Plant Semiochemicals As Mosquito Attractants Dissertation Presented

Plant Semiochemicals as Mosquito Attractants

Dissertation

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

Philip E. Otienoburu. B.Sc.

Graduate Program in Entomology

The Ohio State University

2011

Dissertation Committee:

Woodbridge A. Foster, Advisor

P. Larry Phelan

David Denlinger

Philip E. Otienoburu

2011

Abstract

A new approach to vector surveillance is proposed. Mosquitoes of both sexes are attracted to particular nectar-bearing plants that generate mixtures of volatile chemicals.

Preliminary experiments demonstrate that mosquitoes orient to these mixtures and that they can be used as lures to trap large numbers. Yet virtually nothing is known of their composition, and their use in surveillance traps has never been explored. As widespread early-warning and sampling devices, these attractants have advantages over those currently available or being developed, most of which are based on kairomones from vertebrate hosts. The latter attract only females and only those that have entered the blood feeding mode. Plant-derived attractants on the other hand attract mosquitoes as early as one day after emergence, attract early-emerging localized males, as well as pre- and post-dispersal females, attract females in all gonotrophic stages, in reproductive diapause, and can be slow-released in tiny amounts over extended periods. Therefore phytochemicals provide earlier and more precise information on mosquito mass emergence and population composition, and they can be deployed in simple light-weight traps. The objectives of this project are twofold: to create the most attractive synthetic blends of component volatiles of the plants most frequently used as sources of sugar by two medically important mosquitoes in two different continents – Culex pipiens, a

presumptive primary vector of West Nile virus in the northeastern United States, and

Anopheles gambiae, the main vector of malaria in Africa. The second is to explore the potential use of plant semiochemicals to attract mosquitoes to both natural and artificial kill stations using attractive, toxic sugar baits.

We used a four-pronged approach to achieve these objectives. First was to analyze the components of floral headspace of attractive plants, second was to assay the relative bioactivity of fractions and synthetic blends by mosquito olfactometer, third was to test the most attractive blends in the field at a series of release rates using olfaction traps, and finally we sprayed an attractive toxic sugar bait on plants and artificial bait stations to determine the efficacy of using this as a method of disease vector control.

Both floral extracts and synthetic blends were attractive to mosquitoes in olfactometer experiments, and showed some attraction in the field. Natural toxic stations proved very effective in attracting and killing mosquitoes by using the dominant attractive power of the plant.

iii

Dedication

To Elizabeth Atieno Oure and Caroline Judith Adhiambo

Acknowledgements

I would like to recognize my advisor, Dr. Woodbridge A. Foster, a true gentleman and scholar. I was truly inspired by your views on science and ethics, and the need for scientists to be true to their calling. I appreciated your respect for the world around us, and your role as a scientist to help nurture a better understanding of nature. Thanks for the fifteen minute meetings that ended up running two hours – it was the content of your advice that was enthralling.

To Dr. P. Larry Phelan, I say thank you for pushing me to be a better scientist, and for your patience through the arduous journey. Thanks for all the training you afforded me, invaluable in the field of Chemical Ecology, and the countless hours you spent directing my main research.

Dr. Dave Denlinger, always supportive, always available, always concerned. Thank you for all the letters of recommendation you have written at short notice, and for being the scientist that any graduate students would want to emulate.

To my colleagues in the Foster lab, especially Babak Ebrahimi ―Bobi Ebi‖ who I did a lot of field work with, suffering the torching sun and mosquito bites.

To my family, I say thank you for keeping me sane with random phone calls and e-mails,

I love you all. To great friends I have met along the way – you know who you are!

Vita

April 18, 1979……………………………………. Born-Nairobi, Kenya

1997………………………………………………... Kisumu Boys High School

2000-2001………………………………….. ……... Student Research Assistant Kenya Medical Research Institute

2003……………………………………………….. B.Sc., Jomo Kenyatta University of Agriculture & Technology (JKUAT)

2004-2005…………………………………………. Graduate Research Associate KEMRI/CDC

2006………..………………………………………. United Nations, NY OCHA

2006-2011…………………………………………. Graduate Research/Teaching/ Administrative Associate The Ohio State University

Publications

1. 2007 Otienoburu, PE., Bayoh N., Gimnig, J., Huang, J., Walker, ED., Otieno, MF., Vulule, J. and Miller, JR., Anopheles gambiae oviposition as influenced by type of water infused into black and red soils of Western Kenya International Journal of Tropical Insect Science, 27: 2-5

2. 2006 Huang, J., Walker, ED., Otienoburu, PE., Amimo, F., Vulule, J., Miller, JR., Laboratory tests of oviposition by the African malaria mosquito, Anopheles gambiae, on dark soil as influenced by presence or absence of vegetation. Malaria Journal, 5:88. vi

Field of Study

Major field: Entomology

vii

Table of Contents

Abstract…………………………………………………………………….. ……... ii

Dedication………………………………………………………………….. ……... iv

Acknowledgements………………………………………………………………… v

Vita…………………………………………………………………………………. vi

List of Tables………………………………………………………………………. xi

List of Figures……………………………………………………………………… xii

Chapters

1. Introduction………………………………………………………… ……... 1 Mosquitoes and disease…...……………………………………….. 1 Ecology of Mosquito Vectors…..………………….…………….... 2 Surveillance and trapping of mosquitoes…..……………………… 3 Non attractant-based trapping methods…………………………… 3 Attractant-based trapping methods………………………………... 5 Sugar-feeding behavior of mosquitoes...………………………….. 9 Host plant location by mosquitoes……..………………………….. 12 Research blueprint………...……………………………………….. 15 References………………………………………………………….. 18

2. Analysis and Optimization of a Synthetic Milkweed Floral Attractant for Mosquitoes ……………………………………………. …………...... 34 Abstract…………………………………………………………….. 34 Introduction………………………………………………………… 35 Materials and Methods…………………………………………….. 37 Results……………………………………………………………… 44 Discussion………………………………………………………….. 48 References………………………………………………………….. 54 Illustrations………………………………………………………… 62

viii

3. From Natural Extracts to Synthetic Blends of Canada goldenrod Solidago canadensis: Key considerations and challenges to the effective trapping of mosquitoes in the field ………………………………....………………….. 70 Abstract…………………………………………………………….. 70 Introduction………………………………………………………… 71 Materials and Methods……………………………………………... 72 Results……………………………………………………………… 79 Discussion …………………………………...……………...……... 81 References………………………………………………………….. 85 Illustrations …..………………………………….……………….89

4. A Novel Diffusion-Cage Olfactometer for Measuring Anopheles gambiae s.s (Diptera: Culicidae) Orientation to Plant Volatiles in Semi-Field Enclosures………………………………………………………………….. 98 Abstract…………………………………………………………….. 98 Introduction………………………………………………………… 99 Materials and Methods……………………………………………... 100 Results……………………………………………………………… 103 Discussion………………………………………………………….. 105 References………………………………………………………….. 107 Illustrations ...…...... ………………………….…………………….110

5. Exploiting the attractive power of Parthenium hysterophorus to kill the malaria vector Anopheles gambiae s.s……………………………………………… 112 Abstract…………………………………………………………….. 112 Introduction………………………………………………………… 113 Materials and Methods……………………………………………... 115 Results……………………………………………………………… 121 Discussion………………………………………………………….. 125 References………………………………………………………….. 130 Illustrations .………….………………….….……………………....135

General Conclusions ...…………………………………………………………… 143

Appendix A: Chemicals contained in the headspace of various plants visited by mosquitoes in temperate and tropical regions………………..…………….……….147

Appendix B: Olfactory responses in a gustatory organ of Culex pipiens and Anopheles gambiae to various plant volatiles ……………..…………..…………….……….157

Bibliography ….. ………………………………………………….……………….167

List of Tables

Table Page

2.1 Volatile composition of solvent extracts of the common milkweed Asclepias syriaca and the quantification of extract components.…………………..… 68

2.2 Choice response of Culex pipiens in a flight olfactometer to a six-component synthetic blend of milkweed flower volatiles compared to reducted blends, in which each component is removed individually.…………………………... 69

3.1 Composition and quality of a natural goldenrod extract evaluated at two time points, two years apart (2006 and 2008), highlighting changes in concentration between the two extracts.…………………………………………………………. 90

3.2 Volatile composition of solvent extracts of Canada goldenrod Solidago canadensis and the quantification of extract components ……………….... 92

5.1 Estimated coefficients and standard errors for the multiple regression model predicting response, based on humidity, temperature and plant species in a diffusion-cage olfactometer ………………………………………………... 142

A.1 Headspace volatiles of some plants frequently visited by mosquitoes in tropical and temperate regions …..………………………...…….………….…..…...147

List of Figures

Figure Page

2.1. Overlay of (A) milkweed pentane extract (broken line) and synthetic blend (solid line) total ion chromatograms generated on a Zebron™ ZB-50 capillary column ………………………………………………..……………………………. 62

2.2 Responses of all Culex pipiens released in a dual-port flight olfactometer to: (A) whole milkweed flowers (B) pentane extract of milkweed and to (C) synthetic milkweed blend …………………………………………...……………….. 64

2.3 Response of all Culex pipiens released in a dual-chioce flight olfactometer to a three-component blend ……………………………………………………..65

2.4 Dose-response bioassays of various concentrations of a three-component blend by mosquitoes in a dual-port olfactometer …………………….……………… 67

3.1 Chromatogram of goldenrod extract (2008) showing peaks identified by GC/MS ………………………………..………….………………………………… 89

3.2 Differential responses of Culex pipiens to a pentane extract of goldenrod prepared in 2006 and later tested in 2008 in a dual-port olfactometer…………….....93

3.3 Responses of Culex pipiens to a synthetic goldenrod blend compared to a solvent control (left) and a goldenrod extract (2008) compared to a synthetic blend ……………………………………………………………………………... 94

3.4 Trap catches of Culex pipiens in field traps baited with blends at different release rates and compared to traps baited with goldenrod flowers……..………… 95

3.5 Sex ratio of Culex pipiens mosquitoes caught in field traps with blends at different release rates………………………………………………………………… 96

xii

3.6 Field trapping of Culex pipiens in CFG traps baited with whole goldenrod flowers, goldenrod synthetic blend a pentane extract of goldenrod compared against blank traps. ……………………………………………...... ………97

4.1 Schematic diagram of diffusion-cage olfactometer, showing full assembly and plant placement. ……………………………………..……………………..110

4.2 Side by side comparisons of diffusion-cage and Y-tube olfactometers when tested with four African plants in non-competitive assays.………………………111

5.1 Non-competitive assays of four plants in diffusion-cage olfactometers set up in a small mesocosm …….……………………………………………………...135

5.2 Non-competitive assays of a mango Mangifera indica fruit essence in diffusion- cage olfactometers set up in a small mesocosm. ………………….....….....136

5.3 Competitive assays (dual choice) of Parthenium against Lantana and Senna in diffusion-cage olfactometers set up in a small mesocosm …………...……. 137

5.4 Competitive assays (multiple-choice) of four African plants in diffusion-cage olfactometers set up in a large screenhouse ……………………….………. 138

5.5 Attractive toxic bait associated mortality in screenhouse experiments containing ATSB enhanced Parthenium hysterophorus plants …………..…………… 139

5.6 Artificial bait station for ATSB dispensing in an experimental screenhouse………………………………………………………………….140

5.7 Artificial bait station associated mortality in a sugar rich environment ……141

B.1 Anopheles gambiae maxillary sensillum screening performed with a series of 17 plant odorant compounds………………………...…….………….….….....165

B.2 Culex pipiens maxillary sensillum screening performed with a series of 17 plant odorant compounds ……………………………………..………………… 166

xiii Introduction

Mosquitoes and disease

Mosquitoes are important vectors of disease, transmitting both viral and parasitic diseases to humans. Malaria is the most important mosquito-borne disease, with over 3 million infections worldwide, and up to 800,000 deaths occurring mostly in sub Saharan Africa and affecting mainly children under the age of five, and pregnant women (WHO, 2010).

Malaria is caused by female Anopheles mosquitoes, which transmit the Plasmodium parasite through an infective bite.

Other mosquito-borne diseases include arboviral diseases such as West Nile virus, an important cause of morbidity and mortality in north America, and transmitted by a number of culicine species, including Culex pipiens (Hamer et al., 2008). Transmission of

West Nile virus follows an enzootic cycle (Turell et al., 2005), with birds acting as a natural reservoir infecting mosquitoes that feed on their blood (Kilpatrick et al., 2006;

Hamer et al., 2008). Emerging as an important vector of Chikungunya virus is the rapidly spreading Asian Tiger mosquito, Aedes albopictus, its prevalence greatly enlarged through the international trade in motor vehicle tires (Benedict et al., 2007), since these often hold water that make for a suitable breeding habitat for the mosquitoes. Long established as a vector for dengue, yellow fever and Chikungunya virus, is Aedes aegypti,

1 a container-breeding mosquito closely associated with human habitations (Leake, 1992) and distributed mainly in Asia, Africa and south America.

Ecology of Mosquito vectors

Depending on the species, mosquitoes thrive in different habitat types, and these ecological differences inform the methods for their surveillance and control (Service,

1993; Majambere et al., 2010). The principal malaria vector presents one of the greatest challenges for control, due to their fairly rapid development in small, transient, sun-lit habitats (Gimnig et al., 2001; Mutuku et al., 2006). Anopheles gambiae is predominantly an endophilic and endophagic species, meaning that it prefers to feed and rest indoors, only leaving the domicile to locate an oviposition site (Pates and Curtis, 2005). This means that successful surveillance of this species can be conducted by pyrethrum spray catches (PSC), and effective control is achievable through the use of indoor residual spraying (IRS) and insecticide treated bed-nets (ITNs) (Lengeler and Snow, 1996; WHO,

2010). On the contrary, another member of the Anopheles gambiae complex, Anopheles arabiensis, is a predominantly exophilic (Coetzee et al., 2000) and partly zoophilic

(Mahande et al., 2007) mosquito, and as such indoor surveillance and control interventions would prove unsuccessful for this mosquito. Instead, outdoor trapping and larval control methods are the preferred option. Environmental management and source reduction strategies are generally effective in the control of both Aedes sp. and Culex sp.

(Pates and Curtis, 2005), which easily breed in open containers and drainage ditches, and

Anopheles sp. (Kirton and Spielman, 1989), which almost exclusively breed outdoors. In cases where environmental management and source reduction practices fall short, especially due to the large number of potential breeding sites that are formed after the rainy season in most tropical countries, alternative trapping methods are used in addition to insecticide spraying. Below, we review some of these trapping methods.

Surveillance and Trapping of Mosquitoes

The first step in tracking mosquito-borne disease is knowing about the distribution of the vectors that transmit these diseases. Vector surveillance programs help in developing risk maps, determine vector incrimination, monitoring of entomological risk factors and guide in establishing, monitoring and evaluating vector control programs (Service, 1985;

Nelson, 1994). Several methods have been employed over the past century in mosquito surveillance, and control measures have become more innovative as mosquito ecology and behavior is better understood.

Non-attractant trapping methods

There is always a need in population sampling to have an unbiased sample, and non- attractant-based trap samples are thought to address this need to some extent (Holck and

Meek, 1991), since they passively catch mosquitoes as they randomly fly through an area.

These traps include Malaise traps, sticky traps and electrically powered suction traps

(Tikasingh and Davies, 1972; Service, 1985). The malaise trap is a tent-like structure

3 made from fine, weatherproof netting material which traps flying insects. The insects are stopped by the netting, which acts as a barrier, and their natural tendency to fly upwards when trying to escape leads them into a collecting chamber at the top of the trap that is usually filled with alcohol to preserve the specimens. The trap can be left to operate for several days unsupervised, and the collecting chamber is replaced at the convenience of the collector (Breeland and Pickard, 1965). Sticky traps involve the use of an adhesive tape, which may be hung on branches or on tree trunks or may be used together with ovitraps, and these catch mosquitoes that land on their adhesive surfaces (Facchinelli et al., 2007). The bias when used with ovitraps, however, is that only ovipositing female mosquitoes are trapped, to the exclusion of females in other physiological states and male mosquitoes.

Battery-operated suction traps employ the use of a fan to create suction within a certain range around the trap, thereby trapping mosquitoes flying in the trapping range (Johnson,

1950). Suction traps have undergone a lot of modification and have been adapted for use with visual and chemical attractants.

In summary, non-attractant traps, bear the advantage of giving a balanced representation of the mosquito population in an area, and confer the benefits of low cost and low technology. The sources of bias with these methods are the placement of the traps, the

4 orientation, and the height. In addition, resting mosquitoes, such as gravid females that are hidden in resting sites, are not sampled (Nelson, 1994).

Attractant-based trapping methods

Human landing catches are often used to determine the human biting rate, a key parameter needed to calculate the entomological inoculation rate (EIR), useful when determining the vectorial capacity of disease vectors (Shaukat et al., 2010; Nicodem et al., 2011). This method involves using the natural attractive power of live human subjects to lure biting females to exposed skin surfaces, and collecting the mosquitoes before they have had a chance to bite and is highly effective for anthropophagic mosquitoes such as Aedes aegypti and Anopheles gambiae s.s. Ethical considerations surrounding the use of this method, as important as it may be, make it impossible to justify (Ferguson et al., 2010), thus pushing the drive for alternative kairomone-based attractant strategies.

Light traps

Various mosquito species are attracted to light sources and these have been effectively incorporated in traps, such as the New Jersey light-trap and the CDC light-trap used in several mosquito surveillance programs (Service, 1969, 1985; Govella et al., 2011).

Light traps attract mainly blood-thirsty females, and wavelengths in the lower green (502 nm) spectral range, seem to attract the broadest range of mosquito species (Bentley et al.,

2009). Davis et al. (1995) working in Bagamoyo in Tanzania, compared light trap collections to human landing catches and found them to be as effective in estimating human biting activity.

CO2 Traps

The attractive power of CO2 to mosquitoes was discovered in bioassay experiments by

Rudolfs (1922) who observed that small quantities of the gas led to activation of the insects, and that human expiration, caused the insects to display ―pleasure‖ and a directed flight and probing of the emitting source. It was later observed that different mosquito species are activated by different environmental concentrations of the gas (Reeves, 1990), and this was thought to explain their host predilection. In human hosts, CO2 is considered a very important activator, essential in the receptivity towards host odors

(Dekker et al., 2005).

The role of carbon dioxide as a mosquito attractant has been widely exploited through its use in different traps, including the BG-Sentinel trap (Meeraus et al., 2008), Counter

Flow Geometry trap (Kline, 1999), and CDC light traps (Mboera et al., 2000). A close examination of the odor plumes of four commercial traps, the Encephalitis Virus

Surveillance (EVS), Mosquito Magnet Freedom (MMF), Mosquito Magnet Liberty

(MML) and the Mosquito Magnet-X (MMX) also known as the CFG trap, showed distinctly different plumes structures due to differences in their air velocities at both CO2

6 output and suction inlets (Cooperband and Cardé, 2006a). These differences proved important in the trapping efficiency of these traps, with the CFG traps outperforming the other traps in the study (Cooperband and Cardé, 2006b).

The source of CO2 used in mosquito surveillance/control traps has also influenced their construction, with more commercial traps opting for propane tanks (Kline, 2002;

Cooperband and Cardé, 2006a). The use of blocks of dry ice, where available has been shown to be effective, with the added advantage of being cheap and fairly easy to use

(Higa et al., 2006; Xue et al., 2008). In cases where both propane tanks or dry ice are inaccessible, such as in remote field locations, CO2 can be produced through yeast fermentation, and this has been effectively used in traps (Saitoh et al., 2004; Smallegange et al., 2010b).

Synthetic attractants

The use of synthetic chemicals to lure mosquitoes to traps is a long established practice, with better chemical attractants developed as knowledge of mosquito behavior increases.

The use of synthetic chemicals arose as a need to mimic the different volatiles produced by vertebrate animals, such as ammonia, lactic acid and octanol which are known attractants used by mosquitoes to locate their host (Knols and De Jong, 1996; Qiu et al.,

2011). Jawara et al. (2011) demonstrated the effectiveness of using a synthetic blend consisting tetradecanoic acid, L-lactic acid, ammonia and augmented with CO2, and

7 found that the trap performance was not affected by the presence of human sleepers in huts where these baited traps were used. The discovery of bacterial volatiles produced by normal skin flora led to the identification and use of 3-methyl-1-butanol as an important attractant for mosquitoes, especially when used in concert with ammonia, lactic acid and tetradecanoic acid (Verhulst et al., 2011).

The overall goal of developing these synthetic blends is to find a blend combination that would out-compete the attractive power of humans, thereby leading mosquitoes away from their preferred host and breaking disease transmission (Okumu et al., 2010b). This goal has had varying success, with some authors finding that their optimal blends could not compete with natural skin emanations (Smallegange et al., 2010a), yet other studies have developed a synthetic blend consisting carbon-dioxide, ammonia and carboxylic acids that was more attractive than humans (Okumu et al., 2010c).

Currently, there is a push to develop multimodal lures that combine various properties of the host, including heat, smell and moisture, though studies by Olanga et al. (2010) have indicated that olfactory cues play the biggest part in host location, and that the role of heat and moisture is diminished due to their relatively small range of activity. However, for other hematophagous insects such as bedbugs, this approach has proven effective

(Andersson et al., 2009).

Though human kairomone-based attractants present an important avenue towards the surveillance and control of mosquitoes, they do have their draw-backs. When used in surveillance programs, human kairomone traps sample mainly blood-seeking females to the entire exclusion of males and females in other physiological states.

In this study, we explore the potential for plant-based attractants to trap and kill mosquitoes. These types of attractants potentially confer several advantages, including 1) attracting mosquitoes as early as one day after emergence, 2) attracting early-emerging males as well as pre- and post-dispersal females, and 3) attracting females in all gonotrophic stages and in reproductive diapause. Synthetic plant attractants can be slow released in small amounts over extended periods of time.

Sugar feeding behavior of mosquitoes

Mosquitoes are frequent visitors of plants, obtaining sugar from floral and extra-floral nectar (Patterson et al., 1969; Foster and Hancock, 1994; Burkett et al., 1999; Gouagna et al., 2010), ripening fruits (Joseph, 1970) and honeydew (Foster, 1995; Gary and Foster,

2004). Sugar feeding plays an integral role in the survival and fitness of adult mosquitoes, with both sexes utilizing plant sugars (Sandholm and Price, 1962; Grimstad and Defoliart, 1974). Magnarelli (1977 , 1978) reported that female mosquitoes not only sugar-fed whenever possible, but also obtained a meal at all stages of their gonotrophic cycle. Both male and female mosquitoes have a predilection for sugar (Downes, 1958;

Bindlingmayer and Hem, 1973; Yuval, 1992; Foster, 1995). Plant sugars are essential in ensuring the successful dispersal of mosquitoes from emergence pools to swarming locations for male mosquitoes and for the mated females, to the location of their preferred vertebrate host, providing the critically needed energy to sustain their flight (Hocking,

1953; Nayar and Van Handel, 1971; Harada et al., 1972). Sugar feeding continues throughout the adult life of the mosquito (Reisen et al., 1986; Yuval, 1992) and most female mosquitoes will take sugar before taking a blood meal. Primary follicle development is closely linked to presence of a sugar meal, stimulating follicle progression from Christopher stage I to II in previtellogenic development (Feinsod and

Spielman, 1980).

Although sugar and blood can be interchangeably used by mosquitoes as a source of energy (Foster, 1995), studies have shown that females with access to sugar in addition to blood live significantly longer than females with only access to blood (Thorsteinson and

Brust, 1962; Patterson et al., 1969; Nayar and Sauerman, 1971; Harada et al., 1972; Straif and Beier, 1996; Gary and Foster, 2001). An exception to this is Aedes aegypti, which showed a higher age-specific survival, reproductive output and cumulative net replacement in females fed on human blood and water versus those fed on human blood and sugar (Harrington et al., 2001). Additionally, exclusive blood feeding conferred a fitness advantage to this species (Scott et al., 1997). Blood-feeding behavior continues to dominate the literature to date, partly due to its more direct association with egg

10 development in anautogenous mosquitoes and more significantly, since this is the conduit for the transmission of parasites and arboviruses (Bates, 1949), hence having a direct bearing on vectorial capacity. Indeed part of the reason for this lopsided focus is the historical debate on the actual importance and scope of sugar-feeding activity, with independent field observations favoring either side of the argument - the rare/facultative view (McCrae, 1989; Edman et al., 1992) or the ubiquitous/obligatory view (Hocking,

1953; Downes, 1958). One of the reasons for this conundrum lies in the difficulty of observing and collecting mosquitoes actively feeding on plant sugars in the field, with a majority of studies relying on trap-caught mosquitoes, which more often than not turn up fructose negative through anthrone testing but still, most species have high fructos- positive rates. These results could be explained by reviewing collection procedures, taking cognizance of digestion time of sugar resources and the time after trapping when mosquitoes are actually processed. Indeed factors such as the temperature, mosquito species, meal size and the quality of the sugar ingested will all have a bearing on the amount of sugar quantified in the mosquitoes (Andersson and Jaenson, 1987). Barring all challenges to the data collection methods, there will be great variability in the quantifiable units of fructose based on, but not limited to, the species of mosquito, the quality of plant sugars available in the area, the prevailing weather conditions (Foster,

1995) and season-evoked states (Robich and Denlinger, 2005).

In summary, most of what we know about carbohydrate dependency in mosquitoes has arisen in the past half century, but plenty remains to be elucidated on its relationship to blood feeding and the behavioral choices associated with a selection of a specific food source. It is clear that both happen to some extent, albeit temporally separated, but that there is also some degree of exclusivity of one activity over the other depending on the mosquito species and physiological demands. Both sexes of mosquitoes feed on plant sugars and it is the only food of male mosquitoes, regardless of species. Most mosquitoes need plant sugars to execute various essential functions, including flight and reproduction, both of which have a bearing on vectorial capacity in disease vectors. All ages of adult mosquitoes engage in sugar feeding activity, and females of all gonotrophic stages are not precluded from this activity.

Host plant location by mosquitoes

Studies have demonstrated that several factors are involved in the attraction and orientation of mosquitoes to vascular plants in nature; plant stimuli involved include in varying proportions visual, tactile, gustatory, and olfactory characteristics (Visser, 1986).

All these cues help insects to ―recognize‖ potential sources of reward and involve a complex interaction of physical, chemical and neuronal processes (Bruce et al., 2005).

As aptly stated by Dethier (1982) ―Recognition in the usual sense means that a particular stimulus or configuration of stimuli originating in the external world matches a model in the neural world and that upon the occurrence of congruency a specific relevant behavior 12 usually ensues‖. Among the most prominent drivers of this recognition are floral and nectar scent (Raguso, 2004), the presence of nectar and the color of the flowers

(Sandholm and Price, 1962; Thorsteinson and Brust, 1962; Grimstad and Defoliart, 1974;

Magnarelli, 1977; Jepson and Healy, 1988) communicating the location, abundance and quality of nectar and pollen and resulting in the attraction of insects to plants. Floral morphology is an important determinant of the accessibility of floral nectar by pollinators or nectar thieves, and as such, flower morphometry is useful in explaining flower choice by mosquitoes and other insects (Jürgens, 2006).

