vegetative cues

(n = 6, median duration = 1368 s), had longer duration flights than those animals that only engaged in foraging flights (n = 85, median duration = 190 s; Z = 3.4624, one-sided

P-value < 0.001 (Wilcoxon rank-sum test). A significant difference was not observed in measurements of total distance flown for migratory individuals when compared to foraging flights (Z = 1.2475, one-sided P-value = 0.2122). Unfortunately, one of the longest duration migratory flights (2024 s, with 16 responses to 85 presented cues) had no recorded distance due to a recording error, and so distance comparisons might be affected by the relative strength of outliers in this analysis. Collection source, i.e., trap or field collected, was not a significant factor in either total flight duration (Z = -1.1202, one-sided P-value = 0.262) or total flight distance (Z = -0.9687, one-sided P-value =

0.3327).

Repeated measures of flight duration over time

To analyze the changes in flight characteristics over multiple days, only animals that flew on at least the first 3 consecutive d were included. There was not a significant effect of collection source (field or trap), migratory status, or sex on overall flight 88 duration across all individuals (Table 3). Only when examining the pattern of variation in flight characteristics within individuals over time were certain significant different grouping observed(Table 3). Including the time variable (day) allows a comparison of the effect of these two categories on flight duration over the 3 d for which measurements were made. The interaction between migratory status and day of flight had a statistically significant effect on flight duration (Table 3). Migratory individuals initially flew longer, but there was a significant decline in flight length on subsequent days when compared to non-migratory individuals (Fig. 1A). Likewise, trap caught individuals flew longer than the field sampled group but exhibited significant declines in flight duration by the third day (Fig 1B).

Repeated measures of flight distance over time

Measuring flight over multiple days revealed that migratory status had a significant effect on the distances flown by individuals. It was not migratory status alone that produced the variation in flight distance, but a significant difference was also found when examining the effect of the interaction of migratory status and day on flight distance (Table 4). This finding is consistent with results from the comparison of total flight distances (using the Wilcoxon rank-sum test above.) This indicates that the pattern of flight by migratory individuals changes over time, in a manner different from the pattern of flight recorded for non-migratory individuals. None of the other factors examined (source, sex, or host plant) significantly explained the variation in mean flight distance. As with duration, the repeated measurement of time (measured in terms of flight day in the vertical flight chamber) was found to explain the variance in flight distance as 89 shown by the finding of a significant difference in the variables of intercept and day in the analysis (Table 4, Fig. 2). There was a significant effect of the repeated measure of day in explaining the variation in mean flight distance after accounting for the effects of collection source, migratory status and sex. An overall decrease in flight distance was observed over successive days as shown in Fig. 2a.

Egg production

There was no significant effect of egg production on migratory status over time.

The collection source of tested whiteflies was, however, found to explain variance in egg production over time when examined in a repeated measures analysis of variance model that included the effects of host plant and collection source against daily egg production over days 1 and 2. There was a significant effect of collection source (trap or field) on egg production over time after correcting for the effect of host plant. This was indicated by the significant interaction between collection source and the day of flight (Table 5). It was not possible to include day 3 in this analysis because no field–collected individuals from cotton source fields produced eggs on day 3. It was also necessary to exclude watermelon and tomato host plants from this analysis due to limited sample sizes.

Flight duration measured over the first 2 d was not a significant explanatory variable for the variation observed in egg production over the same time period. This result was found by examining linear regressions of egg production against flight duration. The residual differences for each individual from the mean estimates predicted by the repeated measures analysis were used to account for the effects of multiple day flight, thereby maintaining assumptions of independence. Neither comparisons for day 1 90

2 2 (R = 0.010, t 16 = 0.37, P = 0.71), nor day 2 (R = 0.001, t 16 = -0.14, P = 0.88) produced a significant fit between these two variance measurements from the overall mean estimates created in the above comparisons.

Discussion

The fact that such small insects (less than 40 µg) have the capability for flight over multiple days provides a new perspective on the scale of insect dispersal. B. tabaci survive extreme conditions (maximum temperatures of >40° on collection days) found during the summers of the Southwestern US and are still capable of engaging in flights of greater than 30 min on subsequent days. We did document a reduction in both flight duration and distance over days making it unlikely that repeated migratory flights occur in the context of an individual whitefly’s lifetime.

Although total flight duration was shown to be significantly longer among migratory individuals, the total flight distances measured were not. It is important to note that some of the longest duration flights did not result in the furthest estimated distance.

Distances are accounted for by rates of ascent measured indirectly through the application of downward air current. This could also be an artifact of the sampling methodology. As explained earlier, there were several occasions where the anemometer data was not recorded correctly resulting in a lack of distance data for certain flights.