Plants produce volatile chemicals as a product of their natural metabolic processes, and these tend to be lipophilic compounds, but considerable water solubility has been demonstrated for some oxygenated monoterpenoids and glycosides. Terpenoid compounds are the largest group of plant chemicals (15 – 20K characterized), all having a common biosynthetic pathway in mevalonate (Langenheim, 1994). These compounds generally have molecular weights less than 250 and boiling points below 340° C (Metcalf and Kogan, 1987). Monoterpenes and sesquiterpenes are the main constituents of these volatiles, even though aromatic phenols, ketones, esters and simple alcohols are widely present. Floral odors are assumed to have evolved from defensive compounds, protecting plants from phytophagous insect attack, to attractive compounds that facilitate pollination by guiding insects to flowers through the association of the floral scent with a nectar reward (Fraenkel, 1959; Dethier, 1982; Chapman, 1988). Semiochemical odorants are

13 produced by four main types of structures on the surface of the plant: osmophores of flowers, glandular trichomes of leaves and stems, ducted oil cavities and oil cells of leaves and stems (Metcalf and Kogan, 1987). Osmophores generating odorants are common in the Aristolochiaceae, Araceae, Burmaniaceae, and Orchidaceae and also occur in other plant families. These structures exhibit great morphological diversity, with flaps, cilia, or brushes and are highly complex in the Orchidaceae (Metcalf and Kogan,

1987). Glandular trichomes are highly specialized glandular secretory cells found in the

Labiatae, Solanaceae, Compositae, and Geraniaceae. Trichomes secrete and accumulate a large variety of terpene oils and other essential oils that are generally involved as allomones in the antimicrobial, antifungal, and antiherbivoral protection of plants. Like osmophores, these structures are morphologically diverse, exhibiting pointed, hooked, and lobed hollow hairs.

Oil glands are small pores that secrete essential oils from the surface of the plant as they are produced through metabolic processes. Regardless of the organ that releases these semiochemicals, they reach the environment by various means, including diffusion through the air, leaching through rain and other water, exudation from the plant tissues, plant damage and senescence. Air temperature and diel cycle play an important role in the rate of production and release of these odorants, for instance orchids release their volatiles on warm sunny days (Metcalf and Kogan, 1987) whereas the flowers of Silene latifolia are nocturnal (Jürgens et al., 2002), emitting strong scents during the night and

14 are pollinated by nocturnal Lepidoptera species (Brantjes and Leemans, 1976). Insects encountering these volatiles as they are channeled through the environment may be activated, but must first filter out the background noise (other chemicals in the environment) and make the decisions on whether to orient themselves towards the emitting source. Odors emitted from a point source arrive at downwind points or to the perceiving insect as odor packets or pulses as a result of the natural turbulence of air or water, caused by the Brownian motion of particles and simple wave motion. Hence an upwind orientation is a prerequisite for the location of an odor from a point source

(Chapman, 1988), and it results in an optomotor-mediated anemotaxis (Cardé, 2007).

The frequency of turns is determined by the concentration of the odors in the packet, and the frequency with which the insect encounters theses packets as determined by the air speed.

Research Blueprint

This dissertation addresses various aspects of olfaction and the locating of a sugar meal by mosquitoes. Though various stimuli may be involved in host location and recognition as discussed above, this dissertation focuses mainly on the volatile chemicals released by plants, and I attempt to demonstrate mosquito orientation to these chemicals in the lab, semi-field and field experiments as a clear indication of the importance of olfaction and the potential for olfactory-based control approaches.

Chapter 2 deals with the chemicals that attract Culex pipiens mosquitoes to the flowers of common milkweed Asclepias syriaca. We demonstrate the process of creating and optimizing a synthetic milkweed blend, and show that only a few chemicals in the blend were very important in the overall attractiveness of the blend

Chapter 3 uses a similar approach as the previous chapter in creating a synthetic Canada goldenrod Solidago canadensis blend, but also examines changes that natural extracts undergo over prolonged storage. This chapter also addresses various challenges encountered in deploying a synthetic blend in field trials.

Chapter 4 highlights the development and testing of a new kind of olfactometer known as a diffusion-cage olfactometer, especially suited and specifically designed to test

Anopheles gambiae orientation to plant volatiles in semi-field settings such as screenhouses. This new olfactometer is tested alongside a more commonly used modified Y-tube olfactometer.

Chapter 5 deals with the attraction of Anopheles gambiae s.s. to plant volatiles in a screenhouse, hence building upon the previous chapter and going further to test an attractive toxic bait against the malaria vector by utilizing the dominant attractive power of a weed Parthenium hysterophorus and explores the use of artificial toxic bait stations against mosquitoes.

Appendix A shows some of the volatiles we have identified in the floral headspace of both tropical and temperate plants that mosquitoes visit for nectar.

Appendix B reports on preliminary data of maxillary palp electrophysiological responses of both Anopheles gambiae and Culex pipiens to a series of synthetic compounds of plant origin.

This research makes a good case for using floral odors in traps for surveillance purposes

(Mauer and Rowley, 1999; Jhumur et al., 2006), and builds upon studies that have shown promise to this end (Vargo and Foster, 1982; Geier and Boeckh, 1999; Mauer and

Rowley, 1999; Jhumur et al., 2006).

References

ANDERSSON, F. J., FERRRANDINO, F. J., MCKNIGHT, S., NOLEN, J., and

MILLER, J. 2009. A carbon dioxide, heat and chemical lure trap for the bedbug,

Cimex lectularius. Med. Vet. Entomol. 23:99-105.

ANDERSSON, I. H. and JAENSON, T. G. T. 1987. Nectar feeding by mosquitoes in

Sweden with special reference to Culex pipiens and Cx torrentium. Med. Vet.

Entomol. 1:59-64.

BATES, M. 1949. The natural history of mosquitoes. New York: The Macmillan Co. p

379 pp.

BENEDICT, M. Q., LEVINE, R. S., HAWLEY, W. A., and LOUNIBOS, P. 2007.

Spread of the tiger: global risk of invasion by the mosquito Aedes albopictus.

Vector-Borne Zoonot. 7:76-85.

BENTLEY, M. D., KAUFFMAN, P. E., KLINE, D. L., and HOGSETTE, J. A. 2009.

Response of adult mosquitoes to light-emitting diodes placed in resting boxes and

in the field. J. Am. Mosq. Cont. Assoc. 25:285-291.

BINDLINGMAYER, W. L. and HEM, D. G. 1973. Sugar feeding by Florida mosquitoes.

Mosq. News 33:535-538.

BRANTJES, N. B. M. and LEEMANS, J. A. A. M. 1976. Silene otites (Caryophyllaceae)

pollinated by nocturnal Lepidoptera and mosquitoes. Acta Bot. Neerl. 25:281-295.

BREELAND, S. G. and PICKARD, E. 1965. The malaise trap - an effecient and unbiased

mosquito collecting device. Mosq. News 25:19-21.

BRUCE, T. J. A., WADHAMS, L. J., and WOODCOCK, C. M. 2005. Insect host

location: a volatile situation. Trends Plant Sci. 10:269-274.

BURKETT, D. A., KLINE, D. L., and CARLSON, D. A. 1999. Sugar meal composition

of five north central Florida mosquito species (Diptera: Culicidae) as determined

by Gas Chromatography. J. Med. Entomol. 36:462-467.

CARDÉ, R. T. 2007. Odour plumes and odour-mediated flight in insects. In: Bock GR,

Cardew G, editors. Ciba Foundation Symposium 200 - Olfaction in Mosquito-

Host Interactions. Chichester, UK: John Wiley & Sons, Ltd.

CHAPMAN, R. F. 1988. Odors and the feeding behaviors of insects. In: Grimwade AM,

editor. ISI Atlas Sci-Anim Plant Sci. Philadelphia: Institute for Scientific

Information Inc. p 208-212.

COETZEE, M., CRAIG, M., and LE SUEUR, D. 2000. Distribution of African malaria

mosquitoes belonging to the Anopheles gambiae complex. Parasitol. Today

16:74-77.

COOPERBAND, M. F. and CARDÉ, R. T. 2006a. Comparison of plume structures of

carbon dioxide emitted from different mosquito traps. Medical and Veterinary

Entomology 20:1-10.

COOPERBAND, M. F. and CARDÉ, R. T. 2006b. Orientation of Culex mosquitoes to

carbon dioxide-baited traps: flight manoeuvres and trapping effeciency. Medical

and Veterinary Entomology 20:11-26.

DAVIS, J. D., HALL, T., CHEE, E. M., MAJALA, A., MINJAS, J., and SHIFF, C. J.

1995. Comparison of sampling anopheline mosquitoes by light-trap and human-

bait collections indoors at Bagamoyo, Tanzania. Med. Vet. Entomol. 9:249-255.

DEKKER, T., GEIER, M., and CARDÉ, R. T. 2005. Carbon dioxide instantly sensitizes

female yellow fever mosquitoes to human skin odours. J. Exp. Biol. 208:2963-

2972.

DETHIER, V. G. 1982. Mechanism of host-plant recognition. Ent. exp. & appl. 31:49-56.

DOWNES, J. A. 1958. The feeding habits of biting flies and their significance in

classification. Annu. Rev. Entomol. 3:249-266.

EDMAN, J. D., STRICKMAN, D., and KITTAYAPONG, P. 1992. Female Aedes

aegypti (Diptera : Culicidae) in Thailand rarely feed on sugar. J. Med. Entomol.

29:1035-1038.

FACCHINELLI, L., VALERIO, L., POMBI, M., REITER, P., COSTANTINI, C., and

DELLA TORRE, A. 2007. Development of a novel sticky trap for

containerbreeding mosquitoes and evaluation of its sampling properties to

monitor urban populations of Aedes albopictus. Med. Vet. Entomol. 21:183-195.

FEINSOD, F. M. and SPIELMAN, A. 1980. Nutrient-mediated juvenile hormone

secretion in mosquitoes. J. Insect Physiol. 26:113-117.

FERGUSON, H. M., DORNHAUS, A., BEECHE, A., BORGEMEISTER, C.,

GOTTLIEB, M., MULLA, M. S., GIMNIG, J. E., FISH, D., and KILLEEN, G. F.

2010. Ecology: A prerequisite for malaria elimination and eradication. PLoS Med

7:e1000303.

FOSTER, W. A. 1995. Mosquito sugar feeding and reproductive energetics. Annu. Rev.

Entomol. 40:443-74.

FOSTER, W. A. and HANCOCK, R. G. 1994. Nectar-related olfactory and visual

attractants for mosquitoes. J. Am. Mosq. Cont. Assoc. 10:288-296.

FRAENKEL, G. S. 1959. Raison d'être of secondary plant substances. Science 129:1466-

1470.

GARY, R. E. and FOSTER, W. A. 2001. Effects of available sugar on the reproductive

fitness and vectorial capacity of the malaria vector Anopheles gambiae (Diptera:

Culicidae). J. Med. Entomol. 38:22-28.

GARY, R. E. and FOSTER, W. A. 2004. Anopheles gambiae feeding and survival on

honeydew and extra-floral nectar of peridomestic plants. Med. Vet. Entomol.

18:102-107.

GEIER, M. and BOECKH, J. 1999. A new Y-tube olfactometer for mosquitoes to

measure the attractiveness of host odours. Entomol. Exp. Appl. 92:9-19.

GIMNIG, J. E., OMBOK, M., KAMAU, L., and HAWLEY, W. A. 2001. Characteristics

of larval anopheline (Diptera: Culicidae) habitats in western Kenya. Journal of

Medical Entomology 38:282-288.

GOUAGNA, L. C., POUEME, R. S., DABIRÉ , K. R., OUÉDRAOGO, J. B.,

FONTENILLE, D., and SIMRAD, F. 2010. Patterns of sugar feeding and host

plant preferences in adult males of An. gambiae (Diptera: Culicidae). J. Vector

Ecol. 35:267-276.

GOVELLA, N. J., CHAKI, P. P., MPANGILE, J. M., and KILLEEN, G. F. 2011.

Monitoring mosquitoes in urban Dar es Salaam: Evaluation of resting boxes,

window exit traps, CDC light traps, Ifakara tent traps and human landing catches.

Parasit. Vectors 4.

GRIMSTAD, P. R. and DEFOLIART, G. R. 1974. Nectar sources of Wisconsin

mosquitoes. J. Med. Entomol. 11:331-341.

HAMER, G. L., KITRON, U. D., BRAWN, J. D., LOSS, S. R., RUIZ, M. O.,

GOLDBERG, T. L., and WALKER, E. D. 2008. Culex pipiens (Diptera: 23

Culicidae): A bridge vector of West Nile virus to humans. J. Med. Entomol.

45:125-128.

HARADA, F., MORIYA, K., and YABE, T. 1972. Observations on the survival and

longevity of adult Culex and Aedes mosquitoes fed on flowers of some nectar

plants (II). Jap. J. Sanit. Zool. 23:141-154.

HARRINGTON, L. C., EDMAN, J. D., and SCOTT, T. W. 2001. Why do female Aedes

aegypti (Diptera: Culicidae) feed preferentially and frequently on human blood?

J. Med. Entomol. 38:411-422.

HIGA, Y., HOSHINO, K., TSUDA, Y., and KOBAYASHI, M. 2006. Dry-ice trap and

human bait collection of mosquitoes in the eastern part of Hokkaido, Japan. Med.

Entomol. Zool. 57:93-98.

HOCKING, B. 1953. The intrinsic range and speed of flight of insects. Trans. Roy.

Entomol. Soc. London 104:223-346.

HOLCK, A. R. and MEEK, C. L. 1991. Comparison of sampling techniques for adult

mosquitoes and other nematocera in open vegetation. J. Entomol. Sci. 26:231-236.

JAWARA, M., AWOLOLA, T. S., PINDER, M., JEFFRIES, D., SMALLEGANGE, R.

C., TAKKEN, W., and CONWAY, D. J. 2011. Field testing of different chemical

combinations as odour baits for trapping wild mosquitoes in The Gambia. PLoS

One 6:e19676.

JEPSON, P. C. and HEALY, T. P. 1988. The location of floral nectar sources by

mosquitoes: an advanced bioassay for volatile plant odours and initial studies with

Aedes aegypti (L.) (Diptera: Culicidae). Bull. Entomol. Res. 78.

JHUMUR, U. S., DÖTTERL, S., and JÜRGENS, A. 2006. Naïve and conditioned

responses of Culex pipiens pipiens Biotype molestus (Diptera: Culicidae) to

flower odours. J. Med. Entomol. 43:1164-1170.

JOHNSON, C. G. 1950. The comparison of suction trap, sticky trap and tow-net for the

quantitative sampling of small airborne insects. Ann. Appl. Biol. 37:268-285.

JOSEPH, S. R. 1970. Fruit feeding of mosquitoes in nature. Proc. N. J. Mosq. Exterm.

Assoc. 57:125-131.

JÜRGENS, A. 2006. Comparative floral morphometrics in day-flowering, night-

flowering and self-pollinated Caryophylloideae (Agrostemma, Dianthus,

Saponaria, Silene, and Vaccaria). Plant Syst. Evol. 257:233-250. 25

JÜRGENS, A., WITT, T., and GOTTSBERGER, G. 2002. Flower scent composition in

night-flowering Silene species (Caryophyllaceae). Biochem. Syst. Ecol. 30:383-

397.

KILPATRICK, A. M., KRAMER, L. D., JONES, M. J., MARRA, P. P., and DASZAK,

P. 2006. West Nile Virus epidemics in north America are driven by shifts in

mosquito feeding behavior. Plos Biol. 4:0606-0610.

KIRTON, U. and SPIELMAN, A. 1989. Suppression of transmission of malaria through

source reduction: Antianopheline measures applied in Israel, the United States

and Italy. Rev Infect Dis 11:391-406.

KLINE, D. L. 1999. Comparison of two American Biophysics mosquito traps: The

professional and a new counterflow geometry trap. J. Am. Mosq. Cont. Assoc.

15:276-282.

KLINE, D. L. 2002. Evaluation of various models of propane-powered mosquito traps. J.

Vector Ecol. 27:1-7.

KNOLS, B. G. and DE JONG, R. 1996. Limburger cheese as an attractant for the malaria

mosquito Anopheles gambiae s.s. Parasitol. Today 4:159-160.

LANGENHEIM, J. H. 1994. Higher plant terpenoids: A phytocentric overview of their

ecological roles. J. Chem. Ecol. 20:1223-1279.

LEAKE, C. J. 1992. Arbovirus-mosquito interactions and vector specificity. Parasitol.

Today 8:123-128.

LENGELER, C. and SNOW, R. W. 1996. From efficacy to effectiveness: Insecticide

treated bednets in Africa. Bulletin of the World Health Organisation 73:325-32.

MAGNARELLI, L. A. 1977. Nectar feeding by Aedes sollicitans and its relation to

gonotrophic activity. Environ. Entomol. 6:237-242.

MAGNARELLI, L. A. 1978. Nectar-feeding by female mosquitoes and its relation to

follicular development and parity. J. Med. Entomol. 14:527-530.

MAHANDE, A., MOSHA, F., MAHANDE, J., and KWEKA, E. 2007. Feeding and

resting behaviour of malaria vector, Anopheles arabiensis with reference to

zooprophylaxis. Malar. J. 6.

MAJAMBERE, S., PINDER, M., FILLINGER, U., AMEH, D., CONWAY, D. J.,

GREEN, C., JEFFRIES, D., JAWARA, M., MILLIGAN, P. J., HUTCHINSON,

R. and others. 2010. Is mosquito larval source management appropriate for

reducing malaria in areas of extensive flooding in The Gambia? A cross-over

intervention trial. Am J Trop Med Hyg 82:176-184.

MAUER, D. J. and ROWLEY, W. A. 1999. Attraction of Culex pipiens pipiens (Diptera:

Culicidae) to flower volatiles. J. Med. Entomol. 36:503-507.

MBOERA, L. E. G., KNOLS, B. G., BRAKS, M. A. H., and TAKKEN, W. 2000.

Comparison of carbon dioxide-baited trapping systems for sampling outdoor

mosquito populations in Tanzania. Med. Vet. Entomol. 14:257-263.

MCCRAE, A. W. R. 1989. Differences in sugar feeding activity between tropical and

temperate mosquitoes: field observations and their implications. Vector Ecol.

Newslett. 20:16(Abstr.).

MEERAUS, W. H., ARMISTEAD, J. S., and ARIAS, J. R. 2008. Field comparison of

novel and gold standard traps for collecting Aedes albopictus in northern Virginia.

J. Am. Mosq. Cont. Assoc. 24:244-248.

METCALF, R. and KOGAN, M. 1987. Plant volatiles as insect attractants. Crit. Rev.

Plant Sci. 5:251-301.

MUTUKU, F. M., BAYOH, M. N., GIMNIG, J. E., VULULE, J., KAMAU, L.,

WALKER, E. D., KABIRU, E. W., and HAWLEY, W. 2006. Pupal habitat

productivity of Anopheles gambiae complex mosquitoes in a rural village in

western Kenya Am J Trop Med Hyg 74:54-61.

NAYAR, J. K. and SAUERMAN, D. M., JR. 1971. The effects of diet on life-span,

fecundity, and flight potential of Aedes taeniorhynchus adults. J. Med. Entomol.

8:506-513.

NAYAR, J. K. and VAN HANDEL, E. 1971. The fuel for sustained mosquito flight. J.

Insect Physiol. 17:471-481.

NELSON, M. J. 1994. The role of sampling in vector control. Am J Trop Med Hyg

50:145-150.

NICODEM, J. G., CHAKI, P. P., MPANGILE, J. M., and KILLEEN, G. F. 2011.

Monitoring mosquitoes in urban Dar es Salaam: Evaluation of resting boxes,

window exit traps, CDC light traps, Ifakara tent traps and human landing catches.

Parasit. Vectors 4.

OKUMU, F., GOVELLA, N. J., MOORE, S. J., CHITNIS, N., and KILLEEN, G. F.

2010a. Potential benefits, limitations and target product-profiles of odor-baited

mosquito traps for malaria control in Africa. PLoS One 5:e11573.

OKUMU, F. O., KILLEEN, G. F., OGOMA, S., BISWARO, L., SMALLEGANGE, R.

C., MBEYELA, E., TITUS, E., MUNK, C., NGONYANI, H., TAKKEN, W. and

others. 2010b. Development and field evaluation of a synthetic mosquito lure that

is more attractive than humans. PLoS One 5:e8951.

OLANGA, E. A., OKAL, M. N., MBADI, P. A., KOKWARO, E. D., and MUKABANA,

W. 2010. Attraction of Anopheles gambiae to odour baits augmented with heat

and moisture. Malar. J. 96.

PATES, H. and CURTIS, C. 2005. Mosquito behavior and vector control. Ann. Rev.

Entomol. 50:53-70.

PATTERSON, R. S., SMITTLE, B. J., and DENEVE, R. T. 1969. Feeding habits of male

southern house mosquitoes on 32P-labelled and unlabelled plants. J. Econ.

Entomol. 62:1455-1458.

QIU, Y. T., SMALLEGANGE, R. C., LOON, J. J. A. V., and TAKKEN, W. 2011.

Behavioural responses of Anopheles gambiae sensu stricto to components of 30

human breath, sweat and urine depend on mixture composition and concentration.

Med. Vet. Entomol. 25:247-255.

RAGUSO, R. A. 2004. Why are some floral nectars scented? Ecology 85:1486-1494.

REEVES, W. C. 1990. Quantitative field studies on a carbon dioxide chemotropism of

mosquitoes. J. Am. Mosq. Cont. Assoc. 6:708-712.

REISEN, W. K., MEYER, R. P., and MILBY, M. M. 1986. Patterns of fructose feeding

by Culex tarsalis (Diptera: Culicidae). J. Med. Entomol. 23:366-373.

ROBICH, R. M. and DENLINGER, D. L. 2005. Diapause in the mosquito Culex pipiens

evokes a metabolic switch from blood feeding to sugar gluttony. Proc. Natl. Acad.

Sci. USA 102:1512-1517.

RUDOLFS, W. 1922. Chemotropism of mosquitoes. Bull New Jers Agric Exp Stn 367.

SAITOH, Y., HATTORI, J., CHINONE, S., NIHEI, N., TSUDA, Y., KURAHASHI, H.,

and KOBAYASHI, M. 2004. Yeast-generated CO2 as a convenient source of

carbon dioxide for adult mosquito sampling. J. Am. Mosq. Cont. Assoc. 20:261-

264.

SANDHOLM, H. A. and PRICE, R. D. 1962. Field observations on the nectar feeding

habits of some Minnesota mosquitoes. Mosq. News 22:346-349.

SCOTT, T. W., NAKSATHIT, A., DAY, J. F., KITTAYAPONG, P., and EDMAN, J. D.

1997. A fitness advantage for Aedes aegypti and the viruses it transmits when

females feed only on human blood. Am J Trop Med Hyg 57:235-239.

SERVICE, M. W. 1969. The use of traps in sampling mosquito populations. Ent. exp. &

appl. 12:403-412.

SERVICE, M. W. 1985. Mosquito Ecology: Field sampling methods. Elsevier, London

SERVICE, M. W. 1993. The role of ecology and ecological modeling in vector control.

Bull. Soc. Vector Ecol. 18:85-98.

SHAUKAT, A. M., BREMAN, J. G., and MCKENZIE, F. E. 2010. Using the

entomological inoculation rate to assess the impact of vector control on malaria

parasite transmission and elimination. Malar. J. 9.

SMALLEGANGE, R. C., KNOLS, B. G., and TAKKEN, W. 2010a. Effectiveness of

synthetic versus natural human volatiles as attractants for Anopheles gambiae

(Diptera: Culicidae) sensu stricto. J. Med. Entomol. 47:338-344.

SMALLEGANGE, R. C., SCHMIED, W. H., ROEY, K. J. V., VERHULST, N. O.,

SPITZEN, J., MUKABANA, W. R., and TAKKEN, W. 2010b. Sugar-fermenting

yeast as an organic source of carbon dioxide to attract the malaria mosquito

Anopheles gambiae. Malaria Journal 9.

STRAIF, S. C. and BEIER, J. C. 1996. Effects of sugar availability on the blood feeding

behavior of Anopheles gambiae (Diptera: Culicidae). J. Med. Entomol. 33:608-

612.

THORSTEINSON, A. J. and BRUST, R. A. 1962. The Influence of flower scents on

aggregations of caged adult Aedes aegypti. Mosq. News 22:349-352.

TIKASINGH, E. S. and DAVIES, J. B. 1972. Comparative test of four mosquito traps.

Mosq. News 32:623-628.

TURELL, M. J., DOHM, D. J., SARDELIS, M. R., O’GUINN, M. L., ANDREADIS, T.

G., and BLOW, J. A. 2005. An update on the potential of North American

mosquitoes (Diptera: Culicidae) to transmit West Nile Virus. J. Med. Entomol.

42:57-62.

VARGO, A. M. and FOSTER, W. A. 1982. Responsiveness of female Aedes aegypti

(Diptera: Culicidae) to flower extracts. J. Med. Entomol. 19:710-718. 33

VERHULST, N. O., MBADI, P. A., KISS, G. B., MUKABANA, W., LOON, J. J. A. V.,

TAKKEN, W., and SMALLEGANGE, R. C. 2011. Improvement of a synthetic

lure for Anopheles gambiae using compounds produced by human skin

microbiota. Malar. J. 10.

VISSER, J. H. 1986. Host odor perception in phytophagous insects. Ann. Rev. Entomol.

31:121-144.

WHO. 2010. World Malaria Report 2010. Geneva: World Health Organisation. 204 p.

XUE, R. D., DOYLE, M. A., and KLINE, D. L. 2008. Field evaluation of CDC and

Mosquito Magnet X traps baited with dry ice, CO2 sachet, and octenol against

mosquitoes. J. Am. Mosq. Cont. Assoc. 24:249-252.

YUVAL, B. 1992. The other habit: Sugar feeding by mosquitoes. Bull. Soc. Vector Ecol.

17:150-156.

Chapter 2

Analysis and Optimization of a Synthetic Milkweed Floral Attractant for Mosquitoes

Abstract

A pentane extract of common milkweed, Asclepias syriaca (Asclepiadaceae), flowers elicited significant orientation from both male and female Culex pipiens in a dual-port flight olfactometer. Analysis of the extract by gas chromatography-mass spectrometry

(GC-MS) revealed six major constituents: benzaldehyde, (E)-β-ocimene, phenylacetaldehyde, benzyl alcohol, nonanal and (E)-2-nonenal. Although not all were collected from the headspace profile of live flowers, a synthetic blend of these six compounds, when presented to mosquitoes in the same levels and proportions occurring in the extract, elicited a response comparable to the extract. Subtractive behavioral bioassays demonstrated that a three-component blend consisting of benzaldehyde, phenylacetaldehyde, and (E)-2-nonenal was as attractive as the full blend. These findings suggest the potential use of synthetic floral-odor blends for monitoring or control of both male and female disease-vectoring mosquitoes.

Introduction

The mosquito Culex pipiens is a member of a species complex that includes vectors of lymphatic filariasis and several arboviral diseases, such as St. Louis encephalitis (SLE),

Rift Valley fever (RFV) , and West Nile virus (WNV). In the northeastern quadrant of the

United States, Cx. pipiens recently has become important because of its role in the transmission of WNV, introduced in New York City in 1999 and is currently the most widespread arboviral disease in the U.S. (CDC, 2010). The blood-feeding pattern of Cx. pipiens (shifting from avian to human feeding within the transmission season) (Edman and Taylor, 1968; Kilpatrick et al., 2006), coupled with its high vector competence

(Turell et al., 2005) and ability to carry WNV through the winter, has allowed the virus to spread from its point of origin and maintain transmission throughout the mosquito’s range. It is for this reason that Cx. pipiens is one of the major targets for mosquito vector control and surveillance in the U.S. Current methods of monitoring by abatement districts and health departments include CDC light traps and CO2-baited traps, which mainly trap blood-seeking females, along with gravid traps, which target gravid females.

Plant-derived attractants have the potential to act as surveillance-trap lures and, when combined with poison-laced sugar solutions (Müller et al., 2010c), to control mosquito populations. Mosquitoes are attracted to various plant species, from which they obtain nectar and other juices. Previous studies have shown that sugar feeding by mosquitoes has a significant influence on dispersal (Hocking, 1953; Magnarelli, 1977) and vectorial

capacity (Gary and Foster, 2001; Gu et al., 2011), and contrary to long-held conjecture is required by both male and females throughout the adult stage (Downes, 1958; Yuval,

1992; Foster, 1995). Both sexes typically first visit plants soon after emergence. Males then require sugar at frequent intervals to maintain their energy reserves in order to join nightly mating swarms (Yuval et al., 1992). Females take sugar between blood meals, when they are digesting blood, or when they are gravid (Clements, 1999; Foster, 2008).

Furthermore, females of most temperate-climate Culex and Anopheles species enter reproductive diapause in late summer, after which they no longer are attracted to blood hosts but engage heavily in plant-sugar feeding (Bowen, 1992a).

Bates (1949) reported the successful trapping of anopheline mosquitoes using plant and fruit baits, whereas fruit proved to be an effective attractant in CDC traps for Cx. tarsalis

(Reisen et al., 1986). Sandholm and Price (1962) observed that mosquitoes in the field were attracted to light-colored flowers with distinct fragrances; odor appears to be primarily responsible for long-range attraction, with visual cues playing a role at shorter range (Thorsteinson and Brust, 1962; Healy and Jepson, 1988; Jepson and Healy, 1988).

Orientation to commercially obtained floral extracts and honey has been demonstrated for various mosquito species (Thorsteinson and Brust, (1962); Hancock and Foster, (1997);

Foster and Takken, (2004).

The potential for volatile attractants to lure mosquitos has long been recognized

(Jacobson and Beroza, 1963; Foster, 2008) and has been demonstrated in several

laboratory assays of natural plant extracts (Vargo and Foster, 1982; Jepson and Healy,

1988; Mauer and Rowley, 1999) and single floral compounds (Jhumur et al., 2006).

Schlein and Müller (2008) and Müller et. al. (2008; , 2010c) reported dramatic population reductions of Cx. pipiens and other mosquito species by spraying a fruit-based sugar bait containing insecticide on vegetation surrounding larval habitats. Light-less CDC traps baited with the blossoms of Tamarix jordanis were highly effective in trapping Cx. pipiens, and populations were reduced where these blossoms were treated with the insecticide Spinosad (Schlein and Müller, 2008a). Flower species eliciting the highest attraction probably vary by region, and by volatile profiles as seen in our screen of the volatiles of different temperate and tropical plants (Appendix A). In the northeastern

United States, mosquitoes are observed probing blossoms of common milkweed,

Asclepias syriaca, at rates disproportionate to their abundance relative to other flowers during mid-summer (Sandholm and Price, 1962; Grimstad and Defoliart, 1974; Yee et al.,

1992). The objectives of this study were to compare orientation of male and female Cx. pipiens to extracts of A. syriaca and to identify the odor components responsible.

Methods and Materials

Mosquitoes Experiments were conducted with Cx. pipiens from a colony established in

2009 from a larval development site near Columbus, Ohio, U.S.A. Larvae were identified at L4 by siphonal hair tufts. Colony adults were maintained in 41-liter clear acrylic cages on a diet of 10% sucrose, water, and weekly blood meals from the legs of a

rooster (ILACUC permit No. 2005A0054). Oviposition water was prepared by soaking grass clippings in aged tap water and allowing fermentation over a 3-day period.

Oviposition cups were placed with caged adults 3 days following each blood meal, and eggs were collected the following day. Two hundred first-instar larvae were placed into

22.8 x 33.0 cm aluminum pans with 450 mL of aged tap water. The larvae were fed a daily regimen of 50 mg, 100 mg, 300 mg, 300 mg, and 500 mg of finely ground

TetraMin® flakes, and pupae appeared on the 8th and 9th day post-hatching. Pupae were then counted and transferred to plastic cups and placed in a 41-liter cage supplied with water wicks. Emerging adults were given ad libitum access to water, but were deprived of sugar. Experiments were conducted 36 ± 12 h after emergence. The mosquito rearing and maintenance conditions were 27 ± 1 °C, 85 ± 5% RH, and 16:8 (L:D), with 30-min gradual crepuscular transitions between photophase and scotophase.

Chemicals Phenylacetaldehyde (90+%), benzaldehyde (≥99.5%, purified by redistillation), nonanal (≥95%), (Z)-jasmone (≥97%), (E)-2-nonenal (97%), and an alkane-standards mixture (C8 - C20) were all purchased from Sigma-Aldrich® (Saint

Louis, MO, USA). Benzyl alcohol (99.9%) was procured from Mallinckrodt Baker, Inc.

(Phillipsburg, NJ, USA). β-ocimene (70% (E)-β-ocimene) was synthesized by CHEMOS

GmbH (Regenstauf, Germany). Synthetics were diluted using HPLC grade n-pentane

(Fisher Scientific).

Extract Preparation Flowers of common milkweed, A. syriaca, were collected in early summer from The Ohio State University campus in Columbus, Ohio, U.S.A.

(40º00’18.95’’N 83º02’47.11’’W) and placed in an ice cooler for the 10-min transport to the laboratory. Single florets were plucked from the main milkweed umbel and separated from the calyx and other green parts of the flower, weighed, and placed into a 500 mL narrow-mouth glass Erlenmeyer flask. Milkweed florets were submerged in HPLC-grade n-pentane (Fisher Scientific) in a 1:8 ratio (w/v) and held for 24 hr at room temperature, at which time, the extract was decanted into 21-mL borosilicate glass vials, capped with

Teflon-lined screw caps and stored at -20ºC.

Headspace Analysis For characterization of its volatile profile, a single milkweed floret was placed into a 21-mL borosilicate glass vial with a Teflon-lined rubber septum.

Volatiles were allowed to equilibrate in a 30ºC chamber for 10 min before collections were made. Volatiles were collected using a divinylbenzene/carboxen/polydimethylsiloxane (DVB/CAR/PDMS) (50/30μm) solid phase micro-extraction (SPME) fiber (Supelco, Bellefonte, PA, USA). This mixed- chemistry fiber provides affinity for chemicals with a far broader range of polarity and volatility than PDMS alone. The fiber was introduced to the vial through the septum and exposed for 30 min. The collected volatiles were immediately analyzed using an Agilent

Technologies 6890 series gas chromatograph interfaced to an Agilent Technologies 5973 quadrupole mass selective detector. The SPME fiber was desorbed at 275ºC for 3 min

using a splitless injection onto a Zebron™ ZB-1ms column (30 m x 0.25 mm, 0.25µm phase thickness) (Phenomenex, Torrance, CA, USA). The oven temperature was held at

25ºC for 3 min, then ramped at 15ºC/min to 250ºC, where it was held for 2 min. The carrier gas was He at a flow rate of 1 mL/min. The mass selective detector was operated in EI mode at 70 eV, scanning 19-350 m/z with a quadrupole temperature of 180ºC and source temperature of 240 ºC.

Extract Analysis To guide the construction of a synthetic floral blend, pentane extracts of milkweed flowers were analyzed by gas chromatography/mass spectrometry (GC/MS) using a Finnegan Trace GC (Thermo Fisher Scientific, Inc., USA) coupled with a Trace

MSD system, with low resolution electron ionization operated at 70 eV, scanning 50-500 m/z at 1 scan/sec. A 1-µL volume of floral extract or synthetic blend was injected at

180ºC using a 1:10 split onto a Rtx®-5MS fused silica capillary column (30 m x 0.25 mm ID, 0.25 µm phase thickness) (Restek, Bellefonte, PA, USA). The temperature program was 40ºC for 1 min, ramped at 15ºC/min to 275ºC, with a 2 min final hold time.

The carrier gas was helium at a flow rate of 1 mL/min. Retention indices (RI) of compounds were determined relative to n-alkanes (C8–C20). Peak identifications were corroborated by re-injection on a second column (Zebron™ ZB-50, 15 m x 0.25 mm,

0.25 µm phase thickness) (Phenomenex, Torrance, California, U.S.A.) using the same injector and temperature program. Components of headspace and extract were identified by matching of mass spectra to NIST/EPA/NIH Mass Spectral Library 2005 and

comparison of retention indices to published values (Adams, 1995). Identity of major components was confirmed by comparison of spectra and retention times to authentic standards. Each component was quantified by comparison of its peak area in extracts to that of an internal standard; these quantities were used to create a synthetic odor blend.

Behavioral Assays All behavioral assays were conducted in a clear acrylic dual-port flight olfactometer with three main parts: an introduction chamber, flight chamber, and trapping ports. The introduction/release chamber was located at the downwind end, measuring 30 x 40 cm wide and 30 cm long. A sliding gate separated the release chamber from the main flight chamber (30 x 40 cm wide, 90 cm long), which had two cylindrical glass jar ports (15 cm long by 7 cm diameter) on its upwind end. The ports were fitted with borosilicate glass funnels, with the wide end opening into the flight chamber and the 3 mm diameter end pointing into a glass jar. Thus, mosquitoes entering the wide end of the funnel from the flight chamber were channeled into the jar, where they were retained. The trapping ports were located 11 cm above the flight chamber floor and were separated by 21 cm. The funnel and glass jar were held together by an airtight

Parafilm seal (Pechiney Plastic Company, Menasha, WI), which was replaced between treatments. A 7.5-mm diameter hole in the upwind end of each jar allowed the introduction of purified/humidified air after it had passed from an oil-free air pump through an activated carbon canister, then through a water column. Air flow was maintained at 0.05 L s-1 into each port, providing a velocity of 72 mm s-1 in the center of

the flight chamber. A black cotton cloth, covering a dampened layer of cotton wool, covered the entire floor of the flight chamber, to maintain a level of humidity similar to that within the choice ports. Data loggers (HOBO®, Onset Computer Corporation) recorded temperature and humidity in the olfactometer, which was 25.0-27.5 ºC and 75-

95% RH. An exhaust duct, connected at the downwind end of the release chamber, directed the effluent air out of the building through an exhaust hood.

The 16:8 (L:D) light cycle used during rearing of the mosquitoes was maintained in the bioassay room. Test materials were applied to 15-cm long cotton wicks (TIDI Products,

Neenah, WI) in aluminum weighing boats placed into the trapping jars. Pentane was used as a solvent control in all bioassays. The positions of the treatment and control ports were alternated between bioassays to eliminate possible positional bias. The olfactometer parts were cleaned with 70% ethanol, then water, after each experiment; gloves were used at all times to avoid contamination with human-related kairomones. At 2 h prior to scotophase, approximately 300 mosquitoes of both sexes in similar numbers were released through a sleeve connection, directly from an acrylic plastic cage (30 x 30 x 30 cm) into the holding/release chamber, where they were held for 15 min to acclimatize before the release gate was opened. After 12-h, the numbers of mosquitoes in the treatment port, the control port, and remaining in the flight chamber were recorded.

Subtractive bioassays were conducted on the synthetic blend as choice tests, in which the full six-component blend was presented in one of the paired olfactometer ports alongside a reducted five-component blend, removing a different chemical each time. After establishing the significance of each chemical on mosquito attraction, we compared the minimal attractive blend against the full six-component blend. This was followed by a dose-response study to determine the concentration eliciting the greatest response. Doses were presented in a randomized block design (n = 5) and response was calculated as the number in the treatment port relative to the number in the control port.

Video recording Mosquito flight behavior was recorded using three infrared RCA closed-circuit videocameras (Lancaster, PA) onto which was mounted a TC1824B wide- angle ES 25 mm 1:1.4 lens for flight chamber recordings or a TC1874C ES 75 mm 1:1.8 lens for each of the choice ports. The cameras were controlled in series using a Burle

Security Products TC8108 (Lancaster, PA) eight-channel switcher, with video output to two RCA TC1109 video monitors. Video output was recorded with an Emerson

EWV404 VCR (Parsippany, New Jersey) onto VHS tape in time-lapse mode, 20 seconds each minute.

Statistical analysis Olfactometer response was expressed as a proportion of mosquitoes trapped in the treatment port relative to the proportion collected in the control port.

Percent response data were analyzed by a goodness-of-fit chi-square test using SPSS v.

17 (SPSS Inc, Chicago, IL). In the dose-response experiment, a regression model of mosquito response to log-transformed blend dose was determined using the Fitted Line

Plot module of Minitab v. 16 (Minitab Inc., State College, PA). Experiments with whole milkweed flowers were conducted as in other bioassays, by using 2 g of florets as the test material, paired with blank controls.

Results

Chemical Identification Six compounds comprised >90% of the milkweed floral components in the pentane extract: (E)-β-ocimene, benzaldehyde, nonanal, benzyl alcohol, phenylacetaldehyde, and (E)-2-nonenal (Fig. 1A, Table 1). The volatile profile collected by DVB/CAR/PDMS SPME directly from fresh flowers was dominated by three of these compounds, (E)-β-ocimene, benzaldehyde, and phenylacetaldehyde (>75% of the total), but there were also substantial differences compared to the extract (Fig. 1B).

Most notably, (E,Z)- and (E,E)-alloocimene were collected from the flower headspace, but were almost completely absent from the extract, while nonanal, a major constituent of flower extract, was missing in the headspace. These disparities can be partly explained by variation in volatility among the components, but may also reflect actual differences between the chemical composition in flower tissue and what is released. The extract also contained higher-boiling hydrocarbons and fatty acids, but due to their low volatility, they were not included in behavioral studies. Based on the peak area of the internal standard, the total extractable volatiles of milkweed flower was estimated to be 0.03 mg/g

fresh weight (Table 1); 0.03 mg of total volatiles was estimated to have been applied when the flower extract was tested in the olfactometer bioassays below.

Flight Olfactometer Response Cx. pipens showed a significant response to milkweed flowers in the dual-port flight olfactometer, where 67% of released mosquitoes were captured in the flower-baited port, compared to only 6% in the control ports (Fig. 2A;

χ2=515.02 df= 1, P<0.0001). Despite the chemical differences measured between the headspace and extract profiles of milkweed flowers, mosquitoes similarly showed a greater response to a pentane extract of milkweed flowers (52%) than to pentane alone

(11%) (Fig. 2B; χ2 =61.44, df = 1, P<0.0001). Moreover, mosquitoes were observed in infrared video recordings probing the extract-treated cotton wick during the early scotophase and early photophase, suggesting that the extract also stimulated a feeding response. This behavior was never observed on the control wick.

Based on the positive response to milkweed flower extract, a six-component synthetic blend was formulated to simulate the composition and proportions of major constituents of the solvent extract (Fig. 1A). One discrepancy was caused by the high levels of (Z) - isomer present in synthetic (E)-β-ocimene, resulting in a synthetic blend that far exceeded the levels of this isomer in the floral extract. In the olfactometer, the synthetic floral blend performed very similar to the natural extract, with 48% of released mosquitoes trapped, compared to 16% for the control (Fig. 2(C); χ2=120.6 df= 1 P<0.0001). Again,

there was vigorous probing on the treated wicks, but not on the controls. Males and females showed similar response to milkweed floral odors in all three experiments as the sex ratio of mosquitoes captured in either the treatment or control ports did not deviate significantly from 1:1.

A subtractive bioassay of the synthetic blend indicated a significant role for three compounds in Cx. pipiens response to milkweed: benzaldehyde, phenylacetaldehyde, and

(E)-2-nonenal (Table 2). When any of these compounds was removed, mosquitoes showed a significant preference for the full blend over the reducted blend. When nonanal was removed, mosquitoes showed no preference between the full and the five-component blend, and removal of β-ocimene or benzyl alcohol resulted in a higher response. The activity of benzaldehyde, phenylacetaldehyde, and (E)-2-nonenal was confirmed when the three were combined: the three-component blend captured 31% of mosquitoes compared to only 7% for the control (Fig. 3A; χ2=196.56, df=1, P<0.001) and was as active as the full blend (Fig. 3B 4; χ2 =3.02, df =1, P = 0.082).

Because quantity as well as quality can determine the intensity of chemically mediated behavior, Cx. pipiens response was measured in the flight olfactometer to different doses of the three-component blend compared to a solvent control. There was a significant positive quadratic response to log(dose) (y = -8.4644 + 51.191x - 21.364x2; R2 = 58.1%), with a response maximum predicted at 18 µg. The response flattened out at both the

lowest and highest doses, but remained positive (Fig. 4). Removal of the two most extreme doses produced a stronger fitting regression model (y = -148.6 + 194.1x -

53.55x2; R2 = 93.1%), with a similar optimal dose of 65 µg.

Discussion

The positive upwind response by the mosquito Cx. pipiens to floral volatiles of the common milkweed is demonstrated in this study and suggests a new source of lures for monitoring populations and possibly for control. Males and females responded at similar levels to milkweed flowers, a pentane extract, and a synthetic blend of the extract’s major constituents. Starting with a six-component blend that generally mimicked the level and proportionality of the extract’s major constituents, subtractive bioassays demonstrated that three of the components were required for maximum response. The response to whole flowers was nonetheless stronger than it was to the extract, to the full blend, and to the minimal blend.

The attraction of mosquitoes to plant volatiles has been demonstrated in field and laboratory experiments, yet few studies have attempted to formulate synthetic blends of them. In the only previous study of common milkweed volatiles, Mauer and Rowley

(1999) demonstrated attraction of Cx. pipiens to a methylene chloride extract in a dual port olfactometer, but a synthetic blend of the two dominant compounds trapped from the headspace of the extract on PDMS SPME, benzyl alcohol and 2-phenylethanol, failed to attract the mosquitoes. We did not detect 2-phenylethanol in pentane extracts or volatile collections of milkweed flowers, and benzyl alcohol was not active in our flight olfactometer bioassays. Bowen (1992b) discovered narrowly tuned antennal receptors of

Cx. pipiens to be highly sensitive to the bicyclic terpene thujone, a prominent component

of tansy (Tanecetum) headspace and attractive to moths (Gabel et al., 1992); however, it elicited no response by the mosquitoes in an olfactometer. The only flower-based synthetic blend that has demonstrated attraction for mosquitoes in an olfactometer was developed from Silene otites and consists of phenylacetaldehyde, veratrole, and 2- methoxyphenol, only the first of which elicits a significant response on its own in Cx. pipiens var. molestus (Jhumur et al., 2006). Both Tanecetum (Andersson and Jaenson,

1987) and Silene (Brantjes and Leemans, 1976) flowers attract mosquitoes in the field.

Phenylacetaldehyde was also a prominent component of our milkweed headspace and solvent analyses, and it was an essential part of the minimal synthetic blend.

Syed and Leal (2009) demonstrated nonanal response in ca. 40% of the olfactory receptor neurons of Cx. quinquefasciatus, and they reported low, but significant catch of Cx. quinquefasciatus by nonanal-baited field traps compared to control traps. When nonanal was presented with CO2, there was a 50-66% higher catch than with CO2-baited control traps, demonstrating a synergistic effect. In Anopheles gambiae, nonanal from skin emanations did not induce electrophysiological activity (Meijerink et al., 2000). A similar study showed that nonanal elicited a maximum response in Cx. quinquefasciatus at 0.01 µg and that higher concentrations, though better than a solvent control, led to a drop in response (Puri et al., 2006). In studies with Aedes aegypti at comparable concentrations, it was observed that though nonanal elicited electrophysiological activity, the compound was associated with both reduced flight activity and attraction in a Y-tube

olfactometer, even though probing activity was not adversely affected (Logan et al.,

2008). Our subtractive flight olfactometer bioassay of the synthetic milkweed flower blend did not suggest any contribution to response by nonanal. However, the concentration of nonanal in our synthetic milkweed blend was about 1,000 times the amount reported to be optimal for Cx. quinquefasciatus (Puri et al., 2006). Alternatively, given the interaction with CO2 demonstrated by Syed and Leal (2009), the behavioral and neurophysiological activity of nonanal discussed above may be related to host-finding by blood-feeding mosquito, not by the sugar-seeking mosquitoes in our experiments. The removal of benzyl alcohol from the full blend significantly enhanced its attractiveness

(22% vs. 32% response). Jhumur et. al. (2007) reported that this chemical was not very attractive to Cx. pipiens var. molestus, but Puri et al. (2006) found benzyl alcohol to be attractive to Cx. quinquefasciatus at 10 µg in laboratory assays, which is similar to the amount used in the present study.

We found notable differences between profiles of the pentane extract and the SPME volatile collection of milkweed flower. Extracts were dominated by phenylacetaldehyde,

(E)-β-ocimene, and nonanal, while the headspace profile contained primarily benzaldehyde, ocimene isomers, and phenylacetaldehyde. Disparities between extracts and chemicals collected in the headspace can be explained by differences in the physico- chemical properties of the constituents, deep penetration by solvents to extract compounds that are not normally released by the tissue, and/or selective trapping of

chemicals by SPME. In this study, differences between the methods are largely consistent with chemical differences in vapor pressure. Benzaldehyde, β-ocimene, and alloocimene have the highest vapor pressures of all the chemicals identified from milkweed, and they make up most of the volatile profile. For example, the vapor pressure of benzaldehyde is almost twice that of nonanal, almost four times that of (E)-2-nonenal, and more than six times that of benzyl alcohol. The latter three compounds were either absent or found in very low levels in the headspace analysis. Phenylacetaldehyde is intermediate in its volatility, but was also the component found in the highest levels in the pentane extract.

Volatile collection analyses can also produce misleading results, as they may not accurately reflect the actual proportionality of constituents in the headspace. We do not believe this was a major factor for explaining the differences in the headspace and extract profiles of milkweed due to our choice of SPME phase. Although PDMS has been the most widely used SPME phase, it is actually a poor choice for characterizing plant volatile profiles containing constituents with a range of volatilities and functional groups.

In preliminary studies (not shown), we found a broader array of volatile compounds was trapped from flowers by DVB/CAR/PDMS compared to PDMS. This mixed-bed fiber not only employs a broader range of phase polarity, but also incorporates both adsorption and partitioning as mechanisms of collection (Koziel and Novak, 2002). A number of recent studies have quantified the differential trapping efficiency of SPME phases (Cui et al., 2009; Ferreira et al., 2009; Zhang et al., 2009), and have demonstrated the high

recovery efficiency and linearity by DVB/CAR/PDMS compared to other phases for all of the compound classes we identified from milkweed. For example, Zhang et al. (2009) found that volatile profiles of longan fruit extracted by DVB/CAR/PDMS were 3x higher in terpenes, 5x higher in alcohols, and 14x higher in esters than those produced by a 100

µm PDMS fiber, and revealed volatile carbonyls and acids, which were completely absent from PDMS profiles.

This study points to floral odors as a potential new source of chemical lures useful for mosquito sampling or control. When formulated at the appropriate release rate and proportion, synthetic chemical blends could be used in trapping devices to sample adult populations. Relative to animal-derived odors, floral odors have the advantage of attracting both sexes of mosquitoes and females in all gonotrophic states and in reproductive diapause. Given that Cx. pipiens visit a variety of flowers for nectar- feeding, the three compounds identified here are likely not the only floral components attractive to them, and more effective blends may remain to be discovered. It also remains to be seen whether different mosquito species use similar chemical cues. Future research should seek additional attractants and determine optimal blend release rates, delivery systems, and trap design for maximizing capture in the field.

Acknowledgements – We thank Robert Aldridge for help in rearing mosquitoes and

Ephantus Muturi for helpful suggestions in an early draft of this manuscript. This research was supported through NIH grant R01-AI064506.

References

ADAMS, R. P. 1995. Identification of Essential Oil Components by Gas

Chromatography/Mass Spectroscopy. Allured Publishing, Carol Stream, IL.

ANDERSSON, I. H. and JAENSON, T. G. T. 1987. Nectar feeding by mosquitoes in

Sweden with special reference to Culex pipiens and Cx torrentium. Med. Vet.

Entomol. 1:59-64.

BATES, M. 1949. The natural history of mosquitoes. New York: The Macmillan Co. p

379 pp.

BOWEN, M. F. 1992a. Patterns of sugar feeding in diapausing and nondiapausing Culex

pipiens (Diptera: Culicidae) females J. Med. Entomol. 29:843-849.

BOWEN, M. F. 1992b. Terpene-sensitive receptors in female Culex pipiens mosquitoes:

Electrophysiology and behaviour. J. Insect Physiol. 38:759-764.

BRANTJES, N. B. M. and LEEMANS, J. A. A. M. 1976. Silene otites (Caryophyllaceae)

pollinated by nocturnal Lepidoptera and mosquitoes. Acta Bot. Neerl. 25:281-295.

CDC. 2010. Surveillance for Human West Nile Virus Disease — United States, 1999–

2008. MMWR CDC Surveill Summ 59(No. SS-2) [April 2, 2010].

CLEMENTS, A. N. 1999. The biology of mosquitoes. Sensory Reception and Behavior.

CABI Publishing, New York

CUI, S., TAN, S., OUYANG, G., JIANG, S., and PAWLISZYN, J. 2009. Headspace

solid-phase microextraction gas chromatography–mass spectrometry analysis of

Eupatorium odoratum extract as an oviposition repellent. J. Chromatogr. B

877:1901-1906.

DOWNES, J. A. 1958. The feeding habits of biting flies and their significance in

classification. Annu. Rev. Entomol. 3:249-266.

EDMAN, J. D. and TAYLOR, D. J. 1968. Culex nigripalpus: Seasonal shiftin the bird-

mammal feeding ratio in a mosquito vector of human ecephalitis. Science 161:67-

68.

FERREIRA, L., PERESTRELO, R., CALDEIRA, M., and CÂMARA, J. S. 2009.

Characterization of volatile substances in apples from Rosaceae family by

headspace solid-phase microextraction followed by GC-qMS. J. Sep. Sci.

32:1875-1888.

FOSTER, W. A. 1995. Mosquito sugar feeding and reproductive energetics. Annu. Rev.

Entomol. 40:443-74. 55

FOSTER, W. A. 2008. Phytochemicals as population sampling lures. J. Am. Mosq. Cont.

Assoc. 24.

FOSTER, W. A. and TAKKEN, W. 2004. Nectar-related vs. human-related volatiles:

behavioural response and choice by female and male Anopheles gambiae

(Diptera: Culicidae) between emergence and first feeding. Bull. Entomol. Res.

94:145-157.

GABEL, B., THIÉRY, D., SUCHY, V., MARION-POLL, F., HRADSKY, P., and

FARKAS, P. 1992. Floral volatiles of Tanacetum vulgare L. attractive to

Lobesia botrana DEN. et SCHIFF. females J. Chem. Ecol. 18:693-701.

GARY, R. E. and FOSTER, W. A. 2001. Effects of available sugar on the reproductive

fitness and vectorial capacity of the malaria vector Anopheles gambiae (Diptera:

Culicidae). J. Med. Entomol. 38:22-28.

GRIMSTAD, P. R. and DEFOLIART, G. R. 1974. Nectar sources of Wisconsin

mosquitoes. J. Med. Entomol. 11:331-341.

GU, W., MÜLLER, G., SCHLEIN, Y., NOVAK, R. J., and BEIER, J. C. 2011. Natural

plant sugar sources of Anopheles mosquitoes strongly impact malaria

transmission potential. PLoS ONE 6:e15996. 56

HANCOCK, R. G. and FOSTER, W. A. 1997. Larval and adult nutritional effects on

blood/nectar choice of Culex niagripalpus mosquitoes. Med. Vet. Entomol.

11:112-122.

HEALY, T. P. and JEPSON, P. C. 1988. The location of floral nectar sources by

mosquitoes: the long-range responses of Anopheles arabiensis Patton (Diptera:

Culicidae) to Achillea millefolium flowers and isolated floral odour. Bull.

Entomol. Res. 78:651-657.

HOCKING, B. 1953. The intrinsic range and speed of flight of insects. Trans. Roy.

Entomol. Soc. London 104:223-346.