Despite the overall trend in reductions of flight duration and distance over multiple days, the field sampling techniques resulted in two population subgroups, i.e., those that demonstrated some dispersal from the fields, having been caught by traps at a distance from the field, and those collected within the field, assumed to not have flown 91 previously. Although the multiple day flight analysis did not find significant differences in overall flight distance or duration between these two collection sources, there are still important conclusions to be gained from examining these groups. Animals collected from traps that were found to be migratory in flight chamber testing should be considered remarkable in their flight performance. These individuals could have been trapped either

100 or 200 m away from the field and were capable of engaging in migratory flight efforts on a subsequent day. These insects minimally flew for 100 m plus the distance flown in the chamber. For migratory B. tabaci trapped at least 100m outside of the field, the total maximum flight distance was measured in the chamber at 63.68m. It is important to recognize that estimates of distance alone as exhibited in the flight chamber may not be comparable to field movement, which has been shown to be wind-aided (Byrne et al,

1996).

We must examine behaviours that signify a difference between animals moving within their local environment (foraging) and those that have directed their dispersal beyond their immediate surroundings (migratory individuals). Significant to the definition of insect migratory flight provided by Kennedy (1985) are aspects such as persistent, straightened-out movement as well as the inhibition of station-keeping responses (our vegetative and sky cues). Both of these criteria signify behavioural changes that could represent a tradeoff in terms of fitness risks; risks associated with remaining in a deteriorating habitat versus the risk of not finding a new suitable one.

Whether or not whitefly migration can occur over multiple days was a central hypothesis of this study. Total duration of flight in the flight chamber was found to be 92 significantly increased among those whiteflies that engaged in migratory flight.

Unfortunately, multiple migratory flights were never observed among the insects tested.

These results cast doubt on the overall ability of whiteflies to migrate on multiple days, leading to the conclusion that successive migratory flights by B. tabaci probably do not contribute to meaningful long-range redistribution of their populations. Nevertheless, the findings of this study indicate that multiple day foraging flights occur frequently and should be considered an important aspect in the general dispersal of whiteflies.

Within the 1,500 members of Aleyrodidae (Martin & Mound 2007), polyphagy and dispersal characteristics have been linked in the one species examined for these traits.

Bemisia tabaci has displayed the ability to move readily between host plant patches and is associated with more that 650 plant species (Mound and Halsey, 1978). Since there is no evidence of sex-bias in dispersal (D.N. Byrne, personal communication), the presence of migratory flight in B. tabaci has probably been driven by an adaptation to access ephemeral host plant patches.

There was no evidence that egg production was correlated with flight performance in the repeated measures analysis, indicating that migratory whiteflies in this study were not displaying an oogenesis/flight trade-off. The general trend observed was for a slight increase in egg production over the 2 d for which adequate data was generated. The only factor found to explain differences in mean egg production over time was the collection method used to sample the population. The individuals taken directly from the field were found to increase their egg production over time compared to those caught in traps outside the field. Field-collected individuals also did not reduce their 93 flight duration over time in a manner consistent with the pattern observed for the trap- collected individuals. Comparing these two findings indicates that flight resulting in movement out of the field may have some reproductive costs. Because the reproductive history and age of the whiteflies collected from the source fields is not known, we were not able to distinguish between reproductive cost resulting from migratory flight as opposed to foraging flight. Other studies on whitefly flight and reproductive physiology have found that eggload, as well as egg protein levels, did not differ with respect to distance travelled. Significant differences in these measures of reproductive effort were found in B. tabaci moving at greater than 2 m above the ground (Issacs & Byrne, 1998;

Veenstra & Byrne, 1999), but the migratory status of those animals was not known. A similar examination found differences in reproduction in response to host plant quality in a manner than suggested dynamic control of the physiological state of B. tabaci over a short time period (> 6h) (Veenstra & Byrne 1998). Our results indicating a lack of a relationship between flight distance and egg production are consistent with these studies.

Among these analyses of B. tabaci dispersal and reproductive parameters, we now have a consistent idea of an animal that displays an ability for plasticity in allocation of resources.

It is important to recognize that damage or loss of animals due to handling methods may have affected the overall fitness of these whiteflies, diminishing their capacity or resulting in their elimination from the study. It is therefore suggested that future experiments be performed comparing flight performance with paired non-flying 94 animals to better control for the effects of flight on reproduction with respect to the overall lifetime traits of the whitefly.

There were no indications of host plant differences affecting the duration or distance of flight over multiple days, although the power of this analysis was likely limited by the reduced sample sizes for alternative field host plants. It would be beneficial to further investigate the importance of alternative crops as potential refuges.

The findings of this study suggest that the ability of B. tabaci to move rapidly between and within fields without tradeoffs toward their reproductive abilities contribute to their pest status. 95

References

Asplen, M.K.A (2007) Life history correlates in female parasitoid wasps: exploring old concepts and new frontiers. Ph.D. thesis, University of Arizona, Tucson, Arizona, USA.