JACOBSON, M. and BEROZA, M. 1963. Chemical insect attractants. Science 140:1367-

1373.

JEPSON, P. C. and HEALY, T. P. 1988. The location of floral nectar sources by

mosquitoes: an advanced bioassay for volatile plant odours and initial studies with

Aedes aegypti (L.) (Diptera: Culicidae). Bull. Entomol. Res. 78.

JHUMUR, U. S., DÖTTERL, S., and JÜRGENS, A. 2006. Naïve and conditioned

responses of Culex pipiens pipiens Biotype molestus (Diptera: Culicidae) to

flower odours. J. Med. Entomol. 43:1164-1170. 57

JHUMUR, U. S., DÖTTERL, S., and JÜRGENS, A. 2007. Electrophysiological and

behavioural responses of mosquitoes to volatiles of Silene otites

(Caryophyllaceae). Arthropod-Plant Inte. 1:243-254.

KILPATRICK, A. M., KRAMER, L. D., JONES, M. J., MARRA, P. P., and DASZAK,

P. 2006. West Nile Virus epidemics in north America are driven by shifts in

mosquito feeding behavior. PLoS Biol. 4:0606-0610.

KOZIEL, J. A. and NOVAK, I. 2002. Sampling and sample-preparation strategies based

on solid-phase microextraction for analysis of indoor air Trends Anal. Chem.

21:840-850.

LOGAN, J. G., BIRKETT, M. A., CLARK, S. J., POWERS, S., SEAL, N. J.,

WADHAMS, L. J., (LUNTZ), A. J. M., and PICKETT, J. A. 2008. Identification

of human-derived volatile chemicals that interfere with attraction of Aedes aegypti

mosquitoes. J. Chem. Ecol. 34:308-322.

MAGNARELLI, L. A. 1977. Nectar feeding by Aedes sollicitans and its relation to

gonotrophic activity. Environ. Entomol. 6:237-242.

MAUER, D. J. and ROWLEY, W. A. 1999. Attraction of Culex pipiens pipiens (Diptera:

Culicidae) to flower volatiles. J. Med. Entomol. 36:503-507. 58

MEIJERINK, J., BRAKS, M. A. H., BRACK, A. A., ADAM, W., DEKKER, T.,

POSTHUMUS, M. A., BEEK, T. A. V., and LOON, J. J. A. V. 2000.

Identification of olfactory stimulants for Anopheles gambiae from human sweat

samples. J. Chem. Ecol. 26:1367-1381.

MÜLLER, G. C., JUNNILA, A., and SCHLEIN, Y. 2010. Effective control of adult

Culex pipiens by spraying an attractive toxic sugar bait solution in the vegetation

near larval habitats. J. Med. Entomol. 47:63-66.

MÜLLER, G. C., KRAVCHENKO, V. D., and SCHLEIN, Y. 2008. Decline of

Anopheles sergentii and Aedes caspius populations following presentation of

attractive toxic (Spinosad) sugar bait stations in an oasis. J. Am. Mosq. Cont.

Assoc. 24:147-149.

PURI, S. N., MENDKI, M. J., SUKUMARAN, D., GANESAN, K., PRAKASH, S., and

SEKHAR, K. 2006. Electroantennogram and behavioral responses of Culex

quinquefasciatus (Diptera: Culicidae) females to chemicals found in human skin

emanations. J. Med. Entomol. 43:207-213.

REISEN, W. K., MEYER, R. P., and MILBY, M. M. 1986. Patterns of fructose feeding

by Culex tarsalis (Diptera: Culicidae). J. Med. Entomol. 23:366-373.

SANDHOLM, H. A. and PRICE, R. D. 1962. Field observations on the nectar feeding

habits of some Minnesota mosquitoes. Mosq. News 22:346-349.

SCHLEIN, Y. and MÜLLER, G. C. 2008. An approach to mosquito control. Using the

dominant attraction of flowering Tamarix jordanis trees against Culex pipiens. J.

Med. Entomol. 45:384-390.

SYED, Z. and LEAL, W. 2009. Acute olfactory response of Culex mosquitoes to a

human - and bird - derived attractant. Proc. Natl. Acad. Sci. USA 106:18803-

18808.

THORSTEINSON, A. J. and BRUST, R. A. 1962. The Influence of flower scents on

aggregations of caged adult Aedes aegypti. Mosq. News 22:349-352.

TURELL, M. J., DOHM, D. J., SARDELIS, M. R., O’GUINN, M. L., ANDREADIS, T.

G., and BLOW, J. A. 2005. An update on the potential of North American

mosquitoes (Diptera: Culicidae) to transmit West Nile Virus. J. Med. Entomol.

42:57-62.

VARGO, A. M. and FOSTER, W. A. 1982. Responsiveness of female Aedes aegypti

(Diptera: Culicidae) to flower extracts. J. Med. Entomol. 19:710-718.

YEE, W. L., FOSTER, W. A., HOWE, M. J., and HANCOCK, R. G. 1992. Simultaneous

field comparison of evening temporal distributions of nectar and blood feeding by

Aedes vexans and Ae. trivittatus (Diptera: Culicidae) in Ohio. J. Med. Entomol.

29:356-360.

YUVAL, B. 1992. The other habit: Sugar feeding by mosquitoes. Bull. Soc. Vector Ecol.

17:150-156.

YUVAL, B., HOLLIDAY-HANSON, M. L., and WASHINO, R. K. 1992. Energy budget

of swarming male mosquitoes. Ecol. Entomol. 19:74-78.

ZHANG, Y., GAO, B., ZHANG, M., SHI, J., and XU, Y. 2009. Headspace solid-phase

microextraction–gas chromatography–mass spectrometry analysis of the volatile

components of longan (Dimocarpus longan Lour.). Eur. Food Res. Technol.

229:457-465.

Figure 2.1. Overlay of (A) milkweed pentane extract (broken line) and synthetic blend

(solid line) total ion chromatograms generated on a Zebron™ ZB-50 capillary column

(Peak # 1. Unidentified 2. (Z)-β-Ocimene 3. (E)-β-Ocimene 4. Benzaldehyde 5.

Unidentified 6. Nonanal

7. Benzyl Alcohol 8. Unidentified 9. Phenylacetaldehyde 10. (E)-2-Nonanal 11. 2,6-

Nonadienal

12. Phenylethanol) and (B) total ion chromatogram of the headspace profile of a single milkweed floret captured on a divinylbenzene/carboxen/polydimethylsiloxane SPME fiber and analyzed on a Zebron™ ZB-1 column (Peak # 1. Benzaldehyde 2. Myrcene 3.

Monoterpene 4. Phenylacetaldehyde 5. (Z)-β-Ocimene 6. (E)-β-Ocimene 7. γ-Terpinene

8. Monoterpene 9. Dimethylstyrene 10. Monoterpene 11. Monoterpene 12. (E,Z)-

Alloocimene 13. (E,E)-Alloocimene).

Figure 2.2. Responses of all Culex pipiens released in a dual-port flight olfactometer to: (A) whole milkweed flowers (B)

pentane extract of milkweed and to (C) synthetic milkweed blend compared against a water (A) or solvent (B,C) control.

P<0.0001 (***)

Figure 2.3. A) Response of all Culex pipiens released in a dual-chioce flight olfactometer to a three-component blend, consisting of benzaldehyde, (E)-2-nonenal, and phenylacetaldehyde. (**) indicates significance at P<0.001; B) Choice bioassay showing no difference (P=0.082) in responses of mosquitoes to a full milkweed blend and a minimal (three-component) blend.

Figure 2.4. Dose-response bioassays of various concentrations of a three-component blend (x-axis) by mosquitoes in a dual-port olfactometer (y-axis indicates the blank- corrected percent response).

Table 2.1. Volatile composition of solvent extracts of the Common Milkweed Asclepias syriaca and the quantification of extract components.

Relative Concentration in Synthetic

Retention Flower Blend

COMPOUND Index (µg/g)a (µg/ml)

Benzaldehyde 974 6.9 7.5

(E)-β-Ocimene c 1043 2.9 3.1

Benzyl Alcohol 1052 6.3 4.4

Phenylacetaldehyde 1058 3.4 5.0

Nonanal 1076 10.0 11.3

(E)-2-Nonenal 1171 2.8 2.5

aidentity established by comparison to authentic of each compound, quantity estimated by comparison of peak area to (Z)-jasmone internal standard c synthetic β-ocimene contained a mixture of isomers

Table 2.2. Choice response of Culex pipiens in a flight olfactometer to a six-component synthetic blend of milkweed flower volatiles compared to reducted blends, in which each component is removed individually.

COMPOUND REPS FULL BLEND REDUCTED P

MISSING FROM (mean±SE) BLEND

BLEND (mean±SE)

Benzaldehyde 6 339 ± 18 > 139 ± 7 <0.0001*

(E)/(Z)-β-Ocimeneǂ 5 195 ± 17 < 285 ± 24 <0.0001

Benzyl Alcohol 5 390 ± 9 < 577 ± 18 <0.0001

Phenylacetaldehyde 6 284 ± 13 > 101 ± 5 <0.0001*

Nonanal 5 251 ± 18 = 272 ± 16 0.3585

(E)-2-Nonenal 4 196 ± 24 > 131 ± 13 0.0003 *

*indicates compounds selected for formulation of three component blend

ǂ 70 - 75% (E)-β-ocimene

Chapter 3

From Natural Extracts to Synthetic Blends of Canada goldenrod Solidago canadensis:

Key considerations and challenges to the effective trapping of mosquitoes in the field

Abstract

In this study, we analyzed the headspace profiles of fresh and two-year-old extracts of

Canada goldenrod, Solidago canadensis, flowers, conducted behavioral assays with the northern house mosquito Culex pipiens in a dual-port olfactometer to test the extracts, and created a synthetic blend using the most attractive extract. We then tested, in

Counter Flow Geometry (CFG) traps, floral cuttings of goldenrod, a synthetic blend, and an extract against unbaited traps in field experiments. Our results indicate that prolonged periods of storage may alter the chemical compositions of natural extracts, but we did not observe a drop in the bioactivity of the extract. We further report that a synthetic blend modeled against our most potent extract was significantly more attractive than a solvent control in laboratory-based olfactometer experiments, but in the field, there was no difference in captures by traps baited with floral cuttings, floral extract, or left unbaited.

The present study explores possible explanations for this incongruence and provides recommendations for future studies.

Introduction

Natural extracts are often the first step in creating synthetic attractants used to measure the orientation of insects towards plant-derived volatile organic compounds due to their ease of isolation and storage. These extracts are useful in the quantification of the floral headspace and are often used in behavioral assays to measure the activity of different fractions of plant volatiles. Plant extracts have the added benefit of continuous availability and can be used in assays when live plants are unavailable. Natural and synthetic volatile attractants represent a novel and interesting frontier in the surveillance and control of mosquitoes, with successful lure and kill applications already demonstrated in some areas (Müller and Schlein, 2006; Müller et al., 2008; Müller et al.,

2010b). Traps baited with flowering plants have recorded significantly higher (up to 130 times) catches than control traps (Müller and Schlein, 2006), hence highlighting the potential of using plant-derived volatiles in trapping mosquitoes or luring them to kill stations. The use of attractive synthetic blends for the control of pest species has been successfully used to eliminate tsetse flies from vast regions in Zimbabwe (Vale, 1993;

Torr, 1994), and there is all indication that through further basic research and as part of an integrated approach, similar success can be achieved for mosquitoes. Yet a lot still needs to be done to fully exploit the untapped potential of plant attractants (Foster, 2008) for meaningful levels of vector control to be achieved. Plant attractants, are mainly terpenoid compounds, fatty acids, esters, aldehydes and ketones that were evolutionarily used by plants as defensive compounds against herbivores and microbial attack, but that

have now become attractants to pollinators through co-evolution (Chapman, 1988).

Though limited, our current knowledge about volatile compounds that attract mosquitoes is growing rapidly as better tools for analysis, including volatile capture and analysis

(Appendix A) are developed and greater interest is shown to this approach due to its potential benefits in disease vector control. The current study focuses on volatile organic compounds produced by the flowers of Canada goldenrod, Solidago canadensis, a late summer to early fall bloomer in central Ohio which is frequented by various mosquito species, including Culex pipiens, a major vector of West Nile virus and lymphatic filariasis.

Materials and Methods

Mosquitoes

Experiments were conducted with Cx. pipiens from a colony established in 2009 from a larval development site near Columbus, Ohio, U.S.A. Larvae were identified at L4 by siphonal hair tufts. Colony adults were maintained in 41-liter clear acrylic cages on a diet of 10% sucrose, water, and weekly blood meals from the legs of a rooster (ILACUC permit No. 2005A0054). Oviposition water was prepared by soaking grass clippings in aged tap water and allowing fermentation over a 3-day period. Oviposition cups were placed with caged adults 3 days following each blood meal, and eggs were collected the following day. Two hundred first-instar larvae were placed into 22.8 x 33.0 cm aluminum pans with 450 mL of aged tap water. The larvae were fed a daily regimen of 50

mg, 100 mg, 300 mg, 300 mg, and 500 mg of finely ground TetraMin® flakes, and pupae appeared on the 8th and 9th day post-hatching. Pupae were then counted and transferred to plastic cups and placed in a 41-liter cage supplied with water wicks. Emerging adults were given ad libitum access to water, but were deprived of sugar. Experiments were conducted 36 ± 12 h after emergence. The mosquito rearing and maintenance conditions were 27 ± 1 °C, 85 ± 5% RH, and 16:8 (L:D), with 30-min gradual crepuscular transitions between photophase and scotophase.

Sample preparation

Flowers of Canada goldenrod Solidago canadensis were collected in the late summer and early autumn from a field location West of the Ohio State University main campus

(40º00’18.95’’N 83º02’47.11’’W) in 2006 and placed in an ice cooler for transportation to the laboratory, approximately ten minutes away. Flowers were weighed and placed into a 500ml narrow-mouth glass Erlenmeyer flask (Thermo Fisher Scientific, Inc) then submerged in HPLC grade n-pentane (Fisher Scientific) in 1:8 ratio (w/v), tightly sealed and left at room temperature for a 24 hour extraction period. A yellow to golden extract was then decanted into 6 dram borosilicate glass vials, capped with Teflon lined screw caps and stored at -40º C.

Standards

Analytical grade standards of α-pinene, β-pinene, β-myrcene, α-phellandrene, limonene,

β-ocimene, terpinolene, bornyl acetate and a standard mixture of alkanes (C21-C36) were all purchased from Sigma-Aldrich® (Saint Louis, MO, USA). Dilutions were done using

HPLC grade n-pentane (Fisher Scientific).

Extract and synthetic blend analysis

GC/MS analysis was conducted using an Agilent 6890 series GC (Agilent Technologies,

Little Falls, DE, USA) coupled with a 5973 MSD system equipped with a 7683B series spilt/splitless injector. A 1-µL volume of floral extract or synthetic blend was injected at

180° C and a 1:10 split into a Zebron™ ZB-1 dimethylpolysiloxane fused-silica column

(30 m x 0.25 mm ID, 0.25 µm phase thickness). The oven temperature was held at 25ºC for 3 min, then ramped at 15ºC per minute to 275ºC, where it was held isothermally for 3 min. Peak identifications were corroborated by re-injection on a second column installed in a Finnegan Trace GC (Thermo Fisher Scientific, Inc., USA) coupled with a Trace

MSD system, with low resolution electron ionization operated at 70 eV, scanning 50-500 m/z at 1 scan/sec. A 1-µL volume of floral extract or synthetic blend was injected at

The carrier gas was helium at a flow rate of 1 mL/min. Components of the extracts were

identified by matching of mass spectra to NIST/EPA/NIH Mass Spectral Library 2005 and comparison of retention indices to published values (Adams, 1995). Identity of major components was confirmed by comparison of spectra and retention times to authentic standards. Each component was quantified by comparison of its peak area in extracts to that of an external standard; these quantities were used to create a synthetic odor blend.

Olfactometer Experiments

Behavioral assays were conducted in a dual-port olfactometer as described in the previous chapter. At 2 h prior to scotophase, approximately 300 mosquitoes of both sexes in similar numbers were released through a sleeve connection, directly from an acrylic plastic cage (30 x 30 x 30 cm) into the holding/release chamber, where they were held for 15 min to acclimatize before the release gate was opened. Mosquitoes were supplied with water on soaked cotton wicks placed within the cage, and were used in experiments at 36 ± 12 h after emergence. Test materials were applied to 15-cm long cotton wicks (TIDI Products, Neenah, WI) in aluminum weighing boats placed into the trapping jars/ports. Pentane was used as a solvent control in all bioassays. The positions of the treatment and control ports were alternated between bioassays to eliminate possible positional bias. The olfactometer parts were cleaned with 70% ethanol, then water, after each experiment; gloves were used at all times to avoid contamination with human- related kairomones. After 12-h, the numbers of mosquitoes in the treatment port, the control port, and remaining in the flight chamber were recorded.

Field Experiments

Field experiments were conducted in the early fall in 2008 and 2010 at a wooded farm area adjacent to Don Scott airfield in Columbus, Ohio described in Haramis and Foster

(1983). Briefly, there are two rectangle woodlots in the area; the main woodlot is 13.6 ha

(290 X 470 m2) and the smaller one is 1.6 ha (120 X 135 m2) located 60 m to the east with farm land in between. The main woodlot is surrounded by human dwellings to the west and northwest, and a fenced meadow to the south. The rest of the area is a farm land for cultivating maize or soybeans. There is an animal facility about 500 m southwest of the study area and herds of cattle were allowed to graze on the meadow. A stream runs adjacent to the main woodlot and is connected to an underground canal to the northwest.

The main tree species in both woodlots are beech (Fagus grandifolia), red maple (Acer rubrum), and red oak (Quercus rubra). The early season angiosperms are garlic mustard

(Alliaria petiolata), trillium (Trillium sp.), Japanese honeysuckle (Lonicera japonica), daisy fleabane (Erigeron strigosus ), dogbane (Apocynum medium), and snakeroot

(Eupatorium rugosum). The main flowering plants during the high peak of mosquitoes included common milkweed (Asclepias syriaca), Queen Anne's lace (Daucus carota), and Canada goldenrod (Solidago canadensis) with the later flowering until mid-October.

Release rate experiments – (2008)

In the first year of trapping, we baited sixteen Counter Flow Geometry (CFG) traps

(Kline, 1999) (American Biophysics, North Kingstown, R.I.), four each with cuttings of goldenrod flowers, a synthetic goldenrod blend on a cotton wick, a synthetic goldenrod blend in a slow-release polyethylene tube topped up with glycerine, and control traps.

The positions of the treatments were randomized to balance out positional effects, and we performed four replicates (n=4) using this set-up. The synthetic blend was constituted based on proportions indicated (Table 3.2) at 4 µl total blend volume. In the slow-release tubes, we added 250 µl glycerine to further slow the release of our blend volatiles.

Nylon strip experiments – (2010)

In the second year of trapping, we baited traps with flower cuttings, a goldenrod blend, and a goldenrod extract, and these were compared alongside control traps. The blend and extract were dispensed using strips of nylon fabric cut out of female pantyhose, 98%

Polyamides, 2% Spandex (Walgreens Co, Deerfield, IL), each strip measuring (7.5 cm X

5 cm). Nylon strips were submerged in 6-dram glass vials containing 10 ml of the goldenrod extract and left to evaporate in a fume hood three hours before they were deployed in traps, allowing complete evaporation of the solvent. The same procedure was followed for the control strips, instead submerging them in pentane solvent. We prepared the synthetic blend in the proportions shown (Table 3.2.), six nylon strips per chemical vial were packed into and submerged in 10 ml of solvent in the eight separate

glass vials, and these were left open in a fume hood overnight to allow complete evaporation of the solvent. During deployment, a single nylon strip for each of the eight chemicals was strung through a thin wire and hung on the side of the suction end of the

CFG trap.

Six locations on the edge of the wood were selected, each 5 m from the next, and at each location, the four traps (flower, blend, extract and control) were randomly assigned, 1 m apart from each other. In each of the locations, traps were paired in series to a single 12 volt battery so that the battery strength remained the same for both traps throughout our sampling period. Weakened or depleted batteries were exchanged for charged ones, so that suction was strong through the collection period.

Statistical analysis

We used a chi-square test to analyze the olfactometer experiments by comparing catches in the treatment and control ports. We used a Pearson’s correlation to evaluate the relationship between volatile loss and vapor pressure. Field data were analyzed using a one-way ANOVA to compare mean catches in CFG traps baited with our different treatments. All data was analyzed using Predictive Analytics Software (PASW) version

18, SPSS Inc, Chicago, Illinois.

Results

The goldenrod extract prepared in 2006 (Fig. 3.1) was reanalyzed in 2008 to determine how stable constituent chemicals remained through prolonged storage (Table 3.1). We found that the concentrations of individual chemicals within the extract were altered, with a 14 – 94% drop in concentration; even though the extract was stored at sub zero temperatures. We were unable to show a correlation between the vapor pressures of the chemicals and the proportion lost with time r2= 0.056, p=0.415, indicating that the disproportionate losses might have arisen from handling. When assayed in 2006, the freshly prepared goldenrod extract elicited a significant response over a blank control, with 31.2 ± 8.1% responding to the plant extract versus 1.7 ± 0.6% for the control, and when assayed two years later in 2008, the same extract caught 52.3 ± 5.9% mosquitoes and 8.6 ± 4.2% were caught in the blank port (Fig. 2). Both the fresh and 2-year old extracts were highly attractive to mosquitoes when compared against a blank control (χ2

=318.06, df=1, P<0.001 and χ2 = 565.63, df=1, P<0.001 respectively). Despite drops in concentration in the goldenrod extract, bioactivity remained high in behavioral assays.

We formulated our goldenrod blend by incorporating the main chemical peaks in our extract profiles (Table 3.1). For field concentrations, we used 10 times the concentration of our olfactometer samples to compensate for wind and competition with natural vegetation.

The synthetic goldenrod blend was highly attractive in the dual port olfactometer (Fig.

3.3), attracting 58.1 ± 5.5% versus a blank control that trapped 18.8 ± 4.2% of all mosquitoes released (χ2 df=2, P<0.001). In the dual port olfactometer assays, our natural extract was still significantly more attractive than the synthetic goldenrod extract when these were assayed side by side in a choice test (Fig. 3.3), χ2 df=2, P<0.001.

Release rate experiments

In field experiments, there was no statistical difference among the treatments (Fig. 3.4;

F3,60 = 0.139, p>0.05). There was also no difference in the sex ratios of mosquitoes that were trapped in the CFG traps, regardless of the treatment (Fig. 3.5).

Nylon strip experiments

This method of dispensation has been to shown to be effective when used in CFG traps

(Okumu et al., 2010a). The concept first originated from the observation that foot odors are very attractive to mosquitoes (De Jong and Knols, 1995a; Knols and De Jong, 1996) and that worn socks can be used to measure orientation to human kairomones in CFG traps set up in a screenhouse (Njiru et al., 2006). We however did not observe any differences among treatments when this method of release was employed. There was no statistical difference between the means of the treatments F3,19 = 1.931, p=0.159. There was no difference in the sex ratio of mosquitoes trapped in either treatment (Fig.3.5).

Discussion

Natural extracts and synthetic blends have been used to demonstrate attraction by insects to various plant derived volatile organic compounds (VOCs) in both agricultural

(Jacobson and Beroza, 1963; Loughrin et al., 1998; Agelopoulos et al., 1999) and public health pest systems (Bowen, 1992b; Mauer and Rowley, 1999). One advantage that natural extracts have over whole plants is that they can be made in bulk during periods when their whole plant precursors are available and stored for use in the off season, when the whole plants are unavailable. This provides great flexibility to scientists who can test the next best thing to the natural plants with their insects in colony, since the chemical profiles of the natural extracts closely resemble the profiles of the whole plants.

However, such extracts may not chemically representative the host plant indefinitely. I found that an extract of Canada goldenrod may have undergone some changes in its chemical profile after a prolonged period of storage under sub zero temperatures (Table

3.1), although my methods do not allow me to quantify the changes. Nevertheless, there was no evidence that this extract had lost any activity for Cx. pipiens as it still elicited highly significant reponse in the olfactometer when re-bioassayed after the two years storage (Fig. 3.2).

The synthetic blend was formulated according to calculated estimates made using a cis-

Jasmone internal standard, and based on the commercial availability of chemical components at the time of our study. Certain chemicals, such as germacrene D, which featured prominently in the extract profiles were prohibitively expensive and were

therefore left out of the blend altogether, since this would be unsustainable in a blend system, similar conclusions drawn by Gregg et al. (2010). We were able to demonstrate an attraction to the synthetic blend and natural extract in an olfactometer against blank controls, (Fig. 3.3), however, when the natural extract was assayed in a choice test with the goldenrod blend, mosquitoes oriented to a significant extent to the natural extract, indicating that some of the components excluded from our blend may be critical to the additional attractiveness of the extract. The challenge of translating results obtained in the controlled setting of an olfactometer to the field setting are numerous, since one needs to take into account differences in release rates of chemicals in the two settings, competition with other plants in the field, wind and changes in wind pattern just to name a few. We made all attempts to address these issues, including the use of polyethylene slow release tubes and nylon strips to to control the release rates of the synthetic blend and extracts.

Nylon strips have been shown to be effective in dispensing synthetic attractants in the field setting (Okumu et al., 2010a; Okumu et al., 2010c). We made sure that our field traps were adequately spaced to avoid odor interference, and traps were rotated to eliminate location biases.

Pairwise comparisons did show a statistically significant difference between trap catches from florally baited traps and control traps. In most of our field experiments, we observed that control traps were just as good at trapping mosquitoes as baited traps, leading us to suspect contamination of control traps. However, upon the collection of

headspace volatiles of the blank traps, we were able to overrule this as a possible explanation to our observations. A second hypothesis to this observation was that mosquitoes, in their foraging flight around the trap vicinity could be randomly trapped due to the high trapping efficiency of CFG traps. This could help to explain the lack of discrimination in the fairly distinct treatment groups. We think that greater distances between traps might help to resolve this problem, as would a better and more exhaustive investigation of release rates.

In conclusion, an 8-component synthetic blend of volatiles identified from Canada goldenrod flowers proved highly active in luring Cx. pipiens in a flight olfactometer when tested against a solvent control; however, it was far less effective when tested against the floral extract, indicating that either additional compounds or changes in component proportions are needed for full activity. Further work is also required on the release of floral odors in mosquito traps in the field before this technique can be demonstrated as an effective monitoring tool.

Acknowledgments

We thank Robert L. Aldridge for his help in rearing mosquitoes used in this research, and

Bryan T. Jackson for his help in setting up the field component of this research. This research was supported through an NIH grant # 6001385.

References

ADAMS, R. P. 1995. Identification of Essential Oil Components by Gas

Chromatography/Mass Spectroscopy. Allured Publishing, Carol Stream, IL.

AGELOPOULOS, N., BIRKETT, M. A., HICK, A. J., HOOPER, A. M., PICKETT, J.

A., POW, E. M., SMART, L. E., SMILEY, D. W. M., WADHAMS, L. J., and

WOODCOCK, C. M. 1999. Exploiting Semiochemicals in Insect Control. Pestic.

Sci. 55:225-235.

BAKER, T. C., FADAMIRO, H. Y., and COSSE, A. A. 1998. Moth uses fine tuning for

odor resolution. Nature 393:530.

BOWEN, M. F. 1992. Terpene-sensitive receptors in female Culex pipiens mosquitoes:

Electrophysiology and behaviour. J. Insect Physiol. 38:759-764.

BRUCE, T. J. A., WADHAMS, L. J., and WOODCOCK, C. M. 2005. Insect host

location: a volatile situation. Trends Plant Sci. 10:269-274.

CLYNE, P. J., WARR, C. G., FREEMAN, M. R., LESSING, D., KIM, J., and

CARLSON, J. R. 1999. A novel family of divergent seven-transmembrane

proteins: Candidate odorant receptors in Drosophila. Neuron 22:327-338.

FOSTER, W. A. 2008. Phytochemicals as population sampling lures. J. Am. Mosq. Cont.

Assoc. 24.

FOX, A. N., PITTS, R. J., ROBERTSON, H. M., CARLSON, J. R., and ZWIEBEL, L. J.

2001. Candidate odorant receptors from the malaria vector mosquito Anopheles

gambiae and evidence of down-regulation in response to blood feeding. Proc.

Natl. Acad. Sci. USA 98:14693–14697.

GREGG, P. C., DEL SOCORRO, A. P., and HENDERSON, G. S. 2010. Development of

a synthetic plant volatile-based attracticide for female noctuid moths. II.

Bioassays of synthetic plant volatiles as attractants for the adults of the cotton

bollworm, Helicoverpa armigera (Hübner) (Lepidoptera: Noctuidae). Aust. J.

Entomol. 49:21-30.

HARAMIS, L. D. and FOSTER, W. A. 1983. Survival and population density of Aedes

triseriatus (Diptera: Culicidae) in a woodlot in central Ohio, USA. J. Med.

Entomol. 20:391-398.

JACOBSON, M. and BEROZA, M. 1963. Chemical Insect Attractants. Science

140:1367-1373.

KLINE, D. L. 1999. Comparison of two American Biophysics mosquito traps: The

professional and a new counterflow geometry trap. J. Am. Mosq. Cont. Assoc.

15:276-282.

LOFSTEDT, C. 1993. Moth pheromone genetics and evolution. Phil. Trans. R. Soc. B

340:167-177.

LOUGHRIN, J. H., POTTER, D. A., and HAMILTON-KEMP, T. R. 1998. Attraction of

Japanese beetles (Coleoptera: Scarabaeidae) to host plant volatiles in field

trapping experiments Environ. Entomol. 27:395-400.

MAUER, D. J. and ROWLEY, W. A. 1999. Attraction of Culex pipiens pipiens (Diptera:

Culicidae) to flower volatiles. J. Med. Entomol. 36:503-507.

MÜLLER, G. C., JUNNILA, A., QUALLS, W., REVAY, E. E., KLINE, D. L., ALLAN,

S., SCHLEIN, Y., and XUE, R. D. 2010. Control of Culex quinquefasciatus in a

storm drain system in Florida using attractive toxic sugar baits. Med. Vet.

Entomol. 24:346-351.

MÜLLER, G. C., KRAVCHENKO, V. D., and SCHLEIN, Y. 2008. Decline of

Anopheles sergentii and Aedes caspius populations following presentation of

attractive toxic (Spinosad) sugar bait stations in an oasis. J. Am. Mosq. Cont.

Assoc. 24:147-149.

MÜLLER, G. C. and SCHLEIN, Y. 2006. Sugar questing mosquitoes in arid areas gather

on scarce blossoms that can be used for control. Int. J. Parasitol. 36:1077-1080.

OKUMU, F., BISAWARO, L., MBELEYELA, E., KILLEEN, G. F., MUKABANA, W.,

and MOORE, S. J. 2010a. Using nylon strips to dispense mosquito attractants for

sampling the malaria vector Anopheles gambiae s.s. J. Med. Entomol. 47:274-282.

OKUMU, F. O., KILLEEN, G. F., OGOMA, S., BISWARO, L., SMALLEGANGE, R.

C., MBEYELA, E., TITUS, E., MUNK, C., NGONYANI, H., TAKKEN, W. and

others. 2010b. Development and field evaluation of a synthetic mosquito lure that

is more attractive than humans. PLoS One 5:e8951.

PAWLISZYN, J., editor. 1999. Applications of Solid Phase Microextraction. Waterloo,

Ontario, Canada: Royal Society of Chemistry.

TORR, S. J. 1994. The Tsetse (Diptera: Glossinidae) story: Implications for mosquitoes.

J. Am. Mosq. Cont. Assoc. 10:258-265.

ULRIKE, F., DOTTERL, S., JURGENS, A., and AAS, G. 2007. Inter and intraspecific

variation in floral scent in the genus Salix and its implication for pollination. J.

Chem. Ecol. 33:749-765.

VALE, G. A. 1993. Development of baits for Tsetse flies (Diptera: Glossinidae) in

Zimbabwe. J. Med. Entomol. 30:831-842.

A b u n d a n c e

T IC: G ldrdExt08-2.D \data.ms 8 . 7 8 1 9 0 0 0 0 0 0

8 5 0 0 0 0 0 4 . 3 7 5 8 0 0 0 0 0 0

7 5 0 0 0 0 0

7 0 0 0 0 0 0 3 . 0 9 2 6 5 0 0 0 0 0

6 0 0 0 0 0 0 1 1 . 5 8 2 7 . 3 7 3 5 5 0 0 0 0 0

5 0 0 0 0 0 0 1 5 . 4 9 7 3 . 9 5 0 4 5 0 0 0 0 0

89 4 0 0 0 0 0 0 1 0 . 0 6 7 1 6 . 3 2 1

3 5 0 0 0 0 0 3 . 7 4 7 3 0 0 0 0 0 0

2 5 0 0 0 0 0 1 7 . 5 8 5 1 4 . 3 8 7 1 1 . 1 0 8 1 5 . 3 7 9 2 0 0 0 0 0 0 5 . 1 4 1 1 3 . 7 2 3 1 5 . 1 7 3 1 5 0 0 0 0 0 1 3 . 5 5 2 4 . 4 6 0 1 5 . 2 9 0 1 0 0 0 0 0 0 4 . 1 1 3 1 2 . 2 4 7 1 0 . 9 5 7 1 41 . 5 7 . 3 0 7 2 5 1 7 . 8 9 6 7 . 8 7 . 8 1 2 1 3 9 . 0 5 21 0 . 1 6 8 1 3 . 3 1 7 1 6 . 8 6 8 5 0 0 0 0 0 4 . 6 4 7 1 0 . 2 31 7 1 . 2 8 8 3 . 3 6 4 1 8 . 1 6 4 1 9 . 5 3 9 0 4 . 0 0 6 . 0 0 8 . 0 0 1 0 . 0 0 1 2 . 0 0 1 4 . 0 0 1 6 . 0 0 1 8 . 0 0 T i m e - - >

Figure 3.1. Chromatogram of Goldenrod extract (2008) showing peaks identified by GC/MS. Numbers above spectrum represent retention times.

Table 3.1. Composition and quality of a natural Goldenrod extract evaluated at two time points, two years apart (2006 and 2008) highlighting changes in concentration between the two extracts.

ǂ Concentration of extract in nanograms per microliter aPropotional change in concentration of extract components from 2006 to 2008

Goldenrod Extract 2008 Goldenrod Extract 2006 aConc Δ Vapor Pressure RT % of max ng/ µlǂ Chemical ID % of max ng/ µlǂ Chemical ID 08/06 mm/Hg 25 °C 3.092 100.0% 218 α-pinene 100.0% 312 α-pinene 0.70 3.49 3.364 2.6% 6 camphene 2.7% 8 camphene 0.66 2.50 3.747 27.6% 60 sabinene 58.9% 184 sabinene 0.33 2.63 3.95 32.3% 70 β-myrcene 30.7% 96 β-myrcene 0.74 2.29 4.113 5.8% 13 α-phellandrene 3.8% 12 α-phellandrene 1.08 1.86 4.375 49.5% 108 D-Limonene 66.8% 208 D-Limonene 0.52 1.98 4.46 4.6% 10 β-phellandrene 13.4% 42 β-phellandrene 0.24 1.57 4.647 2.5% 5 β-ocimene 3.6% 11 β-ocimene 0.49 1.56 5.141 5.8% 13 terpinolene 10.7% 33 terpinolene 0.38 1.13 7.373 15.1% 33 bornyl acetate 20.6% 64 bornyl acetate 0.51 0.0959

91 8.781 30.8% 67 germacrene D 28.9% 90 germacrene D 0.75 0.00673

9.052 1.0% 2 β-sesquiphellandrene 5.2% 16 β-sesquiphellandrene 0.14 0.0109 10.067 8.7% 19 Z-bergamotol 13.9% 43 Z-bergamotol 0.44 ? 11.108 4.7% 10 curlone 9.2% 29 curlone 0.35 ? 11.582 15.0% 33 NGM 34.3% 107 NGM 0.31 ? 12.247 2.7% 6 palmitic acid 7.6% 24 palmitic acid 0.25 3.8E-07 13.552 3.4% 7 linoleic acid 17.0% 53 linoleic acid 0.14 3.54E-06 13.723 4.6% 10 linolenic acid 14.9% 46 linolenic acid 0.22 3.1E-06 14.387 5.2% 11 heneicosane 3.8% 12 heneicosane 0.94 6.14E-05 15.173 5.7% 13 unknown n/a 0 n/a 0.00 n/a 15.379 4.2% 9 unknown n/a 0 n/a 0.00 n/a 15.497 21.8% 48 NGM n/a 0 n/a 0.00 n/a 16.321 11.4% 25 straight HC 11.3% 35 NGM 0.71 n/a 17.585 11.2% 25 straight HC 8.5% 26 straight HC 0.93 n/a 17.896 0.0% 0 unknown 15.4% 48 straight HC 0.00 n/a

Table 3.2. Volatile composition of solvent extracts of Canada goldenrod Solidago canadensis and the quantification of extract components.

Chemicala Olfactometer Field NOTES

Concentration Concentration

(µg/ml) (µg/ml)

α – Pinene 5.10 50 98% (1S)-(-)-α Sigma Aldrich

β – Pinene 1.60 16 99% (-)-(β) Sigma Aldrich

β – Myrcene 3.00 30 ≥95% Fluka

α – Phellandrene 0.35 3 ≥95% Sigma Aldrich

Limonene 7.60 75 97% (R)-(+) Sigma Aldrich

β – Ocimeneb 0.19 1.5 70 – 75% (Z) Sigma Aldrich

Terpinolene 1.05 10.5 ≥ 97% Fluka

Bornyl acetate 2.25 22 99% Fluka

aidentity established by comparison to authentic of each compound, quantity estimated by comparison of peak area to (Z)-jasmone internal standard b synthetic β-ocimene contained a mixture of isomers

Figure 3.2. Differential responses of Culex pipiens to a pentane extract of Goldenrod prepared in 2006 and later tested

in 2008 compared against blank controls in a dual-port olfactometer. In both tests, the extract caught significantly more

mosquitoes than the control (***P<0.0001).

Figure 3.3. Responses of Culex pipiens released into a dual-port olfactometer to a synthetic Goldenrod blend compared

to a solvent control (left) and an extract of Goldenrod (2008) compared to a synthetic blend (right) (***P<0.0001)

Figure 3.4. Culex pipiens mosquitoes trapped by either a goldenrod blend under slow release, rapid release, and goldenrod blossoms compared against control (unbaited) field traps in 2008. (N = 16; F3,60 =.139; p>0.05) a Synthetic goldenrod blend rapid release achieved by dispensing the blend on a cotton wick b Synthetic goldenrod blend slow release achieved by dispensing the blend from a polyethylene tube topped-up with glycerine

Figure 3.5. Sex distribution in release-rate experiments showing Culex pipiens mosquitoes trapped by either a goldenrod blend under slow release, a rapid-release blend and goldenrod blossoms, compared to control (unbaited) traps.

SE ± 4

1 Mean catch per trap trap per catch Mean

0 Flower Blend Extract Control

Treatment

Figure 3.6. 2010 Field results showing the trapping of Culex pipiens in CFG traps baited with whole goldenrod flowers, goldenrod synthetic blend, and a pentane extract of goldenrod, compared to blank traps. (N= 4; F3,19 =1.931; p>0.05)

Chapter 4

A Novel Diffusion-Cage Olfactometer for Measuring Anopheles gambiae (Diptera:

Culicidae) Orientation to Plant Volatiles in Semi-Field Enclosures

Abstract

A novel diffusion-cage olfactometer is described for testing the responses of Anopheles gambiae to plant volatiles. Its simple acrylic construction, ease of assembly, and accommodation to whole plants makes it a useful tool for measuring mosquito orientation to plant volatiles within screenhouses. We compared the performance of this diffusion- based olfactometer to that of the more commonly used Y-tube wind-tunnel olfactometer, by testing the orientation of mosquitoes to volatiles of a few prevalent plants of eastern

Africa, Parthenium hysterophorus, Ricinus communis, Lantana camara and Senna occidentalis, reportedly utilized by An. gambiae for sugar. Results indicate that the diffusion-cage olfactometer is a more effective alternative to conventional wind-tunnel olfactometers, to test mosquito orientation to plant volatiles in semi-field enclosures.

Introduction

Mosquito orientation and response to volatiles is most accurately measured under conditions that simulate nature as closely as possible. Laboratory-based experiments that utilize olfactometers face the challenge of recreating these conditions, because the mosquitoes are confined to small spaces. Studies that employ large indoor mesocosms or outdoor screenhouses for studying the behavior of Anopheles gambiae s.s. Giles have gained prominence (Knols et al., 2002; Impoinvil et al., 2004; Ferguson et al., 2008;

Stone et al., 2011), because they allow this malaria vector to behave closer to the way it might in a natural setting. We describe a cage-based olfactometer that is simple in design and ideal for use with An. gambiae inside semi-field environments such as mesocosms and large screenhouse enclosures. We use this olfactometer with whole plants to test the orientation of mosquitoes to plant volatiles that help guide both sexes towards nectar sources. This olfactometer saves time and effort spent on direct observation (Manda et al., 2007) and circumvents unintended and currently unknown effects of cut-plant volatiles that cannot be avoided when plant cuttings must be used in smaller olfactometers.

Materials and Methods

Description of Diffusion Olfactometer

The olfactometer is a modified mosquito cage of clear acrylic plastic, 66 x 50 x 67 cm

(l,w,h), (Fig. 1), placed within a climate-controlled walk-in enclosure in a greenhouse 2.4 by 1.8 by 2.1 m (=9.1 m3), as described by Stone et al. (2009), but without feeders and dark resting sites. The ceiling of the cage consists of two equal pieces (66 x 25 cm), the front one attached only by three 2-cm hinges to the back piece and forming an access door. The front wall has two sliding windows (20 x 22 cm), 8 cm apart. The two windows are covered by solid 25 x 25 cm Styrofoam panels 2 cm thick, mounted inside the cage on four 5-cm-long spacers, attached at the corners with plasticine clay, so that there was a 5-cm gap between the front wall and all four sides of each panel. These window gaps served as the only entrances through which mosquitoes gained entry to the cage’s interior.

Within the cage, humidity was raised above ambient by a pan of water (50 x 30 cm), covered by netting to prevent mosquito access to the water. Air was circulated within the cage by a Sprite® model SP2A2L 115V 50/60Hz 9W PC-fan (Rotron, Woodstock, NY) with an air speed of 1.2 ms-1 measured 5 cm above the fan. The fan lay on its side on top of the netting over the water pan, to facilitate movement of plant volatiles toward the gaps around the window panels. Four cylindrical cups on their sides to provided resting

100

sites for the mosquitoes trapped, and two 2-dram vials with protruding wicks provided

10% sucrose solution.

Description of Y-Tube Olfactometer used for Comparison

We compared the performance of the diffusion olfactometer to that of a modified Y-tube wind-tunnel olfactometer (Geier et al., 1999), also in use in our laboratory. It is clear acrylic plastic, is 49.6 cm in total length, and consists of four sections: a 41-L acrylic cage serving as a holding chamber, a cylindrical downwind flight chamber (20.1 cm long,

6.1 cm i.d.), a stimulus chamber or mixing box (16 x 36.1 x 12.7 cm, l,w,h) and two cylindrical choice arms (13.5 cm long, 6.1 cm i.d.) at the upwind end, leading to funnel traps.

As attractants, we used four common plant species of western Kenya previously reported to be preferred by An. gambiae (Manda et al., 2007): Parthenium hysterophorus

(Asteraceae), Ricinus communis (Euphorbiaceae), Lantana camara (Verbenaceae) and

Senna occidentalis (Fabaceae). The grass control was Phalaris arundinacea (Poaceae).

In diffusion-olfactometer experiments, we used a dual-cage set-up, with one cage acting as the treatment cage, the other as the control. The treatment and control cages were placed back-to-back, with a 10-cm space between the cages and the windows facing in opposite directions. Two hundred An. gambiae s.s. pupae were counted out and allowed to emerge in small 41-liter acrylic cages. Adults were supplied ad libitum with water on

101

soaked cotton wicks placed within the cage. Mosquitoes were released directly into the enclosure 24 ± 12 h after emergence, 2 h prior to scotophase. Y-tube olfactometer experiments were conducted in an indoor bioassay room, with a controlled air flow at 70 mls-1; temperature 27 ± 1ºC and humidity 68% RH. The olfactometer system was connected to an exhaust tube that led the olfactometer effluent into a fume hood.

Between 150 – 250 mosquitoes were released into a holding chamber for Y-tube tests 12

- 24 hrs after emergence, 2 h prior to scotophase. Mosquitoes in the traps and other olfactometer sections were counted 12 h after they were released into the Y-tube olfactometer to determine the response. To verify the symmetry of each system, blank runs were conducted for both types of olfactometers prior to evaluating plants. We also tested two identical plants of the same species in a single test. Treatment and control choice arms were alternated between replications during the experiments.

Statistical analysis

Olfactometer performance was evaluated based on two parameters: overall treatment response, T/N, and discrimination, T/ (T+C). The overall treatment response was based on the number of mosquitoes that were captured in the treatment trap (T) as a proportion of all mosquitoes released into the olfactometer/mesocosm space, while discrimination was evaluated based on the number of mosquitoes in the treatment trap (T) as a proportion of all mosquitoes trapped in either the treatment (T) or control (C) traps, that is, mosquitoes that made a choice one way or other. Data were analyzed by chi-square

102

test. Left and right responses of the olfactometers were analyzed using a paired-sample t- test. All data were analyzed using IBM SPSS Software version 19 (SPSS Inc, Chicago,

Illinois).

Results

With all four plant species, the diffusion-cage olfactometer performed as well or better than the modified Geier Y-tube olfactometer in every replicate (Fig. 4.2.). In combined replicates the overall treatment response was always higher in the diffusion-cage olfactometer than in the modified Y-tube olfactometer for each plant (Lantana 57% vs.

8%, Senna 45% vs. 18%, Parthenium 36% vs. 27% and Ricinus 18% vs. 10%, indicating the proportion of mosquitoes caught in treatment and control traps respectively).

In both olfactometers, we recorded significantly higher responses to the test plants over controls, as follows: Diffusion-cage olfactometer: Lantana (χ2 = 282.75 df=1 p<0.0001)

Senna (χ2 = 57.885 df=1 p<0.0001), Parthenium (χ2 = 181.79 df=1 P<0.0001) and

Ricinus (χ2 = 29.43 df=1 P<0.0001). Y-tube olfactometer: Lantana (χ2 = 85.56 df=1 p<0.0001) Senna (χ2 = 170.67 df=1 p<0.0001), Parthenium (χ2 = 219.57 df=1

P<0.0001) and Ricinus (χ2 = 7.80 df=1 P<0.01).

Discrimination varied in both olfactometers, depending on the plant tested, with the diffusion-cage olfactometer recording the best discrimination for Lantana, Parthenium and Ricinus at 82%, 78% and 72% respectively and the lowest discrimination for Senna,

103

calculated at 68%. The modified Y-tube olfactometer recorded high discrimination levels for Senna, Parthenim and Lantana calculated at 84%, 84% and 82% respectively, and the lowest rate for Ricinus calculated at 59%. There was no difference between left and right responses for either the Y-tube olfactometer (t = 1.26, df = 6, p = 0.256) or the diffusion- cage olfactometer (t = 0.805 df = 6, P = 0.452) when double blanks were run.

104

Discussion

This study demonstrates the high performance of a new simple tool for measuring mosquito orientation to plant-derived volatile organic compounds in semi-field enclosures. Gaining significant prominence in ecological studies (Knols et al., 2002;

Okech et al., 2003), semi-field systems provide a reliable intermediate between laboratory and field trials of these compounds, and they may circumvent the challenges inherent in standardized laboratory methods, whose artifacts may lead to shortfalls in field efficacy (Seyoum et al., 2002). The diffusion-cage olfactometer provides a simple design and set-up, and given its par performance compared to a more precisely controlled and standardized system, it offers a pragmatic and reliable alternative. Furthermore, the higher proportion of mosquitoes responding in the diffusion-cage set-up indicates that a more natural situation, where mosquitoes are less restricted, is more likely to give reliable behavioral information. The diffusion olfactometer gave better sensitivity in tests with

Ricinus communis, which was likely due to the testing of the whole plant, rather than just cuttings, as it was hardly detectable to mosquitoes in the modified Y-tube olfactometer. A notable advantage this system has over the Y-tube or other dual port olfactometers is the high throughput potential, especially when screening a large number of plants. In experiments to be presented in the following chapter, we conducted comparisons of four plant species at the same time, each in a different diffusion-cage in a large screenhouse.

In conclusion, the diffusion-cage olfactometer will not replace wind-tunnel olfactometer systems when precisely controlled conditions are needed, but it will complement them

105

and will be especially useful at the semi-field stage of experimentation with plant-related attractants.

Acknowledgment

We thank Ashley Jackson for her help in rearing mosquitoes used in this research. This research was supported by NIH grant #R01-AI077722.

106

References

FERGUSON, H. M., NG'HABI, K. R., WALDER, T., KADUNGULA, D., MOORE, S.

J., LYIMO, I., RUSSELL, T. L., URASSA, H., MSHINDA, H., KILLEEN, G. F.,

and KNOLS, B. G. 2008. Establishment of a large semi-field system for

experimental study of African malaria vector ecology and control in Tanzania.

Malar. J. 7:158.

GEIER, M., BOSCH, O. J., and BOECKH, J. 1999. Ammonia as an attractive component

for the yellow fever mosquito Aedes aegypti. Chem. Senses 24:647-653.

IMPOINVIL, D. E., KONGERE, J. O., FOSTER, W. A., NJIRU, B. N., KILLEEN, G.

F., GITHURE, J. I., BEIER, J. C., HASSANALI, A., and KNOLS, B. G. J. 2004.

Feeding and survival of the malaria vector Anopheles gambie on plants growing

in Kenya. Med. Vet. Entomol. 18:108-115.

KNOLS, B., NJIRU, B., MATHENGE, E., MUKABANA, W., BEIER, J., and

KILLEEN, G. 2002. MalariaSphere: A greenhouse-enclosed simulation of a

natural Anopheles gambiae (Diptera: Culicidae) ecosystem in western Kenya.

Malar. J. 1:19.

107

MANDA, H., GOUAGNA, L. C., NYANDAT, E., KABIRU, E., JACKSON, R. R.,

FOSTER, W. A., GITHURE, J. I., BEIER, J. C., and HASSANALI, A. 2007.

Discriminative feeding behaviour of Anopheles gambiae s.s. on endemic plants in

western Kenya. Med. Vet. Entomol. 21:103-111.

OKECH, B. A., GOUAGNA, L. C., KILLEEN, G. F., KNOLS, B. G. J., KABIRU, E.

W., BEIER, J. C., YAN, G., and GITHURE, J. I. 2003. Influence of sugar

availability and indoor microclimate on survival of Anopheles gambiae (Diptera:

Culicidae) under semifield conditions in western Kenya. J. Med. Entomol. 40:

657-663.

SEYOUM, A., PÅLSSON, K., KUNG'A, S., KABIRU, E. W., LWANDE, W.,

KILLEEN, G. F., HASSANALI, A., and KNOLS, B. G. J. 2002. Traditional use

of mosquito-repellent plants in western Kenya and their evaluation in semi-field

experimental huts against Anopheles gambiae: ethnobotanical studies and

application by thermal expulsion and direct burning. Trans. R. Soc. Trop. Med.

Hyg. 96:225-231.

STONE, C. M., HAMILTON, I. M., and FOSTER, W. A. 2011. A survival and

reproduction trade-off is resolved in accordance with resource availability by

virgin female mosquitoes. Anim. Behav. 81:765-774.

108

STONE, C. M., TAYLOR, R. M., and FOSTER, W. A. 2009. An effective indoor

mesocosm for studying populations of Anopheles gambiae in temperate climates.

J. Am. Mosq. Cont. Assoc. 25:514-516.

109

66 cm Mosquito flight path Styrofoam barrier Access flap

Section showing barrier opening

67 cm

Resting cups 50 cm

Fan Potted plant in cage Moisture pan

Figure 4.1. Schematic diagram of diffusion-cage olfactometer, showing full assembly and plant placement. Inset is the window section showing Styrofoam barrier placement.

110

Figure 4.2. Side by side comparisons of diffusion-cage and Y-tube olfactometers when tested with four African plants; Lantana camara, Senna occidentalis, Parthenium hysterophorus and Ricinus communis in non-competitive tests.

111

Chapter 5

Exploiting the attractive power of Parthenium hysterophorus to kill the malaria vector

Anopheles gambiae s.s.

Abstract –Parthenium hysterophorus L. (Asteraceae) is a fast spreading invasive weed in the tropics, having been introduced into Africa in the last century from its putative origin in the Gulf of Mexico. This weed is toxic to animals, and its sesquiterpene profile of lactones and phenolics make it highly allelopathic and disruptive to agriculture. Previous studies have shown this weed to be attractive to mosquitoes, despite the very low amounts of sugar recovered from mosquitoes feeding on the plant. We have found through competitive behavioral assays set up in screenhouses with sugar-rich plants that the volatiles emitted by this weed are very attractive to the malaria mosquito Anopheles gambiae even though the mosquitoes do not get a life-sustaining reward from its flowers or leaves. When an insecticide-treated fruit juice bait was sprayed on the leaves and flowers of this weed, very high levels of mortality were recorded as mosquitoes ingested the toxin from exposed plant surfaces. Similar levels of control could not be achieved in untreated plants. This study demonstrates a promising approach to mosquito control by turning naturally available plants into deadly kill stations through a rational use of insecticides.

Key Words – Nectar feeding, Anopheles gambiae, Plant attractants, Oral insecticides.

112

Introduction

Malaria is the leading cause of morbidity and mortality in sub-Saharan Africa, causing close to one million deaths, mainly children under five and pregnant women, and affecting a further 300 – 500 million people worldwide annually (WHO, 2000).

Mosquito population suppression is critical to ensure a low transmission potential, hence interventions that employ cheap, sustainable yet efficacious control methodologies must be implemented. The sugar feeding habit of mosquitoes offers an easily exploitable avenue for control through its direct impacts on vectorial capacity (Gary and Foster,

2001; Gu et al., 2011), especially since both sexes are vulnerable (Downes, 1958) and since sugar feeding is frequent and obligatory in most species (Hocking, 1953). Studies have already demonstrated that mosquitoes ingest and are killed by insecticide laced carbohydrate baits (Xue and Barnard, 2003; Xue et al., 2011) and these, it is argued may provide an alternative to widespread application of insecticides through a more rational point-source targeting of disease vectors (Xue et al., 2011). A new approach for the control of Culex pipiens mosquitoes was demonstrated by spraying attractive blossoms of

Tamarix jordanis trees with a solution of sugar, food dye and an oral insecticide (Schlein and Müller, 2008b) thereby using the dominant attractiveness of the blossoms to achieve mosquito control. Building on this study, and supplementing attractive plant volatiles with a fruit-juice bait to encourage toxin uptake, vegetation around a sewage pond in

Israel was sprayed with a toxic bait to control Culex pipiens from emergence sites

(Müller et al., 2010c). A slight modification of the latter study was successfully

113

employed at a desert oasis in Israel against Anopheles sergentii and Aedes caspius

(Müller et al., 2008) and later in Florida against Culex quinquefasciatus by deploying stand-alone toxic sugar baits set up on the branches of Acacia raddiana and in storm drains in a built up residential area (Müller et al., 2010b) respectively. The manipulation of various aspects of mosquito sugar-feeding behavior can be successfully used to control mosquito populations, either by the use of attractive blend mixtures to lure them to kill stations, or by using toxic baits that the mosquitoes can take up during normal sugar feeding. These strategies may prove especially beneficial in areas that are replete with natural nectar sources accessible to mosquitoes, and where environmental modification by the removal of sugar rich plants is not feasible.