Bellamy, D.E., Byrne, D.N. 2001. Effects of gender and mating status on self-directed dispersal by the whitefly parasitoid Eretmocerus eremicus. Ecological Entomology, 26, 571-577.

Blackmer, J., Byrne, D.N., Tu, Z. (1995a) Behavioural, morphological, and physiological traits associated with migratory Bemisia tabaci (Homoptera, Aleyrodidae). Journal of Insect Behaviour, 8, 251-267.

Blackmer, J., Lindley, V., Byrne, D.N. (1995b) Histological examination of flight-muscle development and breakdown in Bemisia tabaci (Homoptera, Aleyrodidae) - relationship to age and flight behaviour. Journal of Morphology, 226, 213-221.

Blackmer, J.L., Byrne, D.N. (1993a). Flight behaviour of Bemisia tabaci in a vertical flight chamber: effect of time of day, sex, age and host quality. Physiological Entomology, 18, 223-232.

Blackmer, J.L., Byrne, D.N. (1993b) Environmental and physiological factors influencing phototactic flight of Bemisia tabaci. Physiological Entomology, 18, 336-342.

Bush. G.L. (1975) Modes of Animal Speciation. Annual Review of Ecology and Systematics, 6, 339-364.

Byrne, D., Rathman, R.J., Orum, T.V., Palumbo, J.C. (1996) Localized migration and dispersal by the sweet potato whitefly, Bemisia tabaci. Oecologia, 105, 320-328.

Byrne, D.N., Von Bretzel, P. (1987) Similarity in flight activity rhythms in coexisting species of Aleyrodidae, Bemisia tabaci and Trialeurodes abutilonea. Entomologia Experimentalis et Applicata, 43, 215-219.

Campbell, B.C., Steffen-Campbell, J.D., Gill, R.J. (1996) Origin and radiation of whiteflies: an initial molecular and phylogenetic assessment. In: Gerling D, Mayer RT, editors. Bemisia: 1995 Taxonomy, Biology, Damage Control and Management. Andover: Intercept. p 29-52.

David, C.T., Hardie, J. (1988) The visual responses of free-flying summer and autumn forms of the black bean aphid, Aphis fabae, in an automated flight chamber. Physiological Entomology, 13, 277-284.

Elliott, C., Evenden, M. (2009) Factors influencing flight potential of Choristoneura conflictana. Physiological Entomology, 34, 71-78.

Ellsworth, P.C., Martinez-Carrillo, J.L., (2001) IPM for Bemisia tabaci: a case study from North America. Crop Protection, 20, 853-869.

Isaacs, R., Byrne, D.N. 1998. Arial distribution, flight behaviour and eggload: their interrelationship during dispersal but the sweetpotato whitefly. Journal of Animal Ecology, 67, 741-750.

Johnson, C.G. (1969) Migration and Dispersal of Insects by Flight. Methuen and Company, Ltd., London. 763 p.

Martin, J.H., Mifsud, D., Rapisarda, C. (2000) The whiteflies (Hemiptera: Aleyrodidae) of Europe and the Mediterranean Basin. Bulletin of Entomological Research, 90, 407- 448.

Mound, L.A., Halsey, S.H. (1978) Whitefly of the world. A systematic catalogue of the Aleyrodidae (Homoptera). British Museum. 340 pp.

Martin, J. H. and L. A. Mound. 2007. An annotated check list of the world’s whiteflies (Insecta: Hemiptera: Aleyrodidae). Zootaxa 1492: 1–84.

Nolte, K., Brown, J.K. (2007) Cucurbit yellow stunting disorder virus. Yuma Co. Extension fact sheet. 1 p.

Oliviera, M.R.V., Henneberry, T.J., Anderson, P. (2001) History, current status, and collaborative research projects for Bemisia tabaci. Crop Protection, 20, 709-723.

Roitberg, B.D., Mangel, M. (1997) Individuals on the landscape: behavior can mitigate landscape differences among habitats. Oikos, 80, 234-240.

Sandstrom, M. A., Changnon, D., Flood, B. R. (2007) Improving our understanding of Helicoverpa zea migration in the Midwest: Assessment of source populations. Online. Plant Health Progress, doi:10.1094/PHP-2007-0719-08-RV.

Senger, S., Roitberg, B., Thistlewood, H. (2007) Relative flight responses of Rhagoletis indifferens as influenced by crowding, sex, and resources. Entomologia Experimentalis et Applicata, 123, 91-100.

Tonhasca, A., Palumbo, J.C., Byrne, D.N. (1994) Binomial sampling plans for estimating Bemisia tabaci populations in cantaloupes. Researches on Population Ecology, 36, 181-186.

Veenstra, K.H., Byrne, D.N. 1998. Effect of starvation and oviposition activity on the reproductive physiology of the sweet potato whitefly, Bemisia tabaci. Physiological Entomology, 23, 62-68.

Veenstra, K.H., Byrne, D.N. 1999. Does dispersal affect the reproductive physiology of the sweet potato whitefly, Bemisia tabaci? Physiological Entomology, 24, 72-75.

Westbrook, J.K., Isard, S.A. (1999) Atmospheric scales of biotic dispersal. Agricultural and Forest Meteorology, 97, 263-274.

Wiens, J.A. (2001) The landscape context of dispersal. In: Dispersal (ed. J. Clobert, E. Danchin, A.A. Dhondt, & J.D. Nichols), pp 98-109. Oxford University Press, Oxford. 98

Table Legends

Table 1: Trapping dates and host plant types for whiteflies collected from the Yuma

Agricultural Center for flight chamber experiments in this study.

Date Host plant Leaf Disc Host Number of plant whiteflies

18 June 2008 Cantaloupe Cantaloupe 11 25 June 2008 Cantaloupe Cantaloupe 13 1 July 2008 Watermelon Cantaloupe 8 9 July 2008 Cotton Cotton 6 16 July 2008 Tomato Cotton 9 23 July 2008 Cotton Cotton 17 30 July 2008 Cotton Cotton 23 6 August 2008 Cotton Cotton 14 13 August 2008 Cotton Cotton 20 10 September 2008 Melon Melon 17 16 September 2008 Cotton & Melon Melon 21 1 October 2008 Melon Melon 24

Table 2: Distributions of individuals for source populations either collected via trapping

(Trap) or directly from the field (Field). Some individuals exhibited short flights and landed without cue presentation.

Source Individuals with Individuals Individuals Total Population multiple day presented with exhibiting flight individuals flights cue examined

Trap* 26 57 87 110 Field 12 34 59 73 * Of the 5 trap-caught individuals displaying subsequent migratory flight in the vertical flight chamber, 3 were caught in traps ≥ 100m from the field.

100

Table 3. Repeated measures analysis of variance of migratory status, sex and source population in relation to flight duration.

Numerator df, Variance component F Denominator df P

Intercept 41.8093 1,34 < 0.0001 Migratory Status Among individuals 0.9993 1, 34 0.3245 Within individuals Migration*day 6.4984 2, 33 0.0063 Sex Among individuals 0.3323 1, 34 0.5681 Within individuals Sex*day 0.4334 2, 33 0.6519 Source population Among individuals 0.1640 1, 34 0.6880 Within individuals Source*day 5.9310 2, 33 0.0063 Day Within individuals 11.3005 2,33 0.0002

101

Table 4. Repeated measures analysis of variance of migratory status, sex and source population in relation to flight distance.

Numerator df, Variance component F Denominator df P

Intercept 47.5977 1, 24 < 0.0001 Migratory Status Among individuals 0.0142 1, 24 0.9062 Within individuals Migration*day 5.7257 2, 23 0.0096 Sex Among individuals 2.1697 1, 24 0.1537 Within individuals Sex*day 0.1102 2, 23 0.8961 Source population Among individuals 2.0382 1, 24 0.1663 Within individuals Source*day 1.7600 2, 23 0.1944 Day Within individuals 27.0575 2, 23 < 0.0001

102

Table 5. Repeated measures analysis of variance in egg production over days 1 and 2, with effects test for day, host plant (cotton or melon), and source population.

Numerator df, Variance component F Denominator df P

Intercept 477.8578 1, 77 < 0.0001 Host plant Among individuals 2.4955 1, 77 0.1183 Within individuals Host plant*day 0.0188 1, 77 0.8914 Source population Among individuals 1.1421 1, 77 0.2866 Within individuals Source*day 8.2306 1, 77 0.0053 Day Within individuals 4.8774 1, 77 0.0101 103

Figure 1. Repeated Measures analysis of variance analyses for mean flight duration measuring the effects combined in a single model over multiple days (error bars indicate

SE for Least Square Means. These graphs indicate trends, while significant differences are reported on Table 3. Figure 1A, overall mean estimates of flight duration over multiple days including flight duration for foraging and migratory individuals (see

Methods for definitions); figure 1B, estimated mean flight duration of individuals collected from different source populations in the field; figure 1C, estimated mean flight duration of male and female whiteflies over multiple days.

104

105

Figure 2. Repeated Measures analysis of variance analyses for mean flight distance measuring the effects combined in a single model over multiple days. These graphs indicate trends, while significant differences are reported on Table 4. Figure 2A, overall mean estimates of flight distance over multiple days including flight distances for foraging and migratory individuals (see Methods for definitions); figure 2B, estimated mean flight distance of individuals collected from different source populations in the field; figure 2C, estimated mean flight distance of male and female whiteflies over multiple days.

106