One present concern about the widespread application of this method, employing the use of toxic sugar baits in the field, is the undesirable effect of killing non-target organisms such as honeybees which are indispensable pollinators. This can however be circumvented by using screen meshes with gauge sizes that keep out larger insects like honeybees and wasps but allow mosquitoes through. Indeed, the use of highly specific insecticides like the bacterial toxin Bascillus thuringiensis serovar Israelensis (Bti) against adult mosquitoes is currently being investigated and would provide the best solution against non-target insecticidal effects.

114

To date, only exploratory studies have been conducted to examine the potential effectiveness of using attractive toxic sugar baits against the malaria vector Anopheles gambiae (Müller et al., 2010a). The present study aims to provide further evidence of the potential use of natural, attractive plants as toxic stations against malaria mosquitoes.

Methods and Materials

Mosquitoes

All insects used in these experiments came from a colony established in 2001 by staff at the International Centre of Insect Physiology and Ecology from the local population of

An. gambiae s.s. in Mbita Point, Suba District, Nyanza, Kenya, and identified by polymerase chain reaction (PCR). This Mbita strain has been maintained in acrylic cages at 26.6 ± 1°C and 80 ± 5% RH since 2006. Water and 10% sucrose solution were available to colony adults ad libitum. Mosquitoes were reared by transferring 100 newly hatched first instar larvae into 22.8- by 33-cm pans filled with 450 ml of aged tap water and feeding them 0.2 mg of finely ground Tetramin fish flakes per larva during the first 3 d of larval development, 0.4 mg for the next 3 d, and 0.8 mg for subsequent days until pupation. Pupae were allowed to emerge in 41-L acrylic cages for use in the experiments

Plants

Four plant species identified as mosquito host plants (Manda et al., 2007) were used in behavioral assays: Parthenium hysterophorus (Asteraceae), Ricinus communis

115

(Euphorbiaceae), Lantana camara (Verbenaceae) and Senna occidentalis (Fabaceae).

These plants grow naturally in western Kenya and are suspected to be the main nectar hosts for Anopheles gambiae, one of the most important vectors of malaria in Africa.

Experimental Screenhouses

These experiments were conducted in two identical customized walk-in insect cages, the fabric portion manufactured by Megaview LLC and the superstructure and interior constructed by B.T. Jackson and C.M. Stone. The sides of the cage, ceiling, and sleeves were made with white polyester netting (96 x 26 mesh/per sq. in), the floor material was white vinyl. The dimensions of the cage were 5.66 x 4.87 x 3.00 m (l, w, h) for a total of

82.69 m3. Nine cylindrical sleeves (0.45 m diameter and 0.5 m length) where located in the ceiling to allow greenhouse plant growing lights to be positioned inside of the cage.

These growing lights were suspended by chains from the metal structure of the greenhouse. The cylindrical sleeves were then wrapped around the chains and held in place with plastic zip ties. Also located in the ceiling was a 1 m long D-shaped zippered door made from the same material as the ceiling. This was to allow access to mechanical parts above the cage. Located on the floor were two openings, each 0.18 sq. meters with netting to allow water to drain through, and a vinyl flap covering. The position of the drains corresponded to drains in the greenhouse floor. Access to the cage was through a large D-shaped zippered door (1.8 m high and 1.2 m wide) made from clear vinyl, positioned in the middle of one of the short walls. Outside of the door was a small

116

antechamber to prevent mosquito escape. The antechamber measured 1 x 2 x 2 m and had two zippered doors, a vinyl floor, one wall and a ceiling of netting material. Located to the left of the door were two sleeves. The top sleeve was used for the exhaust pipe of the humidifier and the bottom sleeve was for humidistat controls. Temperature was controlled through the greenhouse heating and cooling system. Wall-mounted steam radiators heated the room while louvered roof vents, an exhaust fan, and an evaporative cooling system maintained cooler temperatures all of which were controlled through a thermostat positioned in the middle of the room approximately 1 m from the floor.

Humidity was maintained with an Ocean Mist® MH3 industrial ultrasonic humidifier

(Mico Inc., El Monte, CA) which was controlled with the included humidistat suspended from the room’s thermostat. The entire insect cage structure was supported by a PVC pipe framework and connectors. To help prevent damage to the vinyl floor, a 15 x 20’ white heavy duty Polyvinyl chloride (PVC) tarp cover (Tarpaflex US, Naples, FL) was placed inside of the cage.

Non-Competitive Assays

Behavioral assays were conducted in a small mesocosm enclosure placed within a climate-controlled walk-in room in a greenhouse 2.4 by 1.8 by 2.1 m (=9.1 m3) described by (Stone et al., 2009) but without feeders and dark resting sites. Two diffusion-cage olfactometers (Fig. 3.1) were placed in the middle of the room in a back to back orientation with the Styrofoam barriers opening on opposite directions. One of the cages

117

was baited with a test plant or mango essence and was designated the treatment cage, while the other cage was baited with an artificial plant and was designated the control cage. Both natural and artificial plants were covered with white netting material to balance visual stimuli. A shallow water pan was placed at the bottom of the control cage to balance the humidity in the cage with the natural plant. The cages were rotated after each nine hour assay to balance location effects and each cage was thoroughly cleaned before a new experiment was started to eliminate lingering plant odors. Two hundred mosquitoes were released directly into the enclosure 24 ± 12 h after emergence, 2 h prior to scotophase. Treatments were presented in a randomized block design (n = 6) and response was calculated as the number of mosquitoes caught in the treatment cage relative to the number in the control cage.

Competitive Assays

1. Dual-Choice Experiments

These experiments were set up in the same way as no-choice tests highlighted above, with the only difference being the replacement of the artificial plant with a natural plant.

Two hundred mosquitoes were released directly into the enclosure 24 ± 12 h after emergence, 2 h prior to scotophase. Treatments were presented in a randomized block design (n = 4) and response was calculated as the number of mosquitoes caught in each of the test cages.

118

2. Multiple-Choice Experiments

Four diffusion-cage olfactometers were placed in the middle of the screenhouse (walk-in insect cage) with the dual Styrofoam barrier openings of each cage facing away at 90° from the next cage. All four plants were tested concurrently in a competitive assay/choice test, and the positions of the plants were rotated after each 9-hr assay. Three hundred mosquitoes were released directly into the enclosure 24 ± 12 h after emergence,

2 h prior to scotophase. Treatments were presented in a randomized block design (n = 5) and response was calculated as the number of mosquitoes caught in each of the test cages.

3. Feeding Assays

Eight potted plants, two each of the species listed above, were uniformly spaced on the open floor of the mosquito cage. Each plant was positioned at least 0.5 m from the nearest neighbor, making sure that their leaves or branches did not contact each other.

Plants were watered 2 h prior to the experiment to ensure that they were not dripping wet at the time experiments were set up. In the treatment cage, we sprayed two potted

Parthenium plants with an attractive toxic sugar bait (ATSB), comprising 85% juice of overripe to rotting mango fruits (Mangifera indica), 10% (wt:vol) of table sugar and 5%

(vol) of the oral insecticide Spinosad Conserve® SC (Dow AgroSciences, IN, USA), whereas the control cage had eight untreated plants, two each of the species tested. Three to four hundred mosquitoes were released directly into each cage 24 ± 12 h after emergence, 2 h prior to scotophase and were left for 48 h before survivors and carcasses 119

were retrieved and counted. Treatment and control cages were alternated after each replicate (n = 4), and the positions of the plants in the cage were also randomized to eliminate positional bias.

Bait solution, artificial feeding stations and application of ATSB

The bait solution used in feeding assays consisted of 85% juice of overripe to rotting mango fruits (Mangifera indica), 10% (wt:vol) of table sugar and 5% (vol) of the oral insecticide Spinosad Conserve® SC (Dow AgroSciences, IN, USA). After mixing, the solution was allowed to ferment for about 1 week under refrigerator conditions (4° C).

The attractive toxic sugar bait (ATSB) solution was applied using a 3.7-L RL Flo-

Master® 1985LG Sprayer (Root-Lowell Manufacturing Co., Lowell, MI, USA).

Parthenium hysterophorus plants were lightly sprayed on exposed leaf, stem, and floral surfaces to make sure the solution did not drip or run off the surface of the plants. The solution was allowed to dry off before plants were put into the screenhouses.

Artificial feeding stations were made by dipping one (supply wick) of five 15-cm long cotton wicks (TIDI Products, Neenah, WI) through a 1 cm dia. hole bored through the snap lid of a 20 ml plastic dish (Fig.5.6) The supply wick was soaked in the ATSB solution, allowing the replenishment of the four (distribution wicks) through capillary action. The five wicks were connected above the lid opening using a rubber band that bound 3 cm of the top part of the supply wick to a similar length of the lower ends of the distribution wicks. The top 1 cm ends of the distribution wicks were similarly held

120

together with a twined rubber band. The entire structure was held upright with a 20 cm wooden splint, held both at the top and bottom by the rubber band ties. The splint was then passed the lid opening to the base of the dish.

Statistical analysis

Olfactometer response was expressed as a proportion of mosquitoes trapped in the treatment port relative to the proportion collected in the control port for non competitive and dual-choice assays, and data were analyzed by chi-square test. Multiple-choice experiments in the screenhouse were analyzed by Analysis of Variance (ANOVA), and multiple regression analysis. All data were analyzed using IBM SPSS Software version

19 (SPSS Inc, Chicago, Illinois).

Results

Non-competitive assays

All plants tested elicited a significantly higher response than the control (Fig. 5.1), with

Lantana camara attracting the highest number 47 ± 7% followed by Senna didymobotria

30 ± 6% and Parthenium hysterophorus 26 ± 6% which caught the same numbers and finally Ricinus communis 13 ± 3% which attracted the fewest mosquitoes into the treatment olfactometer. However, all these responses were significantly higher than the control response Lantana χ2 = 29.43, p<0.0001; Senna χ2 = 57.88, p<0.0001; Parthenium

χ2 = 181.79, p<0.0001 and Ricinus χ2 = 282.75, p<0.0001. We tested a mango essence,

121

made from overripe to rotting mangoes, for use in feeding assay experiments and found it to be more attractive than the control, attracting 31 ± 10 % which was significantly more attractive than the control χ2 = 57.88, p<0.0001 (Fig. 5.2).

Competitive Assays

1. Dual Choice Assays

Based on the results obtained in our non-competitive assays, we eliminated Ricinus communis from this phase of testing. Both Lantana camara and Senna occidentalis were compared to our target plant Parthenium hysterophorus (Fig.5.3). Despite having performed the best in no-choice tests, Lantana was significantly less attractive than

Parthenium in side by side comparisons (χ2 = 184.98, p<0.0001) with the latter attracting

56 ± 11% of mosquitoes released, against 15 ± 3% attracted by Lantana.

The same trend held true when Senna was tested with Parthenium, with the latter trapping significantly more mosquitoes than the former, 42 ± 6% versus 22 ± 2%, when the plants were presented together, indicating significant differences in the attractiveness of the two plants χ2 = 50.66, p<0.0001.

2. Multiple-Choice Assays

When all four plants were presented together in these experiments (Fig.5.4), we recorded a significant difference F3,12 = 8.813, p<0.002 and mean separation by Tukeys HSD

122

showed that Parthenium was significantly more attractive than the other plants, confirming results from previous experiments, and that the Lantana, Senna and Ricinus were not significantly different from each other. In these experiments, Parthenium captured 32 ± 5% of all mosquitoes released, Ricinus 8 ± 3%, Lantana 7 ± 3% and Senna

6 ± 5%. Using temperature, humidity and the plant species in the diffusion-cage olfactometer to predict the response of mosquitoes to a given cage, a significant model emerged, F3,16 = 7.392, p<0.01. The model was then used to show that the plant species was the best predictor of response (Table 5.1) and that the temperature and humidity in the cages did not predict the response.

3. Feeding Assays

Natural Feeding Stations

Mosquito mortality in cages with ATSB-treated Parthenium was almost twice that seen in the control cages F1,782 = 587.98, p<0.0001 (Fig. 5.5), with recorded mortality at 42 ±

6% in the treated cages compared to 24 ± 6% in control (untreated) cages. Whereas there was no significant difference in the sex ratio of mosquitoes that consumed and were killed by ATSB χ2 = 0.24, p>0.05, we observed that more males than females died within

48 h in the control cages χ2 = 13.14, p<0.001. When collecting survivors from both treatment and control cages, we found a majority of mosquitoes perched on Parthenium and Lantana, and the fewest were collected on Ricinus.

123

Artificial Feeding Stations (Toxic Bait Stations)

In these experiments, we demonstrated the efficacy of using stand-alone bait stations

(Fig. 5.6) in sugar rich environments to kill mosquitoes, with almost twice the mortality in cages with artificial toxic bait stations 67 ± 9% over control cages 37 ± 16% χ2 =

77.56, p<0.0001 (Fig. 5.7). Both males and female mosquitoes were equally affected by the ingested toxin χ2 = 2.79, p>0.05. However, in the control cages, significantly more males than females died χ2 = 14.75, p<0.001, following the same trend observed in control cages in the previous experimental setup using natural bait stations.

124

Discussion

This study demonstrates a rapid screening method that can be used to identify the most potent attractive plants frequently visited by mosquitoes in the field, with the aim of using their predominant attractive power to develop natural toxic bait stations for mosquito control. A slightly different approach has previously been demonstrated, by baiting mosquito traps with flowering blossoms of candidate plants growing in a

Mediterranean region in Israel, and using this to screen candidate plants for toxic bait stations against Culex pipiens (Schlein and Müller, 2008b). We also demonstrate the effectiveness of attractive toxic sugar bait (ATSB) sprayed on our most attractive plant, against Anopheles gambie s.s. in semi-field experiments set up in a screenhouse as additional confirmation of our screening method. We further show that even in competitive environments, we are able to achieve drastic population reductions by employing the use of artificial bait stations to deploy an oral insecticide. The use of toxic oral baits has recently emerged as an important method of control against mosquitoes

(Xue and Barnard, 2003; Müller et al., 2010c) and sandflies (Schlein and Müller, 2010;

Müller and Schlein, 2011) in arid areas where there is not too much competition from natural plants. The current study demonstrates the feasibility of applying this same method in resource-rich environments by targeting highly attractive plants for ATSB application, as mosquitoes will most often return to these plants (Schlein and Müller,

2008b). The plants we selected for screening in this study have previously been identified as good nectar hosts for mosquitoes in western Kenya (Impoinvil et al., 2004;

125

Manda et al., 2007), but little is known about their individual attractiveness to mosquitoes in the field (Foster, 1995). Our results show that these plants vary significantly in their attractiveness to mosquitoes, and that this attractiveness changes depending on the type of assay; from non-competitive to competitive assays, showing that when mosquitoes do not have too many options in their immediate environment then they are likely to settle for less suitable alternatives. For instance, in our non-competitive assays we recorded the highest attraction to Lantana camara by capturing 47 ± 7% of all mosquitoes released in an olfactometer baited with the plant, compared to 26 ± 6% collected in an olfactometer baited with Parthenium. However, when assayed against Parthenium in both dual-choice and multiple-choice assays, mosquitoes always preferred the later. This is a significant finding as it shows the potential of using environmental modification as a source- reduction strategy to control mosquito populations when these methods become feasible

(Schlein and Müller, 2010). In ATSB application strategies, we recommend the screening of nectar producing plants in an area to establish both their individual and relative attractiveness when in competition for pollinators with other plants in the vicinity as this will provide critical information for effective bait application on the most attractive plants. Parthenium hysterophorus emerges as a very unlikely candidate as a good mosquito attractant plant, since previous studies have demonstrated its relatively low sugar content (Manda et al., 2007), and survival assays on this plant show population crashes within a week when no alternative forms of sugar are provided to mosquitoes caged with the plant, as opposed to a normal survival curve when sugar is supplied in the

126

same assay, indicating that mosquitoes are not killed by the plant, but that observed mortality is as a result of the low sugar content (P.E.O., pers. obs.). Why mosquitoes would be so attracted to the volatiles of this plant when little or no sustenance is derived remains a conundrum, but perhaps the volatiles produced are simply allomonic with pollination benefit derived by the plant, to the detriment of the mosquitoes. Another speculation is that since this plant produces toxic sesquiterpene lactones and phenolic compounds that are allelopathic (Picman and Picman, 1984) and harmful to both grazers and humans (Towers and Mitchell, 1983), it would rarely be disturbed where it grows, and thus provides a safe resting site for mosquitoes. How this ―safety‖ would be encoded in the volatile profile (Table 1.1) is a conundrum. Temperature and humidity both play a role in the close range orientation towards host volatiles in insects (Wright and Kellogg,

1962; De Jong and Knols, 1995b), but in our case, though temperature and humidity were positively correlated, nearly identical ranges in the diffusion-cage olfactometers did not inform the response. Poisoning mosquitoes using ATSB was very effective in these experiments; with close to twice the mortality in treatment over control cages in both natural and artificial bait stations. However, we did observe higher than usual mortality in control cages, which may have been caused by unprecedentedly high summer temperatures, and defective greenhouse cooling systems. Artificial bait stations recorded great success in killing mosquitoes despite competition with natural plants, and this could have been due to the relative ease of obtaining a quick sugar meal from these stations over the more energy intensive search needed when accessing plant sugars. These bait

127

stations bear several advantages including their ease of use, versatility in different environments, and the fact that non-target effects are easier to control. The greatest disadvantage to using artificial bait stations is that they constantly need to be maintained, ensuring that they are adequately supplied with toxic bait and that insecticidal effects are not lost over periods of exposure to the elements. In our experiments, it was encouraging to see the similar sex ratio of mosquitoes that were felled by the oral insecticide in treatment cages, as it confirms that both sexes were attracted to and consumed the poison, and offers great promise for future field trials and population control efforts. The different sex ratio observed in the mortality of mosquitoes in control cages is not surprising, since male mosquitoes tend to have lower energy reserves than females at emergence (Magnarelli, 1983) and as such easily succumb to starvation when unable to quickly locate a sugar meal.

ATSB employs the use of an oral insecticide, such as boric acid (Xue and Barnard, 2003;

Xue et al., 2011) or as we used in this study, Spinosad, which is advantageous as it circumvents the ill effects of the widespread application of insecticides through its more rational targeted delivery (Müller et al., 2010c; Müller and Schlein, 2011). In conclusion, we recommend the use of both natural and artificial bait stations in mosquito control as both methods may prove pragmatic in different settings and since both methods were highly effective in a semi-field setting in the presence of competition.

128

Acknowledgements – We thank Ashley Jackson for help in rearing mosquitoes. This research was supported through NIH grant #R01-AI077722.

129

References

DE JONG, R. and KNOLS, B. G. 1995. Selection of biting sites on man by two malaria

mosquito species. Cell. Mol. Life Sci. 51:80-84.

DOWNES, J. A. 1958. The feeding habits of biting flies and their significance in

classification. Annu. Rev. Entomol. 3:249-266.

FOSTER, W. A. 1995. Mosquito sugar feeding and reproductive energetics. Annu. Rev.

Entomol. 40:443-74.

GARY, R. E. and FOSTER, W. A. 2001. Effects of available sugar on the reproductive

fitness and vectorial capacity of the malaria vector Anopheles gambiae (Diptera:

Culicidae). J. Med. Entomol. 38:22-28.

GU, W., MÜLLER, G., SCHLEIN, Y., NOVAK, R. J., and BEIER, J. C. 2011. Natural

plant sugar sources of Anopheles mosquitoes strongly impact malaria

transmission potential. PLoS One 6:e15996.

HOCKING, B. 1953. The intrinsic range and speed of flight of insects. Trans. Roy.

Entomol. Soc. London 104:223-346.

130

IMPOINVIL, D. E., KONGERE, J. O., FOSTER, W. A., NJIRU, B. N., KILLEEN, G.

F., GITHURE, J. I., BEIER, J. C., HASSANALI, A., and KNOLS, B. G. J. 2004.

Feeding and survival of the malaria vector Anopheles gambie on plants growing

in Kenya. Med. Vet. Entomol. 18:108-115.

MAGNARELLI, L. A. 1983. Nectar sugars and caloric reserves in natural populations of

Aedes canadensis and Aedes stimulans (Diptera: Culicidae). Environ. Entomol.

12:1482-1486.

MANDA, H., GOUAGNA, L. C., NYANDAT, E., KABIRU, E., JACKSON, R. R.,

FOSTER, W. A., GITHURE, J. I., BEIER, J. C., and HASSANALI, A. 2007.

Discriminative feeding behaviour of Anopheles gambiae s.s. on endemic plants in

western Kenya. Med. Vet. Entomol. 21:103-111.

MÜLLER, G. C., BEIER, J. C., TRAORE, S. F., TOURE, M. B., TRAORE, M. M.,

BAH, S., DOUMBIA, S., and SCHLEIN, Y. 2010a. Field experiments of

Anopheles gambiae attraction to local fruits/seedpods and flowering plants in

Mali to optimize strategies for malaria vector control in Africa using attractive

toxic sugar bait methods. Malar. J. 9:262.

131

MÜLLER, G. C., JUNNILA, A., QUALLS, W., REVAY, E. E., KLINE, D. L., ALLAN,

S., SCHLEIN, Y., and XUE, R. D. 2010b. Control of Culex quinquefasciatus in a

storm drain system in Florida using attractive toxic sugar baits. Med. Vet.

Entomol. 24:346-351.

MÜLLER, G. C., JUNNILA, A., and SCHLEIN, Y. 2010c. Effective control of adult

Culex pipiens by spraying an attractive toxic sugar bait solution in the vegetation

near larval habitats. J. Med. Entomol. 47:63-66.

MÜLLER, G. C., KRAVCHENKO, V. D., and SCHLEIN, Y. 2008. Decline of

Anopheles sergentii and Aedes caspius populations following presentation of

attractive toxic (Spinosad) sugar bait stations in an oasis. J. Am. Mosq. Cont.

Assoc. 24:147-149.

MÜLLER, G. C. and SCHLEIN, Y. 2011. Different methods of using attractive sugar

baits (ATSB) for the control of Phlebotomus papatasi. J. Vector Ecol. 36

(Supplement 1):S64-S70.

PICMAN, J. and PICMAN, A. K. 1984. Autotoxicity in Parthenium hysterophorus and

its possible role in control of germination. Biochem. Syst. Ecol. 12:287-292.

132

SCHLEIN, Y. and MÜLLER, G. C. 2008. An approach to mosquito control: Using the

dominant attraction of flowering Tamarix jordanis trees against Culex pipiens. J.

Med. Entomol. 45:384-390.

SCHLEIN, Y. and MÜLLER, G. C. 2010. Experimental control of Phlebotomus papatasi

by spraying attractive toxic sugar bait (ATSB) on vegetation. Trans. R. Soc. Trop.

Med. Hyg. 104:766-71.

STONE, C. M., TAYLOR, R. M., and FOSTER, W. A. 2009. An effective indoor

mesocosm for studying populations of Anopheles gambiae in temperate climates.

J. Am. Mosq. Cont. Assoc. 25:514-516.

TOWERS, G. H. N. and MITCHELL, J. C. 1983. The current status of the weed

Parthenium hysterophorus L. as a cause of allergic contact dermatitis. Contact

Derm. 9: 465-469.

WHO. 2000. WHO Expert Committee on Malaria. Geneva: World Health Organisation

Tech Rep Ser. i-iv;1-74 p.

WRIGHT, R. H. and KELLOGG, F. E. 1962. Response of Aedes aegypti to moist

convection currents. Nature 194:402 - 403.

133

XUE, R. D. and BARNARD, D. R. 2003. Boric acid bait kills adult mosquitoes (Diptera:

Culicidae). J. Econ. Entomol. 95:1559-1562.

XUE, R. D., MÜLLER, G. C., KLINE, D. L., and BARNARD, D. R. 2011. Effect of

application rate and persistence of boric acid sugar baits applied to plants for

control of Aedes albopictus. J. Am. Mosq. Cont. Assoc. 27:56-60.

134

Figure 5.1. Non-competitive assays of four plants in diffusion-cage olfactometers set up in a small mesocosm. Differences represented by (***) are significant at p<0.0001.

135

Figure 5.2. Non-competitive assays of a Mango Mangifera indica in diffusion-cage olfactometers set up in a small mesocosm. Differences represented by (***) are significant at p<0.0001.

136

Figure 5.3. Competitive assays (dual choice) of Parthenium against Lantana and Senna in diffusion-cage olfactometers set up in a small mesocosm. Differences represented by

(***) are significant at p<0.0001.

137

Figure 5.4. Competitive assays (multiple-choice) of four African plants in diffusion-cage olfactometers set up in a large screenhouse. Letters above bars represent mean separations by Tukeys HSD post hoc test.

138

SE SE ± 40 ***

Percent Mortality Mortality Percent 10

0 ATSB Control Treatment

30 ATSB

SE 25 ± 20 Control

5 Percent Mortality Percent

0 Female Male Sex

Figure 5.5. Attractive toxic bait associated mortality in screenhouse experiments containing ATSB enhanced Parthenium hysterophorus plants. Differences represented by (***) are significant at p<0.0001.

139

Wooden support

Cotton wicks (distribution)

Rubber tie

ATSB solution Supply wick

Figure 5.6. Toxic bait station used to dispense attractive toxic sugar bait (ATSB) in screenhouse experiments.

140

SE SE 70 *** ± 60 50 40 30 20

10 Percent Mortality Mortality Percent 0 Bait Station Control Treatment

40 SE SE

± 35 30 25 20 Bait Station 15 Control 10

Percent Mortality Mortality Percent 5 0 Female Male Treatment

Figure 5.7. Toxic bait station associated mortality in a sugar rich environment (A)

Overall mortality (B) Mortality by sex. Differences represented by (***) are significant at p<0.0001.

141

Table 5.1. Estimated coefficients and standard errors for the multiple regression model predicting response, based on

humidity, temperature and plant species in a diffusion-cage olfactometer.

R2 =0.581 *** p<0.0001

Plant in Mean Mean Model B SE b β

Olfactometer Temp (°C) Humidity

142 Parthenium 24.6 ± 0.3 38.0 ± 5.1 Humidity 1.771 0.08 -.482

Ricinus 24.7 ± 0.5 37.0 ± 6.3 Temperature -10.051 0.15 -.245

Lantana 24.7 ± 0.5 33.0 ± 3.6 Plant sp. -30.839 -.818***

Senna 24.5 ± 0.5 48.3 ± 4.7

142

General Conclusions

The studies conducted here demonstrate the process by which chemicals naturally produced by plants can be harnessed and used to for surveillance and possible control of pest species. We highlight the sugar feeding behavior of mosquitoes, a common occurrence in the adult stage, and a critical function for survival that affects dispersal, reproduction and vector competence. We hypothesize that we can use this ubiquitous activity to develop methods of mosquito surveillance and control that are unique in the following ways:

1. Both male and female mosquitoes would be attracted to phytochemical based

attractants since both sexes sugar feed. An important distinction has to be made

from blood feeding because only female anautogenous mosquitoes consume

blood which they need for egg development. Attractants that target males as well

as females would make good early warning surveillance tools since males do not

disperse as far as females, thus high trap catches would indicate foci of

emergence. In addition to this, males tend to emerge a few hours sooner than

females.

2. Female mosquitoes in all gonotrophic stages have been shown to ingest sugar,

hence traps that utilize plant attractants would give a good representation of the

population at an area. Other types of attractants tend to be highly selective, either

trapping only gravid females in the case of skatole baited gravid traps, or blood

143

thirsty females in octanol and carbon dioxide baited traps. Very few males are

caught in these traps.

3. Plant based attractants as easily integrated to existing surveillance and control

programs and require very low maintenance when properly administered. They

can also be used with different types of traps.

The studies listed here demonstrate a logical sequence to creating synthetic blends. First, plants visited by mosquitoes in a natural setting are identified – these are nectar bearing plants that mosquitoes visit during their normal foraging activities. The second step is to perform solvent extractions of the volatile phase of these plants, and this can be targeted to the floral volatiles which normally have sweet fragrances that attract pollinators, or in some cases leaves and other plant organs may produce these volatiles.

Third, the solvent extract is tested in the lab in bioassays, to see if it has a similar activity to the real plant. The fourth step is to develop a synthetic blend using the chemicals identified in the extract as a guide, and testing these to see if similar levels of attraction are achievable. The fourth step is a little challenging as it involves a complicated and tedious process of quantification of plant volatiles and may take some time to achieve.

Behavioral assays are also time-consuming, but their results are conclusive since a hungry mosquito will always orient to food volatiles. Several screening stages are involved in determining the active components of a synthetic blend, and this involves subtractive assays where individual blend components are removed, one by one until you 144

know what effect each has. Electophysiological approaches have been touted as a short cut to this step, but besides telling you whether a mosquito can detect a certain volatile or not, one does not really know whether an action potential is due to an attractant or repellant and as such, assays still need to be done.

The last stage of ―blendology‖ is the most challenging – testing the blend in the field! As we found in our studies, what works in the lab does not necessarily work in the field. We achieved good levels of attraction to synthetic blends in olfactometer experiments in the lab, but had significant challenges in the field. This does not mean that plant attractants are doomed to fail, but that more research has to go into establishing the most effective release rate for phytochemical attractants in a field setting. One of the factors we considered in the course of our experiments was the possibility that blends cause a ―scent garden‖ effect, attracting mosquitoes from afar to the vicinity of the point of release of the blend, and explaining the high numbers caught in control traps. Considering changes in wind speed and direction this is a plausible hypothesis but would need further investigation.

We developed a novel olfactometer device - a response to the need for the simple olfactometer devices that can be used in a not-so-controlled setting that is the hallmark of lab-based olfactometers. We developed a device that is effective for use in a screenhouse, easy in construction and as accurate as the more controlled lab systems, yet

145

more effective in stimulating mosquito flight, hence giving a more natural response. This system also allows the use of whole potted plants, and this is particularly useful in determining field-based orientation because whole-plant headspace is used. The diffusion-cage olfactometer also allows for high throughput, as many devices can be set up and used at the same time.

The most field-ready part of this work is the use of attractive toxic sugar bait (ATSB) to kill mosquitoes. As we were able to demonstrate, even in resource rich environments, we were still able to kill twice the number of mosquitoes using ATSB than those that died by natural attrition in a control setting. This method could easily be integrated into existing vector control programs for that extra kill, and promises to be an effective means of control.

As we learn more about mosquito behavior and the molecular basis for olfaction, newer tools will become available in the arsenal against vectors of disease.

146

Appendix A: Chemicals contained in the headspace of various plants visited by

mosquitoes in temperate and tropical regions

147

Table A.1. Headspace profiles of some plants frequently visited by mosquitoes in tropical and temperate regions

CHEMICAL NAME QAL DFB TB GDR MWD PH SO LC SD RC

Alcohols

1-Octen-3-ol X

3-Hexen-1-ol X X

3-Octanol X

148 Hexan-2-ol X

Cyclopentanol X

Aldehydes

Hexanal X X

2-Hexanal X X

148

Table A.1. CONTINUED CHEMICAL NAME QAL DFB TB GDR MWD PH SO LC SD RC

Octanal

Nonanal X X X

Decanal X X

trans-2-Nonenal X

149

Acids

Benzoic acid X

Nonanoic acid X X X

Dodecanoic acid X X

(CONTINUED)

149

Table A.1. CONTINUED CHEMICAL NAME QAL DFB TB GDR MWD PH SO LC SD RC

Tetradecanoic acid X X

Hexadecanoic acid X X

Pentadecanoic acid X

150 Terpenoids

α-Thujene X

α-Pinene X X X X X

Camphene X X X X

Sabinene X X

(CONTINUED)

150

Table A.1. CONTINUED

CHEMICAL NAME QAL DFB TB GDR MWD PH SO LC SD RC

β-Pinene X X X X

α-Myrcene X

β -Myrcene X X X X X

151 α-Phellandrene X X X X

β-Phellandrene X X X

γ-Terpinene X X X

Limonene X X X X X

2-Carene X

(CONTINUED)

151

Table A.1. CONTINUED CHEMICAL NAME QAL DFB TB GDR MWD PH SO LC SD RC

3-Carene X X

4-Carene X X X

Bornyl acetate X X

152 γ-Elemene

α-Caryophyllene X X X

β-Caryophyllene X X

α-Gurjunene X

Aromadendrene

o-Cymene X

(CONTINUED)

152

Table A.1. CONTINUED CHEMICAL NAME QAL DFB TB GDR MWD PH SO LC SD RC

p-Cymene X

Terpinolene X X X

153 Z-β-Ocimene X X X X X X X

E-β-Ocimene X X X

Allo-ocimene X X

Eucalyptol X X

β-Cubebene X

α-Muurolene X X

(CONTINUED)

153

Table A.1. CONTINUED CHEMICAL NAME QAL DFB TB GDR MWD PH SO LC SD RC

β-Cadinene X

Cedrene X

β-Bourbonene X

154 Cyclosativine X

Copaene X

Z-3-Hexenyl acetate X

3-Hexen-1-ol acetate (Z) X

Z-Sabinene hydrate X

(CONTINUED)

154

Table A.1. CONTINUED CHEMICAL NAME QAL DFB TB GDR MWD PH SO LC SD RC

Z-β-Terpineol X

Aromatics

Benzyl alcohol X

155 Benzaldehyde X X

Phenylethyl alcohol X X

Phenylacetaldehyde X

Anthraquinone X

(CONTINUED)

155

Table A.1. CONTINUED

QAL= Queen Anne’s lace Daucus carota DFB= Daisy Fleabane Erigeron strigosus TB=Tall Boneset

Eupatorium altissimum GDR= Canada goldenrod Solidago canadensis MWD= Common Milkweed

Asclepias syriaca PH= Whitetop weed Parthenium hysterophorus SO= Coffee senna Senna occidentalis

156 LC= Lantana camara SD= African senna Senna didymobotrya RC= Castor Bean Ricinus communis

156

Appendix B: Olfactory responses in a gustatory organ of Culex pipiens and Anopheles

gambiae to various plant volatiles

Abstract

Mosquitoes have three olfactory appendages, the antenna, the proboscis and the maxillary palps, all innervated by olfactory receptor neurons (ORNs). The antenna is the main chemosensory organ in insects, containing the largest quantity of olfactory sensilla, and the maxillary palps are also important chemosensory organs, but are less complex and only only contain a single type of sensillum, the capitates peg. Electophysiological studies are useful in determining the the detection of host volatiles encountered by insects in the environment, and present a useful step in the development of synthetic attractants for pest species.

In this study we examine the maxillary-palp responses of Culex pipiens and Anopheles gambiae to 17 common plant volatiles using sensilla recordings. We report a broad spectrum of detection in both species of mosquitoes of volatiles commonly found in plants.

Introduction

Insect chemosensory organs detect a wide range of volatile organic compounds that are useful in locating both vertebrate (Meijerink et al., 2000; Birkett et al., 2004) and plant hosts (Takken and Knols, 1999; Syed and Leal, 2007) and are often used by insects in

157

mate recognition and finding suitable oviposition sites (Takken and Knols, 1999). The antennae have the largest quantity and variety of olfactory sensilla, but a single type of chemosensory sensillum on the fourth segment of the maxillary palp, the capitate peg, or maxillary palp basiconic sensilla, is important for insect orientation at close range due to its very high sensitivity (Grant and O’Connell, 1996). Two of the most widely used chemostimuli in mosquito research, carbon dioxide and 1-octen-3-ol, are detected by

ORNs located in the peg sensilla on maxillary palps in C. quinquefasciatus (Syed and

Leal, 2007) and Aedes aegypti (Grant and O’Connell, 1996). Studies have also shown sensitivity of these receptors to a number of green-plant volatiles, with a noted preference of (E) – over (Z) – isomers in alcohol groups (Syed and Leal, 2007).

Odor-mediated behavior of mosquitoes has a strong bearing on their vector status, since host preference is dictated by how well mosquitoes can locate their hosts (Zwiebel and

Takken, 2004). In this study, we look at the responses of Culex pipiens and Anopheles gambiae s.s. to some plant volatiles released by nectar bearing plants in regions where the two species exist.

Materials and Methods

Insects

Experiments were conducted with Cx. pipiens from a colony established in 2009 from a larval development site near Columbus, Ohio, U.S.A. Rearing was done as previously

158

described (Chapter 1). Emerging adults were given ad libitum access to water and sugar.

Experiments were conducted 36 ± 12 h after emergence. The mosquito rearing and maintenance conditions were 27 ± 1 °C, 85 ± 5% RH, and 16:8 (L:D), with 30-min gradual crepuscular transitions between photophase and scotophase.

An. gambiae s.s. came from a colony established in 2001 by staff at the International

Centre of Insect Physiology and Ecology from the local population of An. gambiae s.s. in

Mbita Point, Suba District, Nyanza, Kenya, and identified by polymerase chain reaction

(PCR). Water and 10% sucrose solution were available to colony adults ad libitum.

Mosquitoes were reared as previously described (Chapter 5). The colony was maintained in acrylic cages at 26.6 ± 1°C and 80 ± 5% RH.

Chemicals Phenylacetaldehyde (90+%), benzaldehyde (≥99.5%, purified by redistillation), nonanal (≥95%), (Z)-jasmone (≥97%), (E)-2-nonenal (97%), α-cedrene, 1- nonene, hexanal, myrcene, α-humulene, D-limonene, α-phellandrene, 1-coten-3-ol, nonanoic acid, α-pinene, α-gurjunene, bornyl acetate and β-pinene were all purchased from Sigma-Aldrich® (Saint Louis, MO, USA). β-ocimene (70% (E)-β-ocimene) was synthesized by CHEMOS GmbH (Regenstauf, Germany). Synthetics were diluted using

HPLC grade n-pentane (Fisher Scientific). These chemicals were chosen due to their occurance in plants that mosquitoes often visit for nectar, and because we have previously used some of them to make synthetic milkweed and goldenrod blend.

159

Electrophysiology

Electropalpograms were recorded from excised heads of female mosquitoes with Ag-AgCI electrodes, as previously described (Blackwell et al., 1993) with changes to suit the maxillary palps. The indifferent electrode (glass capillary, inner diameter =

0.86mm), filled with Beadle-Ephrussi Ringer (BER) Solution, was inserted into the base of the head and the BER-filled recording electrode was brought into contact with the cut end of a maxillary pulp. To prolong the life of the preparation, high humidity was maintained by placing a damp filter paper immediately below the preparation. Two microlitres of the test solution were applied to a filter paper (1.5 x 8 mm) and inserted into a glass Pasteur pipette. The maxillary palp preparation was continuously exposed to a charcoal-filtered, humidified airstream (600 ml/min), into which stimulus molecules were injected, pulsed at 60-s intervals (1 litre/ min, 0.3-s pulses) and regulated by a stimulus controller (Syntcch). Each maxillary palp was stimulated by a series of test compounds at intervals of 30-60s. Recorded extracellular action potentials (APs) were amplified 1000 and fed into an IDAC4-USB box (Syntech, Hilversum, The Netherlands) via a high-impedance preamplifier and recorded on the hard disk of a PC via a 16-bit analogue–digital IDAC4-USB box and analyzed with the software Auto Spike v. 3.7

(Syntech).

To compensate for palp fatigue, responses were normalized using a standard stimulus

(0.1 µg D-limonene and (E)-β-ocimene for Culex pipiens and Anopheles gambiae respectively).

160

Statistical analysis

Threshold values were calculated by comparing control values with normalized test values (Student’s t tests). Electropalpograms were recorded to the bioassay test compounds and to the additional compounds named below.

Results and Discussion

For Anohpeles gambiae we found two compounds to be significantly different from the control; benzaldehyde (t= -5.210 df = 1 p= 0.035) and α-gurjenene (t= -5.985 df = 1 p=

0.027) (Fig.B.1) whereas Culex pipiens reported (E)-β-ocimene (t= 19.606 df = 1 p=

0.032) and myrcene (t= 119.313 df = 1 p= 0.005) (Fig. B.2) above threshold values. It must be clear however, that low responses towards the other compounds does not mean that mosquitoes were unable to perceive them, but that at the concentration we tested, 0.1

µg, the compounds above produced above background potentials.

Insects, like mammals, have the ability to recognize and discriminate a wide variety of odor molecules. Molecules present in the air are bound to receptors on the dendrites of sensory structures like the antenna and maxillary palps which terminate at sensory neurons that transmit, via axons, the information to higher brain centers and subsequently to motor pathways that control behavior (Keller and Vosshall, 2003). It is in the brain that this information is decoded and discriminated as different odors or blends of odors using a complex matrix of activated receptors (Buck and Axel, 1991; Keller and

Vosshall, 2003) create a sophisticated neural map. It is well established that the

161

concentration of chemicals plays a role in the number and nature of receptor activation

(Hansson and Anton, 2000) and as such completely different behavioral patterns can be observed over a single decadic step.

The next stage of this work will involve making decadic dilutions of these compounds to find out the range of their activity, and to establish the response to these compounds in a blended form.

162

References

BIRKETT, M. A., AGELOPOULOS, N., JENSEN, K.-M. V., JESPERSEN, J. B.,

PICKETT, A. J., PRIJS, H. J., THOMAS, G., TRAPMAN, J. J., WADHAMS, L.

J., and WOODCOCK, C. M. 2004. The role of volatile semiochemicals in

mediating host location and selection by nuisance and disease-transmitting cattle

flies. Med. Vet. Entomol. 18:313-322.

BLACKWELL, A., MORDUE (LUNTZ), A. J., HANSSON, B. S., WADHAMS, L. J.,

and PICKETT, A. J. 1993. A behavioural and electrophysiological study of

oviposition cues for Culex quinquefasciatus. Physiol. Entomol. 18:343-348.

BUCK, L. and AXEL, R. 1991. A novel multigene family may encode odorant

receptors: A molecular basis for odor recognition Cell 65:175-187.

GRANT, A. J. and O’CONNELL, R. J., editors. 1996. ElectrophysioIogicaI responses

from receptor neurons in mosquito maxillary palp sensilla: Wiley, Chichester.

HANSSON, B. S. and ANTON, S. 2000. Function and morphology of the antennal lobe:

New developments. Ann. Rev. Entomol. 45:203-231.

KELLER, A. and VOSSHALL, L. B. 2003. Decoding olfaction in Drosophila. Curr.

Opin. Nurobiol.:103-110. 163

MEIJERINK, J., BRAKS, M. A. H., BRACK, A. A., ADAM, W., DEKKER, T.,

POSTHUMUS, M. A., BEEK, T. A. V., and LOON, J. J. A. V. 2000.

Identification of olfactory stimulants for Anopheles gambiae from human sweat

samples. J. Chem. Ecol. 26:1367-1381.

SYED, Z. and LEAL, W. 2007. Maxillary palps are broad spectrum odorant detectors in

Culex quinquefasciatus. Chem. Senses 32:727-738.

TAKKEN, W. and KNOLS, B. 1999. Odour Mediated behaviour of afrotropical malaria

mosquitoes. Ann. Rev. Entomol. 44:131-157.

ZWIEBEL, L. J. and TAKKEN, W. 2004. Olfactory regulation of mosquito–host

interactions. Insect Biochem. Mol. Biol. 34:645-652.

164

(E)-β-ocimene* Paraffin oil (control) beta-Pinene Nonanal Benzaldehyde Bornyl acetate alpha-Gurjunene (E)-β-ocimene* alpha-Pinene Nonanoic acid 1-Octen-3-ol alpha-Phellandrene D-Limonene alpha-Humulene Myrcene Hexanal 1-Nonene Z-β-ocimene (E)-β-ocimene* alpha-cedrene

0 0.5 1 1.5 2 2.5 Response (mV)

Figure B.1. Anopheles gambiae electroantennogram screening performed with a series of 17 odorant compounds and with parrafin oil as a control. Stimulus dose: 0.1 µg of each compound per filter paper. Each column represents the mean response of 2 females. Error bars indicate 95% confidence interval as based on the T- distribution. (*) standard stimulus.

165

Limonene* Control B-Pinene Nonanal Benzaldehyde (E)-β-ocimene Bornyl acetate alpha-Gurjunene Limonene* alpha-pinene Nonanoic acid 1-Octen-3-ol alpha-Phellandrene Limonene alpha-Humulene Myrcene Hexanal 1-Nonene Z-B-Ocimene alpha-Cedrene Limonene*

0 0.5 1 1.5 2 2.5 3 Response (mV)

Figure B.2. Culex pipiens electroantennogram screening performed with a series of 17 odorant compounds and with parrafin oil as a control. Stimulus dose: 0.1 µg of each compound per filter paper. Each column represents the mean response of 2 females. Error bars indicate 95% confidence interval as based on the T-distribution. (*) standard stimulus.

166

Bibliography

ADAMS, R. P. 1995. Identification of Essential Oil Components by Gas

Chromatography/Mass Spectroscopy. Allured Publishing, Carol Stream, IL.

AGELOPOULOS, N., BIRKETT, M. A., HICK, A. J., HOOPER, A. M., PICKETT, J.

A., POW, E. M., SMART, L. E., SMILEY, D. W. M., WADHAMS, L. J., and

WOODCOCK, C. M. 1999. Exploiting Semiochemicals in Insect Control. Pestic.

Sci. 55:225-235.

ANDERSSON, F. J., FERRRANDINO, F. J., MCKNIGHT, S., NOLEN, J., and

MILLER, J. 2009. A carbon dioxide, heat and chemical lure trap for the bedbug,

Cimex lectularius. Med. Vet. Entomol. 23:99-105.

ANDERSSON, I. H. and JAENSON, T. G. T. 1987. Nectar feeding by mosquitoes in

Sweden with special reference to Culex pipiens and Cx torrentium. Med. Vet.

Entomol. 1:59-64.

BATES, M. 1949. The natural history of mosquitoes. New York: The Macmillan Co. p

379 pp.

167

BENEDICT, M. Q., LEVINE, R. S., HAWLEY, W. A., and LOUNIBOS, P. 2007.

Spread of the tiger: global risk of invasion by the mosquito Aedes albopictus.

Vector-Borne Zoonot. 7:76-85.

BENTLEY, M. D., KAUFFMAN, P. E., KLINE, D. L., and HOGSETTE, J. A. 2009.

Response of adult mosquitoes to light-emitting diodes placed in resting boxes and

in the field. J. Am. Mosq. Cont. Assoc. 25:285-291.

BINDLINGMAYER, W. L. and HEM, D. G. 1973. Sugar feeding by Florida mosquitoes.

Mosq. News 33:535-538.

BIRKETT, M. A., AGELOPOULOS, N., JENSEN, K.-M. V., JESPERSEN, J. B.,

PICKETT, A. J., PRIJS, H. J., THOMAS, G., TRAPMAN, J. J., WADHAMS, L.

J., and WOODCOCK, C. M. 2004. The role of volatile semiochemicals in

mediating host location and selection by nuisance and disease-transmitting cattle

flies. Med. Vet. Entomol. 18:313-322.

BLACKWELL, A., MORDUE (LUNTZ), A. J., HANSSON, B. S., WADHAMS, L. J.,

and PICKETT, A. J. 1993. A behavioural and electrophysiological study of

oviposition cues for Culex quinquefasciatus. Physiol. Entomol. 18:343-348.

168

BOWEN, M. F. 1992a. Patterns of sugar feeding in diapausing and nondiapausing Culex

pipiens (Diptera: Culicidae) females J. Med. Entomol. 29:843-849.

BOWEN, M. F. 1992b. Terpene-sensitive receptors in female Culex pipiens mosquitoes:

Electrophysiology and behaviour. J. Insect Physiol. 38:759-764.

BRANTJES, N. B. M. and LEEMANS, J. A. A. M. 1976. Silene otites (Caryophyllaceae)

pollinated by nocturnal Lepidoptera and mosquitoes. Acta Bot. Neerl. 25:281-295.

BREELAND, S. G. and PICKARD, E. 1965. The malaise trap - an effecient and unbiased

mosquito collecting device. Mosq. News 25:19-21.

BRUCE, T. J. A., WADHAMS, L. J., and WOODCOCK, C. M. 2005. Insect host

location: a volatile situation. Trends Plant Sci. 10:269-274.

BUCK, L. and AXEL, R. 1991. A novel multigene family may encode odorant

receptors: A molecular basis for odor recognition Cell 65:175-187.

BURKETT, D. A., KLINE, D. L., and CARLSON, D. A. 1999. Sugar meal composition

of five north central Florida mosquito species (Diptera: Culicidae) as determined

by Gas Chromatography. J. Med. Entomol. 36:462-467. 169

CARDÉ, R. T. 2007. Odour plumes and odour-mediated flight in insects. In: Bock GR,

Cardew G, editors. Ciba Foundation Symposium 200 - Olfaction in Mosquito-

Host Interactions. Chichester, UK: John Wiley & Sons, Ltd.

CDC. 2010. Surveillance for Human West Nile Virus Disease — United States, 1999–

2008. MMWR CDC Surveill Summ 59(No. SS-2) [April 2, 2010].

CHAPMAN, R. F. 1988. Odors and the feeding behaviors of insects. In: Grimwade AM,

editor. ISI Atlas Sci-Anim Plant Sci. Philadelphia: Institute for Scientific

Information Inc. p 208-212.

CLEMENTS, A. N. 1999. The biology of mosquitoes. Sensory Reception and Behavior.

CABI Publishing, New York

COETZEE, M., CRAIG, M., and LE SUEUR, D. 2000. Distribution of African malaria

mosquitoes belonging to the Anopheles gambiae complex. Parasitol. Today

16:74-77.

COOPERBAND, M. F. and CARDÉ, R. T. 2006a. Comparison of plume structures of

carbon dioxide emitted from different mosquito traps. Medical and Veterinary

Entomology 20:1-10. 170

COOPERBAND, M. F. and CARDÉ, R. T. 2006b. Orientation of Culex mosquitoes to

carbon dioxide-baited traps: flight manoeuvres and trapping effeciency. Medical

and Veterinary Entomology 20:11-26.

CUI, S., TAN, S., OUYANG, G., JIANG, S., and PAWLISZYN, J. 2009. Headspace

solid-phase microextraction gas chromatography–mass spectrometry analysis of

Eupatorium odoratum extract as an oviposition repellent. J. Chromatogr. B

877:1901-1906.

DAVIS, J. D., HALL, T., CHEE, E. M., MAJALA, A., MINJAS, J., and SHIFF, C. J.

1995. Comparison of sampling anopheline mosquitoes by light-trap and human-

bait collections indoors at Bagamoyo, Tanzania. Med. Vet. Entomol. 9:249-255.

DE JONG, R. and KNOLS, B. G. 1995a. Olfactory responses of host-seeking Anopheles

gambiae s.s. Giles (Diptera: Culicidae). Acta Trop. 59:333-335.

DE JONG, R. and KNOLS, B. G. 1995b. Selection of biting sites on man by two malaria

mosquito species. Cell. Mol. Life Sci. 51:80-84.

171

DEKKER, T., GEIER, M., and CARDÉ, R. T. 2005. Carbon dioxide instantly sensitizes

female yellow fever mosquitoes to human skin odours. J. Exp. Biol. 208:2963-

2972.

DETHIER, V. G. 1982. Mechanism of host-plant recognition. Ent. exp. & appl. 31:49-56.

DOWNES, J. A. 1958. The feeding habits of biting flies and their significance in

classification. Annu. Rev. Entomol. 3:249-266.

EDMAN, J. D., STRICKMAN, D., and KITTAYAPONG, P. 1992. Female Aedes

aegypti (Diptera : Culicidae) in Thailand rarely feed on sugar. J. Med. Entomol.

29:1035-1038.

EDMAN, J. D. and TAYLOR, D. J. 1968. Culex nigripalpus: Seasonal shiftin the bird-

mammal feeding ratio in a mosquito vector of human ecephalitis. Science 161:67-

68.

FACCHINELLI, L., VALERIO, L., POMBI, M., REITER, P., COSTANTINI, C., and

DELLA TORRE, A. 2007. Development of a novel sticky trap for

containerbreeding mosquitoes and evaluation of its sampling properties to

monitor urban populations of Aedes albopictus. Med. Vet. Entomol. 21:183-195. 172

FEINSOD, F. M. and SPIELMAN, A. 1980. Nutrient-mediated juvenile hormone

secretion in mosquitoes. J. Insect Physiol. 26:113-117.

FERGUSON, H. M., DORNHAUS, A., BEECHE, A., BORGEMEISTER, C.,

GOTTLIEB, M., MULLA, M. S., GIMNIG, J. E., FISH, D., and KILLEEN, G. F.

2010. Ecology: A prerequisite for malaria elimination and eradication. PLoS Med

7:e1000303.

FERGUSON, H. M., NG'HABI, K. R., WALDER, T., KADUNGULA, D., MOORE, S.

J., LYIMO, I., RUSSELL, T. L., URASSA, H., MSHINDA, H., KILLEEN, G. F.,

and KNOLS, B. G. 2008. Establishment of a large semi-field system for

experimental study of African malaria vector ecology and control in Tanzania.

Malar. J. 7:158.

FERREIRA, L., PERESTRELO, R., CALDEIRA, M., and CÂMARA, J. S. 2009.

Characterization of volatile substances in apples from Rosaceae family by

headspace solid-phase microextraction followed by GC-qMS. J. Sep. Sci.

32:1875-1888.

173

FOSTER, W. A. 1995. Mosquito sugar feeding and reproductive energetics. Annu. Rev.

Entomol. 40:443-74.

FOSTER, W. A. 2008. Phytochemicals as population sampling lures. J. Am. Mosq. Cont.

Assoc. 24.

FOSTER, W. A. and HANCOCK, R. G. 1994. Nectar-related olfactory and visual

attractants for mosquitoes. J. Am. Mosq. Cont. Assoc. 10:288-296.

FOSTER, W. A. and TAKKEN, W. 2004. Nectar-related vs. human-related volatiles:

behavioural response and choice by female and male Anopheles gambiae

(Diptera: Culicidae) between emergence and first feeding. Bull. Entomol. Res.

94:145-157.

FRAENKEL, G. S. 1959. Raison d'être of secondary plant substances. Science 129:1466-

1470.

GABEL, B., THIÉRY, D., SUCHY, V., MARION-POLL, F., HRADSKY, P., and

FARKAS, P. 1992. Floral volatiles of Tanacetum vulgare L. attractive to

Lobesia botrana DEN. et SCHIFF. females J. Chem. Ecol. 18:693-701.

174

GARY, R. E. and FOSTER, W. A. 2001. Effects of available sugar on the reproductive

fitness and vectorial capacity of the malaria vector Anopheles gambiae (Diptera:

Culicidae). J. Med. Entomol. 38:22-28.

GARY, R. E. and FOSTER, W. A. 2004. Anopheles gambiae feeding and survival on

honeydew and extra-floral nectar of peridomestic plants. Med. Vet. Entomol.

18:102-107.

GEIER, M. and BOECKH, J. 1999. A new Y-tube olfactometer for mosquitoes to

measure the attractiveness of host odours. Entomol. Exp. Appl. 92:9-19.

GEIER, M., BOSCH, O. J., and BOECKH, J. 1999. Ammonia as an attractive component

for the yellow fever mosquito Aedes aegypti. Chem. Senses 24:647-653.

GIMNIG, J. E., OMBOK, M., KAMAU, L., and HAWLEY, W. A. 2001. Characteristics

of larval anopheline (Diptera: Culicidae) habitats in western Kenya. Journal of

Medical Entomology 38:282-288.

GOUAGNA, L. C., POUEME, R. S., DABIRÉ , K. R., OUÉDRAOGO, J. B.,

FONTENILLE, D., and SIMRAD, F. 2010. Patterns of sugar feeding and host

175

plant preferences in adult males of An. gambiae (Diptera: Culicidae). J. Vector

Ecol. 35:267-276.

GOVELLA, N. J., CHAKI, P. P., MPANGILE, J. M., and KILLEEN, G. F. 2011.

Monitoring mosquitoes in urban Dar es Salaam: Evaluation of resting boxes,

window exit traps, CDC light traps, Ifakara tent traps and human landing catches.

Parasit. Vectors 4.

GRANT, A. J. and O’CONNELL, R. J., editors. 1996. ElectrophysioIogicaI responses

from receptor neurons in mosquito maxillary palp sensilla: Wiley, Chichester.

GREGG, P. C., DEL SOCORRO, A. P., and HENDERSON, G. S. 2010. Development of

a synthetic plant volatile-based attracticide for female noctuid moths. II.

Bioassays of synthetic plant volatiles as attractants for the adults of the cotton

bollworm, Helicoverpa armigera (Hübner) (Lepidoptera: Noctuidae). Aust. J.

Entomol. 49:21-30.

GRIMSTAD, P. R. and DEFOLIART, G. R. 1974. Nectar sources of Wisconsin

mosquitoes. J. Med. Entomol. 11:331-341.

176

GU, W., MÜLLER, G., SCHLEIN, Y., NOVAK, R. J., and BEIER, J. C. 2011. Natural

plant sugar sources of Anopheles mosquitoes strongly impact malaria

transmission potential. PLoS One 6:e15996.

HAMER, G. L., KITRON, U. D., BRAWN, J. D., LOSS, S. R., RUIZ, M. O.,

GOLDBERG, T. L., and WALKER, E. D. 2008. Culex pipiens (Diptera:

Culicidae): A bridge vector of West Nile virus to humans. J. Med. Entomol.

45:125-128.

HANCOCK, R. G. and FOSTER, W. A. 1997. Larval and adult nutritional effects on

blood/nectar choice of Culex niagripalpus mosquitoes. Med. Vet. Entomol.

11:112-122.

HANSSON, B. S. and ANTON, S. 2000. Function and morphology of the antennal lobe:

New developments. Ann. Rev. Entomol. 45:203-231.

HARADA, F., MORIYA, K., and YABE, T. 1972. Observations on the survival and

longevity of adult Culex and Aedes mosquitoes fed on flowers of some nectar

plants (II). Jap. J. Sanit. Zool. 23:141-154.

177

HARAMIS, L. D. and FOSTER, W. A. 1983. Survival and population density of Aedes

triseriatus (Diptera: Culicidae) in a woodlot in central Ohio, USA. J. Med.

Entomol. 20:391-398.

HARRINGTON, L. C., EDMAN, J. D., and SCOTT, T. W. 2001. Why do female Aedes

aegypti (Diptera: Culicidae) feed preferentially and frequently on human blood?

J. Med. Entomol. 38:411-422.

HEALY, T. P. and JEPSON, P. C. 1988. The location of floral nectar sources by

mosquitoes: the long-range responses of Anopheles arabiensis Patton (Diptera:

Culicidae) to Achillea millefolium flowers and isolated floral odour. Bull.

Entomol. Res. 78:651-657.

HIGA, Y., HOSHINO, K., TSUDA, Y., and KOBAYASHI, M. 2006. Dry-ice trap and

human bait collection of mosquitoes in the eastern part of Hokkaido, Japan. Med.

Entomol. Zool. 57:93-98.

HOCKING, B. 1953. The intrinsic range and speed of flight of insects. Trans. Roy.

Entomol. Soc. London 104:223-346.

178

HOLCK, A. R. and MEEK, C. L. 1991. Comparison of sampling techniques for adult

mosquitoes and other nematocera in open vegetation. J. Entomol. Sci. 26:231-236.

IMPOINVIL, D. E., KONGERE, J. O., FOSTER, W. A., NJIRU, B. N., KILLEEN, G.

F., GITHURE, J. I., BEIER, J. C., HASSANALI, A., and KNOLS, B. G. J. 2004.

Feeding and survival of the malaria vector Anopheles gambie on plants growing

in Kenya. Med. Vet. Entomol. 18:108-115.

JACOBSON, M. and BEROZA, M. 1963. Chemical Insect Attractants. Science

140:1367-1373.

JAWARA, M., AWOLOLA, T. S., PINDER, M., JEFFRIES, D., SMALLEGANGE, R.

C., TAKKEN, W., and CONWAY, D. J. 2011. Field testing of different chemical

combinations as odour baits for trapping wild mosquitoes in The Gambia. PLoS

One 6:e19676.

JEPSON, P. C. and HEALY, T. P. 1988. The location of floral nectar sources by

mosquitoes: an advanced bioassay for volatile plant odours and initial studies with

Aedes aegypti (L.) (Diptera: Culicidae). Bull. Entomol. Res. 78.

179

JHUMUR, U. S., DÖTTERL, S., and JÜRGENS, A. 2006. Naïve and conditioned

responses of Culex pipiens pipiens Biotype molestus (Diptera: Culicidae) to

flower odours. J. Med. Entomol. 43:1164-1170.

JHUMUR, U. S., DÖTTERL, S., and JÜRGENS, A. 2007. Electrophysiological and

behavioural responses of mosquitoes to volatiles of Silene otites

(Caryophyllaceae). Arthropod-Plant Inte. 1:243-254.

JOHNSON, C. G. 1950. The comparison of suction trap, sticky trap and tow-net for the

quantitative sampling of small airborne insects. Ann. Appl. Biol. 37:268-285.

JOSEPH, S. R. 1970. Fruit feeding of mosquitoes in nature. Proc. N. J. Mosq. Exterm.

Assoc. 57:125-131.

JÜRGENS, A. 2006. Comparative floral morphometrics in day-flowering, night-

flowering and self-pollinated Caryophylloideae (Agrostemma, Dianthus,

Saponaria, Silene, and Vaccaria). Plant Syst. Evol. 257:233-250.

JÜRGENS, A., WITT, T., and GOTTSBERGER, G. 2002. Flower scent composition in

night-flowering Silene species (Caryophyllaceae). Biochem. Syst. Ecol. 30:383-

397. 180

KELLER, A. and VOSSHALL, L. B. 2003. Decoding olfaction in Drosophila. Curr.

Opin. Nurobiol.:103-110.

KILPATRICK, A. M., KRAMER, L. D., JONES, M. J., MARRA, P. P., and DASZAK,

P. 2006. West Nile Virus epidemics in north America are driven by shifts in

mosquito feeding behavior. Plos Biol. 4:0606-0610.

KIRTON, U. and SPIELMAN, A. 1989. Suppression of transmission of malaria through

source reduction: Antianopheline measures applied in Israel, the United States

and Italy. Rev Infect Dis 11:391-406.

KLINE, D. L. 1999. Comparison of two American Biophysics mosquito traps: The

professional and a new counterflow geometry trap. J. Am. Mosq. Cont. Assoc.

15:276-282.

KLINE, D. L. 2002. Evaluation of various models of propane-powered mosquito traps. J.

Vector Ecol. 27:1-7.

KNOLS, B., NJIRU, B., MATHENGE, E., MUKABANA, W., BEIER, J., and

KILLEEN, G. 2002. MalariaSphere: A greenhouse-enclosed simulation of a

181

natural Anopheles gambiae (Diptera: Culicidae) ecosystem in western Kenya.

Malar. J. 1:19.

KNOLS, B. G. and DE JONG, R. 1996. Limburger cheese as an attractant for the malaria

mosquito Anopheles gambiae s.s. Parasitol. Today 4:159-160.

KOZIEL, J. A. and NOVAK, I. 2002. Sampling and sample-preparation strategies based

on solid-phase microextraction for analysis of indoor air Trends Anal. Chem.

21:840-850.

LANGENHEIM, J. H. 1994. Higher plant terpenoids: A phytocentric overview of their

ecological roles. J. Chem. Ecol. 20:1223-1279.

LEAKE, C. J. 1992. Arbovirus-mosquito interactions and vector specificity. Parasitol.

Today 8:123-128.

LENGELER, C. and SNOW, R. W. 1996. From efficacy to effectiveness: Insecticide

treated bednets in Africa. Bulletin of the World Health Organisation 73:325-32.

LOGAN, J. G., BIRKETT, M. A., CLARK, S. J., POWERS, S., SEAL, N. J.,

WADHAMS, L. J., (LUNTZ), A. J. M., and PICKETT, J. A. 2008. Identification 182

of human-derived volatile chemicals that interfere with attraction of Aedes aegypti

mosquitoes. J. Chem. Ecol. 34:308-322.

LOUGHRIN, J. H., POTTER, D. A., and HAMILTON-KEMP, T. R. 1998. Attraction of

Japanese beetles (Coleoptera: Scarabaeidae) to host plant volatiles in field

trapping experiments Environ. Entomol. 27:395-400.

MAGNARELLI, L. A. 1977. Nectar feeding by Aedes sollicitans and its relation to

gonotrophic activity. Environ. Entomol. 6:237-242.

MAGNARELLI, L. A. 1978. Nectar-feeding by female mosquitoes and its relation to

follicular development and parity. J. Med. Entomol. 14:527-530.

MAGNARELLI, L. A. 1983. Nectar sugars and caloric reserves in natural populations of

Aedes canadensis and Aedes stimulans (Diptera: Culicidae). Environ. Entomol.

12:1482-1486.

MAHANDE, A., MOSHA, F., MAHANDE, J., and KWEKA, E. 2007. Feeding and

resting behaviour of malaria vector, Anopheles arabiensis with reference to

zooprophylaxis. Malar. J. 6.

183

MAJAMBERE, S., PINDER, M., FILLINGER, U., AMEH, D., CONWAY, D. J.,

GREEN, C., JEFFRIES, D., JAWARA, M., MILLIGAN, P. J., HUTCHINSON,

R., and LINDSAY, S. 2010. Is mosquito larval source management appropriate

for reducing malaria in areas of extensive flooding in The Gambia? A cross-over

intervention trial. Am J Trop Med Hyg 82:176-184.

MANDA, H., GOUAGNA, L. C., NYANDAT, E., KABIRU, E., JACKSON, R. R.,

FOSTER, W. A., GITHURE, J. I., BEIER, J. C., and HASSANALI, A. 2007.

Discriminative feeding behaviour of Anopheles gambiae s.s. on endemic plants in

western Kenya. Med. Vet. Entomol. 21:103-111.

MAUER, D. J. and ROWLEY, W. A. 1999. Attraction of Culex pipiens pipiens (Diptera:

Culicidae) to flower volatiles. J. Med. Entomol. 36:503-507.

MBOERA, L. E. G., KNOLS, B. G., BRAKS, M. A. H., and TAKKEN, W. 2000.

Comparison of carbon dioxide-baited trapping systems for sampling outdoor

mosquito populations in Tanzania. Med. Vet. Entomol. 14:257-263.

184

MCCRAE, A. W. R. 1989. Differences in sugar feeding activity between tropical and

temperate mosquitoes: field observations and their implications. Vector Ecol.

Newslett. 20:16(Abstr.).

MEERAUS, W. H., ARMISTEAD, J. S., and ARIAS, J. R. 2008. Field comparison of

novel and gold standard traps for collecting Aedes albopictus in northern Virginia.

J. Am. Mosq. Cont. Assoc. 24:244-248.

MEIJERINK, J., BRAKS, M. A. H., BRACK, A. A., ADAM, W., DEKKER, T.,

POSTHUMUS, M. A., BEEK, T. A. V., and LOON, J. J. A. V. 2000.

Identification of olfactory stimulants for Anopheles gambiae from human sweat

samples. J. Chem. Ecol. 26:1367-1381.

METCALF, R. and KOGAN, M. 1987. Plant volatiles as insect attractants. Crit. Rev.

Plant Sci. 5:251-301.

MÜLLER, G. C., BEIER, J. C., TRAORE, S. F., TOURE, M. B., TRAORE, M. M.,

BAH, S., DOUMBIA, S., and SCHLEIN, Y. 2010a. Field experiments of

Anopheles gambiae attraction to local fruits/seedpods and flowering plants in

185

Mali to optimize strategies for malaria vector control in Africa using attractive

toxic sugar bait methods. Malar. J. 9:262.

MÜLLER, G. C., JUNNILA, A., QUALLS, W., REVAY, E. E., KLINE, D. L., ALLAN,

S., SCHLEIN, Y., and XUE, R. D. 2010b. Control of Culex quinquefasciatus in a

storm drain system in Florida using attractive toxic sugar baits. Med. Vet.

Entomol. 24:346-351.

MÜLLER, G. C., JUNNILA, A., and SCHLEIN, Y. 2010c. Effective control of adult

Culex pipiens by spraying an attractive toxic sugar bait solution in the vegetation

near larval habitats. J. Med. Entomol. 47:63-66.

MÜLLER, G. C., KRAVCHENKO, V. D., and SCHLEIN, Y. 2008. Decline of

Anopheles sergentii and Aedes caspius populations following presentation of

attractive toxic (Spinosad) sugar bait stations in an oasis. J. Am. Mosq. Cont.

Assoc. 24:147-149.

MÜLLER, G. C. and SCHLEIN, Y. 2006. Sugar questing mosquitoes in arid areas gather

on scarce blossoms that can be used for control. Int. J. Parasitol. 36:1077-1080.

186

MÜLLER, G. C. and SCHLEIN, Y. 2011. Different methods of using attractive sugar

baits (ATSB) for the control of Phlebotomus papatasi. J. Vector Ecol. 36

(Supplement 1):S64-S70.

MUTUKU, F. M., BAYOH, M. N., GIMNIG, J. E., VULULE, J., KAMAU, L.,

WALKER, E. D., KABIRU, E. W., and HAWLEY, W. 2006. Pupal habitat

productivity of Anopheles gambiae complex mosquitoes in a rural village in

western Kenya Am J Trop Med Hyg 74:54-61.

NAYAR, J. K. and SAUERMAN, D. M., JR. 1971. The effects of diet on life-span,

fecundity, and flight potential of Aedes taeniorhynchus adults. J. Med. Entomol.

8:506-513.

NAYAR, J. K. and VAN HANDEL, E. 1971. The fuel for sustained mosquito flight. J.

Insect Physiol. 17:471-481.

NELSON, M. J. 1994. The role of sampling in vector control. Am J Trop Med Hyg

50:145-150.

NICODEM, J. G., CHAKI, P. P., MPANGILE, J. M., and KILLEEN, G. F. 2011.

Monitoring mosquitoes in urban Dar es Salaam: Evaluation of resting boxes, 187

window exit traps, CDC light traps, Ifakara tent traps and human landing catches.

Parasit. Vectors 4.

NJIRU, B. N., MUKABANA, W., TAKKEN, W., and KNOLS, B. G. 2006. Trapping of

the malaria vector Anopheles gambiae with odour-baited MM-X traps in semi-

field conditions in western Kenya. Malar. J. 5.

OKECH, B. A., GOUAGNA, L. C., KILLEEN, G. F., KNOLS, B. G. J., KABIRU, E.

W., BEIER, J. C., YAN, G., and GITHURE, J. I. 2003. Influence of sugar

availability and indoor microclimate on survival of Anopheles gambiae (Diptera:

Culicidae) under semifield conditions in western Kenya. J. Med. Entomol. 40:

657-663.

OKUMU, F., BISAWARO, L., MBELEYELA, E., KILLEEN, G. F., MUKABANA, W.,

and MOORE, S. J. 2010a. Using nylon strips to dispense mosquito attractants for

sampling the malaria vector Anopheles gambiae s.s. J. Med. Entomol. 47:274-282.

OKUMU, F., GOVELLA, N. J., MOORE, S. J., CHITNIS, N., and KILLEEN, G. F.

2010b. Potential benefits, limitations and target product-profiles of odor-baited

mosquito traps for malaria control in Africa. PLoS One 5:e11573.

188

OKUMU, F. O., KILLEEN, G. F., OGOMA, S., BISWARO, L., SMALLEGANGE, R.

C., MBEYELA, E., TITUS, E., MUNK, C., NGONYANI, H., TAKKEN, W.,

MSHINDA, H., MUKABANA, W. R., and MOORE, S. J. 2010c. Development

and field evaluation of a synthetic mosquito lure that is more attractive than

humans. PLoS One 5:e8951.

OLANGA, E. A., OKAL, M. N., MBADI, P. A., KOKWARO, E. D., and MUKABANA,

W. 2010. Attraction of Anopheles gambiae to odour baits augmented with heat

and moisture. Malar. J. 96.

PATES, H. and CURTIS, C. 2005. Mosquito behavior and vector control. Ann. Rev.

Entomol. 50:53-70.

PATTERSON, R. S., SMITTLE, B. J., and DENEVE, R. T. 1969. Feeding habits of male

southern house mosquitoes on 32P-labelled and unlabelled plants. J. Econ.

Entomol. 62:1455-1458.

PICMAN, J. and PICMAN, A. K. 1984. Autotoxicity in Parthenium hysterophorus and

its possible role in control of germination. Biochem. Syst. Ecol. 12:287-292.

189

PURI, S. N., MENDKI, M. J., SUKUMARAN, D., GANESAN, K., PRAKASH, S., and

SEKHAR, K. 2006. Electroantennogram and behavioral responses of Culex

quinquefasciatus (Diptera: Culicidae) females to chemicals found in human skin

emanations. J. Med. Entomol. 43:207-213.

QIU, Y. T., SMALLEGANGE, R. C., LOON, J. J. A. V., and TAKKEN, W. 2011.

Behavioural responses of Anopheles gambiae sensu stricto to components of

human breath, sweat and urine depend on mixture composition and concentration.

Med. Vet. Entomol. 25:247-255.

RAGUSO, R. A. 2004. Why are some floral nectars scented? Ecology 85:1486-1494.

REEVES, W. C. 1990. Quantitative field studies on a carbon dioxide chemotropism of

mosquitoes. J. Am. Mosq. Cont. Assoc. 6:708-712.

REISEN, W. K., MEYER, R. P., and MILBY, M. M. 1986. Patterns of fructose feeding

by Culex tarsalis (Diptera: Culicidae). J. Med. Entomol. 23:366-373.

ROBICH, R. M. and DENLINGER, D. L. 2005. Diapause in the mosquito Culex pipiens

evokes a metabolic switch from blood feeding to sugar gluttony. Proc. Natl. Acad.

Sci. USA 102:1512-1517. 190

RUDOLFS, W. 1922. Chemotropism of mosquitoes. Bull New Jers Agric Exp Stn 367.

SAITOH, Y., HATTORI, J., CHINONE, S., NIHEI, N., TSUDA, Y., KURAHASHI, H.,

and KOBAYASHI, M. 2004. Yeast-generated CO2 as a convenient source of

carbon dioxide for adult mosquito sampling. J. Am. Mosq. Cont. Assoc. 20:261-

264.

SANDHOLM, H. A. and PRICE, R. D. 1962. Field observations on the nectar feeding

habits of some Minnesota mosquitoes. Mosq. News 22:346-349.

SCHLEIN, Y. and MÜLLER, G. C. 2008a. An approach to mosquito control. Using the

dominant attraction of flowering Tamarix jordanis trees against Culex pipiens. J.

Med. Entomol. 45:384-390.

SCHLEIN, Y. and MÜLLER, G. C. 2008b. An approach to mosquito control: Using the

dominant attraction of flowering Tamarix jordanis trees against Culex pipiens. J.

Med. Entomol. 45:384-390.

SCHLEIN, Y. and MÜLLER, G. C. 2010. Experimental control of Phlebotomus papatasi

by spraying attractive toxic sugar bait (ATSB) on vegetation. Trans. R. Soc. Trop.

Med. Hyg. 104:766-71. 191

SCOTT, T. W., NAKSATHIT, A., DAY, J. F., KITTAYAPONG, P., and EDMAN, J. D.

1997. A fitness advantage for Aedes aegypti and the viruses it transmits when

females feed only on human blood. Am J Trop Med Hyg 57:235-239.

SERVICE, M. W. 1969. The use of traps in sampling mosquito populations. Ent. exp. &

appl. 12:403-412.

SERVICE, M. W. 1985. Mosquito Ecology: Field sampling methods. Elsevier, London

SERVICE, M. W. 1993. The role of ecology and ecological modeling in vector control.

Bull. Soc. Vector Ecol. 18:85-98.

SEYOUM, A., PÅLSSON, K., KUNG'A, S., KABIRU, E. W., LWANDE, W.,

KILLEEN, G. F., HASSANALI, A., and KNOLS, B. G. J. 2002. Traditional use

of mosquito-repellent plants in western Kenya and their evaluation in semi-field

experimental huts against Anopheles gambiae: ethnobotanical studies and

application by thermal expulsion and direct burning. Trans. R. Soc. Trop. Med.

Hyg. 96:225-231.

192

SHAUKAT, A. M., BREMAN, J. G., and MCKENZIE, F. E. 2010. Using the

entomological inoculation rate to assess the impact of vector control on malaria

parasite transmission and elimination. Malar. J. 9.

SMALLEGANGE, R. C., KNOLS, B. G., and TAKKEN, W. 2010a. Effectiveness of

synthetic versus natural human volatiles as attractants for Anopheles gambiae

(Diptera: Culicidae) sensu stricto. J. Med. Entomol. 47:338-344.

SMALLEGANGE, R. C., SCHMIED, W. H., ROEY, K. J. V., VERHULST, N. O.,

SPITZEN, J., MUKABANA, W. R., and TAKKEN, W. 2010b. Sugar-fermenting

yeast as an organic source of carbon dioxide to attract the malaria mosquito

Anopheles gambiae. Malaria Journal 9.

STONE, C. M., HAMILTON, I. M., and FOSTER, W. A. 2011. A survival and

reproduction trade-off is resolved in accordance with resource availability by

virgin female mosquitoes. Anim. Behav. 81:765-774.

STONE, C. M., TAYLOR, R. M., and FOSTER, W. A. 2009. An effective indoor

mesocosm for studying populations of Anopheles gambiae in temperate climates.

J. Am. Mosq. Cont. Assoc. 25:514-516.

193

STRAIF, S. C. and BEIER, J. C. 1996. Effects of sugar availability on the blood feeding

behavior of Anopheles gambiae (Diptera: Culicidae). J. Med. Entomol. 33:608-

612.

SYED, Z. and LEAL, W. 2007. Maxillary palps are broad spectrum odorant detectors in

Culex quinquefasciatus. Chem. Senses 32:727-738.

SYED, Z. and LEAL, W. 2009. Acute olfactory response of Culex mosquitoes to a

human - and bird - derived attractant. Proc. Natl. Acad. Sci. USA 106:18803-

18808.

TAKKEN, W. and KNOLS, B. 1999. Odour Mediated behaviour of afrotropical malaria

mosquitoes. Ann. Rev. Entomol. 44:131-157.

THORSTEINSON, A. J. and BRUST, R. A. 1962. The Influence of flower scents on

aggregations of caged adult Aedes aegypti. Mosq. News 22:349-352.

TIKASINGH, E. S. and DAVIES, J. B. 1972. Comparative test of four mosquito traps.

Mosq. News 32:623-628.

194

TORR, S. J. 1994. The Tsetse (Diptera: Glossinidae) story: Implications for mosquitoes.

J. Am. Mosq. Cont. Assoc. 10:258-265.

TOWERS, G. H. N. and MITCHELL, J. C. 1983. The current status of the weed

Parthenium hysterophorus L. as a cause of allergic contact dermatitis. Contact

Derm. 9: 465-469.

TURELL, M. J., DOHM, D. J., SARDELIS, M. R., O’GUINN, M. L., ANDREADIS, T.

G., and BLOW, J. A. 2005. An update on the potential of North American

mosquitoes (Diptera: Culicidae) to transmit West Nile Virus. J. Med. Entomol.

42:57-62.

VALE, G. A. 1993. Development of baits for Tsetse flies (Diptera: Glossinidae) in

Zimbabwe. J. Med. Entomol. 30:831-842.

VARGO, A. M. and FOSTER, W. A. 1982. Responsiveness of female Aedes aegypti

(Diptera: Culicidae) to flower extracts. J. Med. Entomol. 19:710-718.

VERHULST, N. O., MBADI, P. A., KISS, G. B., MUKABANA, W., LOON, J. J. A. V.,

TAKKEN, W., and SMALLEGANGE, R. C. 2011. Improvement of a synthetic

195

lure for Anopheles gambiae using compounds produced by human skin

microbiota. Malar. J. 10.

VISSER, J. H. 1986. Host odor perception in phytophagous insects. Ann. Rev. Entomol.

31:121-144.

WHO. 2000. WHO Expert Committee on Malaria. Geneva: World Health Organisation

Tech Rep Ser. i-iv;1-74 p.

WHO. 2010. World Malaria Report 2010. Geneva: World Health Organisation. 204 p.

WRIGHT, R. H. and KELLOGG, F. E. 1962. Response of Aedes aegypti to moist

convection currents. Nature 194:402 - 403.

XUE, R. D. and BARNARD, D. R. 2003. Boric acid bait kills adult mosquitoes (Diptera:

Culicidae). J. Econ. Entomol. 95:1559-1562.

XUE, R. D., DOYLE, M. A., and KLINE, D. L. 2008. Field evaluation of CDC and

Mosquito Magnet X traps baited with dry ice, CO2 sachet, and octenol against

mosquitoes. J. Am. Mosq. Cont. Assoc. 24:249-252.

196

XUE, R. D., MÜLLER, G. C., KLINE, D. L., and BARNARD, D. R. 2011. Effect of

application rate and persistence of boric acid sugar baits applied to plants for

control of Aedes albopictus. J. Am. Mosq. Cont. Assoc. 27:56-60.

YEE, W. L., FOSTER, W. A., HOWE, M. J., and HANCOCK, R. G. 1992. Simultaneous

field comparison of evening temporal distributions of nectar and blood feeding by

Aedes vexans and Ae. trivittatus (Diptera: Culicidae) in Ohio. J. Med. Entomol.

29:356-360.

YUVAL, B. 1992. The other habit: Sugar feeding by mosquitoes. Bull. Soc. Vector Ecol.

17:150-156.

YUVAL, B., HOLLIDAY-HANSON, M. L., and WASHINO, R. K. 1992. Energy budget

of swarming male mosquitoes. Ecol. Entomol. 19:74-78.

ZHANG, Y., GAO, B., ZHANG, M., SHI, J., and XU, Y. 2009. Headspace solid-phase

microextraction–gas chromatography–mass spectrometry analysis of the volatile

components of longan (Dimocarpus longan Lour.). Eur. Food Res. Technol.

229:457-465.

197

ZWIEBEL, L. J. and TAKKEN, W. 2004. Olfactory regulation of mosquito–host

interactions. Insect Biochem. Mol. Biol. 34:645-652.

198